SENSATION &

PERCEPTION

Stanley Coren
. Lawrence M. Ward
rvoT
do
-
SENSATION &
PERCEPTION
THIRD EDITION
SENSATION &
PERCEPTION
THIRD EDITION

Stanley Coren
University of British Columbia

Lawrence M. Ward
University of British Columbia

Based on previous editions

that included contributions by

Clare Porac
University of Victoria

Oil
Harcourt Brace Jovanovich, Publishers
San Diego New York Chicago Austin Washington, D.C.
London Sydney Tokyo Toronto
Cover photo: Eye, by Geoff Gove/Image Bank

Copyright © 1989, 1984 by Harcourt Brace Jovanovich, Inc.

All rights reserved. No part of this publication may be reproduced or

transmitted in any form or by any means, electronic or mechanical, including
photocopy, recording, or any information storage and retrieval system,
without permission in writing from the publisher.

Requests for permission to make copies of any part of the work should be
mailed to: Permissions, Harcourt Brace Jovanovich, Publishers, Orlando,
Florida 32887.

ISBN: 0-15-579647-X
Library of Congress Catalog Card Number: 88-81033
Printed in the United States of America

We wish to thank the following for permission to reprint photos:

For Chapter 1 opener: New Jersey State Museum Collection, Trenton,
Purchase FA 1968.73; Chapter 2: © George Hall/Woodfin Camp; Chapter 3:
Courtesy of Deric Bownds, Stan Carlson, University of Wisconsin; Chapter
4: Christian Delbert Photography/Picture Cube; Chapter 5: © Larry Lorusso/
Picture Cube; Chapter 6: © Henry Groskinsky; Chapter 7: © Roberta J.
Shefke; Chapter 8: © Margerin Studios/FPG; Chapter 9: Photograph by
L. L. Brown courtesy of The Institute of Psycho-Structural Balancing, San
Diego, CA; Chapter 10: The Soloman R. Guggenheim Museum; Chapter
11: © 1988 M. C. Escher c/o Cordon Art-Baam-Holland; Chapter 12: The
San Diego Opera; Chapter 13: Ed Carlin/Picture Cube; Chapter 14: E.
Nagle/FPG; Chapter 15: Aronson Photographers/Stock Boston; Chapter 16:
Elizabeth Crews/Stock Boston; Chapter 17: © Norman Snyder; Chapter 18:
United Nations Photo 36252. For Figure 10-1: © Margaret K. Porter; Figure
13-5: © Paulette M. Russo; Figure 15-5: © Paulette M. Russo.
 Preface
Take away the sensations of softness, moisture,
redness, tartness, and you take away the
cherry. Since it is not a being distinct from
these sensations; a cherry, I say, is nothing but
a congeries of sensible impressions or ideas
perceived by various senses; which ideas are
united into one thing.
George Berkeley, 1713

Virtually everything we know about the world has have rewritten six of the chapters “from the ground
entered our minds through our senses. We all real¬ up” and reorganized or amalgamated materials
ize that without even one of our senses our experi¬ from other chapters. Every chapter has been up¬
ences would be incredibly limited. Consider the dated to reflect the most recent literature. However,
impossible problem of explaining the difference be¬ we have retained all those features that instructors
tween the color blue and the color green to a person felt made the Second Edition such a useful book.
who has been blind since birth. And how would For instance, concrete examples are used through¬
you explain to a person who has no taste buds how out the text in order to make the subject matter
the taste of chocolate and vanilla differ from each “come alive” for students. Whenever possible,
other? Such aspects of the world will never exist common or natural instances of perceptual phe¬
for these individuals. For the blind person, salt and nomena are described during the discussion of the
pepper differ only in taste; for the person with no concepts underlying them. Each chapter is pre¬
ability to taste, salt and pepper differ only in color. ceded by an outline that serves as a preview to its
For those of us who have the senses of sight, hear¬ contents; the outlines also provide a structure to
ing, taste, touch, and smell, our daily experience is guide students as they review the chapters.
a continuous flow of changing percepts, with each Although terms are defined when they are in¬
new sensation carrying information about the troduced, a glossary is provided at the end of each
world. chapter as well. Any item printed in boldface in
Sensation and Perception, Third Edition, pro¬ the text is also listed in the chapter glossary. Stu¬
vides an introduction to the study of our senses and dents will find that these glossaries serve as a suc¬
how we perceive through them. It has been revised cinct review and chapter summary, and can be used
substantially since the Second Edition, and contains for self-testing and study purposes.
over 45 percent new material. These changes re¬ One special feature of our book is the inclu¬
flect many of the recent findings that have sion of 106 Demonstration Boxes. Each box de¬
emerged, or coalesced into meaningful patterns, scribes a simple demonstration designed to allow
since the completion of the previous edition. We the students to actually experience many of the per-

v
VI Preface

ceptual phenomena described in the text. Most re¬ Each of the chapters has been written so that
quire only the stimuli in the box itself, or it is relatively self-contained and independent of the
commonplace items that can be found in most other chapters, although this is not always com¬
homes or dormitory rooms. The majority of these pletely possible. Therefore, when material from
demonstrations require only a few moments of other places in the book is used in a discussion, the
preparation, which we feel is time well spent in location of that information is always cited. This
improving the understanding of the concepts under has been done to provide users with maximum flex¬
discussion and in maintaining student interest. ibility as far as the sequence of chapter presentation
Some instructors have reported that having students is concerned, thus permitting the instructor to im¬
perform the demonstrations in class has been very press his or her orientation upon the material. A
useful. In such cases, the demonstrations may also brief appendix on some basic aspects of neurophys¬
serve as the focal point for a lecture or for class¬ iology has also been provided for the first time in
room discussion. this edition.
The book is designed to survey the broad The chapter sequence in this Third Edition is
range of topics generally included under the head¬ quite different from that used in the previous edi¬
ing of sensation and perception. The reader will tions. At the request of many individuals who have
notice that no single theory of perception is cham¬ taught from the earlier versions, we have now or¬
pioned. In general, we have attempted to be as ganized the book by sensory systems, with the first
eclectic as we could, describing the various view¬ half of the book covering the basic physiology and
points in areas of controversy and attempting to sensory responses and the second half covering
present a balanced overview so that instructors of those topics involving more complex and cognitive
different opinions might be comfortable using the interactions. Chapters 1 and 2 provide an introduc¬
book. tion to the problems of sensation and perception
The topics in this book were selected on the along with methodological and theoretical aspects
basis of our experience in teaching our own of psychophysical measurement; Chapters 3,4, and
courses; therefore, much of the material has already 5 cover the physiology and basic sensory qualities
been class tested. We have included three chap¬ of vision; Chapters 6 and 7 do the same for audi¬
ters—“Attention,” “Speech and Music,” and tion; and Chapters 8 and 9 cover the chemical and
“Individual Differences”—that are not often seen mechanical senses. These first nine chapters thus
in sensation and perception textbooks. These areas cover the major topics usually grouped together un¬
have attracted a good deal of experimental work in der the heading of sensation. Chapters 10 through
recent years, and they are sufficiently relevant to 15 cover the perception of space, form, speech and
many issues in perception that we felt students music, and time and motion, the perceptual con¬
should study them. stancies, and the perceptual aspects of attention;
In order to keep the book to a manageable and Chapters 16, 17, and 18 discuss how individual
size, we have occasionally been selective in our variables such as age, experience, learning, gender,
coverage. Our first priority was to cover the central culture, drugs, and personality may affect the per¬
concepts of each topic in enough detail to make the ceptual response. These last nine chapters thus
material clear and coherent. To have included all cover the topics most frequently grouped together
the topics ever classified as part of the field of sen¬ as perception.
sation and perception, we would have had to pre¬ Those of you who have encountered earlier
sent a “grocery list” of concepts and terms, each versions of this book should know that Clare Porac
treated superficially. Such an alternative was unac¬ has retired from this project in order to pursue her
ceptable to us. research and other writing projects. Clare’s contri-
 Preface vn

butions were always organized, intelligent, and of Frank, of the University of Cincinnati, Robert
the highest professional quality. Although she did Levy, of Indiana State University, and Lyn Mo-
not directly participate in this revision, we have wafy, of Vanderbilt University, for their helpful
striven to retain the clarity in writing and well- comments and suggestions after reviewing the pre¬
structured discussions that have characterized her vious edition.
contributions to the first two editions. Finally, the reader might notice that there is
In our attempts to collect and interpret the in¬ no dedication page. This is not to say that we do
formation for this book, we have been assisted at not wish to dedicate the book to anyone. It reflects
various stages by our colleagues. Some have read the fact that there are too many people who have
preliminary versions of chapters and made useful been important in our personal and professional
suggestions. We would like to specifically thank lives to list on any single page (no matter how
Ray Corteen, Jim Enns, Ronnie Lakowski, Richard small the print). Perhaps it is best to simply dedi¬
Tees, and Janet Werker, all of the Psychology De¬ cate this book to all of those researchers who have
partment of the University of British Columbia. We provided the knowledge that we have attempted to
would also like to thank the personnel of the Hu¬ organize and review between these covers, and to
man Perception and Psychophysics Laboratories, all of those researchers who will provide further
and especially Wayne Wong and Odie Geiger for insights into sensation and perception for future au¬
assisting with library work and all of the small but thors to collate, review, digest, wonder at, and
necessary chores that eat up innumerable hours of learn from.
a textbook writer’s time.
In addition, we would like to thank Barry An¬ S.C.
ton, of the University of Puget Sound, Robert L.M.W.
 '

»
 Contents
Preface v 5 Color 119
Sr. Color Stimulus 120
1 Sensation and Perception 1
. . . The Physiology of Color
. . . Aspects of the Perceptual . . . Vision 130
. . . Process 9 . . . Color Perception 143
. . . Theories of Perception 10
. . . The Plan of the Book 11
AUDITION

2 Psychophysics 15 6 The Auditory System 151

. . . Detection 16 . . Sound 152

. . . Identification 27 . . The Structure of the Ear 157
. . . Discrimination 33 . . Electrical Activity of the Auditory
. . . Scaling 40 . . Nerve 166
. . The Auditory Pathways 169
. . The Auditory Cortex 172

VISION 7 Hearing 177

3 The Visual System 57 . . . Detection of Sounds 178
. . Light 58 . . . Subjective Dimensions of
. . The Structure of the Eye 58 . . . Sound 195
. . Neural Responses to Light 71
. . The Visual Pathways 74
CHEMICAL AND MECHANICAL
. . The Visual Cortex 78
SENSES

8 Taste and Smell 211

Brightness and Spatial
Frequency 89 . . The Gustatory (Taste) Sense 212
. . The Olfactory (Smell) Sense 222
Photometric Units 90
Factors in Brightness
9 Touch and Pain 237
Perception 92
Brightness Contrast 98 . . The Skin Senses 238
Cognitive Factors in Brightness . . Touch 245
Perception 103 . . Warmth and Cold 252
Visual Acuity 104 . . Kinesthesis 258
Spatial Frequency Analysis 108 . . Pain 261
X Contents

PERCEPTION . . . Lightness or Whiteness

. . . Constancy 419
10 Space 273 . . . Color or Hue Constancy 422
. . Types of Depth and Distance . . . Other Constancies 423
. . Perception 274
. . Pictorial Depth Cues 275
. . Physiological Cues for 15 Attention 427
. . Depth 284 . . . Varieties of Attention 428
. . Physiological Cues for . . . Orienting 429
. . Direction 285 . . . Filtering 433
. . Binocular Depth Perception . . . Searching 440
289 . . . Expecting 447
. . Interaction of Cues 295 . . . Theories of Attention 450
. . Development of Space Perception
299

11 Form 305 PERSONAL FACTORS AND CHANGE

V;_y The Visual Field 306 16 Development 455

. . Contour 307 . . . Perception in Infants 456
. . The Perceptual Object 310 . . . Perceptual Change through
. . Object Recognition and . . . Childhood 468
. . Identification 323 . . . Perceptual Change in Adults
. . Theories of Object Identification . . . 473
331

12 Speech and Music 339 17 Learning and Experience 479

, . . Music 340 . . . Experience and Development

. . Speech 348 . . . 480
. . . Sensory-Motor Learning 487
13 Time and Motion 371 . . . Context and Meaning 494
. . . Environmental and Life History
. . . Time 372
. . . Differences 499
, . . Motion 380

14 The Constancies 403 18 Individual Differences 511

. . The Task of Perception 404 . . . Physiological Differences 512
. . Perceptual Constancies 405 . . . Gender Differences 520
. . Size Constancy 407 . . . Personality and Cognitive Style
. . Shape Constancy 416 . . . Differences 525
 Contents xi

Appendix: Primer of Techniques to Measure Neural

Neurophysiology 529 Function 533
Neurons and the Nervous
References 537
System 530
Author Index 583
The Nature of Neural
Subject Index 597
Activity 531
' CHAPTER
1
Sensation and
Perception
•

ASPECTS OF THE PERCEPTUAL PROCESS

THEORIES OF PERCEPTION
THE PLAN OF THE BOOK

1
2 Chapter 1/Sensation and Perception

C an you answer the following questions? What

color is the sky? Which is warmer, fire or ice?
sea bottom, which may be hidden from our eyes by
hundreds of feet of water. We have spectrographs
Which tastes sweeter, sugar or vinegar? Which has to tell us about the exact chemical composition of
a stronger smell, burning wood or burning rubber? many substances, as compared to the crude chemi¬
Which sounds louder, a chirping bird or the crack cal sensitivity of our noses and tongues.
of a rifle? Such questions probably seem quite triv¬ Although such pieces of apparatus exist, and
ial, and the answers obvious. Perhaps we should measure phenomena not directly available to our
phrase the questions differently. How do you know senses, this does not alter the fact that it is the per¬
what color the sky is? How do you know how hot ception of the scientist that constitutes the subject
fire is relative to ice? How do you know that sugar matter of every science. The eye of the scientist
is sweet? Again, you might feel that the answers presses against the telescope or examines the pho¬
are obvious. You see the color of the sky, you feel tograph of the distant star. The ear of the scientist
the temperatures of a flame and an ice cube, and listens to the sound of sonar tracing out the size and
you taste the sweetness of sugar—in other words, distance of objects, or his eyes read the sonograph.
the answers come through your senses. Although the tongue of the scientist does not taste
Let us push our questioning one step further. the chemical composition of some unknown sub¬
How do you know anything about your world? You stance, his eye, aided by the spectrograph, provides
might say that you learn from books, television, the data for analysis. Really, the only data that
radio, films, lectures, or the actual exploration of reach the mind of the scientist come not from in¬
places. But how do you obtain the information struments but from the scientist’s senses. The in¬
from these sources? Again, the answer is through strument he or she is looking at can be perfectly
your senses. In fact, without your senses of vision, accurate, yet if the scientist misreads a digital read¬
hearing, touch, taste, and smell, your brain, the or¬ out, or does not notice a critical shift in the opera¬
gan that is responsible for your conscious experi¬ tion of a measurement device, the obtained
ence, would be an eternal prisoner in the solitary information is wrong and the resulting picture of
confinement of your skull. You would live in total the world is in error. The minds of the scientist,
silence and darkness. All would be a tasteless, col¬ the nonscientist, our pet dog sniffing about the
orless, feelingless, floating void. Without your sen¬ world, or a fish swimming about in a bowl, in fact,
ses, the world would simply not exist for you. The the minds of all living, thinking organisms, are
philosopher Thomas Hobbes recognized this fact in prisoners that must rely on information smuggled
1651 when he wrote, “There is no conception in in to them by the senses. Your world is what your
man’s mind which hath not at first, totally or by senses tell you. The limitations of your senses set
parts, been begotten upon the organs of sense.’’ the boundaries of your conscious existence.
The Greek philosopher Protagoras stated the same Because our knowledge of the world is depen¬
position around 450 B.c. when he said, “Man is dent on our senses, it is important to know how our
nothing but a bundle of sensations.” senses function. It is also important to know how
You may protest that this is a rather extreme well the world created by our senses corresponds
viewpoint. Certainly, much of what we know about to external reality (i.e., the reality measured by sci¬
the world does not arrive through our eyes, ears, entific instruments). At this point, you are probably
nose, and other sense organs. We have complex smiling to yourself and thinking, “Here comes an¬
scientific instruments, such as telescopes, that tell other academic discourse that will attempt to make
us about the size and the shape of the universe by something that is quite obvious appear to be com¬
analyzing images too faint for the human eye to plex.” You might be saying to yourself, “I see my
see. We have sonar to trace out the shape of the desk in front of me because it is there. I feel my
 Sensation and Perception 3

chair pressing against my back because it is there. ory data. It requires subtle classifications and
I hear my telephone ringing because it contains a comparisons, and myriad decisions before any of
bell that makes sounds. What could be more ob¬ the data in our senses becomes our conscious
vious?” Such faith in your senses is a vital part of awareness of what is “out there.” Contrary to what
existence. It causes you to jump out of the way of you may think, the eyes do not see. There are
an apparently oncoming car, thus preserving your many individuals who have perfectly functioning
life. It provides the basic data that cause you to eyes yet have no sensory impressions. They cannot
step back from a deep hole, thus avoiding a fall and perceive because they have injuries in those parts
serious bodily harm. of the brain that receive and interpret messages
Such faith in our senses is woven into the very from the eyes. Epicharmus knew this in 450 B.c.
fabric of our lives. As the old saying goes, “Seeing when he said, “The mind sees and the mind hears.
is believing.” Long before the birth of Christ, Lu¬ The rest is blind and deaf.”
cretius stated this article of faith when he asked, “So what?” you mutter to yourself. “So
“What can give us surer knowledge than our sen¬ sometimes we make errors in our perceptions, the
ses? With what else can we distinguish the true real point is that the senses simply carry a picture
form from the false?” Perhaps the most striking of the outside world to the brain. The picture in the
example of this faith is found in our courts of law, brain represents our percept. Of course if we mess
where people’s lives and fortunes often rest solely up the brain we will distort or destroy perception.”
on the testimony of the eyes and ears of witnesses. Again, this answer is too simple. If we look outside
A lawyer might argue that a witness is corrupt or and see a car, are we to believe that there is a pic¬
lying, or even that his memory has failed, but no ture of a car somewhere in our brains? If we notice
lawyer would have the audacity to suggest that her that a traffic light is green, are we to believe that
client should be set free because the only evidence some part of the brain has turned green? And sup¬
available was what the witnesses saw or heard. pose that there were such images in the brain, car¬
Certainly no sane person would charge the eye or ried without distortion from the senses, would this
ear with perjury! help us to see? Certainly, images in the brain
The philosophical position that perception is would only be of value if there were some other
an immediate, almost godlike knowledge of exter¬ eyes in the head, which would look at these pic¬
nal reality has been championed not only by popu¬ tures and interpret them. If this were the case, we
lar sentiment but also by philosophers of the stature would be left with the question of how these inter¬
of Immanuel Kant (1724-1804). Unfortunately, it nal eyes see. Thus, we would eventually be forced
is wrong. Look at the drawings shown in Figure to set up an endless chain of pictures and eyes and
1-1. Clearly, they are all composed of outlined pictures and eyes, because the question of who is
forms on various backgrounds. Despite what your perceiving the percept, and how, still remains.
senses tell you, A, B, and C are all perfect squares. If we are to understand perception we must
Despite the evidence of your senses, D is a perfect consider it in its natural context. Sensation and per¬
circle, the lines in E are both straight, and the lines ception are some of the many complex processes
marked x and y in F are both the same length. that occur in the continuing flow of individual be¬
The ease with which we use our senses— havior. There is no clear line between perception
seeing, apparently through the simple act of open¬ and many other behavioral activities. No perception
ing our eyes, or touching, apparently by merely gives direct knowledge of the outside world, rather
pressing our skin against an object—masks the fact such knowledge of the outside world is the end
that perception is an extremely sophisticated activ¬ product of many processes. The wet-looking black
ity of the brain. Perception calls on stores of mem¬ spot on the edge of a desk could be the place where
 4 Chapter 1/Sensation and Perception

FIGURE 1-1 Some instances where the senses tell lies.

 Sensation and Perception 5

ink was spilled. Of course, this percept could be representations of reality. Yet, sensory data are not
wrong. The ink may be dry, or the spot might not always reliable. Sometimes they can be degraded
be there at all. The desk that is seen and touched, or not completely available. There seems to be no
might not really exist. We might be dreaming, clear distinction between perceiving or sensing an
drugged, or hallucinating. Too extreme, you say? object and guessing the identity of an object. In
Consider the following example that actually hap¬ some respects, we can say that all perception of
pened to one of the authors. One night he walked objects requires some guessing. Sensory stimula¬
across the floor of his darkened home. In the dim tion provides the data for our hypotheses about the
gloominess of the night, he saw his dog resting on nature of the external world, and these hypotheses
the floor, clearly asleep. When he bent to touch the form our perceptions of the world.
dog, he found that it was a footstool. He stepped Many human behaviors have been affected by
back, somewhat startled at his stupidity, only to the fallible and often erroneous nature of our per¬
bump against the cold comer of a marble-topped cepts. For example, the most elegant of the classic
coffee table. When he reached back to steady him¬ Greek buildings, the Parthenon, is bent. The
self, he found that the comer of the table was, in straight clean lines, which bring a sense of simple
fact, his dog’s cold nose. Each of these percep¬ elegant grandeur, are actually an illusion. If we
tions, dog, stool, table, and dog again, seemed, schematically represent the east wall of the building
when first received in consciousness, to be accurate as it appears, it is square (as shown in Figure 1-2A).

FIGURE 1-2 (A) The Parthenon as it appears; (B) an illusion that should cause the Parthenon to appear
as C; (D) the way the Parthenon is built to offset the illusion.
6 Chapter 1/Sensation and Perception

Actually, the Parthenon was built in a distorted It is amazing to discover the degree to which
fashion in order to offset a series of optical illu¬ our conscious experience of the world can differ
sions. There is a common visual distortion in which from the physical (scientific) reality. Although
we find that placing angles above a line (much as some perceptual distortions are only slight devia¬
the roof is placed over the architrave) causes the tions from physical reality, some can be quite com¬
line to appear slightly bowed. One form of this il¬ plex and surprising, such as that shown in
lusion is shown as Figure 1-2B where the ends of Demonstration Box 1-1.
the horizontal line appear slightly higher than the Such distortions, in the form of disagreements
center. If the Parthenon were built physically between percept and reality, are quite common. We
square, it would appear to sag as a result of this call them illusions and they occur in predictable
visual distortion. This is shown in an exaggerated circumstances for normal observers. The term illu¬
manner in Figure 1-2C. But the sagging does not sion is drawn from the Latin root illudere, meaning
appear because the building has been altered to “to mock,” and in a sense they do mock us for
compensate for the distortion. Figure 1-2D illus¬ our unthinking reliance on the validity of our sen¬
trates what an undistorted view of the Parthenon sory impressions. Every sensory modality is subject
would look like. The upward curvature is more to distortions, illusions, and systematic errors that
than 6 centimeters on the east and west walls and misrepresent the outside environment to our con¬
almost 11 cm on the longer north and south sides. sciousness. There are illusions of touch, taste, and
The vertical features of the Parthenon (such as hearing, as well as visual illusions. Virtually any
the columns) were inclined inwards in order to cor¬ aspect of perception you might think of can be
rect for a second optical illusion in which the fea¬ subject to these kinds of errors. For instance,
tures of rising objects appear to fall outward at the such basic and apparently simple qualities as the
top. Thus, if we projected all of the columns of the brightness of an object or its color may be percep¬
Parthenon upward, they would meet at a point tually misrepresented, as shown in Demonstra¬
somewhat less than 2 kilometers above the build¬ tion Box 1-2.
ing. Furthermore, the comer columns were made Many perceptual errors are merely amusing,
thicker since it was found that when these columns such as that in Demonstration Box 1-1, or thought-
were seen against the sky, they appeared to be thin¬ provoking, as in Demonstration Box 1-2. Others
ner than those seen against the darker background may lead to some embarrassment or annoyance,
formed by the interior wall. such as might have been felt by the artisan who
These were conscious corrections made by created the picture frame shown as Figure 1-3A.
the Greek architects. To quote one of them, Vi¬ Although his workmanship is faultless, he has been
truvius, writing around 30 B.C.: “For the sight undone because the grain of the wood is too prom¬
follows gracious contours, and unless we flatter inent. Despite the fact that the picture is perfectly
its pleasure by proportionate alterations of these rectangular, it appears to be distorted. Unfortu¬
parts (so that by adjustment we offset the amount nately, some perceptual errors or illusions are quite
to which it suffers illusions) an uncouth and un¬ serious. In Figure 1-3B, we have shown a surgeon
gracious aspect will be presented to the specta¬ probing for a bullet. She is using a fluoroscope,
tors.” In other words, the Parthenon appears to which presents the outline of the patient’s ribs, and
be square, with elegant straight lines, because it her probe is positioned so that it is exactly on line
has been consciously distorted to offset perceptual with the bullet lodged below the rib. As you can
distortions. If it were geometrically square, it see, it appears that she will miss and her probe will
would not be perceptually square. pass above the bullet despite the fact that the probe
 Sensation and Perception 7

DEMONSTRATION BOX 1-1. Gears and Circles

The pattern shown in this box should be viewed in

motion. Move the book around so the motion resem¬
bles that which you would make if you were swirling
coffee around in a cup without using a spoon. Notice
that the six sets of concentric circles seem to show
radial regions of light and dark that appear to move
in the direction you are swirling. They look as though
they were covered by a liquid surface tending to swirl
with the stimulus movement.
A second effect has to do with the center circle that
seems to have gearlike teeth. As you swirl the array,
the center gear seems to rotate, but in a direction op¬
posite to that of the movement of the outer circles.
Some observers see it moving in a jerky, steplike
manner from one rotary position to another and other
observers see a smooth rotation. Of course, there is
no physical movement within the circles, and the
geared center circle is also unchanging, despite your
conscious impression to the contrary.

DEMONSTRATION BOX 1-2. A Subjective Color Grid

The figure in this box consists of a series of thinly

spaced diagonal black lines alternating with white
spaces. Study this figure for a couple of seconds, and
you will begin to see faint, almost pastel streaks of
orange-red and other streaks of blue-green. For many
observers, these streaks tend to run vertically up and
down the figure crossing both white and black lines;
for others, they seem to form a random, almost fish¬
netlike pattern over the grid. These colors are not
present in the stimulus; hence they are subjective, or
illusory, colors.
8 Chapter 1/Sensation and Perception

is angled perfectly. Figure 1-3C shows an even

more disastrous occurrence of an illusion. It repre¬
sents a radar screen with various flight regions
marked across its face. The two oblique streaks
represent jet aircraft approaching the control re¬
gion, both flying at about 950 kilometers per hour.
The information displayed is similar to that which
an air traffic controller might use. From it he might
conclude that if these two aircraft continue in the
same direction they will pass each other with a safe
distance between them. At the moment repre¬
sented here, however, these aircraft are traveling
toward each other on the same line. If they are
flying at the same altitude it is very likely that
they will collide.
These examples illustrate how important dis¬
crepancies between perception and reality can be.
Therefore it becomes important for us to know how
our perceptions arise, how much we can rely on
them, under what circumstances they are most fal¬
lible, and under what conditions our perceptions
most accurately represent the world. An explora¬
tion of these questions is the purpose of this book.

FIGURE 1-3 Some perceptual distortions in common situations.

 Aspects of the Perceptual Process 9

ASPECTS OF THE PERCEPTUAL problem areas, or orientations, that characterize

PROCESS certain groups of investigators, and these seem to
be definable. The study of sensation, or sensory
processes, is concerned with the first contact be¬
The study of perception is diverse. Partly this is the tween the organism and the environment. Thus,
result of the length of time that perceptual problems someone studying sensation might look at the way
have been studied. The Greek philosophers, the in which electromagnetic radiation (light) is regis¬
pre-Renaissance thinkers, the Arabic scholars, the tered by the eye. This investigator would look at
Latin scholastics, the early British empiricists, and the physical structure of the sense organ and would
the German physicists and physicians who founded attempt to establish how sensory experiences are
both physiology and psychology considered issues related to physical stimulation and physiological
in sensation and perception to be basic questions. functioning. These types of studies tend to focus on
When Alexander Bain wrote the first English text¬ less complex (although not less complicated) as¬
book on psychology in 1855 it was entitled The pects of our conscious experience. For instance,
Senses and the Intellect, with the most extensive these investigators might study how we perceive
coverage reserved for sensory and perceptual func¬ brightness, loudness, or color; however, the nature
tions. The major portion of both the theorizing and of the object having a given brightness, sound, or
the empirical work produced by Wilhelm Wundt, color would not make much difference to them.
who is generally credited with the founding of ex¬ Someone who is interested in the study of
perimental psychology, was oriented toward sen¬ perception is interested in our conscious experi¬
sation and perception. In addition to the diversity ence of objects and object relationships. For in¬
caused by a long and varied history, perception has stance, the sensory question would be “How bright
been affected by many “schools” of thought. Each does the target appear to be?” whereas the percep¬
has its own major theoretical viewpoint and its own tual questions would be “Can you identify that ob¬
particular set of methodological techniques. Thus, ject?” “Where is it?” “How far away is it?” and
we encounter psychophysicists, gestaltists, func¬ “How large is it?” In a more global sense, those
tionalists, analytic introspectionists, transactional- who study perception are interested in how we
ists, sensory physiologists, sensory-tonic theorists, form a conscious representation of the outside
“new look” psychologists, efferent theorists, arti¬ environment, and in the accuracy of that represen¬
ficial intelligence experts, and computational psy¬ tation. For those of you who have difficulty
chologists, to name but a few. There are even in drawing a hard-and-fast line between the con¬
theorists (such as some behaviorists) who deny the cepts of perception and sensation, rest easy.
existence of, or at least deny our ability to study, Since Thomas Reid introduced the distinction in
the conscious event we call perception. Despite this 1785, some investigators have championed its use
chorus of diverse voices and viewpoints, there and others have totally ignored the difference,
seems to be a consensus about the important as¬ choosing to treat sensation and perception as a
pects of perceptual study. unitary problem.
Before we look at the major areas of emphasis Cognition is a term used to define a very ac¬
in the study of the perceptual process let us first tive field of inquiry in contemporary psychology.
offer a disclaimer. We recognize that it is difficult, The word itself is quite old, probably first intro¬
perhaps impossible, and most certainly unwise to duced by St. Thomas Aquinas (1225-1274). He di¬
attempt to draw sharp lines separating one field of vided the study of behavior into two broad
inquiry from another. However, there are certain divisions, cognition, meaning how we know the
10 Chapter 1/Sensation and Perception

world, and affect, which was meant to encompass THEORIES OF PERCEPTION

feelings and emotions. Today’s definition of cog¬
nition is equally as broad as that of Aquinas. Al¬ In the same way that there are many aspects of per¬
though many investigators use the term to refer to ception, there are also many theoretical approaches
memory, association, concept formation, language, to perceptual problems. One important approach
and problem solving (all of which simply take the may be called biological reductionism. It is based
act of perception for granted), other investigators on the presumption that for any given aspect of
include the processes of attention and the conscious the observer’s sensation there is a corresponding
representation and interpretation of stimuli as part physiological event. According to this approach,
of the cognitive process. In other words, cognition the main goal of the perceptual researcher is to iso¬
tends to be somewhere between the areas that were late these underlying physiological mechanisms.
traditionally called perception and learning; and it The search for specific neural units whose activity
incorporates elements of both. The similarity be¬ corresponds to specific sensory experiences, char¬
tween many of the problems studied by cognitive acterized by researchers such as David Hubei
psychologists and those studied by perceptual psy¬ and Torston Wiesel (1979), is common to such
chologists is best seen by the fact that both often theories.
publish in the same journals and on similar topics. Other theoretical approaches are often less
Information processing is a relatively new bound to a specific class of mechanism. For ex¬
term. This approach emphasizes how information ample, direct perception involves a set of theories
about the external world is operated on (processed) that begins with the premise that all the information
to produce our conscious percepts and guide our needed to form the conscious percept is available
actions. Information processing is typically as¬ in the stimuli that reach our receptors, or in rela¬
sumed to include a registration or sensory phase, tionships among these stimuli that are invariant
an interpretation or perceptual phase, and a me- predictors of what is “out there” in the environ¬
moric or cognitive phase. Thus, rather than being a ment. This theoretical position is characterized by
separate subdiscipline, the information processing the work of J. J. Gibson (e.g., 1979), who argued
approach attempts to integrate sensation, percep¬ that certain aspects of the environment are imme¬
tion, and cognition within a common framework. diately impressed on the observer and need no fur¬
It relies on a levels-of-processing analysis in ther computation or additional information based
which each stage of processing, from the first reg¬ on inferences or experience.
istration of the stimulus on the receptor to the final Recently, a number of perceptual theorists,
conscious representation entered into memory, is whose thinking has been influenced by develop¬
systematically analyzed. ments in artificial intelligence systems, have
None of these labels should be taken as rep¬ adopted an alternative approach that contains some
resenting inflexible, or completely separate, areas of the same flavor of direct perception. Such theo¬
of study. At a recent professional meeting one ries are usually presented in the form of computer
well-known psychologist lamented, “When I first programs or computational systems that might al¬
started doing research, people said I studied per¬ low machines to directly interpret sensory infor¬
ception. After a while, they said I studied cogni¬ mation in the same manner that a human observer
tion. Now they say I’m studying human might. Typical of such theorists is David Marr
information processing. I don’t know what’s going (1982), who began with the general presumption
on—I’ve been studying the same set of problems made in direct perception that all the information
for the last ten years!” needed is in the stimulus inputs, but added the sug-
 The Plan of the Book 11

gestion that this interpretation might require the de¬ ture, and a city planner might look at the same
tection of fairly subtle dimensions in the stimulus bridge in terms of traffic flow. At first glance there
and might also require a number of computations may seem to be very little overlap between the var¬
and several stages of analysis. This added feature ious views, since the city planner does not care
has resulted in the label computational approach about the specific shape of the bridge structure, and
being applied to such theories. the engineer cares only about the structural aspects
A much older (but still active) theoretical ap¬ of the beams, not their specific alloy constituents.
proach begins with the recognition that our percep¬ Yet each level of analysis is valid for some specific
tual representation of the world is much richer and set of questions. This book addresses the problem
more accurate than might be expected on the basis of how people build a conscious picture of their
of the information contained in the stimuli available environment through the use of information reach¬
at any one moment in time. Theories to explain this ing their senses. We follow the lead of many con¬
fact often begin with the suggestion that perception temporary theorists and try to use data from all
is much like other logical processes. In addition to levels of the perceptual process, and discussions in
the information available to our sense organs at the terms of several different theoretical positions, in
moment, we can also use information based on our order to give an integrated picture of the process of
previous experience, our expectations, and so perception. After all, the label we apply to our ap¬
forth. This means, for example, that a visual per¬ proach is of considerably less importance than the
cept may involve other sources of information, answer itself.
some nonvisual in nature, some arising from our
past history and cognitive processing strategies. THE PLAN OF THE BOOK
The similarity of many of these mechanisms to rea¬
soning leads us to refer to this type of theory as The orientation of this book is implicit rather than
intelligent perception. This approach probably explicit. Although theories are introduced and dis¬
originated with Helmholtz in 1867, and survives cussed in the various chapters, no all-encompassing
today in the work of researchers such as Irving theoretical position has been adopted. We have
Rock (1983) who have a more cognitive orienta¬ chosen to be “militantly eclectic” in our orienta¬
tion. These theories are also called constructive tion. Thus, this text is mostly concerned with
theories of perception, since our final conscious perceptual and sensory processes. In general, the
impression may involve combining a number of presentation of the material follows a levels-of-pro-
different factors to “construct” the final percept. cessing approach, in that the first half of the book
It is quite likely that each of these approaches is concerned with the more basic sensory processes
is useful in describing some aspects of the percep¬ and is organized around specific sensory systems,
tual process (see Coren & Girgus, 1978; Uttal, such as vision or audition, and the second half of
1981); however, different orientations tend to lead the book is concerned with the more clearly per¬
researchers in different directions, searching for ceptual processes that have strong cognitive influ¬
different types of mechanism. Each approach is ences, and are often not bound to any single
likely to be valid for some parts of the problem and sensory modality.
irrelevant to others. This is a common occurrence We have tried to make the individual chapters
in many areas of endeavor. For instance, a metal¬ relatively self-contained. We begin by explaining
lurgist might look at a bridge and consider its ma¬ how sensations and perceptions are measured
terial components, whereas a civil engineer might (Chapter 2). We then proceed with the physiologi¬
look at the load-bearing capacity of the entire struc¬ cal structures and the basic sensory capacities as-
12 Chapter 1/Sensation and Perception

sociated with vision (Chapters 3 through 5), often they demonstrate concepts that are very dif¬
audition (Chapters 6 and 7), and the chemical and ficult to put into words, but which, when experi¬
mechanical senses (Chapters 8 and 9). For those enced, are immediately understandable. You are
who feel a bit “rusty” about some of the very encouraged to try these demonstrations since they
basic physiological facts, we have also included a are an integral part of the book. In the same way
“Primer of Neurophysiology” as an appendix. that perception involves interaction with the world,
Chapters 10 to 15 deal with those problems that these demonstrations allow you to interact with
have traditionally been treated as part of classical your senses in a controlled manner and to gain in¬
perception, our perceptual representation of space, sight into yourself.
time, motion, form, and size. The more cognitive We hope this book will provide you with some
aspects of perception are also introduced here in understanding of the abilities and the limits of your
those chapters that deal with the issues of music, senses. This knowledge should expand your com¬
speech perception, and attention. The last three prehension of many behavioral phenomena that de¬
chapters (16 to 18) deal with perceptual diversity, pend on perception as a first step. Perception seems
which incudes many of the factors that make the to be the final judge of the truth or the falsity of
perceptual experience of one individual different everything we encounter as part of our human ex¬
from that of another. These factors include the perience. How often have you heard the phrase
changes that occur in the developing individual “Seeing is believing” or “I didn’t believe it until
because of the normal aging process, life history, I saw it with my own two eyes”? Yet you have
experience, learning, and personality factors, to already seen in this chapter that such faith in the
name a few. truthfulness of our conscious percepts is often mis¬
You will notice that each chapter includes a placed. In 500 B.c., Parmenides considered how
series of Demonstration Boxes. These are experi¬ perception can deceive us, summarizing his feel¬
mental demonstrations that you can perform for ings in these words: “The eyes and ears are bad
yourself using materials that are easily found witnesses when they are at the service of minds that
around a house or other living quarters. They illus¬ do not understand their language.” In this book,
trate many aspects of the perceptual process. Quite we will try to teach you their language.
 Glossary 13

GLOSSARY

The following definitions are specific to this book.

Biological reductionism The theoretical premise illusions Distortions or incongruencies between per¬
that each sensory experience is associated with particular cept and reality.
physiological events. Information processing The processes by which
Cognition The process of knowing, incorporating stimuli are registered in the receptors, identified, and
both perception and learning. stored in memory.
Computational approach Involves the presumption Intelligent perception The theoretical presumption
that certain perceived qualities require computation and that cognitive processes and experience can affect per¬
that these computations can be precisely described math¬ ception.
ematically. Levels-of-processing analysis Analysis of the con¬
Constructive theories These maintain that percep¬ tribution of each stage of processing to the final percept,
tion may involve the integration of several sources of beginning with the receptor and continuing through cog¬
information, and may be affected by cognitive factors nitive mechanisms.
and experience. ^ Perception The conscious experience of objects and
Direct perception The theoretical position that all object relationships.
the information needed for the final conscious percept is ✓ Sensation Simple conscious experience associated
in the stimulus array. with a stimulus.
 ' CHAPTER
2
Psychophysics
DETECTION
• * Method of Limits
Method of Constant Stimuli
Signal Detection Theory
IDENTIFICATION
Information Theory
Channel Capacity
• • DISCRIMINATION
Weber’s Law
Signal Detection Theory in
Discrimination
Reaction Time
SCALING
• • Indirect Scaling: Fechner’s Law
Direct Scaling
Category Judgment
Magnitude Estimation: Stevens’s Law
Cross-Modality Matching
Adaptation Level Theory

15
16 Chapter 2/Psychophysics

T he ocean liner glides slowly through the thick

stormy night. Somewhere in the distance is
its berth at the dock. The captain of the tugboat
peers from his bridge, carefully judging the dis¬
New York harbor. With the visibility near zero the tance between the ship and the concrete wall of the
captain is forced to rely solely on the ship’s radar pier. He must continually ask himself, “How far
system for information about the position of obsta¬ does the ship appear to be from the pier?” Such
cles impeding the passage of his ship. The ship is questions are part of another sensory problem,
in a heavily traveled trade route, and the crew must “How much of X is there?” This is the problem of
continually be alert for possible collisions with scaling.
other ships. The radar operator is watching her These four problems, detection, identification,
screen intently, searching for a radar echo caused discrimination, and scaling, are the central con¬
by the presence of another ship nearby. Actually, cerns of the area of perceptual psychology called
she is also wrestling with a basic sensory-percep¬ psychophysics. Psychophysics owes its name and
tual problem, that of detection. She is trying to origin to Gustav Theodor Fechner (1801-1887), a
answer the question “Is there anything there?” physicist and philosopher who set out to determine
She is sure she sees an echo. Now the ques¬ the relationship between the magnitude of a sensa¬
tion becomes “What is it?” Is it an echo from an¬ tion registered in the mind and the magnitude of
other ship or just a “ghost,” a false echo often the physical stimulus that gave rise to it. Hence,
encountered in stormy weather? The radar operator the name psychophysics (from the Greek roots psy¬
is facing a second basic problem, identification. che, or “mind,” and physike, which refers to nat¬
We normally solve the detection and identification urally occurring phenomena). Fechner not only
problems quickly and automatically, since we gen¬ established the philosophical rationale for studying
erally encounter stimuli that are so strong, and pro¬ the relationship between sensations and physical
vide so much information, that they pose little stimuli but also developed many of the experimen¬
problem for us. The complex nature of detection tal methods still in use today. These methods of
and identification only emerges in the context of a collecting and analyzing data are employed in
difficult or degraded stimulus situation. every aspect of the study of sensation and percep¬
The echo turns out to be just a “ghost” and tion (see, e.g., Laming, 1986) and of many other
the order is given to maintain the previous heading areas of psychology, including even social, person¬
(compass direction). The helmsman has been given ality, and clinical psychology (Baird & Noma,
the bearing and now holds the ship’s direction so 1978; Grossberg & Grant, 1978; Wegener, 1982).
that the compass needle always points to the correct
place on the dial. At this moment, he is asking
himself, ‘ ‘Has the needle drifted slightly toward the DETECTION
north?” If so, he must compensate by turning the
wheel so that the needle moves back to the desired The basic task for any sensory system is to detect
compass point. He is continuously concerned with the presence of energy changes in the environment.
the problem of whether the compass needle is cen¬ Energy changes may take the form of electromag¬
tered on the desired heading. This task also evokes netic (light), mechanical (sound, touch, movement,
an important perceptual process, called discrimi¬ muscle tension), chemical (tastes, smells), or ther¬
nation. “Is this stimulus different from that one?” mal stimulation. The problem of detection is cen¬
is the general discrimination question. tered around the problem of how much of such a
Finally, through the clearing weather the en¬ stimulus (relative to a zero energy level) is neces¬
trance to New York harbor appears. The ship is sary for an individual to say that the stimulus is
taken in tow by a tugboat and maneuvered toward heard, tasted, smelled, or felt. Classically, this
 Detection 17

minimal amount of energy has been called the ab¬

Threshold
solute threshold. In 1860, Fechner defined a stimulus levelX.
threshold stimulus as one that “lifted the sensation 1.00
co
or sensory difference over the threshold of con¬ CD
CO
c=
sciousness.” The idea is that below some critical o
Q.
CO
value of a stimulus a person would not be expected 0)

to detect that stimulus. As soon as this threshold CO

CD .50
value is exceeded, however, we would expect the >,
observer to always detect its presence. o
c
o
We can represent this relationship by a graph
o
on which we plot the percentage of time an ob¬ CL
o
server would be expected to detect the presence of
a stimulus (values along the ordinate, or vertical
0 1 2 3 4 5 6 7
axis) against stimulus magnitude (values along the
Stimulus intensity
abscissa, or horizontal axis). This has been done in
Figure 2-1 using arbitrary values for stimulus in¬ FIGURE 2-1 Absolute threshold.
tensity. Notice that the percentage of time that the
stimulus is detected takes a sudden step up from 0
to 100 percent when the stimulus reaches a value ble 2-1 is that the absolute threshold for hearing is
of 3.5. The absolute threshold is thus 3.5. not a fixed value as we first proposed, but appears
to vary from trial to trial. For instance, in Trial 6
the observer could no longer detect the stimulus
Method of Limits
when we presented a tone with an intensity of 8; in
How do we measure absolute thresholds? Let us Trial 4, a stimulus intensity of only 5 was detected.
conduct a relatively simple but typical experiment Such data indicate that the absolute threshold is
to measure the threshold of hearing. In this exper¬ anything but absolute. It seems that the threshold
iment an observer sits in a soundproof room wear¬ varies from measurement to measurement, or mo¬
ing headphones. The experimenter presents a very ment to moment. As early as 1888, Joseph Jastrow
faint, undetectable tone of a particular and constant speculated on the reason for this variability in the
frequency and increases its intensity in small steps threshold over time. He theorized that lapses of at¬
until the observer reports, “I hear it.” On alternate tention, slight fatigue, and other psychological
trials the experimenter starts with a tone that can changes could cause the obtained fluctuations.
easily be heard and decreases the intensity until the Demonstration Box 2-1 shows how you can expe¬
observer reports, “I no longer hear it.” This rience this threshold variability for yourself.
method of determining a threshold is called the We can compute an estimate of the average
method of limits. Kraepelin gave it that name in absolute threshold from the tabled data simply by
1891, because a stimulus series always ends when taking the average stimulus intensity at which a re¬
the observer reaches a limit or a point of change in sponse shifted either from an “I hear it” to an “I
his judgments. The two modes of presenting the don’t hear it,” or from an “I don’t hear it” to an
stimulus (increasing or decreasing) are usually “I hear it.” This gives us a threshold value of 6.65
called ascending or descending stimulus series. A intensity units. These computations are shown at
sample of the kind of data such an experiment the bottom of the table. Table 2-1 also shows that
might generate is shown in Table 2-1. there is a slight difference in the threshold value
The first thing we notice about the data in Ta- depending on whether it was computed from an as-
18 Chapter 2/Psychophysics

Table 2-1. Determination of the Absolute Threshold of

Hearing by the Method of Limits

Trials

Sound intensity f I f 1 T I
(scale units) 1 2 3 4 5 6

16 +
15 +
14 + +
13 + +
12 + + +
11 + + +
10 + + +
9 + + +
8 + + + +
7 - + +
6 +a +
b +
5
4
3
2
1
Threshold for series 5.5 7.5 7.5 4.5 6.5 8.5
computations

Mean descending threshold = 7'5 + 4-5 + 85 = 6.8

3
5.5 + 7.5 + 6.5
Mean ascending threshold = o.5
3
Mean absolute threshold = 6.65 sound units

a. “I hear it.”
b. “I don’t hear it.”

cending or a descending series of stimuli. Such dif¬ ries, and we begin each series of the same kind at
ferences may arise from observers continuing to different stimulus intensities.
report yes in a descending series and no in an as¬ The method of limits has been modified to
cending series, a tendency called the error of per¬ produce a different method for measuring absolute
severation. It is also possible to have an error of thresholds called the staircase method. Here the
anticipation. Here an observer feels that she has experimenter attempts to capture the absolute
said yes too often and decides that it is time to say threshold by changing the direction of the steps
no even though she still faintly hears the tone. To whenever the observer changes her response. Thus,
balance out such possible constant errors we use we might increase the intensity of a tone, step by
alternating ascending and descending stimulus se¬ step, until the observer reports that she hears it, and
 Detection 19

DEMONSTRATION BOX 2-1. The Variability of the Threshold

For this demonstration you will need a wristwatch or distance from the watch you will just begin to hear
an alarm clock that ticks. Place the clock on a table the source of the sound. This is your momentary
and move across the room so that you can no longer threshold. Now hold this position for a few moments
hear the ticking. If the tick is faint, you may accom¬ and you will notice that occasionally the sound will
plish this merely by moving your head away some fade and you may have to step forward to reach
distance. Now gradually move toward the clock. Note threshold, whereas at other times it may be noticeably
that by doing this, you are actually performing a louder and you may be able to step back farther and
method of limits experiment since the sound level still hear it. These changes are a result of your chang¬
steadily increases as you approach the watch. At some ing threshold sensitivity.

then start to decrease it, one step at a time, from generated fluctuating background. It is as if every
that level until it is no longer heard, and then again stimulus to be detected is superimposed on a back¬
start to increase it by steps. Notice that in this way ground of noise generated within the observer.
the value of the test stimulus flips back and forth As this endogenous, or internal, noise level
around the threshold value. The advantage of this changes, so does our measured threshold, in the
procedure is that it allows the experimenter to same way that a person standing in the midst of a
“track” the threshold, even if sensitivity is contin¬ noisy crowd must talk louder in order to be heard.
ually changing, such as after administration of Some experimenters have resorted to the introduc¬
some drugs, or during adaptation to different back¬ tion of experimentally controlled background noise
ground stimuli (Bekesy, 1947; Jesteadt, 1980). in order to achieve more constant conditions than
Why does the threshold seem to vary from would be possible if they relied on the constancy
moment to moment? First we must recognize that of internally generated noise. Under these circum¬
we have been assuming that the only stimulus pres¬ stances, the experimenter has a better idea of the
ent is the stimulus we are asking our observer to noise level with which the stimulus is competing.
detect. This is quite false. A constantly present and Many of the experiments we will discuss have em¬
ever-changing background of stimulation exists for ployed such a controlled background noise level.
any signal we present. If you place both your hands By noise we mean any background stimulus other
over your ears to block out the room noises, you than the one to be detected. Of course, if we define
will hear a sound one observer poetically called noise in this way we may have visual, chemical,
“the sound of waves from a distant sea” and an¬ mechanical, and thermal, as well as auditory noise.
other, somewhat less poetically, “the faint hissing
of radio static.” Similarly, if you sit in a com¬
Method of Constant Stimuli
pletely lightproof room in absolute darkness, you
do not see complete blackness. Your visual field Discussion of another method will allow us to see
appears to be filled with a grayish mist (which has more clearly the nature of the absolute threshold.
been termed “cortical gray”) and occasionally you This method is preferred when the threshold must
can even see momentary bright pinpoint flashes be measured precisely, but it is much more time-
here and there. Any stimulus we ask an observer to consuming to use because it requires so many stim¬
detect must force itself through this spontaneously ulus presentations and responses. Suppose we take
20 Chapter 2/Psychophysics

a set of stimuli ranging from clearly imperceptible strength. The strengths of various behaviors can be
to clearly perceptible and present them, one at a represented by numbers, which indicate their rela¬
time, to our observer. We present each stimulus tive magnitudes. The measure that has found most
many times in a prearranged irregular order. The favor among contemporary workers in the field is a
observer is simply required to respond yes when numerical estimate of the likelihood that the partic¬
she detects the stimulus and no when she does not. ular response in question will occur. We call this
This procedure is called the method of constant likelihood the response probability. We can esti¬
stimuli, a name derived from the fact that a fixed mate the response probability for detecting the
or constant set of stimuli is chosen beforehand and stimulus, or more exactly for saying “Yes, I see
presented a fixed or constant number of times to it,” by using the formula
each observer. Some typical data obtained with this
method are presented graphically in Figure 2-2.
Number of “Yes” Responses
We see in the figure that as the stimulus en¬ p( Yes) = ---
Number “Yes” + Number “No”
ergy increases, the relative number of times the ob¬
server says yes (meaning the stimulus was
perceived) gradually increases. It is not the single The data in Figure 2-2 make it clear that the
jump we might have predicted from the definition probability that an observer will detect a stimulus
of absolute threshold illustrated in Figure 2-1. is not an “all or none” affair (as in Figure 2-1),
These S-shaped curves, called ogives, are obtained but rather changes gradually as the stimulus inten¬
commonly with the method of constant stimuli in sity increases. Then where is the absolute thresh¬
all sensory systems. old? Here, as in many places, we must make a
What does the proportion of yes responses in¬ somewhat arbitrary decision. The point usually
dicate in such experiments? One basic assumption taken as the absolute threshold is that value
made by psychophysicists is that any type of be¬ where the probability of saying yes is the same as
havior, such as saying “Yes, I see it,” has some the probability of saying no. This is simply the
stimulus intensity that the subject claims she de¬
tected 50 percent of the time. In Figure 2-2 we
have indicated this threshold value by dotted
lines. The threshold is about 3.5 energy units for
this observer.
Here are some examples of approximate
threshold values as measured by these methods.
The visual system is so sensitive that a candle
flame can be seen from a distance of more than 48
kilometers on a dark clear night. In the auditory
system, we can detect the ticking of a wristwatch
in a quiet room at a distance of 6 meters—sensitiv¬
ity beyond this point would allow us to hear the
sound of air molecules colliding. As for our other
senses, we can taste 1 teaspoon of sugar dissolved
Stimulus intensity
in 7VS liters of water and smell 1 drop of perfume
FIGURE 2-2 Typical data from method of diffused through the volume of an average three-
constant stimuli in detection. room apartment (Galanter, 1962).
 Detection 21

Signal Detection Theory Somehow their behavior was reasonable, although

it was not clear what they were doing.
Some of you may have been bothered by one as¬ If we now change our classical absolute
pect of the psychophysical measurement techniques threshold experiment so that we can study not only
we have been discussing. We are supposedly the observer’s ability to detect a stimulus when it
studying an observer’s sensory capacities, yet we is there but also his guessing behavior as reflected
have not been talking about the probability that an in a yes response when no signal is present, we
observer detects a stimulus, but rather the probabil¬ have entered the domain of signal detection theory
ity that he says “Yes, I hear (or see, or whatever) (see Baird & Noma, 1978; Egan, 1975; Green &
it.” We can imagine that if an observer feels that Swets, 1966). It is a mathematical, theoretical sys¬
this is a “test” of some sort, where it would be tem, which recognizes that the observer is not
good for him to appear to be quite sensitive, he merely a passive receiver of stimuli but is also en¬
might say yes on almost every trial. What is to pre¬ gaged in the process of deciding whether or not he
vent this from happening? Although we might ar¬ is confident enough that the stimulus was present
gue that people are basically honest, and would not to say “Yes, I detected it.”
lie about whether or not they heard a stimulus, this For the purposes of the following discussion
is not the sort of guarantee on which scientists we shall list all the possible behaviors in a new
would like to rest their conclusions. We are not type of detection experiment and give them names.
criticizing the reliability of observers in psycho¬ Table 2-2 is a schematic representation of the stan¬
physical experiments, for most are quite sincere dard signal detection experiment. There are two
and honest. Rather, we are pointing out that at the types of “stimulus” presentations (at the left of the
very low stimulus energies used in most detection table). A signal absent presentation is like a clas¬
experiments an observer may be unsure about sical catch trial in which no stimulus is presented
whether a sensation has been experienced. This and the observer sees or hears only the noise gen¬
may result in being unsure as to whether or not to erated by the sensory system. Signal present is a
respond yes on any particular trial. Thus, on some trial in which the experimenter actually presents the
trials a “guess” response is made. Therefore, in target stimulus (which is, of course, superimposed
order to assess sensory capacities accurately, we on the endogenous noise in the sensory system).
must take into account the observer’s decision¬ There are also two possible responses in the exper¬
making behavior. iment (at the top of the table). Yes indicates that
Experimenters became aware of this problem the observer thinks a stimulus was presented on a
early in the history of psychophysics. They first at¬ particular trial (that is, signal present), and no in-
tempted to cope with it by inserting catch trials,
which were trials in which no stimulus was pre¬
sented. They reasoned that if observers were honest Table 2-2. Outcomes of a Signal Detection
in reporting what was detected, they would respond Experiment
no on these catch trials. If the yes response came
Response
too frequently, the observer was warned by the ex¬
perimenter. Alternatively, an attempt was made to
Signal Yes No
adjust the calculated threshold to account for the
guesses, or the data were simply discarded. Over
Present Hit Miss
many experiments, however, it became clear that Correct negative
Absent False alarm
the observers were not trying to fool anyone.
22 Chapter 2/Psychophysics

dicates that the observer thinks the signal was ab¬ nonsensory aspects of the situation might influence
sent. The combination of two possible stimulus his pattern of responding. Consider the effect of his
presentations and two possible responses leads to expectations. If the observer knows that the signal
four possible outcomes on a given trial (the four is present on almost every trial he might find him¬
cells of the table). When the signal is present and self responding yes to even the faintest or most am¬
the response is yes the observer makes a hit. But biguous of sensations (perhaps even generated by
if the observer responds yes when the signal is ab¬ endogenous noise in his own nervous system). This
sent, then a false alarm is made. The other cells is sensible behavior if the stimulus occurs most of
are called misses and correct negatives. The rela¬ the time, because on these “doubtful” trials he will
tionships among these responses depend not only quite often be correct. However, if the signal rarely
on the nature of the signal but also on the decision occurs, he would be less tempted by ambiguous,
processes occurring within the observer. faint sensations and might want to wait until he ex¬
Consider a typical experiment as an example. perienced a stronger sensation before saying yes.
Suppose we want to measure an observer’s ability If our description of what the observer is
to detect a tone. The tone for a given experiment doing is correct, then we should be able to change
will be constant in intensity and frequency. After a his response pattern by changing his expectations,
ready signal, the observer is required to respond by even though his sensitivity remains the same. Typ¬
pushing one button to indicate “Yes, the signal ical results from the same observer are presented in
tone was present” and a different button to signify Table 2-4. In one case the signal was present in 90
“No, it was not.” Let us also consider some dif¬ percent of the trials and in the other only 10 percent
ferent experimental conditions that might be intro¬ of the trials. Notice that when the signal is occur¬
duced. The first is one in which the signal was ring frequently the observer says yes often. This
presented in 50 percent of the trials, and no signal gives him many hits, but also many false alarms.
was presented for the remaining 50 percent. A typ¬ When he expects the signal only occasionally he
ical set of data for one observer, expressing the pro¬
portions of trials on which the four possible out¬
Table 2-4. Outcome Matrices for Two Different
comes occurred (the outcome matrix), is shown in
Conditions
Table 2-3.
Notice that on 25 percent of the trials when Stimulus present 90 percent of the time
the signal was absent the observer responded “Yes,
the signal was present.” Why should the observer Response
report that a signal was present when it was not?
First, clearly he is not always sure that whatever he Signal Yes No

heard was actually the signal. Because of this many

Present 0.95 0.05
Absent 0.63 0.37
Table 2-3. Outcome Matrix When Stimulus Is
Present 50 Percent of the Time Stimulus present 10 percent of the time

Response Response

Signal Yes No Signal Yes No

Present 0.75 0.25 Present 0.35 0.65

Absent 0.25 0.75 Absent 0.04 0.96
 Detection 23

says no more often, thus reducing the number of

false alarms, but also reducing the number of hits.
How, then, do we measure the observer’s sensitiv¬
ity? By our former definition of threshold (the point
at which a signal is detected 50 percent of the
time), the tone is clearly above threshold in the first
instance, whereas in the second it is clearly below
threshold. This does not make sense, since neither
the tone’s strength nor the observer’s sensitivity has
changed. We need some way of separating the ob¬
server’s sensitivity from his decision strategy.
We can approach such a method of analysis
by exploring how the observer’s responses change
for a particular signal strength if we vary only his
expectations by varying the relative frequency with
which the signal occurs. We will obtain proportions
of hits and false alarms for each different signal FIGURE 2—3 ROC curves. Notice how the shape
probability, as we discussed above. If these pro¬ of the curve changes for different levels of
sensitivity. The black dots on the white curve
portions of hits and false alarms are plotted against
represent results with indicated probability of
each other as in Figure 2-3, we obtain a receiver signal presentation.
operating characteristic curve (frequently abbre¬
viated ROC curve), which displays the relation¬
ship between proportions of hits and false alarms bow in the curve can serve as a measure of the
as the likelihood of the signal changes. The termi¬ perceived signal strength. An alternative way to in¬
nology was inherited from the communications en¬ terpret an ROC curve is in terms of variations in
gineers who first developed signal detection theory. the sensitivity of an observer to a signal of a partic¬
A more descriptive term for those interested in per¬ ular strength. Thus, the two curves in Figure 2-3
ception would be isosensitivity curve, since the could also be interpreted as reflecting two different
curve represents the range of possible outcome sensitivities of a single observer (the more bowed
matrices for one level of sensitivity. As in the pre¬ the curve the more sensitive) or the curves of two
vious example, Figure 2-3 shows that when the different observers with different sensitivities to the
signal is rare, the observer frequently says no even same signal strength.
when the signal is presented. At the high end of the We may also vary the observer’s response pat¬
curve, where the signal occurs frequently, the ob¬ tern, while holding the signal intensity constant, by
server says yes quite often even when the signal is varying the importance or the payoff for a given
not there. response. For instance, if we pay 10 cents for every
An ROC curve (in any modality) reflects an correct detection of the stimulus and do not penal¬
observer’s response pattern for one signal strength. ize the observer for false alarms, the optimal strat¬
If we increase the strength of the signal, we find egy is to guess yes on every trial. This will
that the curve has a more pronounced bow, as maximize the amount of money that can be earned
shown by the curved black line in Figure 2-3. If in the test situation. Contrast this to a situation
we decrease the signal strength, the curve becomes where we deduct 10 cents for each false alarm and
flatter and approaches the 45-degree line, which do not reward for correct detections. Here a reason¬
represents chance responding. Thus, the amount of able observer would minimize the losses by saying
 24 Chapter 2/Psychophysics

no on every trial. Actually, most situations fall caused by the operation of physiological, atten-
somewhere between these two extremes. For in¬ tional, and other variables on the sensory and per¬
stance, we might pay our observer 10 cents for ceptual systems of the observer. Signal detection
every correct response and deduct 5 cents for every theorists represent these fluctuations in the form of
wrong response. This situation may be represented a probability distribution, which is graphed in
in a matrix of numbers as shown in Table 2-5. Figure 2-4 as the “signal absent” curve. The ab¬
Such a set of rewards and penalties is called the scissa is the amount of sensory activity (or sensa¬
payoff matrix. Changing the payoff matrix causes tion level), and the ordinate can be thought of as
changes in an observer’s response pattern in much the likelihood of occurrence of any particular level
the same way that varying an observer’s expecta¬ of sensation over a great many trials. This means
tions concerning stimulus frequency would, so an that even in the absence of any external signal, the
observer’s motives as well as expectations affect re¬ observer experiences some level of sensation that is
sponses during the detection experiment. Thus, by represented by a particular location along the ab¬
systematically varying the payoff matrix of an ex¬ scissa. This level is experienced with a relative fre¬
periment, we can vary an observer’s numbers of quency represented by the height of the curve at
hits and false alarms and produce an ROC curve that point.
similar to that generated by varying the relative fre¬ When a signal is actually presented it occurs
quency of signals. Note that it is the observer’s re¬ against this background of sensory noise. Of
sponse pattern (e.g., the overall number of yes course, the signal produces some sensory response
responses) that varies as the ROC curve is pro¬ of its own, which then adds to whatever amount is
duced, not the sensitivity to the stimulus. Because already present. The effect of this is the creation of
the manipulation of motivation in this case is done a new distribution of sensory activity, the “signal
by varying the payoff matrix, and thus the amount present” curve. On average, the level of activity
of money paid to an observer, this type of experi¬ elicited by the signal added to the sensory noise is
ment has been given the snide name “sweatshop more intense than that of the noise alone. This is
psychophysics.” shown by the fact that the mean of the signal pres¬
Perhaps the theoretical and methodological ent curve is shifted toward higher values of the sen¬
bases for signal detection will become clearer if we sory activity axis in Figure 2-4. When the signal is
look at the detection problem from a different con¬ weak, however, it will not add enough sensory ac¬
ceptual angle. We have said that even when no tivity to make the two distributions (signal absent
stimulus is present an observer’s sensory systems vs. signal present) completely distinct. The two
are still active, generating sensory noise. The distributions in Figure 2-4 would overlap if drawn
amount of noise probably varies from moment to on the same set of axes. You can see from Figure
moment. This fluctuation in noise level is probably 2-4 that some levels of sensation could result either
from presentations of a signal or simply from noise
Table 2-5. A Typical Payoff Matrix for a alone.
Psychophysical Experiment Imagine you are an observer sitting inside the
head trying to decide if a signal has been presented.
Response The only information you have is the intensity of
the sensation. Remember, however, that sometimes
Signal Yes No the noise produces a sensation that is just as intense
as that produced by the signal, as shown in Figure
Present 100 -50
2-4. As a rational observer, you would probably
Absent -50 100
solve this problem by setting a criterion, or cutoff
 Detection 25

Criterion level (p)

plus noise
distribution

Sensory activity level

FIGURE 2-4 Illustration of how signal absent and signal present distributions result in hits, misses,
false alarms, and correct negatives for a particular criterion setting. Notice that the two curves are
actually plotted on the same axes—they are separated for clarity. The curves would overlap if plotted
together.

point, for sensation level. This is the value you are If this is what the observer is doing, then we
willing to accept as probably indicating that a sig¬ can specify the proportions of hits and false alarms
nal is present. If a sensation level is below the cri¬ we might expect, depending on where he places his
terion (to the left in Figure 2-4), you respond no; criterion. According to signal detection theory, the
if it is above the criterion, you respond yes. This proportions of the various outcomes observed in an
simplifies the problem greatly, since you must experiment (see Table 2-2) may be represented as
only decide, based on your motives and expecta¬ that proportion of the area under the appropriate
tions, where to put the criterion. From that point probability distribution curve to the right or left of
on, the experienced level of sensation more or less the criterion location. Thus, if Figure 2-4 repre¬
automatically determines the response. The crite¬ sents an actual situation, the proportion of signal
rion value is usually symbolized by the Greek let¬ present trials on which a yes response would be
ter P (beta). given (the proportion of hits) is represented by the
26 Chapter 2/Psychophysics

area under the signal present curve to the right of Criterion (/3)
the criterion, since the observer would say yes
whenever the sensation level was above, or to the
right of, the criterion. Similarly, the proportion of
false alarms is represented by the area under the
signal absent curve to the right of the criterion,
since that is the proportion of trials on which the
sensation level generated by the sensory system in
the absence of a signal exceeded the criterion level
set for the yes response. The other two possible
outcomes are also represented in Figure 2-4.
The motivation and expectation effects on an
observer’s response pattern in a detection experi¬
ment are now interpretable. Essentially, these vari¬
ables affect the placement of the criterion and,
hence, the proportion of hits and false alarms. For
instance, suppose that the observer is a radiologist
"1 Proportion of
looking for a light spot as evidence of cancer in a J false alarms
set of chest X rays (see e.g., Swensson, 1980). If
the radiologist thinks she has found such a spot, Proportion

she calls the patient back for additional tests. The of hits Criterion (0)
penalty for a false alarm (additional tests when no
cancer is present) only involves some added time
and money on the part of the patient, whereas the
penalty for a miss (not catching an instance of real
cancer) might be the patient’s death. Thus, the ra¬
diologist may set a criterion value that is quite low
(lax), not wanting to miss any danger signals. This
means she will have many hits and few misses, but
also many false alarms, a situation shown in Figure
2-5A. Conversely, if the observer is a radar oper¬
ator looking for blips on a screen signifying enemy
missiles, he might be much more conservative.
Here the penalty for a false alarm could be war,
whereas the penalty for a miss might be only a few
seconds lost in sounding the alarm. He would set a
high (strict) criterion in order to avoid false alarms,
but at the penalty of reducing the number of hits. FIGURE 2—5 The effect of motives or
This would be equivalent to the situation shown in expectations on criterion placement and
Figure 2-5B. In this same manner, each point on proportion of hits and false alarms.

any given ROC curve simply represents a different

criterion setting. sitivity of the observer. That is because there is no
Although we indicated that the location of the such effect. In signal detection theory, sensitivity
criterion alters the pattern of response, we did not refers to the average amount of sensory activity
mention the effect of criterion location on the sen- generated by a given signal as compared with the
 Identification 27

average amount of noise-generated activity. This is

similar to the everyday use of the word sensitivity.
Thus, a radio receiver that produces a large electri¬
cal response that allows a weak signal to be heard
above the background static is more sensitive than
one that produces only a small electrical response
to that signal, which may then be obscured by
static and noise.
Within our present framework, the perceptual
analog of sensitivity is the distance between the
centers (means) of the signal absent and the signal
present distributions. This is merely a measure of
the difference in average sensation levels as a func¬
tion of the presence or absence of a signal. We call
this distance measure of sensitivity d' (see Figure
2-4). When the distributions are far apart, and
overlap very little, as in Figure 2-6B, d' is large
and the ROC curve is far from the diagonal and
sharply curved. When the distributions are close to¬
gether, and overlap to a great extent, d' is rela¬
tively small, as in Figure 2-6A. The corresponding
ROC curve is close to the diagonal, which you may
remember represents zero sensitivity. Signal detec¬
tion theory attempts to measure an observer’s sen¬
sitivity to a signal independently of his decision
strategy, while acknowledging that both might af¬
fect the actual responses made in the experimental
setting. Instructions for calculating d' and /3 using
proportions of hits and false alarms obtained from
any typical signal detection experiment (e.g., Ta¬
bles 2-3 and 2-4) can be found in Computation
Box 2-1.
FIGURE 2-6 The effect of sensitivity and signal
This must seem like an unusually elaborate
strength on d'.
procedure for investigating a seemingly simple
problem, namely, the determination of the minimal
amount of energy necessary for stimulus detection. use the d' measure, which provides an index of the
However, an observer is a living organism whose observer’s sensitivity to stimuli, instead of the tra¬
expectations and motives affect his or her percep¬ ditional detection threshold measures.
tual behaviors and judgments nearly as much as
stimulus reception itself does. These nonperceptual
effects must be removed if we are to look at the IDENTIFICATION
pure sensory responses. Our original notion of an
absolute threshold has proved to be too primitive. The doctor listened very carefully, paused for a
The detection threshold is simply a convenient sta¬ moment to adjust the stethoscope to a more com¬
tistically defined point. As an alternative we may fortable position, and listened again to the sounds
28 Chapter 2/Psychophysics

COMPUTATION BOX 2-1 Calculating d' and p

To calculate d' and p, first obtain the outcome matrix Z(HIT) happens to be negative, is equivalent to add¬
from a signal detection experiment. Then find the ing a positive number, i.e., 2 — ( — 3) = 5. The
false alarm rate from the outcome matrix in the HIT/ value of p can be obtained similarly, except that you
FA column of the accompanying table. Read across should use the ORD column (for ordinate, the height
the table to the Z column (Z is the usual label of the of the bell curve) to obtain the values of ORD(HIT)
abscissa of the graph of the standard bell curve). Call and ORD(FA), and then plug those numbers into the
the value tabled there Z(FA) and write it down. Re¬ following equation:
peat these operations for the hit rate, calling the ta¬
bled value Z(HIT) and writing it down. Be careful to p = ORD(HIT)/ORD(FA)
record the sign of the tabled Z values—Z(HIT) will
often be negative. Then to obtain d', plug Z(FA) and If the exact values of the hit or false alarm rate do not
Z(HIT) into the following equation: appear in the table, interpolate between the nearest
surrounding values that do appear, or simply round
d' = Z(FA) - Z(HIT) the hit and false alarm rates to the closest number that
does appear. Your answer shouldn’t be too far from
Remember that subtracting a negative number, if the exact value of d’ or p

HIT/FA z ORD HIT/FA z ORD

.01 2.33 0.03 .50 0.00 0.40

.02 2.05 0.05 .55 -0.12 0.40
.03 1.88 0.07 .60 -0.25 0.39
.04 1.75 0.09 .65 -0.38 0.37
.05 1.64 0.10 .70 -0.52 0.35
.08 1.40 0.15 .75 -0.67 0.32
.10 1.28 0.18 .80 -0.84 0.28
.13 1.13 0.21 .82 -0.92 0.26
.15 1.04 0.23 .85 -1.04 0.23
.18 0.92 0.26 .88 - 1.18 0.20
.20 0.84 0.28 .90 -1.28 0.18
.25 0.67 0.32 .92 -1.40 0.15
.30 0.52 0.35 .95 - 1.64 0.10
.35 0.38 0.37 .96 -1.75 0.09
.40 0.25 0.39 .97 - 1.88 0.07
.45 0.12 0.40 .98 -2.05 0.05
.50 0.00 0.40 .99 -2.33 0.03

emanating from the patient’s chest. The sounds with a problem that does not involve stimulus
were quite clear and distinct. The problem was detection, for the sounds are clearly above the
simply to decide whether they indicated a normal detection threshold. However, it does involve
or a pathological heartbeat. This doctor is wrestling identifying one of a number of possible alternative
 Identification 29

stimuli. To identify a stimulus is one of the major fication of the stimulus corresponds to the actual
tasks the perceptual system is asked to perform. stimulus input will be affected both by the ability
The difficulty of any identification task de¬ of the sensory system to handle the stimulus input
pends, in part, on the number of possible stimulus without distortion and by the complexity of the
alternatives an observer is asked to distinguish input.
among. Consider an observer who claims she can The quantitative system for specifying the
identify her favorite brand of cola. Suppose we characteristics of the input message is known as in¬
gave her two unmarked glasses of cola and asked formation theory. Information theory is not really
her to sample them and try to select her own favor¬ a theory at all, but rather a system of measurement.
ite brand. If she did select the correct brand we The amount of information in a given stimulus dis¬
would not be very surprised, since she would be play is defined so that the nature of the object being
expected to do so 50 percent of the time by chance measured is irrelevant. What, then, do we mean by
alone, even if her taste buds were nonfunctional. If information? We mean what the everyday use of
our “expert” selected her own brand out of 25 the word implies. If you tell us that this week will
brands presented to her we would be much more contain a Sunday morning, you have conveyed
likely to take her claim seriously, since the prob¬ very little information, since we know that every
ability that she would by chance alone find her week contains a Sunday morning. If you tell us that
brand out of 25 alternatives is only 1/25. Mea¬ this Sunday morning there will be a parade in hon¬
sures of the difficulty of the identification task our of Jiffy the Kangaroo, you have conveyed a
must therefore take into account the number of great deal of information because you have speci¬
stimulus alternatives. fied which one out of a large number of possible
alternative events was about to occur.
One way to quantify information is to define
Information Theory
it in terms of the questions a person must ask to
To solve the problem of specifying the difficulty of discover which member of a stimulus set has oc¬
an identification task, psychologists in the early curred. Suppose we had only two possible alterna¬
1950s turned to ideas arising from the efforts of tives, A or B, and you were to search for the target
engineers to assess the performance of radio and among them. You need only ask “Is it A?” to de¬
telephone communications systems. Books by termine unambiguously which alternative had been
Shannon and Weaver (1949) and by Wiener (1961) selected as the target. If you receive an answer of
made it clear that the problems faced by the psy¬ “No” you know immediately that B is correct.
chophysicist and by the communications engineer Similarly, if you had to determine which of four
were quite similar. The engineer deals with a mes¬ stimuli. A, B, C, or D, had been chosen as the
sage that is transmitted through a communication target, you could determine it with two questions.
channel and decoded by someone or something at The answer to the question “Is it A or 5?” reduces
the receiver end. The degree to which the final de¬ your number of possible alternatives to two, since
coded message reflects the original message de¬ a “No” answer reveals that it is either C or D,
pends, in part, on the ability of the system to whereas a “Yes” indicates that it is A or B. We
transmit information without distortion (this is what already know that only one more question is nec¬
is meant by the fidelity of a system), and on the essary in order to identify the correct item. Each
complexity of the input. The psychophysicist has necessary question, structured to eliminate exactly
an analogous problem. Stimulus information is half of the alternatives, defines a bit of informa¬
transmitted to an observer through a sensory sys¬ tion. Bit is a contraction of the words binary digit
tem, and it is then decoded in the central nervous (which can be either a 0 or a 1, that is, there are
system. The degree to which the observer’s identi¬ two possible digits).
30 Chapter 2/Psychophysics

Table 2-6. Log2n for Selected Numbers

Observer-information
transmission channel
Number of stimulus (processing takes
Stimulus Response
alternatives (n) Number of bits (log2n) place here)
input output

2 1
FIGURE 2-7 A human information channel.
4 2
8 3
16 4
32 5 stimuli presented. That is, if the observer cor¬
64 6 rectly identifies a stimulus, and gives the correct
128 7 label as a response, information (the correct label)
256 8 has been transmitted from one end to the other,
through the channel represented by the observer.
If the response matches the stimulus perfectly for
The number of bits of information needed to all stimuli, then the observer is a perfect informa¬
determine exactly one stimulus alternative is the tion transmitter.
logarithm to the base 2 of the total number of pos¬ Consider an example in which we are calling
sible stimulus alternatives. The logarithm of a out alphabetic letters from a set containing eight
number n to the base 2, which is written log2n, is items: A, B, C, D, F, G, H, X. If the observer
merely the power to which the number 2 must be correctly identifies (response) the letter we have
raised to equal n. Thus, if we have four alternatives called out (stimulus) then she has transmitted 3 bits
we must raise 2 to the second power (i.e., 22 = of information (log28). Suppose identification is not
2x2 = 4) and log24 = 2. Similarly, Table 2-6 perfect. This means that only some of the stimulus
gives the corresponding number of bits for n alter¬ information is being transmitted. Thus, if the ob¬
natives (a more detailed table can be found in Gar¬ server hears a faint “eee” sound, with the first part
ner, 1962). Each time the number of stimulus of the letter cut off, she does not know exactly
alternatives is doubled the amount of information which letter was called out. However, she can
rises by 1 bit. Of course, for intermediate values eliminate A, F, H, and X, which have no “eee”
the number of bits will not be a whole number (for sound; hence, she has reduced the number of stim¬
example, seven alternatives gives 2.81 bits). ulus alternatives by half, and we would say that 1
bit of information has been transmitted. In general,
the greater the probability that the observer will
Channel Capacity
identify the stimulus—that is, the more she “picks
It is important at this point to define the concept of up” from the presentations—the more information
information transmission. Let us consider an ob¬ she is capable of transmitting.
server as a channel, in the way communications en¬ Consider a hypothetical experiment in which
gineers do. Our observer may be represented as in each of four stimuli are presented 12 times and ob¬
Figure 2-7. A stimulus is presented to the ob¬ servers are asked to identify which stimulus was
server, who is asked to try to identify it. By iden¬ presented. In Table 2-7, Observer A shows perfect
tification we mean giving a response that is the information transmission because every time Stim¬
correct, agreed-upon label for the particular stimu¬ ulus 1 is presented our observer correctly identifies
lus presented. We can say that information is trans¬ it, and every time 2 is presented it is named cor¬
mitted by the observer to the extent that the rectly. Observer B shows poorer information trans¬
responses given match the actual labels of the mission. Notice here that when Stimulus 2 is
 Identification 31

Table 2-7. Stimulus-Response Matrices for Three formation transmitted in such experiments may be
Observers found in Gamer and Hake (1951).
How many bits of stimulus information can an
Observer A: Perfect information transmission
observer transmit perfectly? Let us first look at a
group of stimuli selected from a one-dimensional
Response
physical continuum, such as sound or light inten¬
Stimulus l 2 3 4 sity. The number of stimuli from one continuum
that a subject can identify perfectly has been found
1 12 to be surprisingly small. For the judgment of the
2 12 pitch of a tone, Pollack (1952) found it to be about
3 12 5 different pitches, which is equivalent to about 2.3
4 12
bits of stimulus information. Gamer (1953) found
much the same result for loudness, around 2.1 bits.
Observer B: Some information transmission
Eriksen and Hake (1955) measured several visual
Response
continua and found information transmission to be
limited to 2.34 bits for brightness, 2.84 bits for
Stimulus l 2 3 4 size, and 3.08 bits for hue. Overall, the number of
stimuli that may be perfectly identified on any sin¬
1 8 4 gle continuum turns out to be approximately seven
2 2 8 2 plus or minus two (7 ± 2), depending on the par¬
3 2 8 2
ticular stimulus continuum being tested (see Miller,
4 4 8
1956).
This limit is called (again using communica¬
Observer C: No information transmission
tions theory terminology) the observer’s channel
Response capacity, and typical measurement of channel ca¬
pacity is shown in Figure 2-8. Notice that even
Stimulus l 2 3 4 though we increase the amount of information
available in the display, our subject has reached his
1 3 3 3 3 limit of recognition (about 2.5 bits) and can trans¬
2 3 3 3 3 mit no more information.
3 3 3 3 3 Several theories have been proposed to explain
4 3 3 3 3
this general finding. In the most popular of these,
the limit reflects cognitive or response processes
(e.g., Durlach & Braida, 1969; Gravetter & Lock-
head, 1973; Luce, Green & Weber, 1976; Marley
presented, the observer calls it Stimulus 2 most of 6 Cook, 1984). A less popular view is that the
the time; but sometimes he calls it Stimulus 1 and limit is set by the response characteristics of sen¬
sometimes he calls it Stimulus 3. When he does say sory neurons, and is thus an absolute limit for a
that it is Stimulus 2, however, there is a fair like¬ single sensory continuum (Norwich, 1981).
lihood that it is Stimulus 2. He is much better than Seven seems to be a very small number of
Observer C, who seems to be responding without stimuli to be able to identify. Each of us knows that
reference to the stimulus presented. Observer C is singers, for example, seem able to identify (indeed
transmitting none of the available stimulus infor¬ sing) hundreds of different songs. Every one of us
mation. Formulas for computing the amount of in¬ can certainly identify dozens of faces and thou-
32 Chapter 2/Psychophysics

capacity is so limited for any single stimulus di¬

mension, how can this occur? The answer involves
the number of dimensions along which the stimulus
varies.
For example, Pollack (1953) found that if he
varied only pitch information transmission aver¬
aged about 1.8 bits, whereas if he varied only loud¬
ness information transmission was about 1.7 bits.
When both dimensions were varied simultaneously,
however, information transmission was 3.1 bits.
This is more than was obtained for either dimen¬
sion separately, although not the 3.5 bits expected
if the information transmission on the separate di¬
mensions were simply summed. Nonetheless, the
Stimulus information available (bits) more dimensions the stimulus varies along, the bet¬
ter recognition performance is. Certain ways of
FIGURE 2-8 Channel capacity. The straight combining dimensions seem to produce better per¬
diagonal represents perfect information
formance, by making stimuli “stand out” more
transmission. The curve represents typical
performance. The dotted horizontal line is clearly, or capitalizing on the small gains obtain¬
channel capacity. able by familiarity (Fockhead, 1970; Monahan &
Lockhead, 1977). Thus, by proper selection of
sands of words. How can this be, in light of our stimulus dimensions, Anderson and Fitts (1958)
inability to transmit more than about 3 bits of in¬ were able to obtain information transmission levels
formation per stimulus dimension? of 17 bits on a single flashed stimulus. This means
You might think that if the stimuli were more that their observers could perfectly identify 1 stim¬
widely spaced and discriminable, channel capacity ulus out of more than 131,000 alternative stimuli!
would be higher. The discriminability of stimuli, The importance of stimulus dimensions and
however, has very little effect on identification per¬ how they are combined has led modem investiga¬
formance (Pollack, 1952). Another possible expla¬ tors to place less emphasis on the quantity of infor¬
nation is that our everyday performance is mation available and more emphasis on the quality,
explained by practice or repetition. Except in ex¬ or kind, of information and the characteristics of
treme cases, where a person might have years of the information processor (see Cutting, 1987; Gar¬
intensive practice on a single dimension, the prac¬ ner, 1974; Neisser, 1967). The basic ideas of in¬
tice effect is also not large enough to explain our formation theory, especially those associated with
everyday performance. For instance, you hear a the number of stimulus alternatives, have been
new word today and now you can identify that important in calling attention to critical issues in
word with ease, even though you have only en¬ identification. They have taken their place as
countered it once. You can also recognize that this foundation concepts, almost assumptions, and
word is different from every other word in your modem researchers build on them rather than
vocabulary. We are not at all surprised at such a study them for their own sake (although see Nor¬
performance, yet this type of identification may in¬ wich, 1981, 1984, 1987). Chapter 11 considers
volve the transmission of some 16 bits of informa¬ some of these modern extensions of and alterna¬
tion (or more, depending on the total number of tives to information theory in the study of ob¬
words in your vocabulary). Given that our channel ject identification.
 Discrimination 33

DISCRIMINATION standard. We are also measuring a threshold here,

only this is a threshold for the perception of a dif¬
The artist glances at his model’s hair and then back ference between the standard and the other stimuli.
down at the paint on his palette. He mutters to him¬ It is called a difference threshold.
self, “Still not the same.” He daubs a bit more As psychophysicists worked with the measure¬
black, mixes the color through, and glances up ment of difference thresholds for various stimulus
again. “That is a perfect match,” he grunts. This dimensions, it became clear that the “same” re¬
artist is engaging in an act of discrimination. He is sponse category was being used by observers
determining whether two colors are the same or whenever they were unsure or unwilling to state
different. He does not care what the color actually that there was a perceptible difference. Experi¬
is, it can be burnt sienna or just plain brown, he menters soon resorted to using only two response
cares only whether or not the paint matches his alternatives instead of three. For example, in a
model’s hair color. Discrimination problems ask weight judgment experiment, the observer would
the question, “Is this stimulus different from that only be permitted to respond that the comparison
one?” stimulus is either “heavier” or “lighter.” If he
The study of discrimination has focused on the feels that the comparison and the standard are the
question, “By how much must two stimuli differ same, he is still forced to indicate (by guessing) in
in order to be discriminated as not the same?” Sup¬ which direction they appear to differ. The advan¬
pose the melody “Oh! Susanna” were played on a tage of this procedure was demonstrated by the
piano once in the key of C and once in the key of painstaking work of Brown (1910). He showed that
G. Are these two musical stimuli the same or dif¬ in a weight judgment experiment, stimulus differ¬
ferent? The answer to this question depends on the ences as small as 0.2 gram (which is about 0.008
stimulus dimension being judged. If we are judging ounce) produced more correct than incorrect judg¬
whether the melodies are the same or not we would ments, even when the observer felt that the stimuli
answer differently than if we were judging the key were the same and that he was merely guessing.
in which the melodies are played. The results from such an experiment are easy
To avoid such confusions, the standard dis¬ to display. In the weight judgment experiment, for
crimination experiment involves variation of stim¬ instance, the standard was presented with each
uli along only one dimension. Thus, in a study of comparison stimulus many times. We can plot the
the discrimination of weights we might hold the proportion of the presentations on which any given
size and shape of our stimuli constant and vary stimulus was judged heavier than the standard.
only the weight. In the earlier studies, observers Such a plot is illustrated in Figure 2-9. This plot
were presented with pairs of stimuli and asked to is similar to results from the classic experiment on
make the response “heavier,” “lighter,” or weight judgment by Brown (1910). He used a 100-
“same,” or some similar set of judgments appro¬ g standard and a set of comparison weights ranging
priate to the stimulus dimension being judged. One from 82 to 118 g in 1-g steps. Each comparison
of these stimuli was designated the standard. This stimulus was judged 700 times against the standard
is a stimulus that appears on every trial and is com¬ stimulus. Notice that the shift from reports of
pared to a graded set of similar stimuli differing “lighter” to reports of “heavier” is not very
along the dimension being studied. These graded abrupt, as it would be if the threshold were always
stimuli make up the set of comparison stimuli. a single, unique value. Rather, we find a gradual
This is simply a variant of the method of constant change in the probability of a “heavier” response
stimuli (which, you may remember, is used to de¬ as the stimulus changes from much lighter than the
termine the absolute threshold) to which we add the standard to much heavier. Since the change is grad-
34 Chapter 2/Psychophysics

2 to give us a value that we call the just noticeable

difference, or jnd. The jnd computed for the data
in Figure 2-9 is about 4 g. This means that when
a pair of stimuli are separated by 4 g, the subject
will be able to detect the difference between
them about half the time. You can probably see
that the jnd is simply the average of the threshold
for “greater than” and the threshold for “less
than.” In other words, it represents the threshold
for “different” averaged across the direction of
the differences.
If discrimination were good, we would expect
103 - 95 very small differences between stimuli to be no¬
Difference threshold (Jnd) =---= 4 g
ticed. This corresponds to a small jnd. In Figure 2-
Point of subjective equality = 99 g 10 the black line shows a good discriminator with
a jnd of 0.5 units, whereas the white line shows a
FIGURE 2-9 Typical data from the method of
constant stimuli in discrimination with poor discriminator with a jnd of 2 units. As the jnd
calculations of difference threshold (jnd) and point increases in size and discrimination ability de¬
of subjective equality. creases, the curve begins to flatten. The extreme of
no discrimination at all would be represented by a
ual, we again must make some decision regarding horizontal line parallel to the abscissa at p (heavier)
how we will define the difference threshold. equal to 0.5.
Clearly, the point where p (heavier) is equal You may have noticed an interesting aspect of
to 0.5—that is, where the stimulus was called the data pictured in Figure 2-9. The point of sub¬
“lighter” 50 percent of the time and “heavier” 50 jective equality is not equal to the standard in these
percent of the time—is not appropriate. This 50- data. The stimulus that appears to be equal to the
percent point probably represents the stimulus that standard of 100 g is actually 1 g lighter. This is a
appeared most like the standard, since the choices
are evenly divided on either side of it. Therefore,
it has been called the point of subjective equality.
The lightest stimulus for which p (heavier) is equal
to 1.0 represents perfect discrimination (because
here a physically heavier stimulus is judged heavier
100 percent of the time). The stimulus where p
(heavier) is equal to 0.5 represents no perception of
difference. Therefore, the point where p (heavier)
is equal to 0.75 (halfway between these values)
represents a value where the difference is noted 50
percent of the time. Following similar reasoning, p
(heavier) equal to 0.25 is the point at which a stim¬
ulus difference in the lighter direction is noted 50
percent of the time. By convention, we take the
interval from the 0.25 point to the 0.75 point, FIGURE 2-10 Difference thresholds (jnds) for
called the interval of uncertainty, and divide it by observers of different sensitivity.
 Discrimination 35

typical result in many psychophysical experiments

involving the presentation of stimuli that are sepa¬
rated in time. The stimulus presented first (gener¬
ally the standard) is judged to be less intense than
the later stimulus. This effect has been named the
negative time error. It is negative because the
standard is judged as less intense than it should be.
Fechner (1860/1966) and Wolfgang Kohler (1923)
thought this error was caused by the fading of the
image or the memory trace of the sensation of the
standard with the passage of time. However, work
done with auditory stimuli has shown that with Magnitude of standard stimulus
proper selection of a time interval the error can be
FIGURE 2-11 Effect of intensity of standard on
positive rather than negative (Kohler, 1923). Such
difference threshold (jnd).
errors are probably the result of particular cognitive
or judgmental factors closely related to the adapta¬
tion level (see later), which, as we have seen be¬ it would be necessary to add ten candles in order
fore, tend to influence even the most apparently to detect the same apparent increase in illumina¬
simple perceptual tasks (Hellstrom, 1979, 1985). tion” (p. 133). This relation between the size of
the jnd and the size of the standard intensity is
called Weber’s law after its discoverer.
Weber’s Law
Weber’s law is simply written as
Is the jnd a fixed value for any given sense modal¬
ity, or does it vary as a function of the nature of M = KI
the stimulus input or the state of the observer? Fol¬
lowing the lead of Ernst Heinrich Weber (1834), where A/ is the size of the jnd, I is the intensity of
Fechner (1860/1966) conducted an experiment in the standard stimulus, and K is a constant. The
which he measured the jnds for lifted weights using constant K is always a fraction and is equal to AIII.
standard weights of different magnitudes. We may It indicates the proportion by which the standard
plot the size of the jnd, that is, the amount by stimulus must be increased in order to detect a
which we must increase the stimulus so that it is change. This fractional value is called the Weber
discriminable as different from the standard 50 per¬ fraction. Thus, if the Weber fraction is 0.02, it
cent of the time, against the magnitude of the stan¬ means that we must increase the intensity of a stim¬
dard. This has been done for some illustrative data ulus by 2 percent for a difference between it and
in Figure 2—11. First, notice that the jnd is not a the original stimulus to be detected. This propor¬
constant value. It appears to increase in a linear tion (the Weber fraction) is the same regardless of
fashion with the size of the standard. In other the intensity of the standard stimulus. In a weight
words, as the stimulus magnitude increases so does judgment experiment, for example, to discriminate
the size of the change needed for discrimination to a stimulus as different from a 2-g standard the
occur. The intuitive force of this relationship is weight must be increased by only 0.04 g (2 x 0.02
well illustrated in an example proposed by Galanter = 0.04). To discriminate a stimulus as different
(1962): “If in a room with ten candles you had to from a 200-g standard, it must be increased by 4 g
add one more in order to detect an increase in il¬ (200 X 0.02 = 4). A simple demonstration of
lumination, then if the room contained one hundred Weber’s law is given in Demonstration Box 2-2.
36 Chapter 2/Psychophysics

DEMONSTRATION BOX 2-2. Weber’s Law

It is easy to demonstrate Weber’s law for the percep¬ The weight difference should be almost impercepti¬
tion of heaviness. You will need three quarters, two ble. In the first instance the targets differed by the
envelopes, and your shoes. Take one quarter and put weight of the quarter and the difference was discrim¬
it in an envelope and put the remaining two quarters inated easily. In the second instance, although the
in the other. If you now lift each envelope gently and weight differential was the same (one quarter), the
put it down (use the same hand), it is quite easy to overall stimulus intensity was greater because shoes
distinguish the heavier envelope. Now insert one en¬ weigh much more than the envelopes and the quarters
velope into one of your shoes and the other envelope alone.
into your second shoe, and lift them one at a time.

Conceptually, consider the Weber fraction to Table 2-8. Typical Weber Fractions (A///) (Based
be a measure of the overall sensitivity of a sensory on Teghtsoonian, 1971)
system to differences along a stimulus continuum.
The larger the Weber fraction, the larger will be Continuum Weber fraction

the jnds for any stimulus dimension, hence the

Brightness 0.079
larger the change needed for discrimination. Note
Loudness 0.048
that K has no units (such as grams), so that it does
Finger span 0.022
not depend on the physical units used to measure I
Heaviness 0.020
and AI. Thus, we can compare Weber fractions Line length 0.029
across different stimulus dimensions without hav¬ Taste (salt) 0.083
ing to worry about how the stimulus values were Electric shock 0.013
measured. The Weber fraction simply represents Vibration (fingertip)
the average ratio of jnd size to the size of the stan¬ 60 Hz 0.036
dard level at which the jnd was measured, over an 125 Hz 0.046
entire range of standard values. Table 2-8 presents 250 Hz 0.046
typical Weber fractions for a variety of continua.
As you can see, some of the Ks are relatively large
(for example, those for brightness and loudness), data from loudness discrimination experiments by
and some are quite small (for example, electric Miller (1947) and Riesz (1928), and we see a con¬
shock). siderable deviation from the expected constancy at
How well does Weber’s law fit the data? For both extremes. Although these deviations at the ex¬
many years there was considerable argument about tremes look very large, this is only because we
this issue. Measurements were taken in many sense have plotted the stimuli in logarithmic units. The
modalities to check the relation. The clearest pic¬ flat part of the curve actually exceeds 99 percent of
ture of the results is given by plotting the value of the total range of intensities used. Thus, Weber’s
the Weber fraction, AHI, against the standard stim¬ law is a useful summary in spite of the deviation
ulus intensity. If the Weber fraction is actually con¬ of the data from a perfect fit. Moreover, even the
stant, we should see a horizontal line, parallel to deviations are beginning to be understood (see
the abscissa. Figure 2-12 shows a composite of Green 1976; Norwich, 1987).
 Discrimination 37

tally relabel the two distributions Signal 1 and

Signal 2. Two stimuli can give rise to a variety of
different sensation levels, with different probabili¬
ties. Since the curves cover the same general area
of the sensation axis, there is no way to be certain
which stimulus elicited a given sensation level on
any one trial. The best the ideal observer can do is
to place a criterion somewhere on the sensation
axis, and simply determine whether the sensation
level experienced is above or below that criterion.
If above, the appropriate response would be that
Logarithm of magnitude of standard stimulus (log I) the presented stimulus was a 2; if below, a 1. Just
FIGURE 2-12 Typical data for test of Weber’s as in the absolute detection situation, where the ob¬
law. The dotted line is predicted by Weber’s law: server places the criterion will greatly affect the
M/I = K. proportions of different responses he gives. In turn,
criterion placement will be affected by the observ¬
er’s expectations as to the relative frequency of
Signal Detection Theory in presentation of the two stimuli, and the observer’s
Discrimination present motivational biases.
As in the detection experiment, different cri¬
Although signal detection theory was presented terion placements will define an isosensitivity curve
(and first developed) in the context of the detection when we plot the proportion of hits against the pro¬
problem, it can be extended to the discrimination portion of false alarms. The measure of sensitivity
situation. Certainly there are decisional components to the difference between the two stimuli is still
that influence whether or not an observer discrimi¬ called d' and is still unaffected by changes in the
nates a difference between two stimuli. To use the criterion. Actually, d' is determined by the physical
signal detection procedure to assess discrimination, difference between the two stimuli and the sensitiv¬
we must redesign the method of constant stimuli ity of the observer’s sensory system; both are fac¬
experiment so that the observer is asked to say tors that determine the difference between the
which of two very similar stimuli was actually pre¬ average levels of sensation evoked by the stimuli.
sented on a given trial. This is like an identification Thus, d' represents a measure of just how discrimi¬
experiment with only two stimuli. nate two very similar stimuli are. As such, it is
The signal detection analysis of this experi¬ closely related to the difference threshold and to the
ment is quite similar to that used for detection. In¬ Weber fraction (Treisman, 1976; Treisman &
stead of trying to ascertain whether the sensation Watts, 1966).
experienced on a given trial came from the signal
present or the signal absent distribution, the ob¬
Reaction Time
server must decide whether it is from the Signal 1
or the Signal 2 distribution. If the stimuli were very We have been looking at stimuli that are diffi¬
similar the sensory response curves would overlap cult to discriminate correctly. Even when we are
when plotted on the same set of axes, and an ob¬ working with stimuli well above the difference
server would be faced with a situation very similar threshold, we may feel that some discriminations
to that faced by the observer in the absolute detec¬ are easier to make than others. Red is more easily
tion situation. Look back at Figure 2-4 and men¬ differentiated from green than from orange. When
38 Chapter 2/Psychophysics

we are working with sets of stimuli that exceed the

difference threshold, the frequency methods we
have used up to this point are too crude to measure
interstimulus differences in detectability or discrim-
inability. To provide a more sensitive measure we
must turn to one of the oldest techniques in sensory
psychology: reaction time. Reaction time is de¬
fined as the time between the onset of a stimulus
and the beginning of an overt response. It was first
introduced in 1850 by one of the early giants in
perception and physiology, Hermann von Helm¬
holtz, who used it as a crude measure of the speed
of neural conduction in a limb.
There are two varieties of reaction time. Sim¬ Stimulus intensity (dB)

ple reaction time involves pressing or releasing a FIGURE 2-13 Effect of stimulus intensity on
telegraph key (or making some other simple ste¬ simple reaction time. (Based on Chocolle, 1940)
reotyped response) immediately on detecting a
stimulus. Choice reaction time involves making
one of several responses depending on the stimulus
presented (for example, press the right-hand key action times (Posner, 1978). The classic discrimi¬
for a red stimulus and the left-hand for a green). nation experiment utilizing reaction time was done
Simple reaction times are generally used in detec¬ by Henmon (1906). In this experiment the observer
tion paradigms. We have known for a long time had two response keys—one for each hand. In one
that the more intense a stimulus, the faster the re¬ instance, the observer was presented with pairs of
action time. Figure 2-13 shows typical median re¬ lines differing only in length and told to depress the
action times to the onset of a tone plotted against key corresponding to the side on which the line was
the stimulus intensity (Chocolle, 1940). When the longer. Henmon found that the greater the differ¬
stimulus intensity is low and near the detection ence between the line lengths the shorter the reac¬
threshold (although it is still quite detectable), the tion time. He reported similar results for colors and
reaction times are longer. Thus, when the stimulus tones.
is more difficult to apprehend, reaction time is A striking example of the relationship between
longer. Similar results have been obtained for vi¬ choice reaction time and the discriminability of
sual stimuli (Cattell, 1886; Grice, Nullmeyer & stimuli utilized a slightly different technique. In an
Schnizlein, 1979). Simple reaction time has also experiment by Shallice and Vickers (1964), observ¬
been used to measure discrimination. Here, how¬ ers were required to sort decks of cards into piles
ever, observers had to detect a change in stimulus according to which of two lines on the cards ap¬
intensity. Here we find that the larger the change in peared longer. The time it took to sort the cards
the stimulus intensity (either an increase or a de¬ was the measure of reaction time. This measure is,
crease), the shorter the reaction time (Welford, of course, the sum of a number of reaction times,
1980). where we consider the sorting of each card as a
Choice reaction time has been used in studies single response. The standard stimulus in this ex¬
of discrimination and identification. These reaction periment was a 4.5-cm line. In the data shown in
times tend to be somewhat longer than simple re- Table 2-9, we see that the more difficult the dis-
 Discrimination 39

Table 2-9. Differences in Reaction Time as a Table 2-10. Reaction Time as a Function of
Function of Differences in Line Lengths Measured Number of Stimulus Alternatives (Based on
via Card Sorts (Rased on Shallice &. Vickers, 1964) Merkel, 1885)

Difference in length Sorting time Reaction time

(cm) (sec) Number of alternatives (msec)

1.1 39.4 1 187

0.9 40.0 2 316
0.7 40.1 3 364
0.5 41.0 4 434
0.4 42.0 5 487
0.3 42.9 6 534
0.2 46.5 7 570
0.1 52.4 8 603
9 619
10 632

crimination, the longer the sorting time. A simple

demonstration of this effect is given in Demonstra¬ we increase the number of response alternatives,
tion Box 2-3. and indeed, this result has been known for many
Crossman (1953) has shown that these reac¬ years. Merkel (1885) showed with number stimuli
tion time differences are related both to the dis- that the reaction time increased as the number of
criminability of the stimuli and to the amount of response alternatives increased. The data in Table
information they contain. We are referring to infor¬ 2-10 show this clearly.
mation in the technical sense discussed in the sec¬ Hick (1952) attempted to explain these results
tion on identification. If this is the case, then we by postulating that the observer extracts informa¬
would expect choice reaction times to increase as tion from the stimulus display at a constant rate (cf.

DEMONSTRATION BOX 2-3. Reaction Time and Stimulus Discriminability

Take a deck of common playing cards and select out the decks and begin to sort it into two piles. The first
of it 10 of the picture cards (Kings, Queens, and deck gets sorted into number and picture cards; the
Jacks) and 10 numbered cards from the red suits second gets sorted into spades and clubs. Note the
(hearts and diamonds) to make a new deck of 20 time it takes to sort each deck. You may want to re¬
cards. Compose another deck of 20 by using the num¬ peat the task a couple of times so that you are sorting
bered cards (include the Aces) of the black suits smoothly. Notice that the sorting time for the spades
(clubs and spades). Shuffle each deck separately and and clubs (a more difficult task since it involves mak¬
place it in front of you, face down. Next you need a ing small form discriminations on similarly colored
clock or a watch with a sweep second hand. Wait cards) is longer than the easier discrimination task of
until the second hand reaches the 12, pick up one of sorting picture and number cards.
 40 Chapter 2/Psychophysics

DEMONSTRATION BOX 2-4. Number of Stimulus Alternatives and Reaction Time

Take a deck of playing cards and separate 16 cards second hand as you did in Demonstration Box 2-3.
using only the low numbers Ace, 2, 3, and 4. Next, Notice that the reaction time becomes longer (mea¬
make up another deck of 16 cards using 2 each of the sured by sorting time) as the number of alternative
5, 6, 7, 8, 9, 10, Jack, and Queen. Now shuffle each stimuli that must be recognized and responded to be¬
deck. Measure the time it takes to sort each deck into comes greater. Thus, sorting the 4-stimulus deck is
piles by number (4 piles for the first and 8 for the more rapid than sorting the 8-stimulus deck.
second deck) using a watch or clock with a sweep

Norwich, 1981), so the more information that must different types of representations may be estab¬
be obtained from the display, the longer the reac¬ lished, and each has its own characteristics (see
tion time. In an experimental situation where a dis¬ Luce & Narens, 1987; Narens & Luce, 1986; Ste¬
play of lights served as stimuli and finger pressings vens, 1946). The most primitive and unrestricted
of telegraph keys served as responses, he found a type of scale is a nominal scale. Its etymology
linear function relating reaction time and the loga¬ specifies its nature, since nomin is derived from the
rithm of the number of stimulus alternatives. This Latin word for “name.” When numbers are as¬
relation, called Hick’s law, states that choice re¬ signed in a nominal scale, they serve only as iden¬
action time is a linear function of the amount of tity codes or surrogate names. The numbers imply
information in the stimulus. You may demonstrate nothing more about the quantity of some property
effects of the number of stimulus alternatives on than do the numbers on football jerseys.
reaction time by consulting Demonstration Box 2-4. Whenever we are dealing with something for
which it is possible to say that an object or event
contains more or less of the property than some
SCALING other object or event, we can create an ordinal
scale of that property. An ordinal scale simply
The dog trainer glanced at her new St. Bernard pu¬ ranks items on the basis of some quantity. An ex¬
pil and estimated his shoulder height to be 75 cm ample might be the “Best-Seller” or “Top Fifty”
and his weight to be 80 kg. In so doing she was lists that order books or records on the basis of how
actually engaged in the perceptual act called scal¬ many have been sold. It is clear that although this
ing. Scaling attempts to answer the question, scale may prove to be more useful for measurement
“How much of X is there?” X can be a stimulus than a nominal scale, we are still very restricted in
magnitude, a sensation magnitude, or the magni¬ what we can do with the numbers.
tude of such other complex psychological variables The third type of scale is the interval scale. It
as similarity or even pleasantness. not only answers the questions implied by the la¬
To begin with, a scale is a rule by which we bels more or less but also tells by how much. It
assign numbers to objects or events. The scale at¬ employs not only the sequential properties of num¬
tempts to represent numerically some property of bers but also their spacing, or the intervals between
objects or events (see Michell, 1986). A variety of them. A good example of an interval scale is the
 Scaling 41

scale of temperature represented by the common in the apparent quality rather than the apparent
household thermometer. Here the size of the differ¬ quantity of a stimulus. When we have a stimulus
ence between 10 deg and 20 deg C (50 deg and 68 or experience in which the only question it makes
deg F) is exactly the same as between 40 deg and sense to ask is “What kind?” we are dealing with
50 deg C (104 deg and 122 deg F). Such scales are a metathetic continuum. Thus, a change in the
very useful, since most statistical techniques can be wavelength of a light may cause its appearance to
meaningfully applied to interval scale values. Inter¬ change from red to green. There seems to be no
val scales suffer from one major drawback, however. quantitative difference between these two hues,
They do not have a true zero point; rather, conven¬ they just appear to be different. Occasionally both
ience or convention usually dictates where the zero types of continua will be present in the same sense
will be. Thus, in the centigrade scale of temperature, impressions. For instance, in touch, the amount of
the zero point is the freezing point of pure water. pressure applied is a prothetic continuum, but the
The most numerically powerful scale in gen¬ location of the touch is a metathetic continuum.
eral use is the ratio scale. Creation of this type of Metathetic continua can be dealt with using nomi¬
scale is possible only when equality, rank order, nal scales, but scales that imply order have gener¬
equality of intervals and of ratios, and a true zero ally not been successfully applied to such sensory
point can be experimentally determined. Unfortu¬ qualities (but see Schneider & Bissett, 1981).
nately, ratio scales are more often found in the
physical than in the behavioral sciences. Such
Indirect Scaling: Fechner’s Law
things as mass, density, and length can be mea¬
sured on ratio scales since the zero points are not When the perceptual investigator wishes to estab¬
arbitrary. For example, 0 g represents the complete lish a sensory scale for which numbers will be as¬
absence of mass, and we can meaningfully say that signed to the intensity of sensations, there are two
10 g is twice as massive as 5 g. Negative values of alternative approaches. The first is a direct scaling
mass exist only in the fantasies of dieters. procedure in which individuals are asked to assess
All sensory qualities cannot be scaled in the directly some aspect of the strength of the sensa¬
same way. Some perceptual experiences have an tion. Although this might be the easiest procedure,
underlying aspect of intensity (for instance, bright¬ it is often difficult for the untrained observer. In
ness), whereas others do not (such as hue). When addition, many early psychologists distrusted the
we are dealing with a stimulus or an experience in accuracy of such direct reports because there
which it makes sense to ask “How much?” or seemed to be no easy way to convert them to nu¬
“How intense?” we have a prothetic continuum merical values. For this reason indirect scaling
(Stevens & Galanter, 1957). On prothetic continua methods, based on discrimination ability, formed
changes from one level of sensation to another the basis for the first psychological scales. It must
come about by adding or subtracting from what is not be thought that using an indirect procedure is
present. Thus, when we increase the weight of a necessarily bad. After all, we measure temperature
stimulus, the corresponding psychological sensa¬ indirectly, using the height of a column of mercury
tion of ‘heaviness” increases. Such prothetic con¬ as our indicator.
tinua can be meaningfully measured on scales of The first person to attempt to describe the re¬
any of the types we have discussed (with the pos¬ lationship between stimulus intensity and sensation
sible exception of a ratio scale—this is still contro¬ intensity was Gustav Theodor Fechner. To do this,
versial). In the other type of sensory continuum, he had to invent a way to measure the quantity of
changes in the physical stimulus result in a change the sensory experience. As his starting point, he
42 Chapter 2/Psychophysics

assumed that Weber’s law was correct; as we have

seen, it does hold over a wide range of stimuli. His
next assumption engendered a good deal of contro¬
versy and experimental testing. Fechner assumed
that the subjective impression of the difference be¬
tween two stimuli separated by one just noticeable
difference was the same regardless of the absolute
magnitude of the two stimuli. Thus, if we take two
dim lights that are separated by 1 jnd and we take
two lights that are 30 or 40 times brighter, but
again separated from each other by 1 jnd, we
should perceive the two pairs of stimuli as differing
by equal sensory steps. Finally, Fechner assumed FIGURE 2—14 Fechner’s law. It takes larger and
that sensation differences could be represented by larger differences between stimuli (Is) as stimulus
adding or subtracting jnds. intensity increases to give rise to the same size
If we accept Fechner’s postulate that Weber’s differences between sensations (Ss).
law is true and that the subjective sizes of all jnds
are the same, then only a small physical change is
necessary to achieve a 1 -jnd change for a weak
stimulus, whereas a large change is needed for a physical intensity of the stimulus (/). The constant
1 -jnd change when the physical stimulus is intense. W is different for different sensory continua, since
Perceptually, this means that the intensity of the it depends on the value of the Weber fraction for
sensation grows rapidly for weak physical stimuli the continuum scaled. Remember that this is an in¬
and more slowly as the physical stimulus is made direct scale since the strength of the sensation is
more intense. The relationship between the inten¬ never directly measured. To create this scale the
sity of the sensation and the intensity of the physi¬ only thing that must be measured is the size of
cal stimulus is shown in Figure 2-14. This curve the jnd. Once that has been determined, the rest
is described by the equation, is easy, and requires only counting jnds. Fech¬
ner’s technique was to use this aspect of the dis-
S = W log / criminability of stimuli (the jnd) as the unit of
measurement of the intensity of an observer’s
where S is the magnitude of sensation a stimulus sensory impression.
elicits, I is the physical magnitude of the stimulus
(units above the absolute threshold stimulus mag¬
Direct Scaling
nitude), and W is a constant that depends on the
value of the Weber fraction (A///). This equation is Since Fechner’s time, many psychophysicists have
called Fechner’s law. The actual mathematical insisted that indirect scaling is neither necessary
procedures by which Fechner derived this relation¬ nor preferable. Because we are interested in the ap¬
ship are discussed by Falmagne (1974, 1985) and parent intensity of a stimulus to an observer, why
Baird and Noma (1978), among others. We are us¬ not simply require judgments based on how intense
ing the number of jnds above the absolute threshold a stimulus seems to be? The observer’s responses
as a measure of the strength of the sensation gen¬ could then be used directly to establish a scale of
erated by a given stimulus (5), and we are saying measurement. The first attempt to do this was in
that the equation above relates this number to the 1872 by one of Fechner’s contemporaries, a Bel-
 Scaling 43

gian investigator named Plateau. To test Fechner’s uli. Also, of course, there is no such thing as a
law he had eight artists mix a gray that was half¬ correct or an incorrect answer, since the very na¬
way between a particular black and a white. Notice ture of the experiment implies that we cannot know
that this requires direct relative judgments of three in advance what a correct category assignment
stimuli, black, white, and gray. Fechner’s law pre¬ might be.
dicts that this psychological midpoint should cor¬ If our observer has spaced the category bound¬
respond to the average of the logarithm of the aries equally in terms of the magnitudes of the sen¬
physical intensity of the black stimulus and that of sory differences between them, we can, without
the white stimulus. Unfortunately, the results, al¬ making any other assumptions, mark off equal cat¬
though somewhat similar to the prediction, did not egory intervals (to represent the midpoints of the
fully support Fechner’s law. Rather, the grays categories) along the ordinate of a graph and label
mixed by Plateau’s artists seemed to fall halfway them with the category names. On this graph we
between the cube roots (1/3 power) of the intensi¬ can plot the average category label assigned to each
ties of the black and the white stimuli. This nu¬ stimulus intensity over several trials. The curve ob¬
merical discrepancy suggests that Fechner’s law tained for typical data (Figure 2-15) is concave
may only be an approximation to the relationship downward and closely approximates the curve pre¬
between physical and sensory intensity. We care dicted by Fechner’s law (Figure 2-14). The fact
about such mathematical deviations because a ma¬ that we can predict the shape of the category scale
jor purpose of scaling is to make possible a pre¬ from simple discrimination data is quite an impres¬
cise description of the relationship between the sive feat. To Sanford it seemed to provide support
strength of the physical stimulus and the strength for the contention that a logarithmic relationship
of sensations. exists between physical stimulus intensity and per¬
ceived magnitude.

Category Judgment

Sanford was among the early investigators who at¬

tempted to measure sensation directly. As early as
1898, he had worked out a technique that involved
having observers judge a number of envelopes,
each of which contained different weights. The
subjects were instructed to sort the weights into five
categories. Category 1 was to be used for the light¬
est weights and Category 5 for the heaviest, with
the remaining weights distributed in the other cat¬
egories in such a way that the intervals between the
category boundaries would be subjectively equal.
Thus, the difference in sensation between the upper
and the lower boundaries of Category 1 should be
the same as that for Category 2. In other words, all 0 1C 20 30 40 50 60 70

categories should be the same size. This method Stimulus intensity

has been called category scaling or equal-interval

FIGURE 2-15 Some typical data from a category
scaling. There is a similarity between this method judgment experiment. The dots represent average
and an identification task, except that in category category judgments of the various stimulus
scaling we usually have fewer categories than stim¬ intensities.
44 Chapter 2/Psychophysics

Since Sanford’s work, category scaling has judged one at a time and the only restriction on
been studied in great detail. Category scales have responses is that only numbers larger than zero can
been found to be relatively stable over several dif¬ be used.
ferent manipulations, including the labels applied In a typical magnitude estimation experiment,
to the categories (numbers versus words) and the in which we wish to scale the apparent length of
number of categories used (Stevens & Galanter, lines, we would start by showing a standard stim¬
1957). Other variations in procedure, however, ulus. We might then say: “This stimulus has a
dramatically affect the form of the scale. For ex¬ value of 10. You will be presented with several
ample, how the stimuli presented are spaced along stimuli that differ in length. Your task is to assign
the physical intensity dimension (Carter & Hirsch, numbers to these other stimuli in relation to the one
1955) and how often the various stimuli are pre¬ with a value of 10. Thus, if you see a line that
sented (Parducci, 1965) both affect the relation of appears to be twice as long as that one, you should
the category scale to the stimulus values on physi¬ assign to it the number 20. If you see a line one-
cal scales. Also, the range of stimulus intensities fifth as long, you should assign to it the number 2.
presented is important (Parducci, 1965), and the You may use any numbers you choose as long as
category judgments themselves are affected by the they are larger than zero.” In this task the number
memories of previous stimuli and responses (Ward, assigned to the standard stimulus is called the mod¬
1972, 1987). Several theories have been presented ulus. It serves to keep the numerical estimates of
that describe in detail the various factors that affect different observers within the same general range
where subjects locate their category boundaries (for of values. As you can see, this is a very direct way
example, Kombrot, 1984; Treisman & Williams, to attempt to measure sensation. The very nature of
1984) and how this affects the resulting scales. the task (where a stimulus judged to be m times
These biases indicate that we should use caution in larger than the standard is given a number m times
interpreting scales of measurement formed from as large) implies that the resultant scale might be a
category judgments, although some authors argue ratio scale. There is, however, continuing debate
that they are nonetheless the best type of scales to on this point.
use (e.g., Anderson, 1970). Stevens fully expected the results of such ex¬
periments to confirm Fechner’s law. When he plot¬
ted the data from an experiment in the magnitude
Magnitude Estimation: Stevens’s Larv
estimation of loudness (Stevens, 1956), however,
Although in category judgments observers are di¬ he found that the graph differed from what Fech¬
rectly responding to variations in stimulus magni¬ ner’s law had led him to expect. The equation he
tude, there is still some “indirectness” involved. found that best described the relationship of the
Stimuli that are similar but still discriminably dif¬ median magnitude estimates to the stimulus inten¬
ferent from one another may be grouped into the sities was
same category. Also, responses are limited to a few
category labels. S. S. Stevens popularized a pro¬ L = al° 6
cedure called magnitude estimation that avoids
these problems. The method is so simple and direct where L is the subjective loudness obtained through
that one wonders why it had to be “invented” at the observer’s magnitude estimates, a is a constant,
all. In this procedure, observers are simply asked I is the physical intensity of the sound, and 0.6 is
to assign numbers to stimuli on the basis of how a power to which I is raised. In succeeding years,
intense they appear to be. Stimuli are usually Stevens and a host of others produced magnitude
 Scaling 45

Table 2-11. Representative Exponents of the Power

Functions Relating Sensation Magnitude to Stimulus
Magnitude (Based on Stevens, 1961)

Continuum Exponent Stimulus conditions

Loudness 0.6 Both ears

Brightness 0.33 5° target—dark
Brightness 0.5 Point source—dark
Lightness 1.2 Gray papers
Smell 0.55 Coffee odor
Taste 0.8 Saccharine
Taste 1.3 Sucrose
Taste 1.3 Salt
Temperature 1.0 Cold—on arm
Temperature 1.6 Warmth—on arm
Vibration 0.95 60 Hz—on finger
Duration 1.1 White noise stimulus
Finger span 1.3 Thickness of wood blocks
Pressure on palm 1.1 Static force on skin
Heaviness 1.45 Lifted weights
Force of handgrip 1.7 Precision hand dynamometer
Electric shock 3.5 60 Hz—through fingers

estimation scales for a multitude of sensory con- ably standard, and the same measures of physical
tinua. All these scales seemed to be related to the stimulus intensity are used (Myers, 1982), the av¬
physical stimulus intensities by the general rela¬ erage exponents produced by different groups of
tionship observers for the same continuum are quite similar.
Some of them are small fractions (0.3 for bright¬
5 = aln ness), some are close to 1 (for line length), and
others are quite large (3.5 for electric shock). Some
where S is the sensory intensity and n is a charac¬ typical exponents are given in Table 2—11.
teristic exponent that differs for different sensory If we plot some of the relationships between
continua. Since this relation states that the magni¬ judged sensory intensity and physical stimulus in¬
tude of the sensation is simply the intensity of the tensity, we find that the curves for power functions
physical stimulus raised to some power, this rela¬ with different exponents (n) have dramatically dif¬
tionship is often called the power law or, after its ferent shapes. This can be seen in Figure 2-16.
popularizer, Stevens’s law. With exponents of less than 1 (for example, bright¬
In the power law the magnitude of the sensa¬ ness), the curves are concave downward, meaning
tion change, given a change in stimulus intensity, that as the stimulus becomes more intense, greater
depends on the size of the exponent. In general, the stimulus changes are needed to produce the same
exponent for any one continuum is quite stable. As degree of sensory change. When exponents are
long as the experimental situation is kept reason¬ greater than 1 (for example, shock), the curves are
46 Chapter 2/Psychophysics

FIGURE 2-16 Power functions for brightness, length, and electric shock. Notice how the shape of the
curve changes as the exponent changes.

concave upward, meaning that as stimuli become Thomas (1983). Demonstration Box 2-5 (p. 48)
more intense, the same physical stimulus change allows you to perform a magnitude estimation ex¬
produces an even larger sensory change than at periment for yourself.
lower stimulus intensities. Since each sensory con¬ We mentioned earlier that category judgments
tinuum might give a different curve describing the are subject to several sources of bias, and thus cat¬
relationship between sensory and physical inten¬ egory scales should be constructed and used with
sity, it is fortunate that a simple procedure exists caution. Unfortunately, although magnitude esti¬
that allows us to estimate the power function from mations have proved to be quite useful, and aver¬
any set of data. If we plot the logarithm of the av¬ age magnitude estimations behave quite lawfully,
erage magnitude estimates (the average numbers they nonetheless are also subject to a variety of
observers assign to their sensations) against the biases. The particular stimulus used as the stan¬
logarithms of the stimulus intensities, any curve of dard, the modulus used, the range of stimuli pre¬
the general form S = aln will appear as a straight sented, the clarity of the stimuli, how people use
line. In Figure 2-17 the curves in Figure 2-16 numbers, and previous stimuli and responses can
have been replotted in this way. We can now esti¬ all affect observers’ magnitude estimations (see
mate n from the curve by measuring the distances e.g., Baird, Lewis & Romer, 1970; Poulton, 1979;
marked Ay and Ajc in the figures and computing Ward, 1973, 1979). Relatively bias-free scales may
Ay/Ax. The constant a is the point at which the be produced by carefully choosing stimuli and pro¬
line crosses the ordinate. More sophisticated cedures (Poulton, Edwards & Fowler, 1980), but
methods of estimating the parameters in Stevens’s as we have noted before, it is impossible to elimi¬
law and those in Fechner’s law are described by nate completely the observers’ judgmental inclina-
 Scaling 47

FIGURE 2-17 The same power functions as in Figure 2-16 plotted on logarithmic axes. In such “log-
log” plots, all power functions become straight lines, with the slope of the straight line determined by the
exponent (n) of the power function.

tions, habits, or strategies from any such perceptual ample, if my 10-kg dog and my 100-kg brother
task. Because of this difficulty, some (e.g., Schnei¬ both gain 1 kilogram in weight, we may ask,
der & Parker, 1987) have argued that we should “Have they both gained the same amount?” If we
consider abandoning numerical estimation tech¬ are making an equal interval judgment (analogous
niques. However, few are willing to give up the to that required for category scaling) the answer is
simplicity and direct appeal of these techniques, so yes, since both have increased by 1 kilogram. If
the debate continues. we are making an equal ratio judgment (magnitude
You might wonder why category judgments estimate), my dog has increased his body weight
seem to give a logarithmic relationship that sup¬ by 10 percent and my brother by only 1 percent.
ports Fechner’s law, whereas magnitude estimates Thus, the weight gain is far from the same. Both
are related to stimulus magnitude by a power law. judgments require estimates of the magnitude of a
Actually, Stevens and Galanter (1957) found that single event, and both are useful, but the scales
category judgments only approximately fit a loga¬ (and resultant stimulus-sensation curves) are differ¬
rithmic relationship. Since then several investiga¬ ent (see also Marks, 1979b; Popper, Parker & Gal¬
tors (Gibson & Tomko, 1972; Marks, 1968, 1974; anter, 1986).
Ward, 1971, 1972, 1974) have shown that category
judgments also fit the power law, but with expo¬
Cross-Modality Matching
nents in) that are about half the size of those pro¬
duced by magnitude estimation. Marks (1974) and If the size of the exponent varies with the nature of
Torgerson (1961) have suggested that these differ¬ the response, you might wonder whether these
ent results reflect different but equally valid ways scales tell us more about how humans use numbers
of judging the same sensory experience. For ex¬ than they do about how sensation varies with
DEMONSTRATION BOX 2-5. Magnitude Estimation of Loudness

To produce a graded set of sound intensities for this Do this for two or three runs through the stimuli, and
demonstration you will need a long ruler, a coin then average your magnitude estimates for each
(we’ve designed the demonstration for a quarter), an height.
empty tin can or water glass, a soft towel, and a To determine if these judgments follow a power
friend. Place the can on the folded towel and have law, plot them on the log-log coordinates provided in
your friend drop the coin from the designated height the accompanying graph. The vertical axis is the log¬
so that the coin hits the can on its edge only once and arithm of the magnitude estimate and the horizontal
then falls onto the towel (silently, we hope). You axis is the logarithm of the sound intensity, based on
should sit with your back to the apparatus. the height of the coin drop. Draw the straight line that
At the start, your friend should drop the coin from best fits the data points. Usually, the data points fall
a 70-cm height. Try to remember how loud that close to such a line and do not curve significantly.
sounds, and assign it a value of 10. If you feel that a You can compute directly the exponent (n in the
test sound is twice as loud as the first sound, call it power law S = aln) by computing the slope of your
20, if it’s half as loud call it 5, and so on. You may straight line. Simply pick two points on the line and
use any numbers you feel are appropriate as long as measure Ax and Ay for these points with a ruler, as
they are greater than 0. Your friend should then drop pictured in Figure 2-17. Now divide Ay by Aa and
the coin from heights of 1, 10, 70, 100, and 200 cm, you should get a value somewhere around 0.3. This
in some mixed order, while you call out the number exponent means that the sensation of loudness in¬
corresponding to its apparent loudness and your friend creases less rapidly than does the actual sound inten¬
records your judgment for each stimulus (height). sity.

300

200

100
80
60

40
03
15 30
E
<n
<D
20
03
T>
3
C
10
03 9
<0 8
7
6
5
4
3

1
1 10 70 100 200
Centimeters of height (sound energy)
 Scaling 49

stimulus intensity (see Baird, 1975; Baird, Lewis modality matching experiments on log-log axes (as
& Romer, 1970). In order to counter such criti¬ we did for magnitude estimation experiments), we
cism, Stevens invented a scaling procedure that find that the average matches fall onto a straight
does not use numbers at all. In this technique, an line. Despite the fact that we no longer use numer¬
observer adjusts the intensity of a stimulus until it ical estimates from the observers, the data still
appears to be as intense as another stimulus from a obey the power law for sensory intensities. Figure
different sensory continuum. Thus, you might be 2-18 shows this for a number of modalities
asked to squeeze a handgrip until the pressure felt matched with handgrip pressure.
as strong as a particular light was bright. This pro¬ Cross-modality matching is often more diffi¬
cedure is called cross-modality matching, since cult to use than direct magnitude estimation be¬
the observer is asked to match sensory magnitudes cause the subject must adjust one of the sensory
across sensory modalities. Actually, magnitude es¬ continua in order to give a response, rather than
timation can also be viewed as a form of cross¬ simply reporting a number or category label. Fur¬
matching in which the number continuum is thermore, in spite of Stevens’s hopes, cross-modal¬
matched to a stimulus continuum (Oyama, 1968; ity matches too are affected by a variety of biases
Stevens, 1975). When we plot the data from cross¬ (see e.g., Baird, Green & Luce, 1980; Ward,

Relative intensity of criterion stimulus

FIGURE 2—18 Cross-modality matching data for nine different stimulus continua with force of handgrip
as the response continuum. Because the values on both axes are logarithmically spaced, all the straight
lines indicate power function relationships between stimulus and response magnitude. The dashed white
line has an exponent of 1.0. (From s. S. Stevens, 1961, in W. A. Rosenblith, ed., Sensory Communication, Cambridge, MA: MIT
Press. Copyright 1961 by the MIT Press.)
50 Chapter 2/Psychophysics

1975, 1979), and depend in important ways on the (and all the dissenting views), the power law still
context within which they are made (Marks, seems to be a reasonable first approximation to the
Szczesiul & Ohlott, 1986; Ward, 1987). description of how sensory intensity is related to
Because psychophysicists want to obtain as stimulus intensity.
accurate an estimate of the exponent as is possible
in order to describe the relationship between the
Adaptation Level Theory
physical stimulus and its perceived magnitude with
maximum precision, new refinements of the clas¬ Part of the circus strongman’s job was to carry var¬
sical scaling procedures are always being intro¬ ious members of the animal cast onto the circus
duced. For instance, a recent modification of the train. One visitor watched in amazement as one af¬
cross-modality matching technique makes the task ter another he lifted the dancing ponies and placed
somewhat easier for the observer, and seems to them in their railroad car. “Aren’t they heavy?”
give somewhat more reliable results. This proce¬ asked the visitor. “Not if you’ve just carried three
dure is called mixed-modality scaling (Ward elephants,” came the reply. The essence of this
1982a). In mixed-modality scaling, observers apocryphal tale is that no stimulus can be appreci¬
don’t actually match sensory magnitudes. Instead, ated in isolation. Stimuli are always seen in the
they judge two different sets of sensory stimuli (for context of the stimuli that precede and surround
instance, lights and sounds), both of which are in¬ them. Thus, sportscasters of average height look
termixed in the same experiment. Observers try to like midgets when interviewing professional bas¬
use the same scale as the stimuli alternate between ketball players, but like giants when interviewing
the modalities. The experimenter later uses some professional jockeys. They have, of course, not
sophisticated mathematical techniques to determine changed size, but their apparent size has changed
the relationship between the two sets of judgments, as a result of the frame of reference provided by
and also to estimate the exponents of the power the heights of those around them. Contextual ef¬
functions for the two sensory modalities. The first fects have long been known to influence judgments
example of mixed-modality scaling, called magni¬ of sensory magnitude in many psychophysical
tude matching because observers made magnitude tasks, even when the context is in another modality
estimates of stimuli from two sensory continua, than the one being judged (see e.g., Marks,
was introduced by J. C. Stevens and Marks (1980). Szczesiul & Ohlott, 1986; Ward, 1987). You can
In other variants, observers make category judg¬ experience this kind of context effect by using
ments (Ward, 1982a) or they make cross-modality Demonstration Box 2-6.
matches of each of the two sensory continua to a Helson (1964) attempted to explain how the
third, more convenient, continuum (such as dura¬ magnitude of one stimulus can affect our judgments
tion—see Ward, 1986). of the magnitude of other stimuli. His theory has
The theoretical importance of such techniques both quantitative and qualitative aspects. In Hel¬
rests in the exponents derived from them for the son’s theory, the organism is thought to accom¬
power law, in this case particularly the cross-mod¬ modate itself to the changing environment around
ality matching functions. Remarkably, the expo¬ it. This accommodation involves establishing a ref¬
nents usually agree regardless of the technique used erence level against which all other stimuli are
to estimate them (Ward, 1986). In addition, they judged. Stimuli below this reference, or adapta¬
generally agree with the values obtained from tra¬ tion level, are judged in one way (to be weak) and
ditional magnitude estimation techniques (Stevens, stimuli above it in another way (to be intense).
1975; Teghtsoonian, 1975). Given all the evidence Stimuli at or near the adaptation level are judged to
 Scaling 51

DEMONSTRATION BOX 2-6. The Effect of Visual Context on Judged Weight

You will need two envelopes for this demonstration. small envelope was considerably heavier although the
One should be rather small (about 7 by 13 cm or so) weights were physically equal. This is an example of
and one should be large (approximately 20 by 28 cm). how a visual context (the envelope size) can alter our
Put 15 nickels in each envelope. With the same hand, perception of heaviness. The same weight in the con¬
lift the large envelope and next lift the small. Which text of a smaller container seems heavier than when
appears to be heavier? You will probably feel that the judged in the context of a larger container.

be medium or neutral. This implies that all judg¬ quantitatively as a weighted product of all three
ments are relative. A stimulus is not simply weak classes of stimuli: focal, which we will designate
or intense, it is weak or intense compared to the F, background, B, and residual, R. The formula for
subjective adaptation level. the adaptation level (AL) is
For Helson, adaptation levels are established
by pooling the effects of three classes of stimuli. AL = Fm Bw2 Rm
The first class is called focal stimuli. These stimuli
are the center of an observer’s attention and are where Wl, W2, and W3 are weighting coefficients
usually the ones being judged. Clearly, the magni¬ that reflect the importance of any one class of stim¬
tude of these stimuli will in some way determine uli in the determination of the overall adaptation
the observer’s judgments, which is the basic as¬ level. Helson (1959) generally wrote this formula
sumption of all scaling procedures. The second as
class of stimuli is called background stimuli.
These are other stimuli that occur closely in space log AL = W\ log F + W2 log B + W3 log R
and/or time to the focal stimulus, providing the im¬
mediate background against which a focal stimulus which shows clearly that we are dealing with a type
is judged. The final set of stimuli is called residual of weighted average.
stimuli. These are stimuli that are not current for This formulation is not arbitrary. It is based
the observer, but are the residue of stimuli the ob¬ on category judgment data similar to those used to
server has experienced in the past. To be more con¬ test Fechner’s law and it provides a surprisingly
crete, consider the example in which we judged the good approximation for a large class of judgmental
height of a sportscaster surrounded by basketball data. Although there have been some interesting
players or jockeys. The physical height of a sports- extensions of adaptation level theory (e.g., Restle,
caster is the focal stimulus. The background stimuli 1971, 1978), there are also a number of other
are the heights of the surrounding athletes. The re¬ quantitative formulations that make slightly differ¬
sidual stimuli are the heights of all persons previ¬ ent assumptions about how stimulus magnitudes af¬
ously encountered, including those of athletes seen fect each other. These formulations result in
in the past. All these stimuli combine to form the different mathematical expressions (e.g., Ander¬
adaptation level. son, 1970, 1975). All these alternatives, however,
Helson (1964) defined the adaptation level still acknowledge the fact that surrounding stimuli,
52 Chapter 2/Psychophysics

stimuli experienced in the past, and patterns of at¬

tention, as well as the actual stimulus judged, can
affect our judgments of stimulus intensity.
Such effects in our judgment of stimulus mag¬
nitude may be observed in many different tasks.
For instance, consider a simple experiment by En-
gen and Tulunary (1956) in which subjects were
required to find a weight that appeared to be ex¬
actly half the weight of a standard. They found that
the mean weight judged to be one-half as heavy as
the standard was consistently lighter for an ascend¬
ing series of comparison weights (starting with the
light weights and moving gradually to heavier) than
for a descending series. The weight judged half as
heavy as a 300-g weight was 179 g for a descend¬
ing series (with the heavier weights of the series
forming the context) but only 140 g for an ascend¬
ing series (with judgments made in the context of
lighter weights). FIGURE 2-19 The circle surrounded by smaller
Similar shifts caused by context can also pro¬ elements appears larger than the circle
surrounded by larger elements, although both are
duce visual illusions (Coren & Girgus, 1978). For
the same size. (Based on Coren &. Girgus, 1978)
instance, consider Figure 2-19. The two black cir¬
cles are physically the same size, although they ap¬
pear to be different. It is easy to understand this
illusory effect if you recognize that when you are There may be no simple relationship between
looking at circle A, the adaptation level is shifted stimulus magnitude and sensory magnitude. What
toward the smaller size of the surrounding ele¬ we perceive is not simply a photographic reproduc¬
ments, whereas when you are looking at circle B, tion of the stimuli in the environment but is af¬
the adaptation level is shifted toward the larger size fected by all the myriad forces that have impinged
of the elements in that region. Because of this, cir¬ on us in the past and that provide the context for
cle A is above the adaptation level established for the perceptual situation in which we presently
its immediate vicinity, and hence appears larger, find ourselves. We will find that these conditions
and circle B is below the adaptation level in its hold for all the sensory modalities we shall study
region, and hence appears smaller. in the following chapters. Perception is an active
Adaptation level influences not only judg¬ process, and processes within the observer can
ments of magnitude but also discrimination be¬ sometimes be more important in determining the
tween stimuli. It is significantly more difficult to sensory experience than are factors in the external
discriminate stimuli that are both on the same side environment.
of the adaptation level than to discriminate stimuli, The psychophysical measurement techniques
equally close together, that appear on different introduced in this chapter will appear in many dis¬
sides of the adaptation level (Streitfeld & Wilson, guises throughout the rest of the book. We will not
1986). The adaptation level forms a reference point usually stop and identify these techniques, or the
both for sensory magnitude and for discrimination. rationale for using them. The methods by which
 Glossary 53

perceptual data are collected, however, are impor¬ Comparison stimuli A graded set of stimuli differ¬
tant because the measurement technique will fre¬ ing along a specific dimension that are to be judged rel¬
quently interact with the phenomena to be ative to a standard stimulus.

measured. We find, for example, that when we Correct negative Signal detection theory term for a
signal-absent trial to which the observer’s response is
look at raw identification scores in one experiment,
“No.”
red and green are equally discriminable from or¬
Criterion In signal detection theory, a sensation
ange because the individual hues are correctly iden¬ level that differentiates “yes” from “no” responses.
tified 100 percent of the time. In contrast, when we See p.
measure reaction times, it takes longer to discrim¬ Cross-modality matching A scaling procedure in
inate red from orange than green from orange, which the observer adjusts the intensity of a stimulus un¬
til it appears to be as intense as another stimulus from a
indicating a difference in discriminability. Is this
different sensory continuum.
a contradiction of fact? Not necessarily. Every
d' In signal detection theory, the distance between
method of measurement is tuned to a different task, the means of the signal-absent and the signal-present dis¬
and often measures a different psychological func¬ tributions.
tion. In everyday language you can say that the an¬ Detection Psychophysical problem involving being
swer you get depends on the question you ask. aware that a stimulus is present.
Difference threshold The minimum amount of stim¬
ulus change needed for two stimuli to be perceived as
GLOSSARY different.
Direct scaling A procedure in which individuals are
The following definitions are specific to this book. asked to assess directly the intensity of a sensation.
Absolute threshold The minimal amount of energy Discrimination Psychophysical problem involving
required to detect a stimulus. noticing a difference between stimuli.
Adaptation level A subjective reference point Equal-interval scaling See Category scaling.
against which stimuli are both quantitatively and quali¬ Error of anticipation In the method of limits, a
tatively judged. change in response before the percept actually changes.
Background stimuli Adaptation level theory term
Error of perseveration In the method of limits,
for stimuli that form a context for a focal stimulus but continuing to give the same response although the per¬
are not judged by an observer. cept has changed.
P (beta) In signal detection theory, the criterion for False alarm Signal detection theory term for a sig¬
sensation level that separates a “yes” response from a nal-absent trial to which the observer’s response is
“no” response. “Yes.”
Bit The amount of information in a stimulus mea¬
Fechner’s law The logarithmic relationship, pro¬
sured by the logarithm to the base 2 of the number of
posed by Fechner, between the intensity of sensation and
stimulus alternatives.
the intensity of physical stimulus, S = W log I.
Category scaling A psychophysical scaling method
Focal stimuli Adaptation level theory term for stim¬
in which stimuli are grouped in a predetermined number
uli at the center of an observer’s attention, usually those
of categories on the basis of their perceived intensity.
being judged.
Also called equal-interval scaling.
Catch trials Trials in which no stimulus is pre¬ Hick’s law A law stating that choice reaction time is
sented. Used in threshold measuring experiments. a linear function of the amount of information in the
stimuli to be differentiated.
Channel capacity The limit to the number of bits of
information an observer can transmit on a single sensory Hit Signal detection theory term for a signal-present
dimension. trial to which the response is “Yes.”

Choice reaction time Reaction time to make differ¬ Identification Psychophysical problem involving
ent responses to different stimuli. naming stimuli.
54 Chapter 2/Psychophysics

Indirect scaling Any method, often based on dis¬ Negative time error In discrimination experiments,
crimination ability, by which sensation intensity is mea¬ when the point of subjective equality is less than the
sured indirectly. value of the standard stimulus.
Information theory A quantitative system for mea¬ Nominal scale A scale in which the values can be
suring the difficulty of an identification task in terms of used only as names of objects or events, thus reflecting
the number of stimulus alternatives that must be distin¬ only identity.
guished. Ordinal scale A scale involving the ranking of items
Information transmission The degree to which the on the basis of more or less of some quantity.
output of an information channel (for example, an ob¬ Outcome matrix In signal detection theory, a matrix
server in an identification experiment) reflects the infor¬ containing the proportions of trials on which the four
mation input to it. possible outcomes occurred.
Interval of uncertainty In a discrimination experi¬ Payoff matrix In signal detection theory, a matrix
ment, the difference between the stimulus intensity describing the set of rewards and penalties given an ob¬
judged greater than the standard 25 percent of the time server based on his performance in a psychophysical ex¬
and that judged greater 75 percent of the time. periment.
Interval scale A scale in which differences between Point of subjective equality The comparison stimu¬
adjacent values are meaningful but that has no absolute lus intensity that appears most like the standard in a dis¬
zero point. crimination experiment.
Just noticeable difference (jnd) The stimulus dif¬ Power law The relation stating that the magnitude of
ference noticed 50 percent of the time, computed as the sensation varies as the intensity of the physical stimulus
interval of uncertainty divided by 2. raised to some power. Also known as Stevens’s law,
Magnitude estimation A psychophysical scaling S = aln.
procedure requiring the observer to assign numbers to Probability distribution A graphic representation of
stimuli on the basis of the intensity of the sensations they the likelihood that a given event will occur.
arouse. Prothetic continuum A psychological continuum
Magnitude matching A technique whereby observ¬ that involves quantitative aspects (how much) of stimu¬
ers make magnitude estimates of stimuli from two differ¬ lation, such as loudness or brightness.
ent sensory continua intermixed in the same experiment. Ratio scale A measurement scale in which the rank
Metathetic continuum A stimulus continuum in¬ order, spacing, and ratios of the numbers assigned to
volving the quality of sensations, such as color or pitch. events have meaning; it also has an absolute zero point.
Method of constant stimuli A method for determin¬ Reaction time The interval between the onset of a
ing thresholds in which each of a number of stimuli stimulus and the beginning of an overt response.
above and below the suspected threshold is presented and Receiver operating characteristic (ROC) curve In
judged repeatedly. signal detection theory, the graph of probabilities of hits
Method of limits A method for determining thresh¬ (ordinate) versus false alarms (abscissa) generated by cri¬
olds in which stimulus intensity is systematically in¬ terion changes.
creased or decreased until a change in response occurs. Residual stimuli Adaptation level theory term for
Miss Signal detection theory term for a signal-pres¬ stimuli that are no longer present but affect the current
ent trial to which the observer’s response is “No.” adaptation level.
Mixed-modality scaling A psychophysical scaling ROC curve See Receiver operating characteristic
procedure in which observers make judgments (including curve.
magnitude estimations, category judgments, or cross-mo¬ Scaling Psychophysical problem involving the mea¬
dality matches) of stimuli from two different sensory surement of how much of something is present.
continua on the same scale. Signal detection theory A mathematical, theoretical
Modulus In magnitude estimation, the standard nu¬ system that formally deals with both decisional and sen¬
merical value assigned to one of the stimuli at the begin¬ sory components in detection and discrimination tasks.
ning of the judgment procedure; it determines the range Simple reaction time Reaction time for simply de¬
of numbers to be used by the observer in the procedure. tecting the onset of a stimulus.
 Glossary 55

Staircase method A method for measuring absolute Weber fraction The proportion by which the stan¬
thresholds in which the experimenter alters the direction dard stimulus must be increased in order to detect
of changes in stimulus intensity each time the observer change, K = AIII.
changes his or her response. Weber’s law The relation stating that the size of the
Standard A stimulus against which the comparison just noticeable difference (jnd) increases linearly with the
stimuli are judged in a discrimination experiment. size of the standard, A/ = K I.
Stevens’s Law See Power law.
' CHAPTER
3
The Visual
System
•

LIGHT
THE STRUCTURE OF THE EYE
The Crystalline Lens
The Retina
The Fovea
Rods and Cones
NEURAL RESPONSES TO LIGHT
Retinal Responses and Receptive Fields
X, Y, and W Cells
THE VISUAL PATHWAYS
The Geniculostriate System
The Tectopulvinar System
THE VISUAL CORTEX
Receptive Fields in the Visual Cortex
Organization of the Primary Visual
Cortex
The Temporal Lobes
* Alternate Visual Maps

57
58 Chapter 3/The Visual System

A lthough perception occurs within the brain,

your only real contact with the external envi¬
Light often acts as if it were a stream of par¬
ticles, but at other times it acts as if it were made
ronment is through your sense organs. Recalling up of waves. James Clerk Maxwell (1873) showed
the old saying, “The eyes are the windows to the that light travels not only in a straight line but also
world,” it is clear that the physical properties of as an oscillating wave. He suggested that if we
this “window” will affect the nature of your per¬ consider the change in the electromagnetic field
ception in the same manner that the physical prop¬ surrounding the train of photons, we can treat light
erties of a glass window will affect your view. If as purely a wave phenomenon, with the wavelength
the window is colored, your perception of the defined as the physical distance between the peaks
world will be tinted. If the window is dark, or of the photon waves.
dirty, your ability to discern objects will be re¬ Electromagnetic energy can have wavelengths
duced. If the window is curved, so as to magnify over a broad range, varying from trillionths of a
the images, your perception of the size of objects centimeter to many kilometers in length. Very short
viewed through the glass may also be distorted. wavelengths are not visible, nor are very long
Thus, it is important for you to know the nature of wavelengths. As you can see in Figure 3-1, very
the “window” through which you look at the short wavelengths include gamma rays, X-rays,
world, or, more simply, to understand the physio¬ and ultraviolet rays. Longer wavelengths vary from
logical makeup of the visual system. those we call electricity through the broadcasting
wavelengths associated with TV and radio (which
may be more than 100 meters in length). The sec¬
LIGHT tion of the electromagnetic spectrum that we see as
visible light is really quite small, extending from
Each of the sensory systems is maximally respon¬ 380 to about 760 nanometers. A nanometer is a
sive to a different form of physical stimulation. billionth of a meter and is usually abbreviated nm.
Taste and smell respond to chemical stimuli, touch The older method of specifying wavelength was
to mechanical pressure, and hearing to the vibration millimicrons, which was abbreviated mp. Perceptu¬
of air molecules. The physical stimulus for sight is ally, variations in wavelengths correspond roughly
electromagnetic radiation. We call the particular to the hue or color of light. In normal eyes, wave¬
form of electromagnetic radiation that produces a lengths of about 400 nm are seen as violet, 500 nm
visual response light. are seen as blue-green, 600 nm are seen as yellow-
In 1704, Sir Isaac Newton advanced the the¬ orange, and 700 nm are seen as red. However, the
ory that any form of electromagnetic radiation, in¬ perception of color depends on much more than
cluding light, acts as if it were a stream of particles wavelength alone, as you will find out in Chapter 5.
traveling in a straight line. Each particle is called a
quantum, and a quantum of light is called a pho¬
ton. The intensity of light is then given by the THE STRUCTURE OF THE EYE
number of photons. Although this conception of
light is extremely useful in physics, it is only im¬ Most vertebrate eyes, from those of fish to those of
portant to the understanding of vision when we deal mammals, have a similar basic structure. A sche¬
with stimuli that are relatively dim. At low levels matic diagram of the human eye is shown in Figure
of light intensities are often described as the num¬ 3-2. The eyes may be found lying in protective bony
ber of photons reaching the visual receptors. The sockets within the skull and are spherical structures
smallest amount of light possible is one photon. about 20-25 mm in diameter. The outer covering,
 The Structure of the Eye 59

Short wavelength Long wavelength

Ultraviolet
Shortwave

Wavelength in nanometers (nm)

FIGURE 3-1 Electromagnetic radiation spectrum, with the region containing visible light enlarged.

which is seen as the “white” of the eye, is a strong When you look at a human eye your attention
elastic membrane called the sclera. Because the eye is usually captured by a ring of color. This colored
is not made of rigid materials, it maintains its shape membrane, surrounding a central hole, is called the
by means of fluid pressure from within. iris. When you say that a person has brown eyes,
The front of the eye contains a region where you really are saying that he has brown irises. The
the sclera bulges forward to form a clear, domelike actual color, which may vary from blue through
window, about 13 mm in diameter, called the cor¬ black, appears to be genetically determined. The
nea (Martin & Holden, 1982). The cornea is the function of the iris seems to be to control the
first optically active element in the eye. It serves as amount of light entering the eye. It may be of some
a simple fixed lens that gathers and focuses light. interest to note that although blue eyes seem to
Because the cornea is extended forward, it actually have been viewed as more appealing by some
allows reception of light from a region slightly be¬ poets, dark irises, such as brown or black, more
hind the observer, as is shown in Demonstration effectively shield the eye from light. The light en¬
Box 3-1. ters through the hole in the iris, which is called the
Behind the cornea is a small chamber filled pupil. The size of the pupil appears to be con¬
with a watery fluid called the aqueous humor. trolled by a light reflex. When the light is bright
This fluid is similar in nature to the cerebrospinal the pupil may contract to as little as 2 mm in di¬
fluid that bathes the inner cavities of the brain. This ameter, whereas in dim light it may dilate to more
is not surprising since embryological evidence has than 8 mm. That is about a sixteenfold change in
shown that the neural components of the eye ac¬ the area of the aperture. Demonstration Box 3-2
tually develop from the same structures that even¬ shows how you may observe the effect of light on
tually form the brain. pupil size.
60 Chapter 3/The Visual System

A B

FIGURE 3-2 Structure of the human eye, with a demonstration of the image formation of two targets (X
and Y).

The constriction of the pupil serves an impor¬ sensitivity obtained by increasing the amount of
tant function. Despite the fact that the eye needs light entering the eye; thus the pupil increases in
light to function there are some advantages to size to let in more light. The pupil size also
viewing the world with a small pupil. Although the changes as a function of emotional and attentional
amount of light entering the eye is reduced, imper¬ variables. Under conditions of high interest the eye
fections in the lens produce fewer distortions with tries to gather more light and the pupil tends to be
a small pupil, and the depth of focus (which is the large, a cue often used by smart traders as an in¬
range of distances over which objects are simulta¬ dex of a customer’s interest in an item. Clever cus¬
neously in focus) is vastly increased. We might say tomers often negate the usefulness of this cue in
that the eye takes advantage of better light by im¬ bargaining situations by wearing dark glasses.
proving its optical response. In dim light, the abil¬ Similarly, the dimness of candlelight dilates the pu¬
ity of the eye to resolve or discriminate details pils and makes lovers appear to be more attentive
(called acuity) is less important than the increased and interested.
 The Structure of the Eye 61

DEMONSTRATION BOX 3-1. Vision “Behind” the Eye

It is easy to demonstrate that the visual field actually your hand slowly forward until the wiggling finger is
extends to a region somewhat behind the eye. In order just barely visible in your peripheral vision. At this
to do this, simply choose a point that is some distance point stop and, with your head as still as possible,
in front of your head and stare at it. Now raise your move your finger directly in toward your head. You
hand to the side of your head as shown in the figure, will notice that your hand will touch a point on your
with your index finger extended upward. Your hand temple somewhat behind the location of the eye, in¬
should be out of view when you stare at the distant dicating that you were actually seeing somewhat “be¬
point. Now, wiggle your finger slightly, and bring hind yourself.”

Move your hand

inward until it
62 Chapter 3/The Visual System

DEMONSTRATION BOX 3-2. The Pupillary Light Reflex

For this demonstration you need a friend. Dim the how the pupil constricts. Removal of the light will
light in the room, but leave enough light so that you cause the pupil to dilate again. The light reflex of the
can still see the size of the pupil. Notice how large pupil was the first reflex ever studied by Whytt
your friend’s pupil appears to be under these condi¬ (1751), who is credited with the discovery of reflex
tions. Now turn on an overhead light or shine the action. It is still sometimes called Whytt’s reflex.
beam of a flashlight into your friend’s eye and note

The Crystalline Lens

zonal fibers connecting the lens to the inside wall
Most vertebrate eyes contain a lens, located di¬ of the eye cause it to flatten. Under these condi¬
rectly behind the pupillary aperture. Since the cur¬ tions, distant objects should be in focus. Contrac¬
vature of the lens determines the amount by which tion of the ciliary muscles, from which the lens
the light is bent, it is critical in bringing an image is suspended, takes some of the tension from the
into focus at the rear of the eye. The process by lens and it regresses to a more spherical shape.
which the lens varies its focus is called accommo¬ When it is rounder, near objects are in focus. The
dation. The lens changes focus by changing its effect of lens shape on point of focus is shown in
shape (Dalziel & Egan, 1982). The natural shape Figure 3-3.
of the human lens tends to be spherical, but when An individual’s age is important in determin¬
the ciliary muscles that control it relax, the pressure ing the focusing ability of the lens. It is interesting
of the fluid in the eyeball and the tension of the to note that the ability to accommodate is not pres-

Lens round

Lens flat

FIGURE 3-3 Accommodation (focusing) of an image by changing the shape of the crystalline lens of the
eye.
 The Structure of the Eye 63

ent at birth. For a newborn infant, until about the As we noted above, the major purpose of the
age of 1 month, only images of objects that are lens is to focus the image in the eye. An eye having
approximately 19 cm away are in focus (Dobson & normal accommodative (focusing) ability is called
Teller, 1978; Teller & Movshon, 1986). Images of emmetropic. Sometimes there is too much or too
targets closer or farther away than 19 cm are pro¬ little curvature in the cornea or, alternatively, the
portionally blurred. However, during the second shape of the eye is too short or too long, so that
month of infancy, the accommodative system be¬ the accommodative capacity of the lens is not suf¬
gins to respond more adaptively (see Chapter 16). ficient to bring targets into focus. If the eye is too
The ability of the lens to change focus de¬ short, or if the light rays are not bent sharply
creases with age after about age 16. This is because enough by the cornea, distant objects are seen quite
the inner layers of the lens die and lose some of clearly, but it is difficult to bring near objects into
their elasticity (Weale, 1986), and it thus becomes focus. The common term for this is farsightedness,
more difficult for the ciliary muscles to change the and the technical term is hypermetropia. If the eye
lens’s curvature to accommodate to a near object. is too long, or if the light rays are bent too sharply
This results in a form of refractive error (light¬ by the cornea, near objects are sharply in focus but
bending or focusing error) called presbyopia, distant objects are blurred. This condition is called
which simply translates to “old-sighted.” Func¬ nearsightedness or myopia. The optical situations
tionally, this condition increases the near point that result from these difficulties are shown in Fig¬
distance. The near point refers to how close an ob¬ ure 3-4.
ject may be brought to the eye before it can no
longer be held in focus and becomes blurry. Thus,
The Retina
older persons without corrective lenses often may
be seen holding reading material abnormally far The large chamber of the eye is filled with a jelly-
from their faces in order to bring the print into like substance called the vitreous humor. This
proper focus. substance is generally clear, although shreds of de¬
Another feature of the lens, which warrants bris can often be seen floating in it. Try steadily
mention, is the fact that it is not perfectly transpar¬ viewing a clear blue sky and note the shadows that
ent. The lens is tinted somewhat yellow, and the move across it; these shadows are from floating de¬
density of this yellow tint increases with age bris in the vitreous humor.
(Coren & Girgus, 1972a). The yellow pigment The image formed by the optical system of the
serves to screen out some of the ultraviolet light eye is focused on a screen of neural elements at the
entering the eye. Animals with clear lenses (such back of the eye called the retina. The term retina
as many birds and insects) can see ultraviolet light, derives from the Latin word meaning “net,” be¬
as can people who have had their lenses surgically cause when an eye is opened up surgically (or its
removed (e.g., Emmerton, 1983; Hardie & Kirsch- interior viewed with an optical device such as an
feld, 1983). The yellow pigment in the lens also ophthalmoscope) the most salient feature is the net¬
screens out some of the blue light, and thus alters work of blood vessels lining the inner cavity. Dem¬
your perception of color somewhat. For example, onstration Box 3-3 shows how you can observe
you may have heard individuals arguing over these blood vessels in your own eyes.
whether a particular color was blue or green. If The sheet of neural elements that makes up the
they are different ages, the source of the argument retina extends over most of the interior of the eye.
may lie in the fact that because the lens yellows In diurnal or daylight-active animals, the retina is
with age, each is viewing the world through a dif¬ backed by a light-absorbing dark layer called the
ferent yellow filter. pigment epithelium. This dark pigment layer
64 Chapter 3/The Visual System

enters and once as it is reflected out), effectively

doubling its intensity. Although the result is a siz¬
able increase in sensitivity, it is obtained at the ex¬
pense of a considerable degrading of the image
through fogging and blur, especially at higher illu¬
mination levels. The existence of this reflecting
surface explains why cats’ eyes seem to glow in the
dark when a flashlight is pointed toward them.
The retina consists of three major layers of
neural tissue and is about the thickness of a sheet
Actual eye shape of paper (see Figure 3-5). It is here that the light is
(too short) changed or transduced into a neural response. The
outermost layer of the retina, closest to the scleral
wall, contains the photoreceptors. There are two
« Rays focus types of photoreceptors that are distinguishable on
< behind eye
the basis of their shapes—long, thin, cylindrical
cells called rods, and shorter, thicker, somewhat
Normal eye shape more tapered cells called cones. The outer seg¬
ments of these cells contain pigments that absorb
Hypermetropic eye (farsighted) the light and start the visual response. The next
level of the retina consists of bipolar cells, which
are neurons with two, long, extended processes.
One end makes synapses with the photoreceptors;
the other end makes synapses with the large retinal
ganglion cells in the third layer of the retina.
In addition to photoreceptors, bipolars, and
ganglion cells, there are also two types of cells that
have lateral connections. Closest to the receptor
layer are the horizontal cells. These cells typically
have short dendrites and a long horizontal process
Myopic eye (nearsighted)
that extends some distance across the retina. The
FIGURE 3-4 Three common refractive second set of cells that are lateral interconnecters
conditions of the eye. are called amacrine cells. These large cells are
found between the ganglion and bipolar cells, and
seem to interact with spatially adjacent units. Ac¬
serves the same purpose as the black inner coating tually, over 30 types of amacrine cells, differing in
in a photographic camera, reducing the amount of size and chemical properties, have been isolated
reflected and scattered light that could blur or fog (Masland, 1986). Both the horizontal and amacrine
the image. In nocturnal or night-active animals, cells serve to modify the visual signal, and allow
where the detection of light is more important than adjacent cells in the retina to communicate and in¬
image clarity, the light that penetrates the retina is teract with one another (Kolb, Nelson & Mariani,
reflected back through the retina by a shiny surface 1981; Naka, 1982; Tomita, 1986).
known as the reflecting tapetum. This permits the Light reception occurs within the rod and cone
light to pass through the retina twice (once as it cells. Contrary to what we might expect, the ori-
 The Structure of the Eye 65

DEMONSTRATION BOX 3-3. Mapping the Retinal Blood Vessels

For this demonstration you will need a pocket pen- generated by the movements of the shadows of your
light and a white paper or light-colored wall. Hold the retinal blood vessels across your retina. By steadily
penlight near the outside canthus (comer) of your eye. shaking the bulb with one hand and tracing the shad¬
Now, shaking the bulb up and down you will see a ows with the other, you can produce a map of your
netlike pattern on the light surface. This pattern is own retinal blood vessels.

entation of rods and cones is inverted, with the pig¬ are many blood vessels in the epithelial layer at the
ment-bearing end pointing toward the rear of the rear of the eye. If the retina were “right-side up,”
eye rather than toward the lens. Thus, the retina so many blood vessels would be needed that the
may be viewed as if it were a transparent carpet light input would be partially blocked. Therefore,
lying upside down on the floor of the room, with the “upside-down” organization is more func¬
the pile of the carpet corresponding to the rods and tional.
cones. The incoming light must therefore pass
through the carpet (the retina) before reaching the
The Fovea
photoreceptors. Although this arrangement might
appear to be somewhat counterproductive, it ac¬ Not all areas of the retina are of equal importance
tually makes good sense. The photoreceptors need in the perceptual process. The most important sec¬
a rich oxygen supply, and to meet this need there tion of the human retina is located in the region
66 Chapter 3/The Visual System

Choroid
Pigment
epithelium

Rod

Pigment in layers

Horizontal'cell

Light

FIGURE 3-5 Schematic diagram of the human retina.

around the origin of the optic axis, an imaginary scope we note a yellow patch of pigment located in
line from the center of the fovea that passes the region of the origin of the optic axis. This area
through the center of the pupil (see Figure 3-7). If is called the macula lutea (or just macula), which
we view a human retina through an ophthalmo¬ simply translates to “yellow spot.” Demonstration
 The Structure of the Eye 67

DEMONSTRATION BOX 3-4. The Macular Spot

Under appropriate conditions it is possible to see the be improved by moving the cellophane in front of and
macular spot in your own eye. In order to do this you away from your eye, so that you have a flickering
will need a dark blue or purple piece of cellophane. colored field. Some individuals can see the spot when
Brightly illuminate a piece of white paper with a desk staring at a uniform blue field, such as a clear summer
lamp. Now, while looking at the paper with one eye, sky. This percept is caused by the fact that the yellow
quickly bring the piece of cellophane between your pigment in the macula absorbs the blue light and does
eye and the paper. Now as you look at the paper you not let it pass. This causes a circular shadow, which
see what appears to be a faint circular shadow in the can be briefly seen. It is often called Maxwell’s spot,
center of it. The sight of the shadow may only last after James Clerk Maxwell, who noticed its presence
for a couple of seconds. Sometimes its visibility can during some color-matching experiments.

Box 3-4 describes a procedure in which you can ter of the foveal depression, the upper layers of
see your own macula. In the center of the macula cells are apparently pushed away so that the light
is a small depression that looks much like the im¬ passes through a much thinner cellular layer before
print of a pinpoint about 1/3 mm in diameter. This reaching the photoreceptors. The photoreceptors
small circular depression is called the fovea cen¬ themselves are very densely packed in this region.
tralis, or translated, the “central pit.” The fovea This section of the retina contains only cones; there
is critical in visual perception. Whenever you are no rods at all. Foveal cones have a different
“look” directly at a target, it means that your eyes shape than the more peripheral cones depicted in
are rotated so that the image of the target falls on Figure 3-5. They are much longer and thinner (of¬
the foveal region. ten only 0.001 mm in diameter), so that they some¬
The fovea is quite unique in its structure, and what resemble rods.
is schematically depicted in Figure 3-6. In the cen¬ In a laborious study, Osterberg (1935) exam-

Light

Neural tissue

Receptors

FIGURE 3-6 Sketch of a cross section through the fovea. Light comes from the direction of the top of
the page.
68 Chapter 3/The Visual System

ined the retina of a human eye that had been re¬ rods and cones. On the basis of these observations
moved as the result of an accident. By fixing the Schultze offered what has been called the duplex
fresh retina in a suitable fluid it is possible to pre¬ retina theory of vision. He maintained that there
serve it indefinitely. He counted the number of rods are two separate visual systems. One is for vision
and cones in this human retina, demonstrating that under dim light conditions and is dependent on the
there are no rods in the center of the fovea at all. rods; the other is for vision under daylight or bright
Outside of the fovea the number of cones rapidly conditions and is dependent on the cones. Vision
decreases. The number of rods, however, rapidly under bright light is called photopic (which trans¬
increases beyond the foveal region, reaching a peak lates to “light vision”), and vision under dim light
concentration at about 20 degrees of visual angle is called scotopic (“dark vision”).
from the fovea and then decreasing again. This Early clinical studies (Kries, 1895) provided
general distribution (which is shown in Figure 3-7) some behavioral data that supports the idea that the
has been verified in recent computer mappings of eye contains two different visual systems. For in¬
the retina (Curcio, Sloan, Packer, Hendrickson & stance, individuals whose retinas contain no rods,
Kalina, 1987). or only nonfunctioning rods, seem to have normal
vision under daylight conditions. However, as soon
as the light dims beyond a certain point (into what
Rods and Cones
you might call a twilight level of intensity) they
The presence of two types of retinal photoreceptors lose all sense of sight and become functionally
suggests the existence of two types of visual func¬ blind. These individuals suffer from night blind¬
tion. In the early 1860s the retinal anatomist Max ness. The implication is that in the absence of rods
Schultze found that nocturnal animals, such as scotopic vision is lost. A quite different pattern is
owls, have retinas that contain only rods. Animals found for individuals lacking in functioning cones.
that are diurnal, or only active during the day, such These people find normal levels of daylight quite
as the chipmunk and pigeon, have retinas that are painful, totally lack color vision, and have very
all cones. Animals that are active in the twilight, poor visual acuity. Under dim levels of illumina¬
or during both day and night, such as rats, mon¬ tion, however, they function normally. Such indi¬
keys, and humans, have retinas comprised of both viduals suffer from day-blindness and provide

FIGURE 3-7 The distribution of rods and cones in the human retina. The left figure gives the locations
on the retina of the ‘angle’’ relative to the optic axis on the right figure. (Modified from Lindsav &
Norman, 1977) J
 The Structure of the Eye 69

evidence that a functioning cone system is neces¬ made up of two parts: retinal, a complex organic
sary for normal photopic vision and also for the molecule derived from vitamin A, and opsin, a
perception of color. The specifics of the perception protein that has the capacity to act as an enzyme.
of brightness and color are discussed in Chapters 4 As is the case for many organic compounds, the
and 5. retinal component can exist in several different
Before a rod or a cone can signal the presence shapes called isomers. When a molecule of rhodop¬
of light, it must first interact with the light in some sin absorbs a photon of light it isomerizes, or
way. Chemically, such interaction involves absorb¬ changes shape, then splits into its two component
ing, or capturing, one or more photons. Any sub¬ parts. A complex sequence of events, outlined in
stance that absorbs light is called a pigment. A Figure 3-8, then begins. This involves the activa¬
substance that absorbs a lot of light would appear tion of several enzymes, resulting in the breakdown
to be darkly pigmented, since most of the photons of the molecule that normally keeps the cell mem¬
hitting it would be absorbed and very few would brane open to allow the flow of sodium ions
be reflected back to the eye of the viewer. As we (Stryer, 1987). (In order to understand what hap¬
noted earlier, the outer segments of both the rods pens next you should know a little about how in¬
and the cones contain visual pigments. If you refer formation is transmitted to and by neurons and
back to Figure 3-5, you will see the pigments ar¬ receptors. If you are a bit unsure in this area you
ranged in layers in the outer segments of the pho¬ should stop and read the “Primer of Neurophysi¬
toreceptors. For rods, the photosensitive pigment is ology’’ we’ve included as an appendix at the back
arranged in a stack of around 2,000 tiny disks, like of the book.) Once the flow of positively charged
coins inside a tube; for cones, the pigment is part sodium ions into the rod stops, the rod cell hyper-
of a single large elaborately folded membrane that polarizes—that is to say that the normally negative
forms the layers of photosensitive material. charge of —40 millivolts across the cell membrane
Rods and cones do not contain the same pig¬ becomes even more negative, perhaps — 70 to — 80
ment. The first successful isolation of the pigment mv. This hyperpolarization indicates that the rod
in rods was made in 1876 when Franz Boll isolated has been stimulated by light (Hubbell & Bownds,
a brilliant red pigment from the frog retina (which 1979; Schnapf & Baylor, 1987).
contains predominantly rods). He noted that this To be ready for another response this process
pigment bleached, or lost its apparent coloration, must be reversed. The rhodopsin regenerates in the
when exposed to light. This reaction indicated that dark from the retinal and opsin with the help of
the substance was photosensitive. He further noted vitamin A and other enzymes. There is some evi¬
that the pigment regenerated itself in the dark. dence indicating that light actually provides the en¬
Thus, it fulfilled the elementary requirements of a ergy to resynthesize rhodopsin, although this
visual pigment in that it responded by changing process is somewhat too complex to discuss here
chemically in the presence of light, yet it still re¬ (Blazynski & Ostroy, 1981; Rodieck, 1973). No¬
mained capable of resynthesizing itself. Kuhne tice that vitamin A is vital to the resynthesis and in
took up the study of this pigment in 1877 and, in the absence of vitamin A rhodopsin cannot be
one extraordinary year, laid the groundwork for our formed. In isolated communities where fish prod¬
understanding of its action. This pigment has been ucts or appropriate vegetables are not available, the
named rhodopsin (which means “visual red’’ absence of vitamin A in the diet shows up in “ep¬
rather than “visual purple’’ as it is sometimes idemics’’ of night blindness, as the rods become
called). In the century since the work by Boll and nonfunctional (Wald, 1968).
Kuhne, we have been able to work out much of the The identification and analysis of cone pig¬
photochemical reaction in rhodopsin. ments has proved to be more difficult and elusive
Basically, rhodopsin is a compound that is than that of rhodopsin. However, we have learned
70 Chapter 3/The Visual System

FIGURE 3-8 The rod response cycle that transforms the capture of a photon into hyperpolarization,
which signals that light has been received.

that a purple-colored pigment iodopsin (“visual the same in rods and cones. Research indicates that
purple’’) is present in the cone cells of some birds. all photopigments, regardless of the animal species
On exposure to light, iodopsin breaks down into studied, are composed of the same retinal and a
retinal and another form of opsin. The opsin found specific protein or opsin that is characteristic of
in cones is often called photopsin to distinguish it each pigment (Dartnall, 1957; Metzler & Harris,
from that in rods, which is called scotopsin. Op¬ 1978). The biochemical events that lead to hyper¬
sins are large complex protein molecules, and the polarization in the presence of light, as we outlined
opsin protein in cones is slightly different from that above, are believed to be similar for rods and cones
found in rods. The retinal, however, appears to be (Bridges, 1986; Schnapf & Baylor, 1987). This hy-
 Neural Responses to Light 71

perpolarization serves to stimulate the bipolar cells. they form the nerve to exit the eye has led anato¬
The bioplar cells in turn stimulate the ganglion mists to refer to the blind spot as the optic disk.
cells. At the same time, complex interactions occur You may easily demonstrate the absence of vision
between neighboring bipolar and ganglion cells via in this region of the retina by referring to Demon¬
the horizontal and amacrine cells that connect to stration Box 3-5.
them. The axons of the ganglion cells then carry The output of the retina is transmitted to the
the resulting neural signals out of the eye toward brain via the optic nerves. The nerve impulses
the brain. transmitted via the ganglion cell axons that make
up the optic nerve are not “raw” sense data, but
are the result of a large amount of neural process¬
NEURAL RESPONSES TO LIGHT ing that has already taken place in the retina itself.
In order to understand how much processing has
In order to get the information about the retinal im¬ occurred, one might consider that there are some
age out of the eye and up to the brain, the axons of 120 million rods and another 5 million cones in
the retinal ganglion cells extend in a transverse each human eye. There are only about 1 million
fashion across the retina and gather together to exit axons in each optic nerve. Clearly, then, each re¬
from the eye by means of a hole through the retina ceptor cell does not have its own private pipeline
and the scleral wall. The resulting bundle of axons to the brain, but rather the responses of a very large
forms the optic nerve. Through the center of the number of photoreceptors may be represented in
optic nerve come the blood vessels that sustain the one optic nerve fiber. This comes about when the
metabolic needs of the eye. Since the bundle of combined activity of the 125 million rods and
axons must exit through the retina, there are no cones, plus the output of several million more in¬
photoreceptors in this region. Because of this there tervening bipolar, horizontal, and amacrine cells,
can be no visual response to light striking this por¬ converge on the much smaller number of ganglion
tion of the retina and it is appropriately called the cells. We will soon see how the information is
blind spot. The circular pattern of neural axons as modified as it is collected.

DEMONSTRATION BOX 3-5. The Blind Spot

The region of the retina where the optic nerve leaves only does the square seem to disappear, but also the
the eye contains no photoreceptors and thus is blind. line appears to run continuously through the area
You may demonstrate this for yourself by using the where the square should be. This indicates that we
figure here. Close your left eye and with your right automatically “fill in” missing information. We fill it
eye look at the X in the figure. Keeping your eye on in with material that is similar to nearby visible ma¬
the X, move the page toward you. At some point the terial. This accounts for why you are not normally
little open square will seem to disappear. At this point aware of the blind spot. You are simply supplying the
its image is falling on your blind spot. Notice that missing information to fill in this “hole” in the visual
when you have the page at the correct distance, not field.

X
72 Chapter 3/The Visual System

Since the information carried to the brain by a Stimulus

single ganglion cell can represent the combined ac¬ Off On Off

tivity of a large number of rods and cones, a single

ganglion cell may respond to light from a sizeable
region of the retina. Such a region, or area of the
Time —
retina, in which light alters the firing rate of a cell,
is called that cell’s receptive field. Thus, a single
ganglion cell serves as a clearing house for infor¬ i i ..hi i ii i i
mation coming from a substantial zone of receptor On response
cells in the retina. In order to understand how vi¬
sual information is processed, we would want to
1 1 1 nniiiiiiiiiniiiiiiiiiiiiliiit
know how specific ganglion cells respond to var¬
Off response
ious forms of light stimuli. Before attempting to do
this, however, it is important for you to understand
the nature of neural responses in general, and how 1 1 1 millllill 1 1 1 1 1 1 1 HIWUII 1 INI
On-off response
they are measured experimentally. If you have not
already done so, this might be a good time to read FIGURE 3-9 On, off, and on-off neural
the “Primer of Neurophysiology” that we’ve in¬ responses in the retinal ganglion cells.
cluded as an appendix to the book.

When investigators use very small stimulus

Retinal Responses and Receptive
lights (about 0.2 mm in diameter) the nature of the
Fields
retinal ganglion cell response tends to vary from
Most contemporary studies of the response of ret¬ on, through on-off, to off, depending on the loca¬
inal ganglion cells to light have followed the lead tion of the stimulus. A map of the shape of the
of Hartline (1940) and Kuffler (1953), who inserted overall receptive field of the retinal ganglion cell
an electrode through the eye of an anesthetized cat (the region of retinal stimulation to which the cell
and recorded from single ganglion cells in the ret¬ responds) shows that the responses are distributed
ina. Generally, when a single small spot of light is circularly with two distinct zones within each re¬
displayed on a screen, thereby stimulating the ret¬ ceptive field. Typically, the receptive field has a
ina of the animal observing it, three different types relatively circular center that gives on responses
of responses from a cell may be elicited, depending when stimulated. That is, the ganglion cell re¬
on the location of the spot in the field. The first sponds with an on response to the onset of a light
type of response is the one most typically expected stimulus. The outer portion of the receptive field
when a neuron is excited, a burst of neural impul¬ gives the opposite result. That is, the onset of a
ses immediately following the onset of a stimulus. light does not produce a response, but its offset
This response has been dubbed an on response. does. Between these two regions, roughly at the
Alternatively, the cell can give a burst of impulses boundary between the on and off regions, is a nar¬
coincident with the termination of a stimulus. Such row region where on-off responses occur. Typical
a response is termed an off response. Some re¬ receptive fields are shown in Figure 3-10, where on
sponses are hybrids because both the presentation response regions are marked by + and off by —
and the removal of a stimulus causes a burst of (remember, on-off responses occur at the border
neural impulses. These are designated on-off re¬ between these regions).
sponses. Typical examples of these responses are As Figure 3-10 indicates, some receptive
shown in Figure 3-9. fields have the opposite organization, with the cen-
 Neural Responses to Light 73

On responses

zone between on and off areas

On-center Off-center
receptive receptive
field field

FIGURE 3-10 Circular center-surround retinal receptive fields of two types.

tral region giving off responses and the surrounding Rodieck, 1979; Tomita, 1986). To begin with,
region showing on responses. There are approxi¬ these retinal ganglion cells differ in the speed with
mately equal numbers of off-center cells and on- which neural impulses travel along their axons.
center cells in the retina. The ganglion cells that Overall, the Y cells are the fastest, with neural
show the on and off center responses are visibly conduction speeds of about 40 meters per second,
different. Apparently, the off-center cells make whereas the X cells respond with conduction
contact with their respective bipolar and amacrine speeds of one-half that of the Y cells. The most
cells at a more peripheral level in the retina (closer sluggish are the W cells, which are physically very
to the photoreceptors) than do the on-center cells tiny and conduct very slowly (10 m/sec). W cells
(Kaneko, Nishimura, Tachibana, Tauchi & Shimai, may have complex receptive field responses of
1981; Nelson, Kolb, Robinson & Mariani, 1981). unique types that differ from those we showed in
Furthermore, there is evidence suggesting that spe¬ Figure 3-10, with some responding only to stimuli
cific amacrine cells, with different neurotransmit¬ moving in a particular way and others having on-
ters, may shape particular receptive field properties off centers (Schiller, 1986). The sluggish manner
in ganglion cells (Dacey, 1988; Masland, 1986). in which W cells respond, however, has led most
people to assume that the basic analysis of the vi¬
sual scene is mostly done by the X and Y cells
X, Y, and W Cells
(e.g., Sherman, 1985).
We now know that there are several different types X and Y cells differ along a number of impor¬
of ganglion cell responses, which differ along a tant dimensions. Both have the center-surround,
number of important characteristics. Three specific on-off arrangement we have described. However,
classes of cell, which have been given the com¬ X cells are physically smaller than Y cells and have
pletely atheoretical names of W, X, and Y cells, smaller center-surround receptive fields. Propor¬
have attracted the most research (Lennie, 1980; tionally, there are many more X than Y cells, and
74 Chapter 3/The Visual System

they differ in terms of their distribution across the stationary patterns (Kruger, 1981). However, sev¬
retina. Virtually no Y cells are found in the foveal eral other lines of data suggest that the attributes to
region, and the number of Y cells increases as we which each type of cell responds may be more
move outward into the peripheral retina. The char¬ complex (e.g., Sherman, 1985).
acteristic neural response pattern of X and Y cells
also differs. X cells, when stimulated, respond in a
rather sustained manner, continuing their neural ac¬ THE VISUAL PATHWAYS
tivity as long as the stimulus remains or for some
time after it ends. Y cells, however, have a much As we noted in an earlier section, the axons of the
more transient response. They tend to give only a retinal ganglion cells gather together and exit from
brief burst of activity when the stimulus comes on the eye at the blind spot. This bundle of axons,
or goes off, falling silent quickly thereafter. which forms the optic nerve, is the beginning of
Another difference between X and Y cells is the pipeline of information that eventually ends in
illustrated in Figure 3-11. Part A shows a sche¬ the brain. However, there are two distinct anatom¬
matic drawing of the receptive field of a retinal ical routes that lead to the common end point, and
ganglion cell in which half the field is evenly illu¬ each carries somewhat different information.
minated with light and the other half is dark. Sup¬ The primary visual pathway is the geniculo-
pose that we now switched the illumination to the striate system; the secondary pathway is the tec-
pattern shown as B or C. If we were stimulating an topulvinar system. Both begin in the same
X cell it would continue to respond exactly as it fashion, with the information traveling out of the
had been responding. In other words, as long as the eyes along the optic nerves. As can be seen in Fig¬
same amount of illumination is present in the center ure 3-12, the first significant event occurs where
and surround, the X cell does not distinguish be¬ the two optic nerves come together at a point that
tween the different locations of illumination. How¬ looks like an X. This point is called the optic
ever, any switch in the pattern of illumination will chiasm (from the Greek letter which is called
provoke a vigorous response in a Y cell. Since such chi). In lower animals, the optic nerve from the
changes in the distribution of illumination across right eye crosses completely to the left side of the
the receptive field are usually caused by movement head and vice versa. In many mammalian species,
of an object, the usual interpretation of this fact is particularly those who seem to use combined input
that Y cells may be specialized for movement de¬ from the two eyes to obtain better depth perception
tection and X cells specialized for the analysis of (this is discussed more fully in Chapter 10), some
of the fibers do not cross. In primates, such as
man, approximately one-half of the optic nerve fi¬
bers cross to the opposite side of the head. These
are the fibers that represent the two inside or nasal
retinas. Those from the outside or temporal halves
of each retina do not cross but continue on the
same side. Such an arrangement implies that the
ABC two halves of the visual field will be projected to
the opposite sides of the brain. You should also
FIGURE 3-11 If the illumination pattern on a
center-surround receptive field was half light and
remember what is happening optically in order to
half dark as shown in A, and then was shifted to a keep the situation straight. Since the crystalline
new orientation (either B or C), an X cell would lens in the eye reverses the image up-down and
not respond to the change but a Y cell would. right-left, the right visual field is projected onto the
 The Visual Pathways 75

Thalamus

Superior colliculus

Lateral geniculate nucleus

Optic radiations

Visual cortex (occipital lobe)

Optic chiasm

Optic nerve

Left visual field Center Right visual field

Temporal retina ^Temporal retina

Optic nerve—

Optic chiasm

Optic tract

Lateral geniculate nucleus

Nuclei of the thalamus
[ Pulvinar nucleus -

Superior colliculus-

Optic radiations

Visual cortex (occipital lobe)

FIGURE 3-12 The visual pathways from the eye to the visual cortex.
76 Chapter 3/The Visual System

nasal half of the retina of the right eye and the tem¬ rate over the baseline activity level) or inhibitory
poral half of the retina of the left eye. The axons (cause a decrease in activity relative to the resting
from the left half of each retina (or left hemireti- response rate).
nas) terminate in the left side of the brain. Thus, As is the case with the retinal ganglion cells,
information from the right side of your field of lateral geniculate cells do not respond to visual
view ends in the left side of your brain, and vice stimuli unless the stimulation occurs within their
versa as shown in Figure 3-12. receptive fields. Thus, the activity of a particular
As we begin to trace out the visual pathways lateral geniculate neuron provides information
we will need a few maps to help us. A general map about the location of an object in space because it
is given in Figure 3-12. We’ll provide you with a responds only to those objects projected onto the
few additional ones as we progress. patch of retinal receptors that define its receptive
field. These receptors, in turn, respond only to ob¬
jects in a particular region of the visual field. Gen¬
The Geniculostriate System
erally, the receptive fields of the lateral geniculate
Beyond the optic chiasm the pathway is no longer cells are similar to those of the retina. If we map a
called the optic nerve, but rather the optic tract. lateral geniculate cell by projecting points of light
The central relay center for most of the sensory in¬ onto a screen in the visual field in front of an ani¬
formation reaching the brain is a large body in the mal, we find that the cell response appears similar
midbrain known as the thalamus. For most pri¬ to that of a retinal ganglion cell. For instance, most
mates the major termination for the optic tracts is such cells have an on center and an off surround,
reached when the fibers synapse in the lateral ge¬ or the reverse. One interesting finding is that the X
niculate nucleus of the thalamus. The lateral ge¬ and the Y cells go to different regions of the lateral
niculate is arranged in six layers, each of which geniculate, hence acting much like parallel and in¬
seems to receive input from only one eye, although dependent pathways (Schiller, 1986). In addi¬
there is some interaction between the layers. Dif¬ tion, lateral geniculate cells, which receive X and
ferent layers may process different types of visual Y inputs from the retina, show X-like and Y-like
information (Livingston and Hubei, 1988). Each responses to moving and stationary stimuli
layer contains a map of the visual field, in that (Lehmkuhle, Kratz, Mangel & Sherman, 1980;
there is a point in each layer that corresponds to a Lennie, 1980; So & Shapley, 1981).
point in the visual field, or a point in the retina. There is one feature that makes the lateral
Later on we will find that there are many maps of geniculate responses extremely interesting. Much
the visual field in the brain (Orban, 1984). evidence suggests that some cells respond
Electrophysiological studies of the lateral ge¬ differentially not only to the location of a light but
niculate nucleus have shown that its neurons are also to its color (DeValois & Jacobs, 1984; Hurv-
spontaneously active. This means that these cells ich, 1981). For instance DeValois and DeValois
are always producing some number of neural im¬ (1975) reported that one type of cell in the lateral
pulses, even in the dark. Although this may seem geniculate of monkeys responds with an increase in
a bit surprising, spontaneity is a characteristic of firing rate when the center of the receptive field is
brain cells. We do not fully understand why this stimulated by a red spot of light, whereas it shows
activity maintains itself. It may simply be because an off response (a decrease in activity relative to
the neurons are alive and announce this by occa¬ the spontaneous firing rate) when the center is stim¬
sional random responses. This continuing train of ulated by a green light. This finding suggests the
responses does augment the information coding ca¬ possibility of encoding both color and location in¬
pacity of the cells, however, since signals may now formation in the same cells, a concept we will en¬
be either excitatory (cause an increase in the firing counter again in Chapter 5.
 The Visual Pathways 77

FIGURE 3-13 The principal visually responsive regions of the cortex, and associated regions, with
Brodmann’s numbering of the areas and several alternate labeling systems.

When the axons of the lateral geniculate neu¬ 20 such separate maps in the primate cortex (e.g.,
rons leave the geniculate, they spray out and form Essen, 1984). Recently researchers have begun to
a large fan of fibers called the optic radiations. name these maps separately. Thus, Areas 17, 18,
These fibers eventually synapse with cells in the and 19 are referred to as VI, V2, and V3, desig¬
cortex in the rear (or posterior) portion of the brain, nating the first 3 visual maps in the cortex. Figure
known as the occipital lobe. Several alternate la¬ 3-13 shows these areas, and some others we will
bels have been used to refer to this region of the discuss shortly.
brain. The most popular system for locating parts
of the cortex is the numbering system devised by
Brodmann (1914) based on the appearance of cells.
The Tectopulvinar System
In Brodmann’s numbering system this primary area
of visual function, that is, the place where the fi¬ There is a second pathway to the visual cortex,
bers from the lateral geniculate terminate, is des¬ which is also indicated in Figure 3-12. This begins
ignated Area 17. when a number of fibers from the optic tract branch
Since when sliced vertically there is a banded off to go toward the brain stem. Most pictures of
appearance to the cortex, this region is also called the brain do not show the brain stem or midbrain
the striate (“striped”) cortex. An alternate, but structures because they are hidden by the cortex. In
informative, label used for this area is the primary Figure 3-14 we have shown schematically where
visual cortex. Some of the geniculate fibers also these structures are located, and we have marked
project to an area adjacent to Area 17, which is off some of the areas that are of concern to us. The
Area 18. It is one of the regions known as the ex- region of the brain stem of interest is, in an evolu¬
trastriate cortex (extra has the meaning here of tionary sense, a much older, more primitive visual
“beyond”), which is usually regarded as part of center known as the tectum. In some lower ani¬
the secondary visual cortex. Actually, Areas 17, mals most visual processing occurs here. The part
18, and 19 each contain a separate representation of the tectum that receives most of the incoming
of the visual field, with each point of the field rep¬ fibers is the upper pair of what appear to be four
resented by a region in the cortex. There are over bumps on the back (or dorsal) surface of the brain
78 Chapter 3/The Visual System

stem; these are known as the superior colliculi. golden hamster left the animal with an inability to
Not all of the retinal ganglion cell types project to recognize patterns, whereas removal of the superior
the superior colliculi, however. Only Y- and W- colliculi left it with the ability to identify patterns
type inputs come here. X inputs seem only to be but an inability to localize them well enough to ap¬
part of the geniculostriate system, whereas Y and proach them. Deficits in spatial localization and
W cells seem to be part of both (Orban, 1984). depth perception have also been shown for cats
From the superior colliculi the pathway con¬ when the tectopulvinar system is blocked (Ogasa-
tinues on to the thalamus. However, rather than wara, McHaftie & Stein, 1984). This type of result
projecting to the lateral geniculate nucleus, the seems consistent with the speculation that the two
pathways go the pulvinar and the lateral posterior anatomical visual pathways serve the different
nuclei, which are located nearby, as you can see functions of localization and identification, al¬
from Figure 3-14. From here, the fibers project to though some overlap in function seems likely.
the cortex. None are destined for the primary visual
cortex (Area 17), but rather they extend to the sec¬
ondary visual areas (Areas 18 and 19). THE VISUAL CORTEX
Do the two different visual pathways serve
different perceptual functions? Basically, the genic¬ There are over 100 million neurons in the visual
ulostriate system seems to be involved in the fine¬ cortex. Only the smallest fraction of these have
grained perception of patterns, and perhaps colors, been thoroughly studied in attempts to discover
whereas the tectopulvinar system seems to coordi¬ their response characteristics. What we do know of
nate the localization of objects in space, the guid¬ these cells is based largely on research done with
ance of eye movements, and gross pattern microelectrodes, employing techniques similar to
perception (Essen, 1984; Ungerleider & Mishkin, those used in the mapping of the receptive fields
1982). Perhaps the most dramatic demonstrations for the retinal ganglion and lateral geniculate cells.
of these separate functions come from studies of Much of the pioneering work was done by David
lower vertebrates. In one study, Schneider (1969) Hubei and Torstein Wiesel, who received the No¬
showed that removal of the lateral geniculate of the bel prize in 1981 for this research effort.

FIGURE 3-14 The visually responsive areas of the thalamus and optic tectum.
 The Visual Cortex 79

We have already mentioned that there are sev¬ majority of the measured receptive fields have
eral places in the cortex where there is a rather di¬ elongated central regions.
rect topological (point-for-point) mapping of the A map of some such cells is shown in Figure
external visual world, with specific points in the 3-15. This type of cell, which Hubei and Wiesel
environment corresponding to specific points in the labeled a simple cell, generally has no spontaneous
cortex. This was initially established for Area 17 activity at all and never seems to respond to diffuse
based on clinical data from cases of accident or war illumination covering the whole screen. Sometimes
injury, where penetrating missile wounds have in¬ such cells respond, although grudgingly, to small
jured specific parts of the cortex. When a piece of spots of light. However, because of the elongation
the occipital cortex is so damaged, the patient is of the central region of the receptive field, the best
blind in part of the visual field. Such a damaged stimulus for such a cell is a dark or light bar or line
area is technically called a lesion, and the blind flashed in the appropriate location in the receptive
patch in the visual field is called a scotoma (mean¬ field. Figure 3-15 shows the receptive fields that
ing “dark spot”). Through the use of such lesions, might be mapped from several simple cells. Be¬
the correspondence between sections of cortex and neath each of them you will see the stimulus that
parts of the visual field can be mapped, although produces the maximal response for each of these
the cortical map does not correspond exactly with receptive fields. Notice that in every case the edge
the external scene in all of its dimensions. For in¬ between the light and the dark must be at a partic¬
stance, the fovea is represented by an inordinately ular orientation in a particular location. If the edge
large quantity of cortex relative to its actual size on of the line is flashed on the receptive field at a dif¬
the retina. This is in accord with its disproportion¬ ferent angle, a greatly reduced response may be ob¬
ate importance relative to other retinal regions. tained or perhaps no response at all. For this reason
When we say that points on the cortex correspond such cortical cells are said to have orientation
to points in the visual field, we do not mean to specificity, which the simple circular center-sur¬
imply that if you are looking at a house there is a round cells do not have.
house-shaped pattern of electrical excitation in the There are other kinds of neurons in the visual
cortex. Rather, it indicates that much of the analy¬ cortex that seem to be tuned to even more compli¬
sis of features of visual input occurs within indi¬ cated pattern properties of a stimulus. These more
vidual cortical neurons and patches of adjacent elaborate feature-analyzing neurons have been la¬
cortex. beled complex cells. They have larger receptive
fields than do simple cells, although their size may
vary tremendously. Like simple cells, complex
Receptive Fields in the Visual Cortex
cells respond maximally to stimuli when they are
Following the methodological procedures of Hubei in a particular orientation. However, they rarely
and Wiesel (1962, 1979), many investigators have respond to any flashing patterns. What they prefer
mapped receptive fields of cortical cells in animals. is a bar or edge moving somewhere within the re¬
For instance, recording electrodes have been placed ceptive field, and its location does not appear to be
(or implanted) in cortical cells in Area 17 and the particularly important. In other words, the complex
cell’s electrical responses to stimuli projected onto cells seem to generalize their response over a wider
a screen in front of the animal measured, as we area of the visual field. Figure 3-16 shows the re¬
illustrate in the Appendix, Figure A-5 (DeValois, sponses of a complex cell to two different moving
Yund & Helper, 1982; Heggelund, 1981a, 1981b). light slits, one in the optimal and the other in a
When cortical cells are mapped in this way, the nonoptimal orientation. Notice that both direction
familiar circular on and off regions of ganglion and of movement and orientation are important factors
lateral geniculate cells are still found. However, the in determining the response.
80 Chapter 3/The Visual System

Dark line detector Edge detector

FIGURE 3-15 Receptive fields of simple cortical cells: + indicates a region in the receptive field that
gives an on response, and — indicates an off response.

These complex cells do not exhaust the types ulate body is Layer 4. In this layer we find the
of cells recorded in the cortex. At a slightly more largest number of simple cortical cells, and we also
sophisticated level are certain special complex cells find cells with the simple circular center-surround
that are often called hypercomplex cells. These re¬ receptive field found at lower levels of the visual
spond not only to the orientation and to the direc¬ system. Above this level, in Layers 2 and 3, we
tion of movement of the stimulus, but also to the find that nearly 90 percent of the cells have strong
length, the width, or other features of shapes, such orientation sensitivity. Layers 5 and 6 contain cer¬
as the presence of comers. Figure 3-17 shows an tain special complex cells, with the receptive fields
example of some hypercomplex cell responses. of cells in Layer 5 being quite large and particu¬
larly sensitive to the direction of stimulus move¬
Organization of the Primary Visual ment and those in Layer 6 being rather long,
narrow, and directionally sensitive. This arrange¬
Cortex
ment is depicted in Figure 3-18.
Simple, complex, and hypercomplex cells are not Cells sensitive to various orientations are not
randomly intermixed in the visual cortex. Instead, randomly distributed either. Instead, cells with a
particular cell types are organized spatially into an particular orientation sensitivity tend to be aligned
incredibly detailed structure. The cortex in Area 17 in a column, as diagramed in Figure 3-19. As we
is arranged in six layers, numbered 1 to 6, begin¬ move across the cortex, the orientation specificity
ning with the outermost (surface) layer. The only shifts by about 10 deg per column. Moving in the
layer receiving direct inputs from the lateral genic¬ other direction, we encounter columns of cells that
 The Visual Cortex 81

Preferred orientation and direction Length detector

7 Mmimi 111111111 n i ■ 11111111111

/z
.

_ _

Preferred orientation and nonpreferred direction

1-

—-

For nonpreferred orientations, direction

does not make a difference
Width detec;tor

1
1I
11

Stimulus Response

FIGURE 3-16 Some typical complex cortical cell

►
Angle detector
responses.

have the same orientational sensitivity, however

each column looks at a slightly different part of the I
visual field, thus forming a sort of “slab” of cells
with a specific orientational tuning.
One final aspect of the organization of the cor¬
tex must be mentioned. Each cortical cell tends to
be more responsive to one particular eye than to the
other, hence showing a relative eye dominance. Stimulus Response
These inputs are also spatially separated, with a
slab of cells responding to one eye located next to FIGURE 3-17 Some typical hypercomplex
cortical cell responses.
one driven by the other eye. These are systemati¬
cally arranged in alternating stripes across the cor¬
tex. A region of cortex containing all 360 deg of square, and 2 mm deep, and is diagramed in Figure
orientational specificity, and including a region re¬ 3-19. Our catalog of cell types and arrangements
sponsive to both the left eye and right eye, forms a does not exhaust all the forms of special receptive
larger unit sometimes called a hypercolumn. Such field properties that may be perceptually important.
a piece of cortex might be between 0.5 and 1 mm We shall mention others when we talk about
82 Chapter 3/The Visual System

tainly, many pattern-recognition machines have

been built using this principle. For instance, the
machine that reads the numbers on your bank check
is tuned to respond to certain, highly stylized, nu¬
merical stimuli by matching the patterns on paper
to internally represented codes. Unfortunately, in
mammalian species the significance of such fea¬
ture-extraction cells for the perceptual process has
not yet been determined. There also is evidence
that removal of large areas of the primary visual
cortex (and hence most of these orientation-specific
cells), although greatly reducing visual acuity, ac¬
tually does not have the massively disruptive ef¬
fects that might be expected if higher-level
perception were directly dependent on extraction of
simple features by these receptive cells (e.g.,
Lehmkuhle, Kratz & Sherman, 1982). Further¬
more, it is now known that if we alter an animal’s
visual experience from birth, we can alter the dis¬
tribution of orientations and features to which cor¬
tical cells are responsive. The significance of such
changes will be discussed later when we deal with
the effects of experience in Chapter 17. As Barlow
(1985) has pointed out, it is very difficult to go
directly from single-cell responses to higher-level
FIGURE 3-18 Location of various visually perceptual phenomena. It is often much more use¬
responsive cells in the layers of the striate cortex
ful to go the other way, from the perceptual phe¬
(Area 17).
nomena to the search for a physiological unit that
might support it.
brightness, spatial frequencies, color, and depth
perception in later chapters. For the moment, how¬
The Temporal Lobes
ever, these give a general idea of the types of
visual analysis that may be monitored at the single¬ Both the geniculostriate system and the tectopulvi-
cell level in the primary and secondary visual cor¬ nar system have inputs to the secondary visual
tices. areas 18 and 19. From here the information seems
Such a detailed architecture, spatial arrange¬ to travel onward to the temporal lobes of the brain,
ment, and specificity at the cellular level is bound which roughly correspond to the regions of the
to elicit certain forms of theoretical speculation. brain directly behind the temples of your skull
Thus, the existence of orientation-specific or fea¬ (Rockland & Pandya, 1981). Here, recent evidence
ture-specific cells in the cortex suggests that pattern suggests, there are additional maps of the visual
perception may take place by decomposing visual field and some very complex visual processing
stimuli into component features or contours. These takes place. In order to localize these tertiary visual
would then be resynthesized at a later point accord¬ areas you might look back at Figure 3-13.
ing to some plan or template (Frisby, 1980). Cer¬ The first part of this region to be studied was
 The Visual Cortex 83

Each slab of cells with

FIGURE 3-19 Diagram of a hypercolumn, which is a small region of visual cortex containing inputs
from both eyes and all visual orientations, separated spatially, as shown.

the lower portion, called the inferotemporal cor¬ mouth. Within a few trials, a normal monkey will
tex. The visual significance of this section of cortex let the metal objects pass by and select only the
was accidentally discovered by Kluver and Bucy in food. For animals with inferotemporal lobe loss,
1937 while observing monkeys who had undergone however, both the food and the inedible object
surgery that removed most of both temporal lobes. were picked up on virtually every trial. The animal
They called the syndrome psychic blindness. The seemed to show no evidence of learning to discrim¬
animals could reach for, and accurately pick up, inate between the targets. Wilson (1957) found that
small objects; hence, they were clearly not blind. such monkeys could discriminate between an in¬
However, they did appear to have lost the ability to verted and an upright L by touch, yet with infero¬
identify objects by sight. An example of this is temporal lesions they could not make the same
shown in what was named the concentration test. discrimination visually.
Here, a piece of food or a metal object is presented This syndrome is similar to a human defect
to the monkey approximately every 30 sec. Gener¬ called visual agnosia (Kolb & Whishaw, 1985).
ally, a normal monkey will eat the food, and dis¬ Such patients can see all parts of the visual field,
card the nail or steel nut after examination by but the objects they see mean nothing to them.
84 Chapter 3/The Visual System

Patients with lesions of the right temporal lobe also side. An animal looking at his own hand would
show deficits on a variety of visual tests. For in¬ most likely see a hand with fingers pointing up¬
stance, they have difficulty in placing pictures in a ward. Such complex response specificity has been
sequence that relates a meaningful story or pattern. observed a number of times in this region of the
They also have difficulty in learning to recognize cortex (Desimone & Gross, 1979).
new faces. Furthermore, such patients make poor Even more startling degrees of stimulus anal¬
visual estimates of the number of dots in an array, ysis seem to be emerging from a region of the tem¬
have difficulty recognizing overlapping figures, poral lobe called the superior temporal cortex.
have poor memory for nonsense forms, and gener¬ Here Bruce, Desimone, and Gross (1981) found
ally have poor picture memory. We’ll have more cells in monkeys that responded selectively to
to say about such agnosias when we consider some drawings of faces. The more realistic, and monkey¬
pathologies of perception in Chapter 18. like, the face, the stronger the response. Distorting
Needless to say, some investigators have be¬ the stimulus, by removing the eyes, scrambling the
gun to map single neurons in the inferotemporal features, or presenting a cartoon caricature, re¬
cortex. Some microelectrode measurements in the sulted in weaker responses. Kendrick and Baldwin
monkey brain have produced startling results, sug¬ (1987) have found similar cells in sheep that re¬
gesting that neurons found in this part of the brain spond preferentially to sheep faces and, interest¬
have amazing response specificities. Although this ingly, to human faces as well. Thus, it seems
research area is quite new, neurons sensitive to possible that in your temporal cortex there might
size, shape, color, orientation, and direction of be a template for the perception of your grand¬
movement have already been discovered in this re¬ mother, your car, or many other familiar stimulus
gion of the brain (Desimone, Albright, Gross & shapes.
Bruce, 1980). There is even a report that one neu¬
ron produced its best response when the stimulus
was the outline of a monkey’s paw. Gross, Rocha-

4f \
Miranda, and Bender (1972) report that one day
they discovered a cell that seemed unresponsive to % \ Most vigorously
any light stimulus. When they waved a hand in ( responded to stimuli

front of the stimulus screen, however, they elicited

a very vigorous response from the previously un¬
responsive neuron. They then spent the next 12
hours testing various paper cutouts in an attempt to
find out what feature triggered this specific unit. Stimuli producing
When the entire set of stimuli was ranked accord¬ moderate response

ing to the strength of the responses they produced,

they could not find any simple physical dimension

♦I
that correlated with this rank order. However, the
rank order of stimuli, in terms of their ability to
drive the cell, did correlate with their apparent sim¬ Stimuli producing
ilarity (at least for the experimenters) to the shadow weak response

of a monkey’s hand. The relative adequacy of a

few of these stimuli in producing a neural response
FIGURE 3-20 The stimuli in the figure were
is shown in Figure 3-20. Interestingly enough, fin¬
used to excite a neuron in the inferotemporal
gers pointing downward elicited very little response cortex of a monkey. Notice that the more handlike
when compared to fingers pointing upward or to the a stimulus is, the more vigorous the response.
 The Visual Cortex 85

Alternate Visual Maps visual system is to identify objects and locate them
in space. A number of investigators (e.g., Cowey,
We mentioned earlier that there were many alter¬ 1981; Livingston and Hubei, 1988; Phillips, Zeki
nate maps of the visual field in the cortex. More & Barlow, 1984) have suggested that each map of
than 20 have been described for primates, with the visual field is set up to extract some subset of
over a dozen of these in the temporal cortex, at properties from the visual image. Thus, one map
least 3 in the occipital region, and others dotted might be relatively specialized for color, another
around the brain (Essen, 1984). The pattern of in¬ for orientation or movement, yet another for dis¬
nervations, in the forms of X, Y, and W cells, and tance or texture, and so forth. The reason for the
the specific pathways to the cortex activated can topographic mapping may be to isolate “an object”
take on quite a complex pattern, as can be seen in with “a property” with “a specific location in
the summary presented in Figure 3-21. space.” We will encounter this concept again when
Why are there so many different maps of the we talk about object and form perception in Chap¬
visual field? To begin with, you must remember ter 11 (see also Treisman, 1986). This means that
that the function of the visual system is not to re¬ the destruction of any one of these cortical maps
create an image of the outside world in the brain, may produce very subtle disruptions of visual pro¬
since there is nobody in there to look at such an cessing, rather than massive global effects, such as
image even if it were there. The function of the the total loss of the ability to read type on paper.

Geniculostriate system

I I I

Tectopulvinar system

FIGURE 3-21 A highly schematic overview of the visual system, indicating the two visual pathways and
their connections with the major subcortical and cortical centers.
 86 Chapter 3/The Visual System

GLOSSARY Ganglion cells Third layer of the retina through

which neural signals travel after the photoreceptor and
bipolar cells.
The following definitions are specific to this book. Geniculostriate system The primary visual pathway
passing through the lateral geniculate nucleus to the
Accommodation The process by which the lens of
striate cortex.
the eye varies its focus.
Horizontal cells Retinal cells with short dendrites
Acuity The ability of the eye to resolve or discrimi¬
and an axonal process that extends horizontally.
nate details.
Hypercolumn A small piece of visual cortex, sensi¬
Amacrine cells Large, laterally interconnecting neu¬
tive to input from both eyes and to a full range of ori¬
rons found in the retina near the ganglion cell layer.
entation specificity.
Aqueous humor The fluid occupying the small
Hypercomplex cells Cortical cells that respond to
chamber between the cornea and the lens of the eye.
complex stimulus features regardless of where they occur
Area 17 According to Brodmann’s numbering sys¬
in the receptive field.
tem, another name for the primary visual cortex, which
Hypermetropia Farsightedness.
is one part of the occipital lobe of the brain receiving
inputs from the geniculostriate system. Hyperpolarization The change of the electrical po¬
tential of a cell toward increasing negativity.
Areas 18 and 19 Secondary visual cortex receiving
inputs from both the geniculostriate and tectopulvinar Inferotemporal cortex A cortical region located in
systems. the temporal lobes of the brain, which may be associated
Bipolar cells Neural cells in the retina between the with recognition abilities and, pathologically, with visual
photoreceptor and ganglion cell layers. agnosia.

Iffind spot Portion of retina through which optic Iodopsin The cone pigment present in some birds.
nerve passes and therefore an area without photorecep¬ «-Jfis The opaque, colored membrane controlling the
tors. This region shows no response to light. amount of light entering the eye by changing the size of
Complex cells Cells in the visual cortex that respond the pupil.
to features such as line orientation and direction of Lateral geniculate nucleus The first major relay
movement. center in the geniculostriate system for optic nerve fibers
Concentration test Method for determining ability leaving the retina. It is in the thalamus in primates.
to discriminate objects visually following inferotemporal Lateral posterior nucleus A visual center in the
lobe loss. thalamus, part of the tectopulvinar system.
Cones Short, thick, tapering cells in the photorecep¬ -''Lens A transparent body in the eye. It can change
tive layer of the retina, used in bright-light and color shape, thus altering the focus of the retinal image.
vision. Macula lutea A yellow pigmented area of the retina
•—'(fornea The transparent, domelike part of the eye centered over the fovea.
formed by the bulged sclera. Millimicron (mp) A thousandth of a micron, or a
^/£)ay blindness Visual difficulty under bright-light millionth of a millimeter. Now usually called a nanome¬
conditions, caused by an absence or nonfunctioning of ter.
cones. Myopia Nearsightedness.
Duplex retina theory The concept of two separate Nanometer (nm) A billionth of a meter (a millionth
visual systems, rod-dependent for dim-light vision and of a millimeter).
cone-dependent for bright-light vision.
Near point distance The nearest point to which an
Emmetropic Referring to an eye with normal ac¬ object may be brought to an eye and still remain in focus
commodative ability. on the retina.
Extrastriate cortex Areas 18 and 19 in the occipital ’■'Night blindness The inability to see under low-light
lobe of the cortex, the secondary visual cortex. (twilight) conditions, caused by an absence of function¬
Eye dominance Refers to the fact that most cortical ing rods.
receptive cells can be driven better by one eye than the Occipital lobe The rear portion of the brain, which
other. serves as the primary visual processing center.
ovea centralis A small depression in the retina that Off response A neural response commencing with
contains mostly cones and where acuity is best. the termination of a stimulus.
 Glossary 87

On response A neural response commencing with Refractive error Light-bending or focusing error.
the onset of a stimulus. '-'Retina The rear portion of the eye containing pho¬
On-off response A burst of neural responses given toreceptors and several types of sensory neurons.
both at the onset and at the termination of a stimulus. Retinal Part of the rhodopsin pigment, similar to vi¬
Opsin Protein part of the rhodopsin pigment. tamin A.
Optic axis Hypothetical line from the center of the Rhodopsin The photopigment found in rods.
pupil to the fovea, used as a reference point for distances ~/ftods Long, thin, cylindrical photoreceptors in the
across the retina. retina, used in low-light vision.
Optic chiasm The point at which the two optic '-''iSclera Strong, elastic outer covering, seen as the
nerves meet and the nasal fibers cross to the contralateral “white” of the eye.
side.
-''Scotopic Referring to vision under low-light condi¬
Optic disk The region of the retina where the optic tions.
nerve leaves the eye. See Blind spot,
Scotoma A localized blind spot.
v^t)ptic nerve The collection of axons from retinal
Scotopsin The protein portion of rhodopsin—a type
ganglion cells as they exit the eye.
of opsin.
Optic radiations The large fans of neural fibers Secondary visual cortex Brodmann’s Areas 18 and
spreading out from the lateral geniculate nucleus to the 19 in the occipital lobe.
occipital cortex.
Simple cell A cortical cell that responds to lines of
Optic tract The path of the optic nerve once it is particular orientation and location.
past the optic chiasm.
Striate cortex Area 17 in the occipital lobe.
Orientation specificity A property whereby cortical
Superior colliculi The termination points of optic
cells respond selectively to visually presented lines at a
nerve fibers in the brain stem.
particular orientation.
Superior temporal cortex An area of the temporal
Photon A quantum of light energy.
lobe whose cells show a high degree of response speci¬
•-^Photopic Referring to vision under bright-light con¬ ficity to visual stimuli.
ditions.
Tectopulvinar system A secondary pathway to the
Photopsin The protein segments of the photochemi¬ visual cortex that includes nuclei in the brain stem and
cal in cones. the thalamus.
Photoreceptors Photosensitive cells in the retina Tectum A primitive visual center in the brain stem.
(rods and cones).
Thalamus A large mass of neural tissue located at
Pigment epithelium The light-absorbing dark layer the base of the cerebrum, which serves as a major
backing the retina in diurnal animals. “switching center” for sensory information.
Presbyopia Farsightedness found in older individu¬ VI Area 17 in primates.
als.
V2 Area 18 in primates.
Primary visual cortex Primary area of visual func¬
V3 Area 19 in primates.
tion in the occipital lobe (Area 17).
Visual agnosia Syndrome in which all parts of a vi¬
Psychic blindness The condition in which animals
sual field are seen, but the objects seen are without
are able to locate objects yet are unable to identify them.
meaning.
Pulvinar nucleus A visual center in the thalamus.
Vitreous humor The jellylike substance filling the
'•''Pupil The opening in the iris of the eye through large chamber of the eye.
which light enters. W cells Slow-conducting retinal ganglion cells
Quantum The smallest amount possible of any form whose function is not yet fully established.
of electromagnetic radiation. See Photon.
X cells Moderate-conduction-speed retinal ganglion
Receptive fields For any particular cell, the region cells that may be important in detailed pattern vision.
of the visual field in which a stimulus can produce a Y cells Fast-conducting retinal ganglion cells that
response. may be important in motion perception.
Reflecting tapetum The shiny surface backing the
retina in some nocturnal animals.
CHAPTER
4
Brightness and
Spatial
Frequency
PHOTOMETRIC UNITS
FACTORS IN BRIGHTNESS PERCEPTION
Adaptation
Retinal Locus
Wavelength
Time and Area
Maximum Sensitivity
BRIGHTNESS CONTRAST
COGNITIVE FACTORS IN BRIGHTNESS
PERCEPTION
VISUAL ACUITY
SPATIAL FREQUENCY ANALYSIS
Spatial Fourier Analysis
Modulation Transfer Function
Neural Spatial Frequency Channels
Selective Adaptation
90 Chapter 4/Brightness and Spatial Frequency

T he following scene must have played countless

times on cinema screens in Grade B horror
specific purpose by some technical or academic
subdiscipline. The result was chaos. Even among
movies. the most scholarly, few can tell you how many nits
It is night, and in the darkness two old ragged there are in an apostilb or a blondel, or how any of
beachcombers can barely be seen moving along the these units are related to a candle or a lambert. In
water’s edge. Suddenly, one stops. 1960, the International Conference on Weights and
“Hey, Charlie, I think there’s something out Measures established the Systeme International
there.” d’Unites, which is a uniform system of measure¬
“W-What is it?” ment (known commonly as the SI System).
“I can’t make it out. It’s some sort of glow. Throughout this book we use these standard units.
It’s too dim to make out what it is.” Should your reading bring you into contactxvith
This scene illustrates the most basic property some of the older forms of photometric measure¬
of vision, namely, that it depends on the presence ment, we can only refer you to some more ad¬
of light. The most primitive visual percepts are vanced texts such as Wyszecki and Stiles (1967) to
simply reactions to the intensity of the incoming try to make sense of the quantities involved.
energy. These responses are represented in con¬ Basically, there are two ways light can reach
sciousness as a brightness or glow. We often sense the eye: (a) directly from a radiating source, such
the presence of light before sufficient energy exists as a light bulb, fluorescent tube, firefly, or the sun;
for us to apprehend shape or form. Thus, the next or (b) by reflection from surfaces that have radiant
line in the scenario above usually goes, “It’s get¬ energy falling on them, such as walls and paper.
ting brighter, ’ ’ and then as the energy becomes suf¬ Different types of measures are used for these dif¬
ficient to apprehend the object itself, “Oh my God! ferent types of light input. All photometric units,
It’s some sort of creature!” As we shall see, the however, are ultimately based on the amount of
perception of brightness is much more complex and light emitted from a single candle. The nature of
surprising than the script of this particular film. this standard candle, its photic energy, and the
specific measures derived from it have been fixed
(albeit somewhat arbitrarily) by an international
PHOTOMETRIC UNITS body called the Commission Internationale de
I’Eclairage, usually known as the CIE.
Electromagnetic energy, or light, can vary along Each different aspect of light is designated by
three dimensions: intensity, wavelength, and dura¬ its own name and requires a different measurement
tion (see Chapter 3). All dimensions are important unit. The amount of energy coming from a light
in the perception of brightness, although brightness source is called its radiance. The unit of radiance
varies most directly with intensity. Of course, to is the standard candle, which produces an energy
make sense of the perceptual effects, we must first of slightly more than 0.001 watt at a wavelength of
be able to specify the physical intensity of the stim¬ 555 nanometers. This quantity of luminous energy
ulus. This is not as simple as it seems. is called a lumen. The amount of light falling on a
Light measurement is based on the visual ef¬ surface is another photometric quantity called illu¬
fects produced by visible radiation and is called minance. The amount of light reflected from a sur¬
photometry. Photometric units are used to describe face is called its luminance, and the percentage of
the stimulus and these units are, by convention, ex¬ light falling on a surface that is reflected is called
pressed in terms of energy. Unfortunately, over the its reflectance. The amount of light reaching the
years a confusing array of photometric units has retina is called the retinal illuminance. Finally,
been developed, most of them designed for some the phenomenal impression of the light intensity of
 Photometric Units 91

a stimulus is called its brightness. Thus, if we parent brightness measured by a direct scaling tech¬
have a slide projector shining on a screen, the nique (such as magnitude estimation) grows
amount of light energy leaving the bulb determines approximately as the cube root of the physical in¬
its radiance, the amount of light falling on the tensity (to be precise, the phenomenal sensation
screen is its illuminance, the amount of light re¬ grows at a rate equivalent to the light intensity
flected from the screen is its luminance, and our raised to the 0.33 power). This means that if you
perceptual or phenomenal impression is the bright¬ had a theater stage illuminated by 8 lights and you
ness of the screen. Table 4-1 summarizes the most wished to increase the perceived brightness of the
common photometric quantities, how they are mea¬ area, doubling the number of lights to 16 would not
sured, the units used, and some of their specific double the perceived brightness but would only in¬
properties. crease it by one-third. If you wanted to double the
Before becoming too involved with measure¬ phenomenal brightness, you would have to increase
ment, however, it is important to remember that, the number of lights to 64!
perceptually, brightness is not explained simply by Figure 4-1 shows the general shape of this re¬
the amount of light reaching the eye. As we noted lationship graphically. Notice that the curve in Fig¬
in Chapter 2, when we plot the magnitude of the ure 4-1 greatly resembles the logarithmic curve of
brightness sensation against the physical stimulus Fechner’s law (remember Chapter 2, and particu¬
intensity we get a nonlinear relationship. The ap- larly Figure 2-14). Although this would not be the

Table 4-1 Photometric Units

Photometric What is
term measured Unit How measured Comments

Radiance or Radiant energy from Lumen A candela is the light Defined in terms of a
luminous flux a light source of a 1-lumen standard candle
source at a distance (candela)
of 1 m shone on a
square meter
Illuminance Light falling on a sur¬ Lux 1 lumen/m2 As the source moves
face farther away illumi¬
nance decreases

Luminance Light reflected from a Candelas per square Lumens reflected Independent of dis¬
surface meter from a surface tance of eye from
surface

Reflectance Proportion of light re¬ Percentage reflectance Luminance Really ratio of re¬
(albedo) flected from surface Illuminance flected to incident
light

Retinal . Amount of light inci¬ Trolands 1 candela/m2 seen Roughly 0.0036 lu¬
illuminance dent on the retina through pupil of 1 mens/m2 through a
mm2 area 1-mm2 pupil

Brightness Phenomenal impres¬ Not yet agreed on, Relative matching and Psychological rather
sion of light inten¬ but bril is best con¬ scaling techniques than physical quan¬
sity tender tity
92 Chapter 4/Brightness and Spatial Frequency

image depends on the sensitivity of the film. An

amount of light that would produce a faint image
on insensitive film may produce an overly bright
image on very sensitive film. You are probably
aware of the effects of your eyes’ changing sensi¬
tivity when you walk from a darkened room into
the bright sunlight, only to find that everything ap¬
pears to be so bright and “washed out” that a few
moments must pass before objects are clearly visi¬
ble. The opposite occurs when you walk from a
bright outside into a darkened room. Now every¬
FIGURE 4-1 The nonlinear relationship between thing appears to be very dark, and objects are dif¬
stimulus intensity and brightness. ficult to resolve in the gloom. After a while you
can discern objects, although adaptation to dark¬
ness takes somewhat longer than adaptation to a
case for power functions with larger exponents brighter environment. We call the process of ad¬
(e.g., electric shock), the resemblance of the power aptation to a darker environment dark adaptation
and log laws for brightness means that the old cus¬ and that to a brighter environment light adapta¬
tom of measuring brightness in terms of logarithms tion. Although we cannot slip off our daylight ret¬
of photometric values still provides a relatively ac¬ ina and put on the twilight one in the way that we
curate representation of the relationship between change film in a camera to deal with changes in
the intensity of the physical stimulus and the sen¬ lighting conditions, the sensitivity of our eyes
sation. For this reason various photometric values, changes through these two adaptation processes.
such as the brightness scales used in television stu¬ We can monitor directly the changes in sensi¬
dios, are frequently presented in logarithmic units, tivity associated with dark adaptation. First we
especially when designed for visual purposes. This adapt an observer to bright light by putting him in
serves to equalize the sizes of sensory changes as a a brightly lit room for a few minutes, then we turn
function of changes in physical intensity. The unit off the lights. Now we test to find the observer’s
of brightness in the graph in Figure 4-1 is the bril, absolute threshold for the detection of light. This
which was suggested by S. S. Stevens. Each bril is done at fixed time intervals after the onset of
represents about one-tenth of a log unit above darkness, using one of the standard psychophysical
threshold, in much the same way that a decibel (see methods outlined in Chapter 2. Such an experiment
Chapter 6) represents one-tenth of a log unit above usually reveals that the observer at first needs rela¬
threshold in audition. tively strong stimuli to reach threshold. However,
the eye rapidly becomes more sensitive over the
first minute or two, at which point it begins to sta¬
FACTORS IN BRIGHTNESS bilize at a level that is about 100 times more sen¬
PERCEPTION sitive (2 log units) than when we initially turned
off the lights. After about 10 min of darkness, the
sensitivity begins to rapidly increase again. During
Adaptation
this second period, the threshold drops quickly for
The perception of brightness depends on the current 5 or 10 min, then again stabilizes, reaching a rela¬
state of sensitivity of your eye, in much the same tively constant level after about 1/2 hour. When we
way that the brightness of the final photographic graph the change in threshold for a typical ob-
 Factors in Brightness Perception 93

Time in the dark (min)

A B C

FIGURE 4-2 (A) The normal time course of dark adaptation, (B) dark adaptation in the cones (or central
fovea), and (C) dark adaptation in the rods (or periphery).

server, as we have done in Figure 4-2A, we can hours (or in conditions of dim illumination),
see a break, or kink, in the sensitivity curve. The whereas cones are found predominantly in the reti¬
kink indicates a change in the rate of dark adapta¬ nas of animals that are active during daylight. It
tion. was suggested that cones provide photopic or day¬
Whenever a sudden transition or break is light vision (including the perception of color) and
found in a curve, it often suggests that a second rods provide scotopic or twilight vision. Humans
mechanism or process has come into operation. have rods and cones; hence, it seems possible that
This is confirmed in the present case by the fact the two segments of the dark adaptation curve rep¬
that a marked change occurs in conscious percep¬ resent separate rod and cone contributions. The
tion near this sharp break in the curve. For in¬ cones seem rapidly to reach their level of maximal
stance, if we used a greenish light (or nearly any sensitivity. The rods take longer to adapt. When
color, for that matter) to measure the threshold, the they do, the threshold begins to drop, but at the
observer would be able to identify the color expense of a loss in color vision. The point at
throughout the first 10 min or so of the test session. which the adaptation of the rods catches up to that
At about the point at which the sensitivity suddenly of the cones is the break in the dark-adaptation
begins to increase again, the test stimulus would curve shown in Figure 4-2A.
seem to lose its color and become grayish. An old We can verify this experimentally. Suppose
proverb is based on this loss of color vision under we return to the experimental situation we used to
dim levels of illumination: “At night, all cats are track the course of dark adaptation, but now change
gray.” the stimulus so that a tiny pencil of light is focused
In Chapter 3 we talked about the differences only on the fovea when taking threshold measure¬
between rods and cones. At that time, we reviewed ments. Since the fovea contains only cones (Chap¬
some evidence indicating that rods are found pre¬ ter 3), this method will allow us to track dark
dominantly in animals active during the twilight adaptation in cones. Such an experiment gives us
94 Chapter 4/Brightness and Spatial Frequency

the data shown in Figure 4-2B. Notice that this mer illumination. There is even some evidence
looks just like the first segment of the curve in Fig¬ suggesting that when bright light is present, and the
ure 4-2A. No second increase in sensitivity occurs, cones are active, they actually inhibit or “turn off”
no matter how long we continue in darkness. To the action of the rods (Drum, 1981).
demonstrate the lower or rod portion of the curve There is still much to be learned about the na¬
we repeat the experiment, only now we focus our ture of the adaptation process. Clearly, any incom¬
pencil of light about 20 degrees from the center of ing light will bleach the available photopigments in
the fovea, where the retina contains predominantly the rods and cones, and time will be needed for
rods. When we do this, we get the curve shown in them to regenerate. As more pigment becomes
Figure 4-2C, in which the first rapid change (attri¬ available, the sensitivity of the eye should increase.
butable to cone action) is almost completely absent. Although such a process does seem to play a role
An even more spectacular way to show the (MacLeod, 1978), dark adaptation actually in¬
separate rod and cone origin for the two portions of volves changes in the sensitivity and responsive¬
the dark adaptation curve was provided by Hecht ness of neural processes as well (Green & Powers,
and Mandelbaum (1938). They placed a normal ob¬ 1982; Shapley & Enroth-Cugell, 1984). Later in
server on a vitamin-A-deficient diet for 57 days. this chapter we shall see that even higher-level cog¬
Since this vitamin is critical for the synthesis of nitive responses may influence our perception of
rhodopsin, the pigment in rods, the diet effectively brightness. Demonstration Box 4-1 allows you to
eliminated the action of these receptors. After 57 see the effects of dark adaptation for yourself.
days, the observer had a dark adaptation curve sim¬
ilar to that in Figure 4-2B. Not only was the rod
Retinal Locus
portion of the curve almost totally absent, but the
individual was almost completely night-blind and As we saw in Figure 3-7, rods and cones are un¬
unable to see dimly illuminated targets. By the evenly distributed across the retina. The fovea con¬
way, the observer completely recovered when he tains only cones, which are less sensitive to weak
went back to his normal diet. Perhaps similar nat¬ stimuli, and the more sensitive rods are more plen¬
urally occurring instances have given carrots, a tiful in the periphery. Suppose the apparent bright¬
vegetable high in vitamin A, their reputation for ness of a light depended directly on the sensitivity
being “good for the eyes.” of the stimulated receptors, as well as on the inten¬
Overall, these experiments indicate that two sity of the light. If that were the case, then moving
separate physiological mechanisms are involved in a constant light stimulus across the retina, stimulat¬
the perception of brightness: the cone system for ing less sensitive cones near the fovea and more
brighter illumination and the rod system for dim¬ sensitive rods in the periphery, should change the

DEMONSTRATION BOX 4-1. Dark Adaptation

To show the dramatic increase in sensitivity associ¬ move the blindfold and compare the sensitivity of
ated with dark adaptation you should first carefully your two eyes by alternately opening one eye at a
blindfold one eye. Use a couple of cotton balls and time. The dark-adapted eye should see quite well in
some tape to do this. After about 30 minutes darken the dim illumination, but the other eye will be vir¬
the room, or step into a reasonably dark closet. Re¬ tually blind.
 Factors in Brightness Perception 95

apparent brightness of the light. This has been ver¬ Purkinje

ified experimentally (Drum, 1980; Osaka, 1981). shift in
sensitivity
Peripheral targets appear brighter. This finding is
H-1
also embodied in a bit of folk wisdom. At some
time in antiquity people noted that looking directly
at a dim object, such as a star, could cause it to
disappear from view. For this reason, early astron¬
omers would often look at a point off to the side of
a star in order to let its image fall on the more
sensitive peripheral retina (containing mostly rods).
This technique allows such a dim target to be per¬
ceived more clearly. If you ever try this yourself,
look at a point about 20 deg of visual angle from
the star you wish to see. This would allow the
star’s image to fall on the part of the retina where
the density of rods is greatest and would give you
maximum sensitivity.

Wavelength
The wavelength of the light stimulating the eye will
also affect our perception of its brightness. For in¬
stance, yellow light (medium wavelengths) almost
always appears to be brighter than blue light (short
wavelengths). The usual procedure for assessing
the relative brightness of lights of different colors (Blue) (Red)
is to use a bipartite target. This is simply a circular Wavelength (nm)
target that has been divided in half. One half con¬
tains the standard color that is to be matched in FIGURE 4-3 Differences in relative sensitivity to
various wavelengths under photopic and scotopic
brightness, the other half is the comparison color
illumination conditions.
that is adjustable. Systematically pairing various
colors and then matching their apparent brightness
provides a set of measures of the relative amounts photopic and represents the results we would obtain
of energy needed to produce equal sensations of from the matching experiment if we were working
brightness for various wavelengths of light. For at daylight levels of light intensity. It shows a peak
convenience, the wavelength requiring the least en¬ sensitivity for wavelengths around 555 nm and the
ergy to equal the brightness of the standard is set apparent brightness falls off rapidly for shorter (to¬
at a value of 1.0. All other wavelengths, being less ward the blue) or longer (toward the red) wave¬
effective, are assigned values less than 1, depend¬ lengths. If we repeat this matching experiment
ing on their relative brightnesses. under conditions of dim illumination, in which only
Once this conversion has been made, a curve rod vision is operating, the observer will not be
can be plotted as in Figure 4-3. Such a curve is aware of the color of the stimuli, and both halves
called a luminosity curve. Notice that we actually of the field will appear gray regardless of their
have two curves in this figure. The first is labeled wavelength. Nonetheless, some wavelengths will
96 Chapter 4/Brightness and Spatial Frequency

still look brighter than others. Thus, we can map grays, while daylight reds change to moonlight
out the luminosity curve that is marked scotopic in blacks. Demonstration Box 4-2 allows you to ex¬
Figure 4-3. Under these conditions the curve is perience this shift in sensitivity for yourself.
somewhat different, with a peak around 505 nm. There is an interesting application of the
This curve is shifted toward the short wavelengths, Purkinje shift. You may remember from watching
suggesting that we may be more sensitive to blue war movies that the briefing rooms next to airstrips
light under dim viewing conditions. or the control rooms of ships and submarines are
The change in the apparent brightness of light often depicted as illuminated by red light. This red
of different wavelengths as the intensity is changed is not used solely for dramatic effect in the film but
was first described by the Czechoslovak phenome- is actually used in such settings. Rods are relatively
nologist Johannes E. Purkinje, and in his honor this insensitive to the red end of the spectrum, hence
phenomenon is referred to as the Purkinje shift. red light is virtually equivalent to no light at all for
He first noticed the change while looking at his the rods. However, the cones still function at these
garden as twilight was falling. As the light longer wavelengths if there is sufficient stimulus
dimmed, the apparent brightnesses of the various intensity, so the cones may be used while the rods
colors began to change. Reds that had been bright are beginning to dark adapt. The dashed line at the
relative to blues and greens began to look darker, upper end of the scotopic curve in Figure 4-3
while the bluer tones appeared relatively brighter. shows a wavelength where this is true. Thus, pilots
Because scotopic vision lacks the sensation of about to fly night missions or sailors about to stand
color, daylight greens or blues change to moonlight the night watch can be briefed or can check their

DEMONSTRATION BOX 4-2. The Purkinje Shift

For this demonstration you will need a dark room and deal of light; hence, control of illumination may be
some way of providing a light whose intensity you improved by dimming the light in the outside room.
can vary without altering its color. A good method is Now, look at Color Plate 1. Here we have two
to use a television set as a light source. This may be colored spots, one blue and one red. When viewed in
done by tuning the set to an unused channel and turn¬ moderate or bright light (the brightness control on the
ing the contrast control to a minimum. This reduces television is set to high, or the closet door is more
the visibility of the random dots that normally appear widely ajar), the blue and the red spot appear to be
on the screen. Now, if you darken the room so that approximately equal in brightness. Now, make the
the television is the only source of illumination, the light very dim (close the door almost completely, or
brightness control on the set will be a means of con¬ turn down the brightness control on the television). In
trolling the room light. An alternate procedure in the the bright light, you were viewing the spots with cone
absence of a television is to turn on a light in a room vision. Now, if you dim the lights sufficiently, only
and enter a closet, shutting the door after you. The rod vision will be activated. After 5-10 min, as your
amount of light entering the closet can be controlled eye dark adapts, the blue spot will appear to be sig¬
by opening the door by differing amounts. Turning nificantly brighter than the red spot. In fact, the red
your back to the door allows for a diffusion of the spot may actually disappear. The effect may be ac¬
light to any target that you wish to be illuminated. centuated by staring at the white spot. This shifts the
Unfortunately, if the outside room is well lit, opening images away from the fovea to an area of the retina
the door by a few centimeters will provide a good containing a greater number of rods.
 Factors in Brightness Perception 97

instruments under red illumination, and then go di¬ means that if the duration is greater than one-tenth
rectly into the dark without waiting the many min¬ of a second, the probability that you will detect a
utes necessary to complete dark adaptation. stimulus is no longer affected by stimulus duration
but depends only on stimulus intensity. Bloch’s
law, like many other things in vision, only holds
Time and Area
under certain circumstances. For instance, it holds
In addition to the intensity, wavelength, and retinal better in the periphery (where there are many rods)
location of the stimulus, our ability to detect a spot than in the fovea (Gottlieb, Kietzman & Bemhaus,
of light depends on other stimulus properties. For 1985) and may also depend on the wavelength of
instance, a photographer knows that when she is the stimuli used (Schwartz & Loop, 1984).
taking a picture under dim illumination she may The size of a stimulus is also important in de¬
have to lengthen the exposure time in order to col¬ termining its detectability. In Chapter 3, we noted
lect enough light to adequately register the image that there is a good deal of convergence in the vi¬
on the film. In bright sunlight a short exposure will sual system. A number of rods or cones may syn¬
usually do. Actually the same amount of physical apse with the same bipolar cell, and several bipolar
energy is necessary to expose the film properly in cells may synapse with the same retinal ganglion
each case; it just takes longer to collect the requi¬ cell. Consider a hypothetical example. Suppose
site amount under dim illumination. In physics this four units of neurotransmitter per second are suffi¬
relationship is known as the Bunsen-Roscoe law. cient to activate a bipolar cell, and that bipolar cell
This law describes the photochemical reaction of has four receptors making synapses with it. If we
any light-sensitive substance, whether it be film or provide a tiny spot of light, which is only strong
visual pigment. There is a similar trade-off between enough to elicit one unit of neurotransmitter per
stimulus duration and stimulus intensity in vision second from the retinal receptor, and the light is
when we are dealing with the problem of the ab¬ only wide enough to stimulate two receptors,
solute threshold for the detection of light. We can clearly the bipolar cell will not respond. If we dou¬
express this relationship using simple algebra. If ble the size of the stimulus so that all four receptors
we define C as the critical amount of light energy are illuminated, however, the bipolar cell will re¬
necessary to reach threshold, / as the stimulus in¬ ceive a total of four units of neurotransmitter per
tensity, and T as the stimulus duration, the relation¬ second and it will become activated. Thus, as the
ship is area of a stimulus increases, even if its intensity
does not change, the likelihood increases that we
T x / = C will recruit enough photoreceptors to begin a chain
of neural activity. An alternative way of concep¬
When applied to vision, this is known as Bloch’s tualizing this is in terms of retinal receptive fields,
law. Thus, we must increase the length of time a such as those illustrated in Figure 3-10. Increasing
dim stimulus is presented in order for it to be de¬ the stimulus size might be thought of as simply
tected, whereas a more intense stimulus can be pre¬ “filling in” the center of the receptive field with
sented for a shorter duration and still be detected. light, thus adding more on responses to the overall
This time-versus-intensity trade-off only works activity.
over stimulus durations less than about one-tenth of For relatively small areas, covering visual an¬
a second. This limiting value may vary a bit, being gles of 10 minutes of arc or less (about 1 mm
somewhat longer if you are completely dark- viewed at arm’s length), there is a direct relation¬
adapted and somewhat shorter if you are very light- ship between area and intensity. If A signifies the
adapted (Montellese, Sharpe & Brown, 1979). This area stimulated, and / and C are stimulus intensity
98 Chapter 4/Brightness and Spatial Frequency

and critical amount of light energy, respectively, as size, duration, and retinal position, and the ob¬
before, we can describe the relationship as server were fully dark-adapted. The classic experi¬
ment to answer this question was conducted by
A x / = C Hecht, Schlaer, and Pirenne (1942). They found
that the threshold for the perception of a brightness
This is known as Ricco’s law. Thus, if we increase sensation occurred when only six quanta of light
the intensity of a stimulus we can decrease its size (photons) were stimulating the retina. Further com¬
and still be able to detect it, and vice versa for a putations showed that at this ultimate threshold it
decrease in stimulus intensity. was probable that each of six different rods was
For stimulus sizes greater than 10 min in vi¬ responding to a different one of the six photons.
sual angle, increasing the area has a reduced effect. This is, of course, the maximum sensitivity that is
The effect of area on detection for larger stimuli is possible theoretically. Even at higher levels of il¬
described by lumination, however, it is possible to show
that fluctuations of only a few photons may affect
Va x i = c our perception of brightness, thus showing the
exquisite sensitivity of the eye as a light detector
In other words, for larger stimuli a greater increase (Krauskopf & Reeves, 1980; Zuidema, Gresnight,
in area is needed to achieve the same compensation Bouman & Koenderink, 1978).
for a decrease in stimulus intensity. This second
area-intensity relationship is known as Piper’s law.
Beyond 24 deg of visual angle no further benefit is BRIGHTNESS CONTRAST
gained by increasing the size of the stimulus, and
the likelihood of detection depends solely on the Strange as it may seem, our perception of the
intensity. brightness of targets often depends more on the lu¬
Because these effects are supposedly a result minance of adjacent objects than on the luminance
of the summation of neural responses converging of the target itself. Figure 4-4 demonstrates this.
on a single retinal ganglion cell, a few additional Here we have four small squares, each of which is
relationships might be expected. For instance, the surrounded by a larger square. The central squares
degree of summation (or convergence) varies as we are actually all printed in the same gray; thus, the
move across the retina, with more convergence on amount of light that reaches your eye from each is
a single ganglion cell found in the periphery. Re¬ the same. Notice, however, that the apparent
ceptive fields are smaller in the foveal region brightnesses of these small squares are not equal.
(Randsom-Hogg & Spillman, 1980), which sug¬ Their brightnesses vary depending on their back¬
gests that increasing stimulus area might have a ground, with the grays printed on dark backgrounds
different effect at different positions across the ret¬ appearing lighter than the grays printed on light
ina, with the area increases facilitating detection backgrounds. This effect is called simultaneous
more for stimuli in the periphery, and this seems to brightness contrast.
be the case (Lie, 1980). Everyday experience tells us that your con¬
scious experience of brightness will increase as the
amount of light reaching your eye increases. Un¬
Maximum Sensitivity
fortunately, our perceptual experiences often defy
After this discussion, you may be wondering just such “common sense.” Despite increases in the
what the ultimate limit of sensitivity might be if the amount of light reaching your eye, the apparent
stimulus were adjusted to the optimal wavelength, brightness of a surface may actually decrease de-
 Brightness Contrast 99

1234

FIGURE 4-4 Simultaneous brightness contrast, showing how the background can alter the perception
of the central gray regions.

pending on the brightness of the background on darker. The response from the part of the retina
which it rests. This helps to explain why the color exposed to Square 4 has been reduced. This sug¬
black on a television screen appears darker than the gests some sort of inhibition of the response as a
apparent darkness of the screen when the set is function of activity in surrounding retinal areas.
turned off and no light is being emitted from it at Actually, physiological evidence indicates that
all! This surprising brightness paradox is more such inhibitory spatial interaction does take place
clearly shown in Demonstration Box 4-3. in the eye. Most of this evidence has been collected
The fact that surfaces vary in brightness as a from Limulus (the horseshoe crab), an animal com¬
function of the intensity of their background sug¬ monly found on the eastern shores of the United
gests that there is some form of spatial interaction States. Limulus has several sets of eyes, but the
present, perhaps between adjacent retinal regions ones that are most important for research purposes
that are illuminated by the light reflected from the are the lateral eyes, which are faceted (as in the eye
surfaces. Notice that surfaces on a light back¬ of a fly). In such a compound eye, a separate op¬
ground, such as Square 4 in Figure 4-4, appear tical system exists for each facet, and each has its

DEMONSTRATION BOX 4-3. The Interaction of Luminance and Background

For this experiment you will need your variable light their apparent brightness depend solely on their back¬
source again (either the closet or the television). Hold grounds. This may seem strange because we tend to
up Figure 4-4 and look at the central squares, with associate black with the absence of light. Since you
your light source providing a low (but not dim) level are already in a room or a place that potentially can
of illumination. As you increase the level of illumi¬ be darkened, turn off all the light sources and close
nation from its lowest value, the center target in your eyes (to eliminate any stray illumination). Notice
Square 1 should grow brighter. Now repeat the pro¬ that what you are seeing is not black, but rather a
cedure while looking at the center target in Square 4. misty gray (often called cortical gray). Thus, the ab¬
Notice that as the luminance level increases, this tar¬ sence of light is gray, not black. Only in fields that
get square actually gets darker. Since all the center contain some areas of bright illumination can real
squares are identical in reflectance, the differences in black be seen.
 100 Chapter 4/Brightness and Spatial Frequency

own primitive retina. Since each eyelet has its own onstrate the inhibitory neural interactions between
optic nerve, this arrangement spreads out the neural nearby receptors using a very simple but elegant
fibers somewhat. experiment (Hartline & Ratliff, 1957).
With skill (and a dissecting microscope) it is First, they monitored the responses from the
possible to separate out a single nerve fiber, drape cell in Limulus that is functionally equivalent to a
it over an electrode, and record its electrical ac¬ ganglion cell, called the eccentric cell, while the
tivity. Much of the work on the visual system of receptor attached to it was stimulated. The fact that
Limulus was carried out in the laboratories of the the onset of the light increased the activity of the
Nobel-prizewinner H. K. Hartline and his frequent cell, of course, indicated that its activity was con¬
collaborator Floyd Ratliff. They were able to dem¬ trolled by stimulation of the particular receptor they

Light at A Light at B
I 1
r\ r\ r\ r\ r\ r\ r\r\ nnnnnrinnnn

Receptors

Stimulation at B:
Response observed E/..} Ganglion cell
at ganglion cell No response
in ganglion cell

Light at B Light at A

i I
n r\ r\ r\ r\ r\ n r\

Stimulation at A and B:
Ganglion cell response
slows when B is stimulated

FIGURE 4-5 Lateral inhibition: Stimulating A produces a response in the ganglion cell, whereas
stimulating B does not. Stimulating B while A is active depresses its response because of lateral inhibition
(From Lindsay &. Norman, 19771
 Brightness Contrast 101

had illuminated. Let us call this ganglionlike cell

and its receptor A. Next, Hartline and Ratliff illu¬
minated a receptor located a short distance away
(call this one B). There was no increase in the ac¬
tivity of A, indicating that there were no excitatory
connections between A and B. Now the researchers
again stimulated A and, while the light remained on
at A, they turned on a light at B. Now they ob¬
served that the stimulation of B actually decreased
the response of cell A. This experiment is shown
diagrammatically in Figure 4-5. The importance of
these results is that they demonstrate that visual
cells may be inhibited by the activity of adjacent
visual units. This process is called lateral inhibi¬
tion because the inhibition acts laterally (sideways)
on adjacent cells. The amount of inhibition any
given cell applies to its neighbors depends on how
strongly it is responding and how close the cells are
to each other. The more a cell is stimulated, and
the closer it is to another cell, the more intensely it c 0 O © @000 ©00
will inhibit the other. ' I

Corresponding retinal cells

It is now easy to understand why the surface
in Square 1 is seen to be brighter than the surface
in Square 4 in Figure 4-4. In the part of the retina
exposed to the bright surround (Square 4) many
cells are active and, as a consequence of this activ¬
ity, they are actively inhibiting their neighbors.
This inhibition from the bright surround should re¬
duce the neural response rate in the central square,
making it appear dimmer. The part of the retina
exposed to the surface with the dark background
FIGURE 4-6 (A) A Mach band pattern, (B) the
does not receive as much inhibition from its less
actual distribution of stimulus intensity, (C)
strongly stimulated neighbors. Since the amount of corresponding retinal cells (see the text), and (D)
stimulation from the central squares is the same, neural response intensity distribution. (A and B
but the cells exposed to Square 1 are undergoing a based on Cornsweet, 1970)
lesser amount of inhibition, Square 1 appears to be
brighter. Thus, lateral inhibition provides a basis
for explaining brightness contrast effects. such a distribution generates (Weale, 1979). In this
Lateral inhibition can also explain more com¬ figure, we have a uniform dark and a uniform light
plex effects observed in other stimulus configura¬ area, with an intermediate zone that gradually
tions. Ever since the 1860s, when Ernst Mach changes from dark to light. However, when we
studied patterns with an intensity distribution like look at the actual stimulus depicted in Figure 4-6A,
that shown in Figure 4-6B, investigators have been we do not see a uniform change in brightness
intrigued by a particular brightness phenomenon flanked by two uniform areas. Instead, two bands
102 Chapter 4/Brightoess and Spatial Frequency

DEMONSTRATION BOX 4-4. Mach Band Patterns

Mach band patterns do not reproduce well in print. on the other side. In between there is a graded
This is probably because the range of luminances pos¬ shadow, the penumbra, which gradually moves from
sible from ink on paper is not very large. It is actually light to dark. Hold the card still and look at the
quite easy to produce your own Mach band pattern brightness pattern—you will easily see the dark and
using a distribution of light. All you need is a card or light Mach bands. You may increase the visibility of
a book that is opaque and has a straight edge, and a the bands by moving the card closer to the surface.
large light source. If you are in a room that has flu¬ This reduces the size of the penumbra and makes the
orescent or large frosted light fixtures in the ceiling, area of gradual change in intensity steeper, as shown
these produce a fine uniform source of illumination. in the diagram. Since this puts the bright and dim
When you hold the card near a surface, you cast a areas nearer one another, it enhances the effect of the
shadow. As shown in the accompanying diagram, inhibitory process.
there is a full shadow under the surface and full light
 Cognitive Factors in Brightness Perception 103

or blurry lines are visible at the points marked by COGNITIVE FACTORS IN

the arrows in the figure. One is darker than any BRIGHTNESS PERCEPTION
other part of the figure and the other is brighter.
They are called Mach bands, in honor of their dis¬ The relatively simple manner in which changes in
coverer Ernst Mach. Their presence can be ex¬ brightness can be explained by spatial interactions
plained by lateral inhibition. between retinal cells is quite exciting, especially
We have indicated the location of some retinal given the apparent precision suggested by some rel¬
cells illuminated by the Mach-band-producing pat¬ atively sophisticated mathematical descriptions of
tern in Figure 4-6C. Cell b is stimulated by bright these interactions (e.g., Arend & Goldstein, 1987;
incoming light, but it is also strongly inhibited by Ratliff, 1965). Unfortunately, brightness perception
the activity of the adjacent cells a and c. Cell d is is much more complex, and there are instances
stimulated to the same extent as cell b. But on one where predictions made from lateral inhibitory or
side it is strongly inhibited by c while on the other excitatory considerations can be wrong. As an ex¬
side it is somewhat more weakly inhibited by e, ample, consider Figure 4-7. The gray under the
which is not receiving as much light. The important white stripes is identical to that under the black
thing to derive from this discussion is the fact that stripes. Notice, however, that the gray under the
cells b and d have the same degree of stimulation, white appears to be lighter than the gray under the
but d is less strongly inhibited. In this case, we black. This is the opposite of the prediction we
might expect that its corresponding response would would make based on the action of lateral inhibi¬
be more vigorous than that in cells like b. This tion, that the white stripes should darken the gray
should cause the region around d to appear rela¬ rather than lighten it. The phenomenal impression,
tively brighter. Next consider cell i. It is not stim¬ then, is the reverse of brightness contrast, and it is
ulated very much, but neither are the nearby cells called brightness assimilation (Shapley & Reid,
h and j. This means that i is not being strongly 1985). This effect seems to depend on a cognitive
inhibited by surrounding units. Cell g is receiving factor. The term cognition is used to cover all men¬
the same small amount of stimulation as i. How¬ tal processes by which we come to know the world.
ever, while g is weakly inhibited on one side by h, Thus, processes such as learning, reasoning, intui¬
it is more strongly inhibited on the other side by /, tion, or attention are all cognitive factors, as we
which is responding more vigorously because of noted in Chapter 1.
the higher intensity of light falling on it. Thus, al¬ Attention seems to be a relevant variable for
though g and i receive the same amount of stimu¬ the appearance of brightness assimilation. The part
lation, g is more strongly inhibited than i. This of the visual field you are attending to shows
means that its response will be decreased, causing greater brightness contrast (Brussell & Festinger,
an apparently darker region to appear there. The 1973; Coren, 1969), whereas regions you are not
relationship between the input and the neural (and attending to show brightness assimilation (Festin¬
perceptual) response is diagramed in Figure 4-6D. ger, Coren & Rivers, 1970). Observers usually de¬
It is quite easy to produce a Mach band pattern for scribe the pattern shown in Figure 4-7A as a gray
yourself, as shown in Demonstration Box 4-4. It field with white lines on it and Figure 4-7B as a set
seems likely that a very wide array of brightness of black lines on a gray background. Festinger et
perception phenomena can be explained by theories al. reasoned that the lines have a “figurelike”
that assume particular patterns of inhibitory and ex¬ quality that captures the attention (this aspect of
citatory interactions between sensory neurons (see perception is discussed more fully in Chapter 11).
e.g., Comsweet, 1985; Grossberg, 1987). Since the gray is then a nonfigural background to
104 Chapter 4/Brightness and Spatial Frequency

FIGURE 4-7 Brightness assimilation, where the gray under the white stripes appears lighter than the
gray under the black.

which we pay little attention, it shows assimilation. nitive factors, namely, the assumptions the ob¬
If this is the case then voluntary shifts in attention, server makes about the nature of the world, or even
so that it is focused on the gray regions, should the way in which regions of the visual field appear
alter the brightness effect from one of assimilation to be arranged (Flock & Nusinowitz, 1984; Gil¬
to one of contrast: the gray under the white stripes christ, Delman & Jacobsen, 1983). We have more
should now appear to be the darker member of the to say about this in Chapter 14 where we discuss
pair. This is exactly what happens (Festinger, the issue of brightness constancy.
Coren & Rivers, 1970). You can demonstrate this
for yourself by focusing your attention on the gray
for a few moments, and soon the grays will appear VISUAL ACUITY
to differ in a contrast direction, rather than showing
brightness assimilation. Visual acuity refers to the ability of the eye to re¬
Another attentional mechanism may account solve details. There are different types of visual
for the fact that we fail to notice recurrent periods acuity, each dependent on the specific task or spe¬
of total “blackout” in the visual system. These pe¬ cific detail to be resolved. The type of visual acuity
riods occur when we blink, which we do about 10 most commonly measured is recognition acuity,
to 15 times a minute. Each blink causes a total which was introduced by Herman Snellen (1862).
blackout for 100 to 150 msec, yet it usually goes This task uses the familiar eye chart found in most
unnoticed. When the observer consciously attends ophthalmologists’ or optometrists’ offices, com¬
to the presence of blinks, some darkening can be posed of rows of letters of progressively smaller
seen, but direct measurements show that observers size. The observer is asked simply to identify the
underestimate the duration of the blackout by 90 letters on the chart and the size of the smallest
percent and also underestimate the amount of actual identifiable letters determines acuity. Acuity is usu¬
darkening by about 75 percent (Manning, Riggs & ally measured relative to the performance of a nor¬
Komenda, 1983; Volkman, Riggs, Aimee & mal observer. Thus, an acuity of 6/6 indicates that
Moore, 1982). This suppression is probably a cog¬ an observer is able to identify letters at a distance
nitive adjustment to maintain continuity in the con¬ of 6 meters that a normal observer can also read at
scious flow of perception. that distance (you may be more familiar with the
Some brightness effects depend on other cog¬ designation 20/20; 6 m is equivalent to 20 ft). In
 Visual Acuity 105

other words, the measured acuity is normal. An visual acuity. The most primitive measure is simply
acuity of 6/9 (or 20/30) would mean that an ob¬ the specification of the smallest target of any type
server is able to read letters at 6 m that are large that can be detected. The relationship between
enough for a normal observer to read at a distance brightness perception and acuity is most apparent
of 9 m. Here, the visual acuity is less than normal. for this task, where the target is a light line or spot
A more general means of specifying the limits against a dark background, or a dark line or spot
of acuity is to use the minimum visual angle of a against a light background. Vernier or directional
detail that can be resolved. The visual angle is a acuity requires an observer to distinguish a broken
measure of the size of the retinal image. Figure 4-8 line from an unbroken line. Resolution or grating
shows what is meant by visual angle and demon¬ acuity is measured by an observer’s ability to de¬
strates a simple computation based on the size and tect a gap between two bars, or the orientation of a
the distance of the object. Generally speaking, a grid of lines. This particular form of acuity task has
normal observer can resolve details of 1 minute of certain theoretical implications, which we discuss
arc (about the size of a quarter seen at a distance in the next section. Figure 4-9 shows examples of
of 81 m, which is nearly the length of a football the above-mentioned acuity targets with arrows
field), although different tasks often produce differ¬ pointing to the crucial detail. Notice that each de¬
ent limits of acuity (Beck & Schwartz, 1979). tail is merely a region of the field where there is a
The identification of letters on a Snellen chart change in luminance.
is not the best way to measure acuity, since letters It is reasonable to expect that the minimum
differ in their degree of identifiability. For instance, resolvable detail size would be determined by the
O and Q, or P and F, are letter pairs that are easily size of the retinal receptors or the size of the retinal
confused, where as L and W or O and I are quite receptive fields. Thus, in order to determine
easy to discriminate. Because these differences whether one or two spots of light are present, we
might affect acuity measurements, Hans Landolt might assume that it would be necessary to have at
(1889) introduced a different task that used circles least one unstimulated retinal receptor (or receptive
with a gap in them as targets (see Figure 4-9). The field) between two light-stimulated retinal receptors
gap can be oriented either up, down, to the right, (or receptive fields). Surprisingly, for tasks such as
or to the left, and the observer’s task is to indicate vernier acuity, people can resolve much finer de¬
the position of the gap. The circles differ in size tails than we would predict on the basis of these
and the smallest detectable gap is the measure of considerations. Acuities of 5 seconds (a second of
acuity. arc is 1/3600 of a degree) or less are possible, de¬
A variety of other tasks is used to measure spite the fact that the smallest receptive fields are

FIGURE 4-8 Computation of the size of the visual angle of the image of a quarter viewed at a distance
of 70 cm (approximately arm’s length), where the observer’s line of sight is perpendicular to the lower
edge of the coin. Tangent of visual angle = size/distance, therefore tan a = S/D = 2.4/70 = 0.034. Thus,
a is approximately 2 deg.
106 Chapter 4/Brightness and Spatial Frequency

Vernier Landolt Resolution Grating Snellen

acuity C acuity acuity letter

FIGURE 4-9 Some typical acuity targets and the details to be discriminated.

around 25 times larger than this (Klein & Levi, by increasing the amount of time the observer
1985; Westheimer, 1979). In fact this is about one- views the stimulus. Although Bloch’s law only
sixth the diameter of the smallest retinal cones. holds for times less than 100 msec for simple light
Resolution of details less than about 10 sec is often detection, the trade-off between time and intensity
referred to as hyperacuity and suggests that some holds for up to 300 msec in acuity tasks in which
fairly complex neural circuitry, and pooling of observers are trying to detect pattern details
neural responses, may be involved in acuity tasks (Kahneman, 1966; Kahneman, Norman & Kubovy,
(Carlson, 1983; Wilson, 1986). 1967).
Because acuity tasks are closely related to Retinal position is also as important for acuity
brightness discrimination, it is not surprising to find as it is for brightness perception (Jennings & Char-
that acuity varies as a function of the many factors man, 1981). The figure in Demonstration Box 4-5
shown to be important in the perception of bright¬ allows you to experience the drastic reductions in
ness. For instance, the adaptive state of the eye de¬ visual acuity for targets that are imaged some dis¬
termines the minimum details that can be tance from the fovea. When we measure relative
discriminated under particular viewing conditions acuity for various locations on the retina, we find
(Lie, 1980). Thus, if you step out of the bright sun¬ that it varies as shown in Figure 4-10. Notice that
light into a dim room, you may find it impossible acuity is best in the central fovea and drops off
to read even the large type of the headlines of a rapidly as we move into the periphery. This curve
newspaper for a few moments. As your eyes adapt looks remarkably like the distribution of cones
to the dim surroundings, however, you can soon across the retina diagramed in Figure 3-7. It also
easily read even fine print. Even a brief flash of looks much like the distribution of X cells in the
light, bright enough to alter an observer’s state of retina (Peichl & Wassle, 1979). Direct physiologi¬
adaptation, markedly reduces an observer’s ability cal measurements of the responsiveness of X and
to detect and identify acuity targets (Miller, 1965). Y cells shows that X cells have smaller receptive
The detection of details in acuity targets also fields and seem to respond better to small stimuli.
shows an interaction between time and stimulus in¬ This has led a number of researchers to suggest that
tensity, very much like that described by Bloch’s the limits of visual acuity are set by the prevalence
law for light detection. This means that you can of X cells, which are best designed for detection
increase the likelihood that a detail will be detected and analysis of small details in stationary visual ar¬
by either increasing the difference between the in¬ rays (Andrews & Pollen, 1979; Robson, 1980).
tensity of the target and that of its background, or Because the part of the retina that is highest in
 Visual Acuity 107

DEMONSTRATION BOX 4-5. Visual Acuity as a Function of Retinal Location

Visual acuity is best in the fovea. The range of clear moving your right eye, you will note that the letter
vision extends less than 10 deg away from the foveal over the 0° mark is relatively clear, and that the letter
center. Lay this book flat on the table and view the at 5° is also legible. However, the letters at 10° and
accompanying diagram from a distance of approxi¬ beyond begin to appear fuzzy, and the letters at 40°
mately 12 cm. Cover your left eye with your left hand and 50° are virtually unreadable.
and look directly at the point marked 0°. Without

K B X M P A S
+ + + + + + +
50° 40° 30° 20° 10° 5° 0°

visual acuity contains mostly cones, which then low, in the scotopic (rod) range, acuity is poor, and
send their signals to X-type ganglion cells, we can it improves only slightly as the light intensity is
predict some further interactions between bright¬ increased. However, as we begin to shift into the
ness and acuity. Since cones are only operative at photopic (cone) range, acuity improves rapidly. Of
higher levels of illumination we would expect bet¬ course, at too high a light level the acuity is re¬
ter acuity at higher illumination levels. When we duced again because of the effects of glare (not
measure the relationship between acuity and illu¬ shown). Demonstration Box 4-6 provides a stimu¬
mination directly we obtain the curve shown in lus figure and instructions for demonstrating the re¬
Figure 4-11. Notice that when the illumination is lationship between acuity and illumination.

Nasal ^_ Foyea __ Temporal

retina retina
Degrees from fovea

FIGURE 4-10 The distribution of visual acuity FIGURE 4-11 The effect of illumination on visual
across the retina. acuity.
108 Chapter 4/Brightness and Spatial Frequency

DEMONSTRATION BOX 4-6. Acuity and Intensity

You will need your variable intensity light source crease the intensity of the light source and notice that
again (either television or closet). Under very dim il¬ you can now begin to resolve lines at a higher point
lumination, view the bundle of converging lines in the bundle. In fact, under normal room illumina¬
shown in the diagram. Notice that at some point, tion, the lines in the area where acuity began to fail
where the lines are relatively close together, you can under lower illumination may now be discriminable.
no longer resolve individual lines and the bundle ap¬ Notice also that the apparent blackness of the lines
pears to be gray. This is the limit of your grating and whiteness of the paper (in other words the psy¬
acuity under these conditions. Place your finger or the chological contrast) also improves under higher light
point of a pencil at the point where the individual conditions.
lines are no longer resolvable. Now gradually in¬

One aspect of the relationship between acuity retinal receptors per eye, each receiving an amount
and illumination has important implications for of light ranging from zero up to many millions of
some common situations. In 1789 Lord Maskelyne, units. Pity the poor perceptual researcher who must
Director of the Royal Greenwich Observatory, no¬ now find a method of describing all of this activity
ticed that he became noticeably nearsighted at (not to mention the poor brain that must interpret
night. This common tendency to accommodate the it). If we had to catalog every point of light and its
eye inappropriately near, even when the object of intensity before we could understand the major
interest is far away, is called night myopia (Lei- phenomena associated with brightness and acuity,
bowitz. Post, Brandt & Dichgans, 1982). Practi¬ our information about these topics would be limited
cally, it degrades the sharpness of the retinal indeed. Many researchers realized this and began
image, interfering with the ability to see details un¬ to look for some reasonably small set of relation¬
der twilight and nighttime observation conditions. ships among the variables that affect brightness
This reduction of acuity may be an important com¬ perception that could be used to describe visual ar¬
ponent in many nighttime driving accidents (Lei- rays, hoping that such a simplified description
bowitz & Owens, 1977). might yield deeper insight into these phenomena.
In some ways, the most successful attempt to
summarize brightness and acuity data to date has
SPATIAL FREQUENCY ANALYSIS
involved the use of a mathematical technique based
on Fourier’s theorem. This theorem states that it
Spatial Fourier Analysis
is possible to analyze any periodic pattern into a
A complete description of the relationship between series of sine waves. In Chapter 6 we apply this
brightness perception and acuity must take into ac¬ theorem to complex sound waves, in which sound
count a great deal of information. Imagine any test pressure level at some point in space varies over
pattern of light. Next, realize that when this pattern time in an irregular but repeated pattern. For our
stimulates the eye there are 125 million or more current problem we are concerned with how light
 Spatial Frequency Analysis 109

intensity varies across space, namely across the ret¬ to light distributions, Fourier’s theorem states that
inal image. According to Fourier’s theorem we can by adding together (synthesizing) a number of such
analyze any such complex pattern of light intensity gratings we can produce any specified light distri¬
across space into a series of simpler sine wave pat¬ bution. Moreover, although individual sine wave
terns, each of which would be seen as a regularly patterns have only gradual changes in intensity, by
varying pattern of light and dark, if seen alone. adding many of them together we can even produce
You might recall from trigonometry that a sine light distributions that contain sharp comers, such
wave is simply a regular, smooth, periodically re¬ as that shown in Figure 4-12B. This pattern is
peating function that can be precisely specified called a square wave grating, since the light
mathematically. Figure 4-12A shows a graph of a changes are sharp and give a boxlike intensity pat¬
sine wave, and beside it a distribution of light that tern. Successive addition of the appropriate fre¬
varies in the same way, growing more intense quencies of sine waves (or more accurately, the
where the function rises and less intense where it sine wave gratings they represent) gradually gives
falls. The pattern in 4-12A is called a sine wave a better and better approximation of the sharp cor¬
grating, because the intensity of reflected light ners of the square wave grating. Figure 4-13 shows
from the page varies sinusoidally as we move hor¬ this graphically.
izontally across the figure and the whole pattern If we take this approach to describing the pat¬
forms a sort of blurry grating or grid. As applied terns of light that act as stimuli to our visual

FIGURE 4-12 (A) A sine wave spatial distribution of light, and (B) a square wave spatial distribution of
light. (Based on Cornsweet, 1970).
110 Chapter 4/Brightness and Spatial Frequency

Resultant distribution
Sine wave • through addition

A+B

c/WWWWXA

d/wwv\/\/wv\y\/\/N

FIGURE 4-13 Gradually adding higher-frequency sine waves to the distribution leads to better
approximations of a square wave through the process of Fourier synthesis.

systems, we no longer have to catalog the intensity to use a series of gratings, such as those shown in
of every point in the pattern. Now we can describe Figure 4-12. Some of the gratings will have very
a light pattern precisely with a relatively compact broad bars and spaces. In such gratings the light
mathematical expression indicating the particular intensity rises and falls slowly as we move across
sine wave gratings to be added together to repro¬ the spatial extent of a surface, hence they are said
duce it. Even if the mathematics sometimes be¬ to have low spatial frequencies (the frequency of
come complex, the resulting description is still far changes in light intensity across space is low).
simpler than a catalog of the light hitting 125 mil¬ Other gratings will have narrow bars and spaces. In
lion or more individual retinal receptors. (See Lev¬ these gratings the light intensity changes rapidly (at
ine & Shefner, 1981, or Weisstein, 1980, for a a high rate) as we move across space, hence they
more complete introduction to Fourier analysis.) are said to have high spatial frequencies. Now we
photograph each grating to see how well it is repro¬
duced. At some point, when the bars and spaces
Modulation Transfer Function
become quite narrow, the system will reach its
Fourier analysis (breaking up a pattern into its limit. The lens will no longer be able to resolve the
component sine waves) and Fourier synthesis (add¬ individual bars, and all of the bars and spaces will
ing together a set of sine wave variations to create merge into a gray blur. This is exactly the same
a more complex pattern) provide more than a sim¬ type of task we would use to measure the resolu¬
ple shorthand for the description of light patterns. tion acuity or grating acuity of human observers,
They serve as powerful tools for analyzing how the except that here we are looking at the resolution
visual system responds to stimuli. Consider for a acuity of an optical system.
moment how we might test the fidelity of a photo¬ When photographic engineers do this type of
graphic system. The simplest way of doing this is analysis for an optical system, they measure its res-
 Spatial Frequency Analysis 111

olution in terms of the maximum number of lines spaces), the perceived contrast (the apparent differ¬
per inch that can be resolved. Since very finely ence between light and dark regions) is much less
packed arrays of lines, corresponding to high spa¬ for the higher frequency with the black looking a
tial frequencies, cannot be resolved and are simply bit lighter and the white a bit darker in C than in
blurred, we say that optical systems attenuate the B. You can increase this difference by propping the
high-frequency components of the pattern. A book up and stepping back a foot or two. In a con¬
graphic or mathematical description of the way cer¬ trast matching task, observers would be asked to
tain spatial frequencies are lost because the system match the apparent contrast of such targets (or
can not resolve them, whereas others are retained more usually sine wave gratings) by adjusting the
because they are within the system’s resolution ca¬ intensities of the light and the dark regions until the
pacity, is called the spatial modulation transfer two patterns matched. In this way, we could map
function. It measures the system’s ability to the differences in visibility of various spatial fre¬
“transfer” to the final image the spatial modulation quencies.
(or intensity change over space) present in the tar¬ An alternative method of measuring sensitivity
get stimulus. to various spatial frequencies involves measuring
To assess the human visual system’s limita¬ the contrast threshold. This is the amount of con¬
tions in resolving changes in light intensity over trast needed for you to detect that there is a grating
space, contrast matching is used to measure the present, rather than a uniform gray. Either of these
modulation transfer function. Consider gratings A techniques will give us a representation of how
and B in Figure 4-14. Although both are square sensitivity changes as we change the spatial fre¬
wave gratings, they differ in terms of their physical quency of the stimuli, which is the modulation
contrast, the ratio of the reflectances of the light transfer function where the stimulus modulations
and the dark areas; grating A has a lower contrast (or intensity changes in the environment) are being
than grating B. Now consider the difference be¬ transferred to (detected in) the observer’s conscious
tween B and C. Both are square wave gratings but experience of the pattern.
B has a lower spatial frequency than C. Despite the When we measure a typical modulation trans¬
fact that the physical contrast is the same (both are fer function for a human observer, it looks like the
the same black ink with the same white inter¬ solid line shown in Figure 4-15. As the graph

A B C
FIGURE 4-14 The effect of spatial frequency on the apparent brightness and contrast of patterns.
Notice that the higher frequency pattern has less apparent contrast.
112 Chapter 4/Brightness and Spatial Frequency

convenient basis for predicting apparent bright¬

nesses in many types of stimulus configurations and
the spatial frequencies we can detect serve as a
measure of our visual acuity. Thus, the modulation
transfer function serves as a sort of a summary of
our visual resolution ability and responsiveness to
light. For example, there are changes in the mod¬
ulation transfer function as we grow older, with a
general reduction in sensitivity to frequencies
higher than 4 cycles/deg. These changes can accu¬
FIGURE 4-15 The modulation transfer function, rately predict reductions in visual acuity and certain
which shows the relative visibility of targets of aspects of our depth perception (stereopsis) mea¬
various spatial frequencies. The solid line
sured by other techniques (Greene & Madden,
represents the normal transfer function; the
dotted line represents the transfer function after 1987).
selective adaptation to a 6 cycle/deg stimulus.

Neural Spatial Frequency Channels

shows, human observers are quite sensitive to spa¬ Imagine an extremely self-assured scientist sitting
tial frequencies around 6 cycles per degree (6 cy¬ at his personal computer, complete with all the pro¬
cles of the sine wave over each degree of visual grams necessary to do Fourier analyses of any light
angle). Sensitivity decreases rapidly the higher the patterns that might happen to be of importance or
spatial frequency. This loss at higher frequencies is interest, muttering to himself, “If I find Fourier
probably a result of the fact that the eye is an op¬ analysis so useful in analyzing patterns of light,
tical system, containing a lens, and any such sys¬ maybe the visual system does too. Perhaps the vi¬
tem has a high-frequency cutoff. Notice also that sual system is set up to conduct some sort of spatial
there is some loss of resolution in the lower spatial frequency analysis for any given pattern of light.
frequencies (less than 6 cycles/deg). This loss re¬ Certainly, if it did, it would benefit from the same
sults from the fact that as the bars and spaces be¬ sort of concise description of the incoming light
come wider, the lateral inhibitory interactions that pattern that I obtain, and could thus also avoid hav¬
sharpen the contours and increase apparent bright¬ ing to deal individually with the millions of re¬
ness differences between adjacent regions become sponses of millions of photoreceptors.”
less effective. In much the same way, Demonstra¬ Actually, this suggestion is not as strange as it
tion Box 4-4 showed that Mach bands were accen¬ might seem. At a general level, the first stage of
tuated when the intensity changes were near one spatial frequency analysis can be accomplished by
another. Thus, intensity changes are most effective mechanisms that we know exist and have already
in producing the phenomenal impression of a discussed—the circularly organized retinal recep¬
brightness difference when these changes occur at tive fields described in Chapter 3. Recall that each
intermediate spatial frequencies (as determined by of these has an excitatory, or on, region that when
the Fourier analysis). When intensity changes occur stimulated by light causes an increase in neural re¬
too frequently within the visual image they are dif¬ sponse rate, and an inhibitory, or off, region that
ficult to resolve. Similarly, when the physical causes a decrease in the neural response rate when
changes are too infrequent, there is no perception stimulated by light (and a burst of responses on the
of brightness differences. light’s termination). Before we discuss how such
The modulation transfer function provides a an arrangement can do a spatial frequency analysis,
 Spatial Frequency Analysis 113

we must first introduce a bit of terminology. Every responses tend to cancel each other out. Thus, the
cycle of a sine wave grating has both a dark and a total response of the ganglion cell with this recep¬
light phase, as we saw in Figure 4-12. This means tive field is low. Now consider the other extreme,
that the dark stripe (or the light stripe) is one-half where the spatial frequency is very high, and there
of the sine wave cycle. Now, every circular recep¬ are many stripes falling across the field. The on and
tive field is “tuned” to a sine wave frequency off regions of the field are each stimulated by about
whose half cycle is equal to the size of its central equal proportions of light and dark, again produc¬
excitatory or inhibitory region. To visualize this ing little or no net response. Finally, consider a
type of structure, consider Figure 4-16. spatial frequency in which the half cycle width is
Suppose we have an on-center receptive field approximately the same as the central region of the
of the size illustrated in the figure. If the spatial receptive field. If the bright stripe now covers the
frequency is too low, that is to say the stripes are central region of the on center cell, there will be a
too wide, the stripes of illumination will fall on vigorous on response. There will be little inhibition
both the center and the surround. Even though the from the surrounding off region, which lies mostly
central on region of the field is stimulated, there is in darkness from the dark half of the cycle. Thus,
an equal degree of stimulation of the inhibitory sur¬ the net response to this grating would be relatively
rounding off region of the receptive field. Because stronger than to any other size grating. Notice that
of lateral inhibitory interactions the two types of the same sort of analysis of spatial frequency can

Spatial Tuned or Spatial

frequency optimal frequency
too low frequency too high

On-center
receptive
field

Off-center
receptive
field

Little or no High response Little or no

response response

FIGURE 4-16 A demonstration of how a circular receptive field organization of a particular size can
perform a crude spatial frequency analysis.
114 Chapter 4/Brightness and Spatial Frequency

occur in the off-center cell, only here the optimal occurs in the visual system, it at least suggests that
response is obtained when the dark half of the cycle the equipment to perform such an analysis does
is over the central region of the receptive field. exist.
Each receptive field is maximally responsive to a
specific spatial frequency of light intensity changes.
Selective Adaptation
This crude analysis of spatial frequencies
could serve as the first step of a Fourier analysis of Many of the findings supporting the idea of spatial
an incoming stimulus pattern if a few additional re¬ frequency channels in the visual system arise from
quirements were met. For example, there must be selective adaptation studies. (The use of the word
a broad range of receptive field sizes so that “tun¬ adaptation here involves the concept of neural sa¬
ing” would be fine enough to approximately deter¬ tiation, or fatigue, which makes it quite different
mine sufficiently many of the spatial frequencies from the dark or light adaptation discussed earlier
that make up the pattern. This requirement seems in the chapter.) In such a study, an observer is ex¬
to be easily fulfilled since, as we noted in Chapter posed to a specific spatial frequency for a moder¬
3, X and Y cells differ quite a bit in the speed and ately long time (perhaps several minutes). If there
nature of their responses. They also differ in their is a specific group of neurons tuned to that partic¬
ranges of receptive field size, with X cells tuned ular frequency, they will, of course, immediately
for higher spatial frequencies than Y cells. Thus, start responding when their optimal stimulus ap¬
there may be a number of different channels in the pears. If the stimulus remains in view for a long
visual system, each tuned to a different range of period of time, these neurons will continue to re¬
spatial frequencies and each composed of cells with spond but at an ever-decreasing rate, until eventu¬
a different receptive field size. There is some evi¬ ally they respond only weakly. Since this weak
dence that the Y-like lower spatial frequency chan¬ response might last for a minute or two after ex¬
nels interact with, and can inhibit, the X-like posure to the adapting stimulus, we have then tem¬
higher spatial frequency channels (Hughes, 1986; porarily disabled a particular group of spatial
Olzak, 1986). Moreover, a model assuming only frequency channels and this should have detectable
six such channels (or receptive field sizes) can ex¬ perceptual effects.
plain some of the remarkable feats of acuity people As an example of how this technique works,
are capable of, namely the hyperacuity we dis¬ suppose we begin by measuring the modulation
cussed above, which shows resolution abilities bet¬ transfer function of an observer, just as we did to
ter than would be predicted on the basis of the produce the solid line in Figure 4-15. Then, we
physical size of the retinal receptors (Bradley & have an observer stare for a while at a grating of
Skottun, 1987; Wilson, 1986). about 6 cycles/deg (the adapting stimulus), in order
Of course, for such Fourier analysis to be of to adapt the spatial frequency channels associated
value perceptually there must be higher-level cells, with this middle range of frequencies. When we
perhaps in the visual cortex, that preserve the spa¬ next measure the observer’s transfer function we
tial frequency information extracted by the tuned now get the results shown as the dotted line in Fig¬
receptive fields of the retinal ganglion cells. There ure 4-15. Notice that there is a depression in sen¬
is evidence that such cells exist in the cortex. These sitivity around the adapted spatial frequency. Put
cells have not only preferred edge orientations to simply, this means that it is now harder to detect
which they respond maximally but also preferred gratings in this range of spatial frequencies, and
ranges of spatial frequency (Derrington & Fuchs, larger amounts of physical contrast are needed to
1981; DeValois, Albrecht & Thorell, 1982; De- produce the same perceptual effects. This is what
Valois & DeValois, 1987). Although the existence we would expect if the channels tuned to the adapt¬
of such cells does not prove that Fourier analysis ing stimulus frequency have been fatigued, and
 Spatial Frequency Analysis 115

hence no longer are responding as effectively. Of quency channels. With only the high frequency
course, if we used a different adapting stimulus, channels operating, they would be the major deter¬
the region of reduced sensitivity would be differ¬ minant of our responses to any stimuli, since the
ent, depending on its spatial frequency (Graham, other channels are responding only weakly. Since
1980; Harris, 1980). the action of these higher-frequency channels usu¬
There is one particularly interesting perceptual ally signals the presence of higher spatial frequen¬
effect that can be produced using this technique. cies, we might expect that the pattern would appear
Remember that spatial frequency roughly corre¬ to be dominated by high-frequency (smaller) ele¬
sponds to the size of elements in a pattern. Thus, ments compared with a situation where all channels
low spatial frequencies correspond to large ele¬ were operating normally. Demonstration Box 4-7
ments, or in our gratings to wide stripes, and high allows you to demonstrate this effect for yourself.
frequencies correspond to small elements. Suppose Although spatial frequency analysis seems to
we had somehow disabled all of the low spatial fre¬ provide a useful approach to the problems of

DEMONSTRATION BOX 4-7. Selective Adaptation of Spatial Frequency Channels

If you look at the figure, you will see that one of the quencies from the lower part of your visual field are
squares on the left has broad bars (low spatial fre¬ fatiguing, or adapting. Now if you transfer your gaze
quency) and the other has narrow bars (high spatial quickly to the dot between the identical gratings on
frequency). The pattern on the right contains two the right, you will notice that they no longer seem to
gratings, both of which have the same spatial fre¬ be the same. The top part of the grating now appears
quencies, but they are neither as high nor as low as to be spaced more finely, with thinner stripes than
the ones on the left. Hold the illustration about 80 cm those on the bottom. The low frequency channels
away from you. Now look at the horizontal bar be¬ have been disabled in the upper region of the visual
tween the upper and lower patterns on the left for field. With more high spatial frequency channels ac¬
about 20 to 30 sec. Move your gaze from one portion tive, the percept is shifted toward higher frequencies;
of the bar to another, but keep your eyes on the bar. hence, the stripes are seen as smaller and more dense.
As you look steadily at the bar, the channels tuned to The opposite effect is occurring in the lower region
low spatial frequencies from the upper part of your of the field.
visual field and these tuned to the high spatial fre-
116 Chapter 4/Brightness and Spatial Frequency

brightness and acuity, it does not provide us with Lateral inhibition The process of adjacent neurons
complete answers, any more than does lateral in¬ inhibiting one another.
hibition. The final perceptual response that emerges Light adaptation The progressive decrease in visual
sensitivity after a change from a lower to a higher level
in consciousness, what we perceive as a particular
of illumination.
brightness, size, or detail, involves the operation of
Lumen The unit of radiance equal to the light ema¬
all levels of the perceptual system. In later chapters nating from a standard candle, which is slightly more
(e.g., Chapter 11) we shall see how spatial fre¬ than 0.001 watt at a wavelength of 555 nm.
quency analysis is useful in understanding some as¬ Luminance The amount of light reflected from a
pects of form perception, and how cognitive factors surface.
interact with the basic sensory mechanisms we Luminosity curve A plot of the relative brightnesses
have discussed so far to determine our perception of light of different wavelengths.
of complex stimuli in our environment. Mach bands The perception of dark and light lines
at regions near abrupt changes in an intensity gradient.
Neural satiation A process in which specific groups
of neurons fatigue in response to optimal and continuous
GLOSSARY stimulation. The presumed cause of selective adaptation.
Photometry The measurement of light.
The following definitions are specific to this book.
Photopic A term for high-light (daylight) visibility
Bloch’s law The trade-off relationship between stim¬ conditions and vision under these conditions.
ulus duration and stimulus intensity in their effect on ab¬ Piper’s law The trade-off relationship between area
solute threshold: T X 7 = C. and intensity in the detection of stimuli between 10 min
Brightness The phenomenal impression of the inten¬ and 24 deg of visual angle in size: s/A x I = C.
sity of a light stimulus. Purkinje shift The change in the apparent bright¬
Brightness assimilation The reverse of simultaneous nesses of light of different wavelengths as an observer
brightness contrast. Here, added light elements lighten a goes from a light-adapted to a dark-adapted state.
stimulus and added dark elements darken it. Radiance The amount of energy emitted by a light
Bril A unit for measuring the apparent brightness of source.
stimuli. Recognition acuity A type of visual acuity com¬
Bunsen-Roscoe law The physical law describing the monly measured by means of letter identification and
photochemical reaction of any light-sensitive substance scaled relative to a norm of identification at 6 m distance
as a function of the intensity and duration of light expo¬ from the observer (6/6).
sure. Reflectance The proportion of incident light that a
CIE The Commission Internationale de l’Eclairage, surface reflects.
an international organization responsible for light mea¬ Resolution acuity The observer’s ability to detect a
surement. gap between two lines, or the orientation of a grid of
Dark adaptation The progressive increase in visual lines.
sensitivity after a change from a higher to a lower level Retinal illuminance The amount of light reaching
of illumination. the retina.
Directional acuity See Vernier acuity. Ricco’s law The trade-off relationship between area
Fourier’s theorem States that any periodic wave can and intensity in the detection of stimuli smaller than 10
be mathematically analyzed into a series of simple sine min of visual angle in size: A x / = C.
or cosine waves. Scotopic A term for low-light (night) visibility con¬
Grating acuity See Resolution acuity. ditions and vision under these conditions.
Hyperacuity Resolution of details that are smaller Selective adaptation Weakening the response of
than the diameter of one retinal receptor; usually, any particular spatial frequency channels by exposing the ob¬
acuity less than 10 sec of visual angle. server to those spatial frequencies for several minutes.
Illuminance The amount of light falling on a sur¬ See Neural satiation.
face. Simultaneous brightness contrast Perceptual phe-

/’
 Glossary 117

nomenon in which a target area of a given luminance Square wave grating Sharply alternating light and
appears brighter when surrounded by a darker back¬ dark stripes.
ground than when surrounded by a lighter background. Standard units Internationally agreed-on measures
SI system The Systeme International d’Unites, a of photic energy.
uniform system of measurement. Vernier acuity The measure of an individual’s abil¬
Sine wave grating A pattern of light intensity that ity to distinguish a broken line from an unbroken line.
varies from light to dark following sinusoidal gradations. Visual acuity The ability of the eye to resolve de¬
Spatial modulation transfer function A graphical tails.
description of the way an optical system’s ability to re¬ Visual angle A measure of the size of the retinal im¬
solve spatial modulations (intensity changes across age.
space) varies with spatial frequency.
CHAPTER
5
Color
COLOR STIMULUS
Color Appearance Systems
Color Mixture
CIE Color Space
THE PHYSIOLOGY OF COLOR VISION
Trichromatic Color Theory
Color Vision Defects
Physiological Basis of Trichromatic Theory
Opponent-Process Theory
Physiological Basis of Opponent-Process
Theory
Models of Color-Coding
COLOR PERCEPTION
Intensity and Duration
Spatial Interactions
Age and Physical Condition
Cognitive Factors in Color Perception
Memory Color
Culture and Color
Color Impressions
120 Chapter 5/Color

( /Ty dad was color-blind, but didn’t find out ‘‘leaps out.” Thus, color provides an important
JA/Auntil he was nearly fifty. He was always stimulus dimension that aids in the localization and
doing strange things. He couldn’t be trusted to pick identification of objects, which explains why some
tomatoes from the garden because he was always occupations, such as air traffic control, require nor¬
mixing up the ripe and the green ones. We finally mal color vision (Kuyk, Veres, Lahey & Clark,
suspected that something was wrong when he com¬ 1986). For some species, color vision is a matter
mented that he really admired cherry pickers for of life and death. For instance, if bees lacked color
their ability to recognize shapes. ‘After all,’ he vision, their task of locating the nectar-bearing
said, ‘the only thing that tells ’em it’s a cherry is flowers hidden among shrubs, grasses, or leaves
the fact that it’s round and the leaves aren’t. I just would be almost impossible. The survival of this
don’t see how they find ’em in those trees!”’ species may well depend on the ability to spot a
Like this student, you may be surprised to find glint of color that indicates the presence of blos¬
out how important a factor color is in determining soms.
your ability to acquire information about the world.
For instance, consider Figure 5-1. Although the
figure appears to be a random collection of gray COLOR STIMULUS
shapes, there is a word hidden in it. Each letter is
spelled out by a series of similar shapes. If you The human eye registers as light wavelengths be¬
study the figure for a moment, you will begin to tween 360 and 760 nanometers. Sir Isaac Newton
see how difficult it is to pick out the word (if you was able to show that stimuli of different wave¬
can do it at all), despite the fact that the shape and lengths within this range produce different color
brightness information are there. In this task you sensations. Newton’s experiment was quite simple.
are much like the color-blind person trying to pick He took a glass prism and allowed some sunlight
out bunches of cherries, among the leaves, by to pass through it from a slit in a window shade.
shape alone. Now flip to Color Plate 2, where we When he held a sheet of white on the other side of
have added the dimension of color to the figure. the prism, the light no longer appeared to be white;
Notice that in this color plate the hidden word rather, it took the form of a colored spectrum,

FIGURE 5-1 Can you find the hidden word? If not turn to Color Plate 2.
 Color Stimulus 121

looking much like the arrangement of light in a Table 5-1. Wavelengths of Light and
rainbow (see Figure 5-2A). Newton knew that light Associated Color Sensations
bends when passing through a prism, and that the
amount of bending (called refraction) depends on Color Name Wavelength (nm)

wavelength. There is less refraction of the longer

wavelengths (600-700 nm) and more bending of Violet 450 ■
Blue 470
the shorter (400-500 nm). Thus, a prism takes the
Cyan 495
various wavelengths of light that make up sunlight
Green 510
and separates them according to wavelength. The
Yellow- 560
fact that we see this spread of light as varying in Green
hue shows that color perception depends on the Yellow 575
wavelength of the light. Table 5-1 shows some typ¬ Orange 600
ical color names associated with some selected Red 660
wavelengths of light. Purple Not a spectral color
Newton also inserted another prism (in the op¬ but a mixture of “red”
posite orientation) so that the light was now re¬ and “blue”
fracted in the direction opposite to the effect of the
original prism. This, of course, recombined all of
these wavelengths into a single beam. Now when he peared to be white, with no hint of the original col¬
placed a piece of paper into this beam it again ap- ors that went into the combination. This indicates
that the sensation of white results from a mixture
of many different wavelengths (see Figure 5-2B).
An important technical distinction should be
made here. Figure 5-2 does not describe how white
light is broken up into “colored light.” Colored
light does not exist; rather, what does exist is visi¬
ble radiation of different wavelengths. If there were
no observer there would be no color. Newton
pointed this out when he said, “For the rays, to
speak properly, are not coloured. In them is noth¬
ing else than a certain Power and Disposition to stir
up a sensation of this or that Colour.” When we
talk about the color stimulus, we should actually
speak of radiation of different wavelengths, since
the sensations of red, green, blue, or any other
color reside in the observer. Having made this
technical distinction, we must admit that it is ex¬
tremely convenient to talk about red light or green
light, and for the sake of brevity we will not hesi¬
tate to do so in some of our later descriptions. Re¬
member, however, that when we refer to a “blue
FIGURE 5-2 Newton’s experiments: (A)
light” we are referring to those wavelengths of
separation of white light into its various
wavelengths gives the color spectrum; (B) light that elicit the sensation of blue, namely, the
recombination of spectral lights gives white light. shorter wavelengths in the visible spectrum.
122 Chapter 5/Color

Objects appear to be colored because they re¬ you would end up with a pile of red stones, another
flect to our eyes only selected wavelengths of light. of green stones, and so forth. Once you have your
Consider a common object, such as an apple with piles of stones, you would next have to look for
white light falling on it. It appears to be red. We some meaningful arrangement for the piles. For in¬
have already seen that white light, such as sunlight, stance, you might notice that orange seems to fall,
is a combination of all wavelengths. Since the light in terms of appearance, somewhere between red
stimulus that reaches your eyes produces the sen¬ and yellow. The yellow-greens, of course, seem to
sation of red, all of the wavelengths except the fall between yellow and green. Once you reach the
longer (red-appearing) ones must have been ab¬ blue end of your line of stones, however, you
sorbed by the surface of the apple. Colored objects might find yourself running into a bit of a problem.
or surfaces contain pigments that selectively absorb The purple stones seem to fall somewhere between
some wavelengths of light while the rest are re¬ the blues and the reds. This means that a straight
flected and thus reach your eye. It is this selective line arrangement is not adequate. Instead, you
“subtraction” of some wavelengths from the in¬ might arrange the pebbles as shown in Figure 5-3.
coming light that gives an object its color. If a sur¬ This crude color arrangement scheme is cir¬
face does not absorb any of the wavelengths cular in form. You have probably seen it before in
reaching it but reflects them all uniformly, it ap¬ books on art, decorating, or design, where it is
pears white rather than colored. Colored filters usually called the color circle or color wheel. In
work in much the same way, that is, by absorbing this arrangement, you have separated the colors ac¬
some wavelengths of light. For instance, if a white cording to hue, which is the psychological dimen¬
light is projected through a green filter, the result¬ sion that most clearly corresponds to variations in
ing beam is green. This means that the filter has wavelength. Very often when we use the word
absorbed most of the long and short wavelengths, color, in everyday life and in this chapter, we are
allowing only the medium-range, or green-appear¬ actually referring to hue. Let us consider the effect
ing, wavelengths to reach the eye. of wavelength on sensation by looking at the ef¬
You should be alerted to the fact that simply fects produced by pure or monochromatic stimuli.
specifying the wavelength, or wavelengths, in a A monochromatic stimulus contains only one
stimulus does not seem to fully describe the way
the color appears to an observer. For instance, a
stimulus with a dominant wavelength of 570 nm
may appear yellow while another with the same
wavelength composition might appear brown. For
this reason, additional factors other than wave¬
length are used to classify colors.

Color Appearance Systems

Suppose you were marooned on a desert island that
had a beach covered with many colored pebbles.
Lacking anything else to do, you set about the task
of classifying the colors of all the pebbles in some
meaningful way. The first classification scheme
that might come to mind would involve grouping FIGURE 5-3 A primitive color circle for encoding
stones together on the basis of their hues. Thus, the colors of pebbles.
 Color Stimulus 123

wavelength (from the Greek mono meaning “one” Nonspectral purples

and chroma meaning “color”). These stimuli are
similar to those found in the spectrum generated by
Newton’s prismatic separation of light and there¬
fore are often called spectral colors. Such mono¬
chromatic stimuli do not produce all the hues found
in the color wheel. For instance, we find that there
is no single wavelength that produces the sensation
of purple. This sensation requires a mixture of blue
and red wavelengths. Similarly, there is no place
in the spectrum where we can find a red that
doesn’t appear to have a tinge of yellow. In order
to achieve such a hue, we must add a bit of blue
(short-wavelength) light.
Meanwhile, back on the beach, it has become
clear that our color wheel classification scheme
FIGURE 5-4 The color circle modified to allow
based only on the psychological attribute of hue the encoding of both hue and saturation. Spectral
seems incomplete. A close look at the piles of peb¬ colors are on the outer rim, white is in the center.
bles reveals marked color differences. For instance,
among the red pebbles you might find that some
are deep red color and others are pink; another color. It is quite possible to have two colors match
group may be almost pure white with only a hint in both of these attributes but still appear to be dif¬
of red coloration. This observation corresponds to ferent. For instance, a blue spot of light projected
the physical dimension of purity. Clearly the pur¬ onto a screen would not appear to be the same as
est color you could get would correspond to a mon¬ another spot identical in all respects except that it
ochromatic or spectral hue, and as you add other has been dimmed by putting a light-reducing filter
wavelengths, or white light, the color would appear in front of it. Thus, the sensory quality of bright¬
to become “washed out.” This psychological at¬ ness (which we discussed in Chapter 4) must be
tribute of color appearance is called saturation. It worked into our system of describing colors. Since
is quite easy to integrate saturation into the color we have already used the two dimensions capable
circle by simply placing white in the center. Now of being reproduced on a flat piece of paper, it is
imagine that the various degrees of saturation cor¬ clear that the addition of a third color dimension
respond to positions along the spokes or radii em¬ (brightness) forces us to use a solid instead of a flat
anating from the center of the wheel. The center representation.
represents white (or gray) and the perimeter repre¬ The shape of the three-dimensional color
sents the purest or most saturated color possible. “space” can be derived from common observation
Figure 5-4 shows the color wheel now modified to if we recognize that at high brightness levels colors
include saturation. Notice that the point corre¬ appear to be “washed out,” whereas at low bright¬
sponding to pink (a moderately desaturated red) is ness levels colors seem “weak” or “muddy,”
plotted near the center along the line connecting red meaning that they are of low saturation. Thus, the
and white, whereas a crimson is plotted further hue circle must shrink at these extremes, since sat¬
away from the center along the same line. uration seems to vary over a confined range, and
To the average observer, hue and saturation do very high degrees of saturation are never observed
not completely describe all the visible nuances of at very high or low levels of brightness.
124 Chapter 5/Color

If we combine the three psychological attri¬ way typical observers arrange color stimuli. To ac¬
butes of hue, saturation, and brightness, we get tually classify colors you can use a color atlas, in
something that looks like Figure 5-5. It appears to which each page represents a horizontal or a verti¬
be a pair of cones placed base to base. This is usu¬ cal slice through the color solid. Color samples il¬
ally called the color spindle or the color solid. The lustrating colors found in varying locations in the
central core as we move up or down represents color solid are given in such atlases, allowing the
brightness and is comprised of all the grays running observer to identify and label any given test color.
from white (at the top) to black (at the bottom). We
can imagine that at each brightness level, if we
Color Mixture
sliced through the color solid in the direction
shown in the diagram, we would get a color circle Pure colors of a single wavelength usually are pro¬
in which the hue would be represented along the duced only under precise laboratory conditions.
perimeter. Totally desaturated colors (the grays) are Most of the light reaching your eye is composed of
at the central core, as we’ve already noted; hence, a mixture of many different wavelengths. Gener¬
saturation is represented by moving from the center ally, the dominant wavelength will determine
outward. This is the basic representation used in what hue you see, although this is not always the
many color appearance systems. Probably the most case. When we combine two or more wavelengths
popular in use among psychologists is the one de¬ of light a new color, with a different psychological
veloped by Munsell (1915) and modified by New- hue and saturation, is often seen. Once the colors
hall, Nickerson, and Judd (1943) to agree with the are mixed, the eye can no longer discriminate the
individual wavelengths that make up the mixture.
Thus, you can have a pure yellow made up of 570-
White nm light and another yellow that matches it, com¬
posed of a mixture of a 500-nm green and a 650-
nm red. You will not be able to distinguish be¬
tween these hues, nor will you be able to isolate
the red and the green that went into the mixture.
Colors that appear to be the same but are made up
of different wavelengths of light are called meta-
meric colors.
There are two types of color mixtures. The
first, and the simplest to describe, is called addi¬
tive color mixture. Additive mixtures occur when
we mix light. For instance, if we project a red cir¬
cle on a screen, the light reaching the eye from the
projected circle is red. If we project a blue circle
on the screen so that it partially overlaps with the
red circle, the light reaching the eye from the re¬
gion where the circles overlap contains both blue
and red light. Thus, each new wavelength projected
onto the same region of the screen adds to the mix¬
ture of wavelengths reaching the eye. Figure 5-6A
shows a situation that might occur if we used three
FIGURE 5-5 The color solid. projectors with the first projecting a red beam, the
 Color Stimulus 125

Additive color mixture happens where the red and green overlap—we get
using three beams of light
the sensation of yellow. Where all three beams
overlap we get the sensation of white. Demonstra¬
tion Box 5-1 shows another way to get additive
color mixtures.
These results cannot be duplicated using
paints. If you tried mixing all your paints together
to get white, you would get a shade of gray or
black instead. This is because pigments do not
work in the same way as lights. Something that
appears red, such as a tomato, will have a surface
pigment that absorbs most of the short and medium
wavelengths, reflecting to your eyes only the long
(red) wavelengths. A pigment that gives you a
color similar to grass green might absorb most of
the long wavelengths and the short wavelengths,
reflecting to your eye mainly middle wavelengths.
Thus, when you mix the red and the green paints
together, you end up with a mixture in which only
the middle wavelengths are reflected by the green,
yet these are absorbed by the red pigment. Hence
you are essentially subtracting all the wavelengths,
leaving only a muddy gray.
Since pigments work by subtracting or absorb¬
ing wavelengths of light, a mixture of pigments is
called a subtractive color mixture. Such pigment
mixtures produce colors that are considerably less
predictable than mixtures of lights because the
wavelength-absorbing property of pigments is com¬
plex. For example, Figure 5-7 shows the wave¬
lengths reflected by some typical pigments. Notice
how irregularly they reflect the light and imagine
FIGURE 5-6 Color-mixture systems: (A) additive;
(B) subtractive. the problems in predicting what the resultant mixes
might reflect and absorb.
Suppose, however, that we are dealing with
second a green beam, and the third a deep blue relatively simple pigments, where the yellow re¬
(almost violet) beam. If the circles of light were flects only middle and long wavelengths, the cyan
arranged so that they partially overlapped with one only long and middle wavelengths, and the ma¬
another, we would get a series of additive mix¬ genta only long and short wavelengths. If we now
tures. Where red and the deep blue overlap we get painted circles of these pigments so that they over¬
a reddish-purple generally called magenta. Where lapped, we would get a crude representation of
the deep blue and the green overlap we get a lighter what would generally be expected in subtractive
hue that is greenish blue, usually called cyan. color mixtures. Since the yellow pigment works
Something that some people find quite surprising by absorbing all the short wavelengths, and the
126 Chapter 5/Color

DEMONSTRATION BOX 5-1. Color Mixture

There is a simple way to obtain additive color mix¬ glass and hold it up to the screen, you will see that
tures without using projected beams of light. Consider each region is made up of a series of tiny dots. When
Color Plate 4A, in which you see a checkerboard of you sit at normal viewing distance, you can no longer
tiny red and green squares. In Color Plate 4B you see resolve the individual dots. They have combined
a yellow disk. Prop up the book so that you can see within the eye to give you an additive color mixture.
the color plates when you move across the room. A similar technique was used by the French painter
Now, standing at a distance, look back at the figures. Georges Seurat, who replaced the traditional irregular
What formerly appeared to be red and green now ap¬ brush stroke used in painting with meticulously
pears to be yellow and should match the yellow disk. placed dots of color. Thus, instead of mixing paints
At a distance, the optics of the eye can no longer on his palette, he allowed the mixture to be accom¬
resolve the individual squares. The light from each of plished optically within the eye of the onlooker view¬
them smears, or blurs, across the retina giving rise to ing the painting from an appropriate distance.
the color mixture effect.
This technique is similar to the technique used in
your color television set. If you take a magnifying

cyan pigment absorbs all the long wavelengths, the long or red-appearing wavelengths remain. In a
their mixture absorbs both the short and the long similar fashion, overlapping cyan and magenta
wavelengths, leaving us only with the middle or leaves us with only the blue wavelengths, because
green-appearing portion of the spectrum. When all others are subtracted, and the mixture produces
we combine yellow with magenta, we find that the blue. Clearly, when all three pigments overlap
yellow subtracts the short wavelengths and the ma¬ everything is absorbed and we get black, as shown
genta subtracts the middle wavelengths, hence only in Figure 5-6B. In general, it is more convenient
to deal with additive color mixtures since they are
easier to conceptualize.
The color circle, which we have already dis¬
cussed, provides a convenient means of predicting
the appearance of additive color mixtures. But note
that the color circle describes, rather than explains,
how colors interact. For instance, the spacing
around its circumference corresponds to the way
the various hues appear to an average observer
rather than to a regular spacing according to wave¬
length. To use the color circle to predict a color
mixture is actually quite simple. Suppose we mix a
spectral red (about 650 nm) with a spectral yellow
(about 570 nm). We can depict this as in Figure
FIGURE 5-7 The relative wavelength 5-8, where the resultant mixture is represented by
composition of a blue, a yellow, and a green the line connecting these two colors. If we combine
pigment. the yellow and the red in equal proportions we will
 Color Stimulus 127

Nonspectral purples sensation is given by the center of a triangle pro¬

duced by connecting the three colors. If the amount
of each hue differs, the center point of the triangle
shifts toward the dominant hue.

CIE Color Space

An interesting effect occurs when we mix two col¬

ors that are exactly opposite to each other on the
color circle. For example, mixing a violet with a
yellow along the line shown in Figure 5-8 results
in a colorless gray. This is because, when the pro¬
portions are correct, this mixture lies in the center
of the circle. Colors whose mixture produces such
an achromatic gray are known as complementary
colors.
FIGURE 5-8 Using the color circle to predict
One of the most important facts about color
color mixtures.
mixtures emerged in the 1850s. The German phy¬
sicist and physiologist Hermann von Helmholtz
get a color that corresponds to the dot in the center (1821-1894) and the Scottish physicist James
of the line. We can determine what this color will Clerk Maxwell (1831-1879) carried out a set of
look like by simply drawing a line from the center color matching experiments. They reported that by
of the color circle through the dot to the perimeter. combining an appropriate set of three monochro¬
When this is done, we find that we get an orange matic light sources in appropriate amounts, they
corresponding to about a 600-nm spectral stimulus. could match any other hue. These three wave¬
Increasing the amount of yellow shifts the point lengths were to be known as primaries. Actually,
along the line in the direction closer to the yellow the choice of primaries is rather arbitrary. Primary
hue. Adding more red shifts the point along the line colors need only be reasonably far apart, with the
in the other direction. You will notice that we requirement that the mixture of any two of them
started out with two spectral, or pure, hues (marked alone will not match the third.
on the perimeter), however, the resultant mixed Wright (1929) made a set of measurements in
color is no longer on the perimeter but is closer to which observers matched the hue of the various
the center of the color circle. The purest colors spectral sensations. He selected as his primary col¬
possible (the spectral colors) are placed on the pe¬ ors a red of 650 nm, a green of 530 nm, and a blue
rimeter of the color circle; more desaturated colors of 460 nm. The observers matched the color of two
are found closer to the center (nearer white or patches of light, where the first was the test color
gray). From this we can conclude that any color and the second could contain any combination of
mixture is less saturated than either of the two the three primaries. Wright’s results are shown in
component colors that went into it. No mixture of Figure 5-9, which indicates the relative amount of
colors can ever be quite as saturated as a mono¬ each of the three primaries needed to match any
chromatic or spectral color. given wavelength. You might notice that some of
Mixing more than two hues (or hues contain¬ the values are negative. This indicates that some of
ing more than a single wavelength) is a little more the particular primary had to be added to the test
complex. If we mix three colors, the resultant color sample in order to reduce its saturation to the point
128 Chapter 5/Color

lected primary colors suggests an alternate way of

specifying the hue of a stimulus, namely, in terms
of the proportion (sometimes “negative”) of the
three primaries needed to reach this match. Geo¬
metrically this suggests a triangular space with a
primary color at each comer. Color mixtures may
then be represented in the same way as they are on
the color circle. Thus yellow, which is a mixture
of red and green, is represented by a point on the
line between red and green. If we add more red the
point moves toward the red primary, and if we add
more green it moves toward the green. As in the
FIGURE 5-9 The proportion of each primary (a color circle, white is represented by a point in the
460-nm blue, a 530-nm green, and a 650-nm red) middle, and is comprised of an equal proportion of
needed to match any spectral color. the three primaries. Also, as in the color circle, a
red of lower saturation (the whitish red or pink)
where it could be matched by a mixture of the two would be represented by a point moved inward to¬
remaining primaries. In other words, some matches ward the center. Such a diagram is shown in Figure
could be made only if one of the three primaries 5-10.
was not included in the mixture but was added to In 1931, a special body of the Commission
the test sample. Internationale de l’Eclairage (CIE) standardized the
The fact that any selected color can be procedure for specifying the color of a stimulus.
matched by a mixture of three appropriately se¬ They decided to use a color space created by the

Primary blue Primary red

FIGURE 5-10 Specifying colors using a color triangle.

 Color Stimulus 129

mixing of three primaries as described above. Un¬ in Figure 5-11. These imaginary primary colors are
fortunately, if we select any three spectral primary more saturated than any real colors can be. (Re¬
colors, a number of perceptual and mathematical member, this is done so that all colors can be rep¬
problems result. The major perceptual problem is resented within the space.) Notice that we have
the fact that there are other spectral colors that can¬ labeled the horizontal and vertical axes of the tri¬
not be represented within the triangle. For instance, angle with the labels x and y.
a pure spectral yellow cannot be represented (un¬ We can now represent any color as a point in
less it is one of the primaries, which creates other the color space. The reason that we can plot a mix¬
problems), since any color mixture can never be as ture of three colors by using a point that has only
saturated as the pure spectral color itself. To solve two spatial coordinates is because the CIE chro-
this perceptual problem, the CIE selected three maticity space has been arranged so that y repre¬
imaginary primary colors. They arranged the imag¬ sents the proportion of green in the mixture and x
inary primaries at the comers of the triangle shown represents the proportion of red in the mixture.

FIGURE 5-11 The CIE chromaticity space, which is a variant of the color triangle system using three
imaginary “super” primaries.
130 Chapter 5/Color

Clearly the proportions of red and green and blue tors. Let us consider these in light of the two major
in any mixture must sum to a total of 1.0 (you can theoretical positions that have emerged during the
think of these as representing percentages, where last 150 years.
all the items must sum to 100 percent of the light).
If we know the proportions of green and red in the
ichromatic Color Theory
mixture, we need only subtract these from 1.0 to
find the proportion of blue. The actual colors that Much research has gone into the search for the
can be perceived do not fill the full triangle (re¬ physiological basis of color vision. One of the ear¬
member the primaries we are using are imaginary liest findings was discussed in Chapter 4, where it
“super” colors). Instead, they fill a horseshoe¬ was reported that under scotopic levels of illumi¬
shaped area with the spectral colors forming the nation, when only the rods are active, no color vi¬
outside boundary. We have labeled this area on the sion is found. On the basis of these observations,
figure so that you can see the regions filled by var¬ it was concluded that cones are the retinal receptors
ious colors. Thus, if we have a color at ,2x and that provide the first stage of the color response.
.6y, we would know that it is comprised of 20 per¬ Therefore, how the cones provide information
cent red, 60 percent green, and (subtracting from a about the wavelength of the incoming light be¬
total of 100 percent) 20 percent blue. Looking at comes the first question to be answered.
the figure you can find the point described by these Most normal people can discriminate among
x and y coordinates and see that this color would thousands of colors under a myriad of conditions;
look green. however, holding brightness and saturation con¬
Notice that brightness is not represented any¬ stant the average human observer can discriminate
where on this diagram. As in the color space we about 200 different hues. Suppose that we wished
discussed earlier, brightness requires a third dimen¬ to create an artificial eye with this same ability.
sion. The color space in the figure can be thought The simplest procedure might seem to require a
of as a single slice through a three-dimensional separate cone that responds to each of the discrim-
color space, just as we demonstrated in Figure inable hues. Unfortunately, such a scheme is not
5-5. This third dimension would be called z. Using practical. For any given colored stimulus we would
the CIE color system we can specify any color only have 1/200 of the cones active, which means
stimulus by its tristimulus values, which are sim¬ that our visual acuity would be much poorer than
ply the x and y coordinates for the hue of the stim¬ research has shown it to be. In addition, such a
ulus, and a z coordinate for the brightness of the system would mean that our acuity measured under
stimulus. white illumination would be many times better than
our acuity measured under monochromatic stimu¬
lation. This also is not found.
An alternate scheme would be to have only
THE PHYSIOLOGY OF COLOR VISION one type of retinal cone with 200 different code
signals by which it could indicate the discriminable
To this point, we have dealt with the physical color hues. This could be done via a sort of neural Morse
stimulus, some methods of specifying the appear¬ code. Although it may be the case that such a
ance of a color, and some aspects of combining neural Morse code can play a part in some aspects
various wavelengths of light. None of the foregoing of color vision (see “Models of Color-Coding”
descriptions specify how a particular color sensa¬ section), the evidence for such a mechanism is still
tion arises. In order to understand this, we must controversial. Research suggests that each cone
deal with both physiological and psychological fac¬ contains only one pigment. If so, how would the
 The Physiology of Color Vision 131

cone itself “recognize” the wavelength of the ness. This term is much too strong, since only a
light? The only thing that the cone “knows” is the very small percentage of individuals are totally in¬
amount of pigment that has been bleached. Al¬ capable of discriminating colors. According to a
though different wavelengths of light may bleach trichromatic theory of color vision, we can predict
more or less pigment, simply increasing or decreas¬ five different varieties of color abnormality. The
ing the intensity of the stimulus could also cause first, and most drastic, would be found in those
the same variations in degree of bleaching. It thus who have no functioning cones. Since all their
seems unlikely that a single cone would be able to seeing would be done only with the rod system,
discriminate 200 hues. they would be expected to have no color discrimi¬
An explanation was suggested almost 200 nation ability. In addition, they should find pho-
years ago by Thomas Young (1773-1829). Young topic, or daylight, levels of illumination to be quite
suggested that only a few different retinal recep¬ uncomfortable. A slightly less drastic malady is
tors, operating with different wavelength sensitivi¬ one in which only one variety of cone is function¬
ties, would be necessary to allow humans to ing in addition to the rods. With this problem, vi¬
perceive the number of colors they do. He further sion should be possible under both photopic and
suggested that perhaps as few as three would do. scotopic conditions but there would still be a lack
His theoretical notion was revived in the 1850s by of any color discrimination ability. Any wavelength
Helmholtz, as we have already noted, when he and of light hitting one of the functioning rods (or the
Maxwell were able to show that normal observers single-cone system) would produce some bleaching
need only three primaries to match any color stim¬ of the pigment. Even though different wavelengths
ulus. These data were taken as evidence for the might bleach different amounts, this is not enough
presence of three different receptors in the retina. to allow color discrimination since the response
Since the usual color matching primaries consisted produced by any one wavelength of light can be
of a red, a green, and a blue, it was presumed matched by merely adjusting the intensity of any
that there were three types of receptors, one re¬ other. In other words, the individual with no func¬
sponsive to red, one to green, and one to blue. Since tioning cones, or the one with only one function¬
these receptors are cones, and cones operate by ing cone type, responds to light in much the way
the bleaching of pigment, we may suggest three that a sheet of black and white film does. All col¬
hypothetical pigments. The first we would call ors are recorded simply as gradations in intensity
erythrolabe (translated from the Greek this means of the response. Such individuals are called
“red-catching”), another chlorolabe (mean¬ monochromats.
ing “green-catching”), and the third cyanolabe We might also suppose that some individuals,
(meaning “blue-catching”). This trichromatic rather than lacking two or three sets of cones as
theory (from the Greek tri meaning “three” and does the monochromat, might only have one mal¬
chroma meaning “color”) finds some very con¬ functioning cone system. Given two functioning
vincing support in the study of defective color cone systems, they should have some color percep¬
vision. tion, though it would differ from that of a normal
observer. In effect, they should be able to match
Color Vision Defects
all other colors with a mixture of only two primar¬
Virtually all individuals differ from what is usually ies (rather than the three required by color-normal
called “normal” color vision in one way or an¬ observers). Such individuals are usually called di-
other. However, some show drastic deficiencies in chromats (from the Greek di meaning “two” and
their ability to discriminate colored stimuli and, in chroma meaning “color”). The existence of such
popular speech, are said to suffer from color blind¬ individuals has been known since the 1700s. The
132 Chapter 5/Color

English chemist John Dalton (1766-1844) was or malfunction of the blue cone system. Although
such a dichromat, a fact he learned rather late in a name existed for this phenomenon, tritanopia
his life. Supposedly, it first came to his attention (from the Greek tritan, for the “third” primary),
when he wore a scarlet robe to receive his Ph.D. there was no confirmed report of this difficulty until
degree. Since he was a Quaker, a sect that shuns about 1950, when a magazine article containing a
bright colors, this caused quite a stir, until it be¬ color-vision test plate appeared as part of an inten¬
came clear that woolen yam dyed crimson or yam sive search throughout England. This national
dyed dark blue-green appeared to be the same color search resulted in the discovery of 17 tritanopes
to him. (Wright, 1952). These individuals, instead of
There are three predictable forms of dichro- seeing the spectrum as composed of blue and yel¬
macy, depending on whether it is the red-, green-, low as do other dichromats, see the longer wave¬
or blue-responding cones that are inoperative. The lengths as red and the shorter ones as bluish green.
specific confusions are predictable from the color¬ The discovery of this last class of individuals pro¬
matching curves of normal observers shown in Fig¬ vides strong support for a trichromatic theory of
ure 5-9. Dalton’s type of color defect is usually color vision.
referred to as protanopia (the Greek prefix proto Color defects are a fairly common problem.
means “first,” and red light is generally designated Some instances of it are relatively mild and result
as the first primary). A protanope is insensitive to in what is called anomalous trichromatism. Color
long wavelengths normally perceived as red light. matches of individuals with this problem require
If a red light is made very much brighter than a more red (protanomaly) or more green (deuter-
green light, a protanope could easily confuse them, anomaly) than do color matches of normal observ¬
whereas, a color-normal observer would perceive ers. If we count all individuals with any form of
both that the red light was brighter than the green color deficiency, we find that just over 8 percent of
and that they differ in hue. Dalton described his all males show color weaknesses, whereas slightly
subjective experiences when viewing a spectrum less than .05 percent of all females show similar
such as that produced by Newton’s prism. Most in¬ deficits. Color defects are genetically transmitted
dividuals perceive six different colors, blending and recent studies have conclusively mapped the
one into another. Dalton reported: “To me it is pattern of this transmission (Nathans, Piantanida,
quite otherwise. I see only two, or at most three Eddy, Shows & Hogness, 1986).
distinctions. These 1 should call yellow and blue, What colors does a dichromat actually see? It
or yellow, blue, and purple. My yellow compre¬ is really not possible to know how the colors of a
hends the red, orange, yellow, and green of others dichromat compare with those seen by a color-nor¬
and my blue and purple coincide with theirs.” mal observer. However, a glimpse into the visual
The most common form of dichromacy is world of the color-defective has been provided by
called deuteranopia (the Greek prefix deuteros a rare person who was deuteranope in her left eye
means “second,” and green light is by convention but color-normal in her right eye. Graham and Hsia
the second primary). Individuals with deuteranopia (1958) had this observer adjust the color seen by
presumably have a malfunction in the green cone her normal eye so that it appeared to be the same
system. With deuteranopia, they are still able to hue as the color seen by her defective eye. The
respond to green light; however, they cannot distin¬ results of her matches are shown in Figure 5-12.
guish green from certain combinations of red and As can be seen from this figure, the colors over the
blue. entire range of red to green (from about 750 to 502
Trichromatic theory also predicts that there is nm) all appeared to have the same yellow hue
a third form of dichromacy caused by the absence (about 570 nm), and all the colors from green to
 The Physiology of Color Vision 133

Color-defective eye
Normal eye
(deuteranopic)

Equivalent
to 470 nm

Equivalent
to 570 nm

FIGURE 5-12 Color matches of a normal eye to a “color-blind” eye (deuteranopic) in the same observer.
(Based on Graham &, Hsia, 1958)

violet appeared to be blue (matching a 470-nm lengths are passed through the cone the amount of
stimulus). The region that appears to be blue-green light absorbed at each wavelength is measured. The
to the normal observer (around 502 nm) was per¬ more light of a given wavelength that is absorbed
ceived as being a neutral gray in the defective eye. by the cone pigment, the more sensitive is the cone
Knowledge of the nature of the color confusions to light of that particular wavelength. Such mea¬
among dichromats allowed Coren and Hakstian surements were taken using cones from the retinas
(1987, 1988) to develop a simple questionnaire that of goldfish, monkeys, and finally from humans.
assesses whether individuals are likely to be color¬ Although researchers are still refining the de¬
blind. You can test yourself with this questionnaire tailed description of the cone pigments (e.g.,
using Demonstration Box 5-2. MacNichol, 1986), the general pattern of the re¬
sults is unambiguous. There are three major groups
of cones. A typical set of measurements, taken
Physiological Basis of Trichromatic
from a human eye that had to be surgically re¬
Theory
moved (Bowmaker & Dartnall, 1980), shows max¬
Although the data from color mixing and color de¬ imum absorptions in the ranges of 420, 534, and
fects seem to support a trichromatic theory of color 564 nm, respectively (rods have a maximum ab¬
vision, direct physiological evidence for the three sorption of 498 nm measured on this same eye, us¬
cone pigments did not appear until the 1960s. The ing the same technique). Figure 5-13 shows the
measurement procedure involved is conceptually relative absorption of these three pigments (where
simple but technically quite difficult (Bowmaker & 1.0 is the maximum amount absorbed by the pig¬
Dartnall, 1980; Brown & Wald, 1964; Marks, Do- ment). Clearly, on the basis of their sensitivity
belle & MacNichol, 1964). It involves a device peaks, we should call the short-wavelength-absorb-
called a microspectrophotometer. With this de¬ ing pigment “violet,” the middle “yellow-green,”
vice a narrow beam of monochromatic light is fo¬ and the long “orange” if we wish to be more pre¬
cused on the pigment-bearing outer segment of a cise than the blue, green, and red labels we have
cone. As tiny amounts of light of various wave¬ been using.
134 Chapter 5/Color

DEMONSTRATION BOX 5-2. Color Vision Screening Inventory

To see if you may have a color-vision deficit simply 6. Do you have difficulty discriminating between
take this test, which is the Color Vision Screening pale green and pale red?
Inventory* developed by Coren and Hakstian (1987,
1988). For each question you should select the re¬ 7. Do you have difficulty discriminating between
sponse that best describes you and your behaviors. blue and purple? N S O F A
You can select from among the following response
alternatives: Never (or almost never). Seldom, Occa¬ 8. Do the color names that you use disagree with
sionally, Frequently, Always (or almost always). those that other people use? N S O F A
Simply circle the letter corresponding to the first letter
of your choice. 9. Are the colors of traffic lights difficult to distin¬
guish? N S O F A

1. Do you have difficulty discriminating between 10. Do you tend to confuse colors? N S O F A
yellow and orange? N S O F A Scoring Instructions: Responses are scored 1 for
Never, 2 for Seldom, 3 for Occasionally, 4 for Fre¬
2. Do you have difficulty discriminating between quently, and 5 for Always. Simply add together your
yellow and green? N S O F A scores for the 10 questions. If your score is 17 or
higher, you have an 81 percent likelihood of failing a
3. Do you have difficulty discriminating between standard screening test for color vision. If your score
gray and blue-green? N S O F A is in this range you might want to get your color vi¬
sion tested by your doctor or in a perception labora¬
4. Do you have difficulty discriminating between tory.
red and brown? N S O F A
*The Color Vision Screening Inventory is copyrighted by SC
Psychological Enterprises Ltd., and is reprinted here with
5. Do you have difficulty discriminating between
permission.
green and brown? N S O F A

Rushton (1962, 1965) introduced a similar most strongly, hence it would, with continued ex¬
technique for measuring the photopigments in liv¬ posure, be “bleached out.” When he remeasured
ing human observers that does not require a micro¬ the amount of light absorbed at each wavelength,
spectrophotometer. First, he sent a beam of light the difference between light reflected back by the
into the eye and then took measurements on the “bleached” and “unbleached” retinas gave ab¬
amount of light reflected back out of the eye. By sorption curves similar to those shown in Figure
taking the difference between the amount of light 5-13 for pigments marked “red” and “green” in
sent and the amount reflected, an estimate was ob¬ the fovea. Rushton reasoned further that protanopes
tained of the amount of light at each wavelength and deuteranopes, according to trichromatic theory,
absorbed by the photopigments in the intact human should be missing one or the other of the two
eye. Next, he flooded the eye with light of a partic¬ longer wavelength pigments. When he used his
ular distribution. Thus, red light might be expected procedure with color-defective observers, he found
to activate the long-wavelength-catching pigment that they were missing the appropriate pigments.
 The Physiology of Color Vision 135

"Blue” “Green” “Red”

cones Rods cones cones

FIGURE 5-13 The relative absorption of various wavelengths of light by the three different cone types
and the rods of a human. (Based on Bowmaker & Dartnall, 1980)

Rushton could not find evidence for cones contain¬ creasing distance from the fovea and disappears
ing blue-matching pigment in the fovea, which completely at about 40 degrees from the fovea. A
suggests that all observers are dichromats, specifi¬ similar pattern holds for sensitivity to red and yel¬
cally tritanopes, for targets seen in central vision. low light, with color responses disappearing in the
This conclusion has been verified using psycho¬ order green, red, yellow, and blue as distance from
physical techniques as well (Bomstein & Monroe, the fovea increases. In the far periphery of the ret¬
1978; Williams, MacLeod & Hayhoe, 1981). Re¬ ina, you are totally color-blind. The exact distance,
cent evidence based on destruction of blue cones in however, depends on the size of the stimulus—you
the monkey retina by prolonged exposure to short- can discriminate the colors of larger stimuli farther
wavelength light has confirmed the fact that there out on the peripheral retina (Johnson, 1986). To
are no blue cones in a circular region 25 minutes see how your own color discrimination varies
in diameter in the central fovea (e.g., Sperling, across the retina try Demonstration Box 5-3.
1986). The relative rarity of blue cones probably
also explains why blue contributes less than red or
Opponent-Process Theory
green to many aspects of the visual process (e.g.,
Kaiser & Boynton, 1985). The German physiologist Ewald Hering (1878/
Because the cones are differentially distributed 1964) was not completely satisfied with a trichro¬
across the retina (e.g., Sperling, 1986), your color matic theory of color vision. It seemed to him that
response is different over different portions of the human observers acted as if there were four, rather
eye. The central foveal region is relatively blue- than three, primary colors. For instance, when ob¬
blind, and sensitivity to blue light first increases servers are presented with a large number of color
then decreases with increasing distance from the samples and asked to pick out those that appear to
fovea. Sensitivity to green light diminishes with in¬ be pure (defined as not showing any trace of being
136 Chapter 5/Color

DEMONSTRATION BOX 5-3. Color Sensitive Zones on the Retina

Color perception is best in the central region of the Eventually you will reach a point where the orange
retina (excluding the small central region of the will look yellowish, meaning that you have now im¬
fovea, which is blue-blind). You can observe the aged it beyond the red sensitive zone. If you continue
changes in color discrimination for different parts of moving your eyes outward you may even hit a point
the retina by taking a small orange piece of paper and where the orange no longer looks colored at all, but
placing it on a gray surface. Now keeping your head merely appears gray. Your eye will have to move far¬
fixed, look off to the side of the orange target. If you ther to get these changes in color appearance if the
keep moving your eyes outward (away from the tar¬ orange patch is larger (see Johnson, 1986).
get), you stimulate more peripheral parts of the retina.

a mixture of colors), they tend to pick out four, at it than they did at a wavelength selected from
rather than three, colors. These unique colors al¬ the same hue category. The infants acted as if stim¬
most always include a red, a green, and a blue, as uli in the same hue category were more similar than
trichromatic theory predicts (Fuld, Wooten & those from different categories; hence it seems they
Whalen, 1981); however, they also include a yel¬ were categorizing hues into the same four groups
low (Bomstein, 1973). that the adults do.
Boynton and Gordon (1965) showed that with Hering looked at another aspect of the subjec¬
the color names red, yellow, green, and blue, En¬ tive experience of hue. He noted that certain color
glish-speaking observers can categorize the entire combinations are never reported by observers, for
range of visible hues (some stimuli seem to require instance, a yellowish blue or a greenish red. This
a combination term containing two primaries, such led Hering to suggest hypothetical neural processes
as yellow-green). The way adult observers distrib¬ in which the four primaries were arranged in op¬
ute their hue names is shown in Figure 5-14, which posing pairs. One opponent process would signal
indicates four overlapping hue name categories cor¬
responding to red, green, blue, and yellow. These
results cannot be attributed simply to learning or
language use. For example, Bomstein, Kessen, and
Weiskopf (1976) showed that 4-month-old infants
tend to see the spectrum as if it were divided into
four hue categories. They did this by repeatedly
presenting a given wavelength of a light until the
infants became visually bored and stopped looking
at the light (a process called habituation). They
next monitored how much time an infant spent
looking at a second wavelength of light. They
found that when the second wavelength was se¬
lected from another hue name category (based on FIGURE 5-14 The relationship between color
the adult data) the infants spent more time looking names and wavelengths.
 The Physiology of Color Vision 137

the presence of red or green, and a separate process of light could cause opponent effects in neural re¬
would signal blue or yellow. An example of such a sponse was offered by Svaetichin (1956), who in¬
process would be a single neuron whose activity serted an electrode into the retina of a goldfish.
rate increased with the presence of one color (red) When he recorded the responses to light transmitted
and decreased in the presence of its opponent color by the horizontal cells (units at the first cellular
(green). Since the cell’s activity cannot increase layer beyond the cones, as noted in Chapter 3), he
and decrease simultaneously, one could never have found that responses varied depending on the
a reddish green. A different “opponent-process” wavelength of the light reaching the cones. These
cell might respond similarly to blue and yellow. A neural responses were not in the form of the typical
third process was suggested to account for bright¬ action potential found in most neurons but rather
ness perception. This was called a black-white op¬ were graded shifts in the electrical polarization of
ponent-process, after the fact that black and white the cells. Svaetichin found not only that the
are treated psychologically as if they were “pure strength of response varied as the wavelength
colors” (see Quinn, Wooten & Ludman, 1985). changed but, more importantly, that the electrical
But we need not limit the discussion to speculation sign of the response was different for long and
based on color appearances alone, since physiolog¬ short wavelengths.
ical evidence exists that bears directly on the issue Figure 5-15 shows the pattern of responses re¬
of opponent-process coding of color information. corded by Svaetichin and MacNichol (1958). No¬
tice that the spectral sensitivities of the first two
units are exactly what we would need for a blue-
Physiological Basis of Opponent-
yellow cell and a red-green cell. For instance, the
Process Theory
cell marked red-green would respond with a large
At the time Hering first suggested an opponent-pro¬ positive signal if the unit is stimulated with a long-
cess mechanism for the neural encoding of hue in¬ wavelength light (around 675 nm). This positive
formation, there was no physiological evidence to response could signal red. If the unit is stimulated
support such a speculation. Perhaps the single most with a greenish hue (around 500 nm), it would give
important finding of 20th-century sensory physiol¬ its peak negative response, thus signaling the pres¬
ogy was that neural responses are subject to both ence of green. If we simultaneously stimulated this
excitatory and inhibitory influences caused by in¬ unit with both a red and a green stimulus, the pos¬
teraction between neighboring units. We introduced itive and negative responses would cancel each
you to several such systems in Chapters 3 and 4. other and no signal would result. Thus, red and
In fact, in Chapter 4 you saw that many brightness green oppose each other, and the same unit can
phenomena can be explained by the presence of a never simultaneously signal both red and green.
spatially opponent mechanism on the retina, where Such graded potentials are usually called S poten¬
excitation in one region might cause inhibition in tials after their discoverer, Svaetichin, and the cells
another. If we could also find spectrally opponent that give these responses are called C-type horizon¬
organization, where stimulation by one wavelength tal cells (where the C stands for color). Also notice
of light causes excitation in a cell, and stimulation that there is another form of cellular response
by a wavelength in another region of the spectrum shown in the figure; it is marked luminance. This
causes inhibition of that cell’s neural response, then type of cell responds to the intensity of the light
we would have a physiological unit that corre¬ regardless of the wavelength. These L-type hori¬
sponds to the mechanism postulated by Hering. zontal cells could be the basis of the black-white
The first evidence that different wavelengths response hypothesized by Hering.
138 Chapter 5/Color

reduction of response rate as the spot begins to en¬

ter the inhibitory region of the receptive field. If
we repeat the experiment with a green spot, we find
that the cell appears unresponsive when the green
spot is in the center of the receptive field; however,
as the spot becomes larger, or it is moved into the
surround field, the resting level of activity is re¬
duced. Thus, we have a cell that has the property
of being exited by red and inhibited by green, if
the stimulus is the appropriate size and in the ap¬
propriate location on the retina. Of course, an equal
number of cells with the opposite organization
(green excitatory center, red inhibitory surround)
plus cells in which the centers are inhibitory and
the surrounds are excitatory are also found. The vi¬
sual system, once having come upon a particular
organizational scheme, seems to like to exhaust all
possible combinations (see Gouras & Zrenner,
1981; Jacobs, 1986).
Further along in the visual system at the lateral
geniculate nucleus, this particular arrangement can
easily produce cells that generate a spectrally op¬
ponent signal with appropriate stimulus arrange¬
FIGURE 5-15 Graded response of retinal cells to ments. In Chapter 3, we discussed some of the
various wavelengths of light. (Based on Svaetichin work of De Valois and his co-workers (De Valois &
& MacNichol, 1958)
De Valois, 1980). They found that cells in the lat¬
eral geniculate of monkeys were also color-coded,
When we reach the level of the retinal gan¬ similar to the color-coded retinal ganglion cells.
glion cells, there is clearly an opponent-process These units showed a resting level of activity (in
coding; however, there is also a spatial distribution terms of neural responses per unit time) even in the
to these responses (Boynton, 1979; De Monasterio, absence of any light stimulation. When the eye was
1978). The general form of this encoding involves stimulated by large spots of light the response pat¬
the center-surround organization of receptive fields tern changed. Some cells responded more vigor¬
that we discussed in Chapter 3. Suppose we shine ously when the eye was stimulated with short
a tiny red spot on the eye while recording from a wavelengths of light, and decreased their response
retinal ganglion cell. In some cases, as the size of rate below their spontaneous (dark) activity level
the spot increases, the vigorousness of the neural for long wavelengths of light. Other cells acted in
response increases up to a point. After that, further exactly an opposite manner. As with the S poten¬
increases in the size of the red spot have no further tials, two different classes of cells were reported.
influence on the cell’s response, Notice that this is Each had different patterns of response as a func¬
very different from the type of response seen when tion of wavelength, similar to what is needed for a
white light is used (as in Chapters 3 and 4), where red-green cell and a blue-yellow cell. Since the lat¬
increasing the size of the spot starts to produce a eral geniculate receives its input directly from the
 The Physiology of Color Vision 139

retinal ganglion cells, this is exactly the pattern of simply to the amount of luminance reaching the
results that we would expect. Thus, returning to eye. The spectrally tuned cells code both chromatic
our example above, if we have a red excitatory and spatial information in the responses. This
center in a receptive field, we should get increased means that whether a given wavelength will pro¬
response for a large area red light while the green duce an increase or a decrease in neural response
inhibitory surround would completely ignore its may also vary as the spatial position of the stimulus
presence. Conversely, a large green spot would spot is varied within the receptive field of the cell.
cause an inhibitory response and be ignored by the The understanding of the processing of color
red excitatory center, and so forth. information in the visual cortex has now become
Typical responses from lateral geniculate cells somewhat clearer, although much remains to be
can be seen in Figure 5-16. There are three cell learned. Hubei and Wiesel’s (1968) pioneering
types. One responds differentially to short and study, which recorded electrical responses from
moderately long wavelengths (blue-yellow), one single cortical cells, showed that some cells in the
responds differentially to moderately short and long monkey cortex respond differently as a function of
wavelengths (green-red), and one does not show the color of the test stimulus. Since then, evidence
different opponent processing, but rather responds suggests that perhaps 50 percent or more of cortical
cells are color-sensitive (Jacobs, 1976). This re¬
sponsiveness to color appears in addition to the re¬
sponsiveness to orientation, motion, and so forth,
found in both simple and complex cortical cells
(DeValois & Jacobs, 1984; Kruger & Gouras,
1980; Michael, 1981). Thus, in one complex cell,
we might find that the largest response is for a red
line of a particular orientation moving in a specific
direction, whereas another cell might show its
maximum response for a green line. An interesting
demonstration of the way orientation and color
responsiveness are intertwined can be seen behav-
iorally in the McCollough aftereffect (Harris,
1980; McCollough, 1965; Stromeyer, 1978). Dem¬
onstration Box 5-4 allows you to see this effect for
yourself.
As we saw in Chapter 3, visual coding in the
cortex seems to be highly organized, both across
and down into the cortical surface. Similar organi¬
zation is found for color-coding. Penetrating into
the cortex, there are vertical columns, or slabs,
where all the cells will be color-responsive,
whereas in others no color responsiveness appears
(Michael, 1981). Within a color column all cells
FIGURE 5-16 The neural response rate for cells
show color sensitivity, however each might be
in the lateral geniculate relative to their resting
response rate, for stimulation by lights of different tuned to different colors, orientation, or eye of
wavelengths. (Based on DeValois & DeValois, 1975) input.
140 Chapter 5/Color

DEMONSTRATION BOX 5-4. McCullough’s Demonstration of the Interaction

between Color and Form

The idea behind this demonstration is that through re¬ to Color Plate 6 and notice that there are two colored
peated exposure to colored stimuli at a particular ori¬ grids there, one containing vertical green lines and the
entation, the cortical cells that are tuned to that com¬ other horizontal red lines. To selectively adapt the
bination of color and orientation will become cortical cells, simply look at the green grid for about
fatigued. When we next show a set of white stimuli 5 seconds, then shift your gaze to the red grid for
at the same orientation, to which all color systems another 5 sec, then back to green grid again for 5 sec,
typically respond equally, the fatigued cells respond continuing this alternation for about 2 or 3 minutes.
more weakly, causing a color aftereffect. The partic¬ When that inspection period has passed, look back at
ular color seen is complementary to the adapting the figure in this box and you will find that it appears
color. Thus, for instance, if you fatigue the green re¬ to be colored. Now the vertical white bars appear red¬
sponse, normally white light will appear tinged with dish (the green response to white vertical lines has
red, fatigue of the blue response will produce a yel¬ been weakened) and the horizontal bars appear green¬
low aftereffect, and so forth. Of particular interest ish (the red response to white horizontal lines is di¬
here is the fact that these aftereffects are also orientation- minished). Notice also that turning the book side¬
specific and seem to be caused by the selective fatigue ways, or tilting your head so that the orientation of
of cortical cells tuned to color and orientation, al¬ the lines changes on your retina, will change the col¬
though other explanations of this phenomenon have ors of the lines. This demonstration indicates the in¬
been suggested (see Skowbo, 1984). timate relationship between color and form predict¬
To perform this demonstration, first notice that the able from our knowledge of the cortical coding of
figure in this box is completely achromatic. Now turn color and orientation.
 The Physiology of Color Vision 141

Penetrating the cortex vertically, we find types that have a double opponent process. Such a cell
of cells that show an increase in response when the might increase its activity when the center of its
eye is stimulated with some colors and a decrease receptive field is stimulated with red light, but ac¬
in response when stimulated with others. Once tually decrease its firing when the surround is stim¬
again, there is a spatial factor in this response. For ulated with red light. The opposite organization is
example, stimulating the center of the cell’s recep¬ seen for responses to green light in the same cell,
tive field with red light would cause the cell to in¬ with a green spot on the center of the receptive
crease its activity, whereas stimulating the flanks field producing a decrease in response and a green
with green light would cause the cell to decrease spot on the surround producing an increase in firing
its activity. In Layer 4 of the cortex, there are cells (Michael, 1985). Typical receptive fields of these
types are shown in Figure 5-17. Such double op¬
ponent-process cells would respond most vigor¬
ously to contrasting colors placed next to each
other, such as a red surrounded by a green. We still
have much to learn about color processing in the
central nervous system. However, a schematic out¬
line of the arrangement of color processing in a
section of striate cortex, as we currently know it,
is shown in Figure 5-18.

Models of Color-Coding

How can a four-primary, opponent-process (or

“push-pull”) system exist when we already have
provided physiological and psychophysical evi¬
A dence indicating that the retina operates with a

Broadband Broadband
(non-color selective) (non-color selective)

FIGURE 5-17 Receptive fields of typical color-

opponent (A) and double-opponent (B) cells, FIGURE 5-18 Arrangement of color-responsive
recorded in the cortex. cells in the striate cortex.
142 Chapter 5/Color

three-color pigment system? Hurvich and Jameson it required considerably less information to be
(1974) have suggested a neural wiring diagram that transmitted through each channel, thus providing
indicates the way cones, each containing only one good fidelity and increased economy. Perhaps sim¬
of three pigments, could produce opponent re¬ ilar considerations of economy and fidelity underlie
sponses at the postretinal level. An example of how the organization of our visual systems.
such a wiring diagram might work is shown in Fig¬ An alternative neural coding theory for color
ure 5-19. It requires only that certain cones excite was first suggested by Troland (1921) and recently
cells further along in the system, and that other revived. It claims that separate neural channels may
cones inhibit the response rates of those cells. En¬ not be needed for the various primary hues. In¬
gineers hit on a similar system when they designed stead, information may be sent through common
color television transmission. The color in the orig¬ channels, with the color information carried via a
inal scene is first analyzed into its red, green, and sort of neural Morse code. Here, specific patterns
blue components by the camera, and then trans¬ of neural responses creating specified time intervals
formed into two color-difference (or opponent-pro- between bursts of firing could signal various colors.
cess) signals (plus an intensity signal). After If this is the case, then it should be possible to
reception at its distant location, the signals are re¬ create the subjective impression of color by flick¬
converted into red, green, and blue signals by the ering white light on and off in a pattern that mimics
television set. This technique was selected because the usual neural code. Several investigators have
been able to do exactly this, namely, generate the
appearance of colors by pulsing white lights on and
off (Festinger, Allyn & White, 1971; Jarvis, 1977;
Piggins, Kingham & Holmes, 1972). Young (1977)
has even managed to do this by pulsing tiny electric
currents into the eye to simulate the supposed
neural code. One phenomenon that this theory can
explain very well is the appearance of subjective
colors. These are colors perceived in the absence
of the appropriate wavelengths of light, which can
be made to appear in certain flickering black and
white displays. A procedure for creating subjective
colors for yourself is shown in Demonstration Box
5-5. Interestingly, people who show color defects
for real colors also show the same color defects for
subjective colors (White, Lockhead & Evans,
1977).
Most researchers today accept the fact that
FIGURE 5-19 Schematic diagram indicating how several mechanisms may be involved in color per¬
a three-pigment system might be connected to ception. It seems that there is an initial trichromatic
produce opponent-process neural responses. The
coding at the level of the cones, followed by op¬
lines represent the connections. The round and
flat connections differ in that one is excitatory and ponent processing at the higher neural levels with,
the other is inhibitory (which is which is perhaps, some additional information provided by
arbitrary). Numbers indicate the wavelength of the time relationships in the overall patterns of
maximum sensitivity. neural response.
 Color Perception 143

DEMONSTRATION BOX 5-5. Subjective Colors

You have already encountered subjective colors in spinning it clockwise, the inner bands should be
Demonstration Box 1-2, where colors appeared in a slightly red, the next yellow, then green, and the last
stationary stimulus. A more powerful set of subjective blue or violet. The order of the colors should reverse
colors, produced by flickering black-and-white pat¬ if you spin the pattern counterclockwise. The color
terns, began as a toy invented by C. E. Benham in effects arise because of the specific patterns of flick¬
1894. It was painted on a top and meant to be spun; ering white and black set up by each band. These
hence the pattern is often referred to as Benham’s patterns mimic the flashing on-and-off light patterns
top. The pattern is shown in the figure. Cut out this used to study subjective colors in a laboratory setting.
pattern (or carefully reproduce it), and mount it on a If you alter the adaptive state of your eye by staring
piece of thin cardboard. Punch a hole in the marked at a white surface for a minute, you will notice that
center region and insert a nail or a round pencil. Now the perceived colors on each line will be different
spin the pattern as shown. Colors should appear when (Karvellas, Pokomy, Smith & Tanczos, 1979).
the pattern is spun at a moderate speed. If you are

COLOR PERCEPTION Intensify and Duration

The wavelengths of light that are present are not Increasing evidence suggests that color and bright¬
the only factors that determine our perception of ness information are carried by different visual
hue, as was demonstrated by the existence of sub¬ channels (Boynton, 1978, 1988; Favreau & Cavan-
jective colors. A number of factors, such as stim¬ agh, 1981). Perhaps the best evidence for this is
ulus intensity and duration, as well as the the fact that their time courses are much different—
characteristics of surrounding stimuli, can also alter responses to color changes are much slower than
the perceived color. responses to brightness changes (Bowen, 1981).
144 Chapter 5/Color

However, it is also clear that the perception of hue Prolonged exposure to colored stimuli also
may interact with the intensity of the stimulus produces a shift in the perception of hue. For in¬
(e.g., Emmerson & Ross, 1986). If intensity levels stance, if you viewed the world through a deep red
are low, only rods will be active and no color will filter for a sufficient period of time, you would find
be seen. But even beyond the cone threshold the that when the filter was removed the world would
perceived hue of a stimulus will change depending take on a blue-green tint. This fatiguing of a spe¬
on the stimulus intensity. Specifically, if we in¬ cific color response is called chromatic adapta¬
crease the intensity of red or yellow-green stimuli, tion. It is believed that these adaptation effects are
they not only appear brighter but begin to take on due either to selective bleaching of one particular
a more yellow hue. Similarly, blue-greens and vi¬ photopigment or to the fatigue of one aspect of the
olets begin to appear bluer when the intensity is neural response of an opponent-process system
increased. This phenomenon is called the Bezold- (e.g., Vimal, Pokomy & Smith, 1987). Suppose
Brucke effect, in honor of its two discoverers. It you look through the red filter for a long period of
is quite easy to demonstrate, as is shown in Dem¬ time. The red-catching pigment becomes bleached,
onstration Box 5-6. Although the basis of the Be- or the red response in the red-green opponent-pro-
zold-Brucke effect is not yet fully understood, it is cess cells becomes fatigued. Now, when you view
clearly neural in its origin (Coren & Keith, 1970; a white surface, the absence of red pigment (or the
Nagy, 1980). This effect may come about because weakness of the red response) causes the blue and
the red-green opponent-process cells are slightly green systems to account for a greater proportion
more sensitive than the blue-yellow cells. Thus, we of the total activity. This gives the white a cyan
can discriminate between red and green at lower (blue-green) tint. When such fatigue effects due to
intensity levels. Because the blue-yellow units be¬ prolonged stimulation are localized (that is confined
come more active at higher intensity levels, hues to one region of the retina), they are called after¬
tend to be dominated by these colors when stimuli images. Demonstration Box 5-7 provides a stimu¬
are bright (Hurvich, 1981). lus for the production of color afterimages. You

DEMONSTRATION BOX 5-6. The Bezold-Brucke Effect

For this demonstration you will need three pieces of demonstration of the hue shift, associated with in¬
colored cellophane, glass, or celluloid to serve as creasing stimulus intensity, called the Bezold-Brucke
color filters. One should be a red, the other a green, effect.
and the last a yellow. Take a white sheet of paper Another way to see this effect is to simply look at
that is brightly illuminated with room lighting, and an incandescent light bulb (60-100 W) through the
cast a shadow over one-half of the paper. Looking red or the green filter. You will notice that the light
through the red filter, you will notice that the hue of bulb appears to be yellow, despite the presence of the
the red seen on the bright half of the paper is notice¬ filter. Since the red filter only allows the long (red)
ably yellower than the hue seen on the shadowed por¬ wavelengths of light to pass, and the green only al¬
tion. When you peer through the green filter, you lows the middle (green) wavelengths through, no yel¬
should experience the same effect. However, looking low is reaching your eye. The yellow appearance of
through a yellow filter should not cause an apparent the bulb is caused by the Bezold-Brucke hue shift that
change in hue. Thus, the brighter you make a red or occurs when the intensity of the stimulus is high.
a green, the more yellow it will appear. This is a
 Color Perception 145

DEMONSTRATION BOX 5-7. Color Afterimages

You can easily demonstrate negative or complemen¬ plement of the pattern originally viewed. Where the
tary color afterimages using Color Plate 7. Here you red patch was, you will see green; where the green
see four square patches of color: red, green, blue, and patch was, you will see red; where the blue patch
yellow. Notice that there is a black x in the middle of was, you will see yellow; and where the yellow patch
this pattern. Stare at the black x for about 2 minutes was, you will see blue. These are the complementary
while keeping the plate under reasonably bright illu¬ color afterimages caused by the fatiguing of the var¬
mination. At the end of this period, transfer your gaze ious color responses during the time you were staring
to the black x to the right of the figure. You should at the color patches.
see a pattern of colored squares that is the exact com¬

will notice when performing this demonstration that we have a situation where the red response systems
the hue of the afterimage tends to be a complemen¬ exposed to the surround are highly activated. In
tary hue of the stimulus producing the afterimage. turn these active neurons will inhibit the red re¬
sponse in the neurons exposed to the central gray
patch. Since the red and green responses are usu¬
Spatial Interactions
ally in balance, inhibition of the red response re¬
In Chapter 4 you learned that the brightness of a sults in the emergence of the complementary, or
stimulus could be affected by the intensity of adja¬ opponent, green response in this region. A tinge of
cent stimuli. The general nature of the interaction green hue is then seen in the gray. As one might
was inhibitory, so that a bright surround made a expect, if this is truly an inhibitory interaction such
central area appear dim. Inhibitory interactions be¬ as we observed in brightness effects, it is even pos¬
tween adjacent color systems can also occur and sible to produce colored Mach bands (Ware &
they result in hue shifts. The phenomenon is called Cowan, 1987).
simultaneous color contrast. Consider Color Plate Contrast-induced colors act very much like
3. Notice that this figure has four brightly colored real colors in their ability to produce other percep¬
patches, each of which surrounds a small central tual effects. For instance, Anstis, Rogers and
square. The square on the red patch appears to be Henry (1978) induced very strong contrast colors
slightly green, and that on the green appears to be on surrounded gray patches (as in Color Plate 3)
slightly red. The square on the blue patch appears and found that observers developed negative after¬
to be slightly yellow, and that on the yellow patch images to the contrast colors, just as though they
appears to be slightly blue. However, each square had been viewing real colors!
is exactly the same gray. You might be able to in¬
crease the strength of this effect by viewing Color
Age and Physical Condition
Plate 3 through a sheet of tracing paper or thin tissue.
Jameson and Hurvich (1964) suggested that si¬ An individual may have normal color vision when
multaneous color contrast arises from mechanisms tested at one stage in the life span, but may show
similar to those that cause brightness contrast, color discrimination defects when tested at a later
namely, an active neuron tends to inhibit the re¬ stage. Aging seems to alter color vision. Perhaps
sponding of adjacent neurons. In the case of the this is because the crystalline lens of the eye grows
gray square on the red background, for example, more yellow as an individual ages; hence we look
146 Chapter 5/Color

through a gradually darkening yellow filter (Coren are shown color samples and later asked to match
& Girgus, 1972a). Other effects, such as the loss them from an array of colored chips, systematic er¬
of cone pigment with age (Kilbride, Hutman, Fish¬ rors are made. Observers tend to pick chips of
man & Read, 1986), may also account for changes greater brightness when asked to remember bright
in color vision. Generally speaking, aging seems to colors and greater darkness when asked to remem¬
bring about a gradual deterioration of blue vision ber dark colors (Bartleson, 1960; Newhall, Bum-
(Lakowski, 1962; Verriest, 1974). Most individuals ham & Clark, 1957). When asked to remember and
are unaware of such changes because the onset is match colors of familiar objects with characteristic
quite slow; however, the effect gradually accumu¬ hues, we remember apples or tomatoes as being
lates. Since the perception of hue is subjective, you more red than the actual objects, bananas are more
seldom have opportunities to assess whether your yellow in memory than in the bunch, and remem¬
hue perception agrees with that of others. Does bered grass is greener than it is on the lawn. Be¬
your red appear to be the same as that of your cause of this memory effect, many film
friends? Clearly, this is an unanswerable question. manufacturers have chosen to modify the spectral
Physical conditions can also result in losses in reproduction ability of color film so that the repro¬
the ability to discriminate colors. Such acquired duced colors are richer than they are in nature.
color vision losses are called dyschromatopsias. Since television engineers have not made a similar
There are several diseases or physical conditions correction, color memory distortions may account
that lead to such dyschromatopsias. The most com¬ for part of our feeling that the picture reproduced
monly observed losses are for sensitivity to blue on a color television set is an unfaithful reproduc¬
(see Pokomy & Smith, 1986), which are observed tion of real color.
in diabetics (Lakowski, Aspinall & Kinnear, 1972), Memory color effects tend to creep into cer¬
individuals with glaucoma (Lakowski & Drance, tain other matching tasks. For instance, if you are
1979), and alcoholics (Reynolds, 1979). These asked to match the color of a Valentine’s Day heart
color losses can be aggravated by a number of fac¬ or an apple, both of which have been cut out of
tors. For instance, diabetic women who take oral orange paper, you will match them with a redder
contraceptives show significantly greater discrimi¬ hue than you would use to match an oval or a tri¬
nation losses in the blue range (Lakowski & Mor¬ angle cut out of the same material. A banana¬
ton, 1977). Acquired problems with the red-green shaped figure, or one labeled lemon, is matched
system are rarer and usually are associated with with a yellower hue. It seems as if the remembered
cone degeneration or optic nerve diseases (Pinkers color blends with the observed stimulus, altering
& Marre, 1983). the percept toward the ideal, or prototypical, color
of an object (Bruner, Postman & Rodrigues, 1951;
Cognitive Factors in Color Perception Delk & Fillenbaum, 1965; Harper, 1953; White &
Montgomery, 1976). The color you remember is
Although color is a basic sensory experience, there
probably “better” than the color that is present;
are also nonsensory factors that affect the perceived
however, the color you see now may be tinged by
color of an object. In addition, color may interact
your memory’s hue.
with other nonperceptual behaviors.
Culture and Color
Memory Color
As we noted earlier, an English speaker is content
The remembered color of familiar objects often dif¬ to describe hue differences using four basic cate¬
fers from the objects’ actual color. When observers gories: red, yellow, green, and blue. This is not the
 Color Perception 147

case for many other languages, some of which have quences. It delights and depresses. It makes hu¬
no separate names for green and blue, or the same mans feel warm or cold, tense or relaxed. For
name for yellow and green, or red and yellow. instance, a manufacturer of detergent found that the
There are some languages that only distinguish red color of the detergent box made a difference in how
as a separate color and have no names for the other the user evaluated the strength of the detergent.
hues. It is often argued that there is an interaction Women were given the same detergent in three
between language and perception, and that when boxes, which differed only in color. When the de¬
separate names for separate sensory experiences tergents were rated after use, the women felt that
exist, these labels make discriminations easier. In the detergent in a yellow-orange box was too
other words, the Lakuti tribe, who only have a sin¬ strong, and supposedly had ruined some of the
gle term for blue and green, may see the two colors clothes. The detergent in a blue box was too weak,
as being more similar to each other than English whereas the one with both blue and yellow-orange
speakers, who have separate words for these stim¬ flashes seemed to be most effective (Kupchella,
uli (Whorf, 1956). The suggestion that different 1976).
language terms for colors indicate different percep¬ Color can even produce sensory impressions
tual abilities has been presented in many different that are characteristic of other senses. It is almost
forms. For instance, Robertson (1967) suggested universal to call the short-wavelength (blue) colors
that there has been an evolutionary development in “cool,” and the longer wavelengths (yellow)
both the color-perceiving ability of humans and in “warm.” Perhaps these labels arise because the
the color terms encoded in the language. He sug¬ cool of the night is first broken by the red of the
gested that the first discriminations were between dawn, with midday characterized by the yellow of
red and green, then the discrimination ability for sunlight and warmth. As the yellow begins to dis¬
yellow evolved, and finally that for blue. He ana¬ appear and the blue of twilight begins to predomi¬
lyzed a number of ancient languages and found nate, temperatures again grow cool. Many years
such evolutionary trends. One could conclude from and many generations of such an association might
such evidence that the ancient Greeks were rela¬ stamp this warm-cool relationship into our lan¬
tively weak in their ability to perceive colors be¬ guages. In an era when the conservation of energy
cause their language had only a limited set of color is important, it is interesting to note that people
names. will turn a heat control to a higher setting in a blue
Actually, when the ability of individuals to room than they will in a yellow room, as if they
match, discriminate, or reproduce colors (rather are trying to compensate thermally for the coolness
than just to name colors) is measured directly, the that has been visually induced (Boynton, 1971).
picture changes. It seems as if the number of color Similarly, Alexander and Shansky (1976) have
names in a language does not affect the ability to shown that dark saturated colors are perceived as
make such discriminations (Berlin & Kay, 1969; being associated with a greater sensation of
Bomstein, 1973, 1975). These findings indicate the “weight” or “heaviness.” All these findings em¬
danger, in the absence of perceptual measurements, phasize that color is a psychological achievement,
of assuming that language usage directly reflects not simply a direct effect of the physical variation
perceptual abilities. of wavelengths of light. If you still doubt this state¬
ment it will probably be instructive to turn back to
Color Impressions Demonstration Box 1-2 or 5-5 to see colors de¬
Color does more than provide us with additional velop in your mind where no physical variations in
information about stimuli, it has emotional conse¬ wavelength exist.
148 Chapter 5/Color

GLOSSARY Dichromats Individuals whose color vision is defec¬

tive, allowing all hues to be matched with two rather
than three primaries.
The following definitions are specific to this book.
Dominant wavelength The wavelength of a mono¬
Additive color mixture A color mixture resulting chromatic stimulus that best approximates the hue of a
from the addition of light of one wavelength to light of color mixture.
another, e.g., the projection of a blue light on top of a
Dyschromatopsias Acquired color-vision losses.
red light on a screen to produce magenta.
Erythrolabe Red-sensitive cone pigment.
Afterimage A visual sensation that appears after an
Habituation The process by which an observer
intense or prolonged exposure to a stimulus.
ceases to respond, or reduces the magnitude of a re¬
Anomalous trichromatism A defect in color vision
sponse, to a repeated stimulus.
in which color matches made by an individual are sys¬
Hue The term denoting the psychological dimension
tematically different from normal, although the three pri¬
most clearly corresponding to the wavelength of light and
mary color systems are still functioning.
most often termed color in common language.
Benham’s top A black-and-white pattern that when
Metameric colors Colors that appear to be the same
rotated produces subjective colors.
but are composed of different wavelengths.
Bezold-Brucke effect The shift in the apparent hue
Microspectrophotometer A device for measuring
of a color as the intensity is changed.
the amounts and wavelengths of light emanating from
Brightness The psychological impression of light in¬
microscopic target areas.
tensity.
Monochromatic stimuli Stimuli that contain only
Chlorolabe Green-sensitive cone pigment.
one wavelength of light.
Chromatic adaptation A weakened response to a
Monochromats Individuals who see color as mere
color stimulus due to previous exposure to other chro¬
gradations of intensity, because of the absence of any
matic stimuli.
functioning cones.
CIE chromaticity space A variant of the color tri¬
Opponent process Neural process that signals the
angle system, using three imaginary “super” primaries.
presence of one color by increasing its activity and of an
Color atlas A book in which each page represents a opposing color by decreasing its activity.
horizontal or a vertical slice through a color space.
Primaries Three monochromatic light sources that
Color blindness A condition in which individuals when combined in appropriate amounts can match any
lack the ability to make discriminations on the basis of other hue.
wavelength of light.
Protanopia A form of color blindness resulting in
Color circle See Color wheel.
the confusion of reds and greens due to insensitivity in
Color solid See Color spindle. the red system.
Color spindle A three-dimensional model of color Protanomaly A condition in which an individual’s
appearance in which the relationships between hue, color matches require more green than those of a color-
brightness, and saturation are depicted. normal individual.
Color wheel A circular scheme in which colors are Purity A spectrally pure stimulus is composed of
represented according to hue, with complementary colors only one wavelength.
placed directly across from each other.
S potentials Graded electrical retinal-cell responses
Complementary colors Color pairs whose mixture that vary in direction and strength depending on the
produces an achromatic gray or white. wavelength of the stimulus.
Cyanolabe Blue-sensitive cone pigment. Saturation The psychological attribute of a color as¬
Deuteranomaly A condition in which an individu¬ sociated with “how much” of a hue is present.
al’s color matches require more green than those of a Simultaneous color contrast A process in which in¬
color-normal individual. hibitory interactions between adjacent color systems
Deuteranopia A form of color blindness associated cause hue shifts.
with the confusion of reds and greens due to insensitivity Spectral colors Pure monochromatic stimuli, such
in the green system. as those in a prismatic spectrum.
 Glossary 149

Subjective colors Colors that are consciously expe¬ Trichromatic theory The theory that color vision is
rienced, but not associated with any wavelength change based on three primary responses.
in the physical stimulus. Tristimulus values The combination of the stimulus
Subtractive color mixture A color mixture resulting hue and brightness used in the CIE color system for de¬
from the subtraction or absorption of light of various termining any color stimulus.
wavelengths, e.g., the mixture of yellow and blue pig¬ Tritanopia The color defect in which yellows and
ments to produce green. blues are confused due to reduced blue sensitivity.
' CHAPTER
6
The Auditory
System
•

SOUND
A Simple Sound Wave
THE STRUCTURE OF THE EAR
Evolution and Anatomy of the Ear
Physiology of the Human Ear
Wave Motion on the Basilar Membrane
Mechanisms of Transduction
ELECTRICAL ACTIVITY OF THE AUDITORY
NERVE
THE AUDITORY PATHWAYS
Electrical Activity of the Lower Auditory
Centers
THE AUDITORY CORTEX

151
152 Chapter 6/The Auditory System

I t is one of those strange historical occurrences

that while studying the physiology of the ear in
lide with others, causing air compression as the
string moves forward and rarefaction as it moves
order to help the deaf deal with the world of sound, back. This movement results in a wave of mechan¬
Alexander Graham Bell developed the telephone. ical energy, as is shown in Figure 6-1.
Perhaps the leap was not too great however, for if Sound waves are alternations of rarefaction
the eye is our window to the world then the ear and compression of an elastic medium (e.g., water,
must be our microphone. Just as an understanding air, or walls) in which they travel and are created
of the physical nature of light and of the anatomy by rapid movements of a source in mechanical con¬
and physiology of the eye is important to under¬ tact with the medium. The medium acquires some
standing the psychology of vision, an understand¬ of the movement energy of the source and transfers
ing of the physical nature of sound and of the it to other parts of the medium by means of colli¬
anatomy and physiology of the auditory system is sions between the molecules of the medium. Sound
crucial to our understanding of hearing. waves can be transmitted for great distances, al¬
The ear is a remarkable physical instrument. though the individual air molecules simply move
An engineer trying to duplicate the function of the back and forth over very small distances. The col¬
ear would have to compress a sound system of im¬ lisions of molecules, of course, are not perfectly
mense complexity into a space of approximately 2 efficient in transferring the original collision en¬
cubic centimeters (Stevens & Warshovsky, 1965). ergy, so a sound wave tends to become less intense
For example, the major functional part of the ear, as it moves farther away from the original source.
the cochlea, has over 1 million moving parts (Hud¬ Consequently, its ability to move or vibrate other
speth, 1985). So much complicated apparatus sug¬ objects decreases. Since sound involves the vibra¬
gests that the path of sound waves, from their tion of parts of the medium through which it trav¬
origin in the external environment to the final stage els, it cannot pass through a vacuum. The necessity
where we “hear” a sound, involves several differ¬ of a medium for the existence of sound waves was
ent mechanisms and stages of information pro¬ demonstrated by Robert Boyle in 1660, when he
cessing. pumped the air out of a jar and then failed to hear
the sounds made by a watch suspended by a thread
in the jar.
SOUND The speed of sound varies according to the
medium in which it travels. The elasticity and den¬
Sound is a form of mechanical pressure. If you sity of the medium are important, with sound trav¬
have ever attended a rock concert, you probably eling faster in a denser or more elastic medium
have felt the actual mechanical pulsations, espe¬ where molecules are closer or more strongly con¬
cially those from the bass instruments, that may nected to each other. The speed of sound is rather
cause the floor, seat, or the air about you to vi¬ slow compared to that of electromagnetic waves
brate. You are feeling the results of the movements such as light, being approximately 340 m/sec in air
of air molecules being pushed forward in waves by and about 1,360 m/sec in water.
the cones of the speakers. It may be easier to un¬
derstand this phenomenon if we consider what hap¬
A Simple Sound Wave
pens when we pluck a guitar string, causing a
sound. We can see the string vibrate, moving rap¬ Any sound wave may be described by specifying
idly back and forth in space. This movement causes certain values. The simplest wave is sinusoidal,
the strand of steel or nylon to collide with the air so-called because the trigonometric sine function
molecules around it. These molecules in turn col¬ describes it mathematically. Figure 6-1 shows the
 Sound 153

Guitar string
(exaggerated)

Direction of sound

3 ••• • • • • • • • •••••••••••••• • • • • am

Compression Rarefaction

4 • • • ••••••••• •••••••• ••••••••• • • • • •

FIGURE 6-1 The nature and description of a simple sound wave in air.
154 Chapter 6/The Auditory System

alternate compressions and rarefactions of air mole¬ measures. When dealing with sound in terms of en¬
cules over time from the vibrating string of a gui¬ ergy units, we can speak of the difference between
tar. When we plot the air pressure as it varies over two levels by asking by how many powers of 10
time we get the wave shown in the figure. Its (the logarithm) one energy level exceeds another.
wavelength (represented by the Greek letter If one energy level is 1 million times greater than
lambda, \) is the distance from one peak to the another (106 times greater), we say that it is 6 bels
next peak, representing a single cycle of the wave. greater (a measurement unit named after Alexander
The frequency (f) of the wave, by convention, is Graham Bell). It is more convenient, however, to
the number of cycles the wave is able to complete speak in terms of sound pressure level (SPL),
in one second. The unit used to measure frequency which represents the actual force against the ear,
is the hertz (Hz), named after the German physi¬ than in terms of sound energy. Because of the na¬
cist Heinrich R. Hertz, one hertz being equivalent ture of the mathematical relationship between en¬
to one cycle per second. The frequency of a sound ergy and pressure amplitude, the number of bels is
wave is important to our later discussion, since it doubled when we speak of ratios of sound pressure
is related to (although not the same as) the per¬ levels. Since a bel is a rather large unit relative to
ceived pitch of the sound. The range of frequencies normal hearing levels, the most common unit used
that seem to have pitch for most people is from is the decibel (dB), which is one-tenth of a bel.
about 20 to 20000 Hz. Sounds below 20 Hz are The formula for decibels is
sensed as vibration, whereas sounds above 20000
Hz are not heard at all except by young children. Number of dB = 20 log (P/P0)
Other animals, such as bats or dolphins, can hear
frequencies several times as high as the upper limit where P is the sound pressure level we wish to ex¬
for humans. press in decibels, and P0 is the standard reference
The intensity of a sound wave is related to its level. The standard reference level is psychologi¬
pressure amplitude. This is a measure of cally meaningful, since it is the average value of
compression or rarefaction in the air or other me¬ the threshold for sound (0.0002 dynes/cm2) mea¬
dium at the peaks or valleys of the sound wave. sured in young adults at 1000 Hz. Decibels are par¬
For a sine wave in air, the pressure amplitude is ticularly suited to express the relationships between
the maximum amount by which the wave causes sound pressure levels since they compress the large
the air pressure (force per unit area expressed as range of possible pressures into more manageable
dynes/cm2) to differ from the normal atmospheric units. Table 6-1 gives typical values of sound pres¬
pressure (which is about 1,000,000 dynes/cm2). sure levels expressed in decibels for some represen¬
The maximum pressure variation the ear can tol¬ tative sounds. The table shows that as the measured
erate is about 280 dynes/cm2 above or below at¬ intensity of a sound increases, subjective loudness
mospheric pressure, whereas the minimum pressure also increases. Intensity is related to (but not iden¬
variation detectable (average for young adults) is tical to) loudness.
about 0.0002 dynes/cm2. For these threshold-level A final important parameter of sound waves is
sound waves, the air molecules are displaced (on phase, which is important when two or more sim¬
average) about 0.0000000001 cm, which is about ple waves are compared. Phase refers to the partic¬
one-tenth the diameter of an average air molecule. ular part of the compression-rarefaction cycle a
Obviously the ear is an extremely sensitive organ wave has reached at one instant of time. If two
with a broad response range. waves are at exactly the same part of their respec¬
In order to express conveniently this wide tive cycles at the same instant (so that their peaks
range of sound sensitivity, we use some special and valleys coincide), they are said to be in phase.
 Sound 155

Table 6-1. Sound Pressure Levels (intensity ing. Only a few sound sources, such as tuning
levels) of Various Sound Sources forks or electronic instruments, produce “pure”
sounds, which are composed of a simple sine wave
Source Sound Level (dB) variation of compression and rarefaction. Sounds
produced by musical instruments, the human voice,
Manned spacecraft launch 180
automobiles, waterfalls, and so on, have enor¬
(from 45 m)
mously complex cycles of compression and rare¬
Loudest rock band on record 160
faction. These complexities result from the
Pain threshold (approximate) 140
Large jet motor (at 22m) 120
interaction of many different waves of different fre¬
Loudest human shout on quencies and phases. Such complex waveforms
record 111 produce the timbre of sounds (see Chapter 7 for a
Heavy auto traffic 100 discussion of how this occurs). We can differen¬
Conversation (at about lm) 60 tiate among the sounds of a trumpet, a clarinet, a
Quiet office 40 piano, and a violin quite easily, because the wave
Soft whisper 20 forms they produce, even when they are playing
Threshold of hearing 0 the same musical “note,” are quite different. You
can see in Figure 6-2 that instruments produce very
complicated waveforms. Demonstration Box 6-1
If their peaks and valleys do not coincide, the two provides a method by which you can experience
waves are out of phase. How much they are out of our extraordinary ability to recognize complex
phase is expressed in terms of phase angle. A sin¬ sounds through their timbre.
gle cycle is assigned 360 degrees (as in circular Complex sounds, such as those in Figure 6-2,
motion); thus, a portion of a cycle can be specified can be described most usefully by analyzing each
by number of degrees from 0 to 360 (see Figure 6-
1). If one wave is at its 90-deg point (its peak)
when another wave is at its 180-deg point (crossing
the zero pressure-difference line), then the two Clarinet
waves are 90 deg out of phase.
If we remember that these waves simply de¬
scribe increases and decreases in mechanical pres¬
sure at various moments in time, it should be clear
that different sound waves (patterns of pressure)
that occur at the same time can interact with each
other. If two same-frequency waves are perfectly
in phase (0 deg out of phase), their peak and min¬
imum pressures coincide; hence, they add strongly
to each other’s intensities. When two waves of the
same frequency are 180 deg out of phase, one
reaches its minimum when the other is reaching its
maximum and they cancel each other’s effects;
therefore, we would not be able to hear the inter¬
acting sound waves.
Typically, everyday sounds are more complex FIGURE 6-2 Complex sound waves produced by
than the simple sine waves we have been discuss¬ three musical instruments.
156 Chapter 6/The Auditory System

DEMONSTRATION BOX 6-1. Perception of Timbre

Perhaps the most primitive musical instrument is the

human hand, used (usually in pairs) to clap rhythms.
As in any musical instrument, the shape and orienta¬
tion of the surrounding parts will alter the complex
components of the resultant sound, hence the timbre
that we hear. There seem to be only a few basic ways
of clapping, and people have the remarkable ability
to distinguish which is occurring from the sound
alone (Repp, 1987). Try holding your hands in the
configuration shown in part A of the figure so that
your hands are aligned and flat. Clap a few times,
listening closely to the sound. Now hold your hands
in configuration B, with your hands oblique and
slightly cupped. Clap a few times, again listening
closely to the sound. Position B generates more low
frequency sounds in the mixture than position A. You
should be able to distinguish the different claps quite
clearly. You can also hear differences if the palms are
crossed while clapping, or if you clap with your fin¬
gers around your palm, and so forth. It might be fun
to have someone else now clap while you are not
looking, and see if you can approximate what their
hand positions are. If you can, you are responding to
the timbre of the sounds, and performing some form
of analysis of the complex sounds actually present
into their constituent components, which then allows
you to recognize their source.

one into a set of simpler sine waves, which when mathematical ratios. Figure 6-3 shows an example
added together would produce the more compli¬ of the decomposition of a complex waveform into
cated waveforms. This method was invented by the its Fourier components. Speech sounds may also
French scientist Jean B. J. Fourier in the course of be analyzed into their Fourier components, with re¬
his studies of heat conduction. Fourier proved a sults that are very useful for the understanding of
mathematical theorem that states, in essence, that speech perception (see Chapter 12). The ear itself
any waveform that is continuous and periodic can acts as a sound analyzer, decomposing complex
be represented as the sum of a series of simple sine sounds into their individual components. This fact
waves with appropriate wavelengths, phases, and is known as Ohm’s acoustical law after the phy¬
amplitudes. As it turns out, these simple waves of¬ sicist George Ohm. You may demonstrate this ef¬
ten have frequencies related to each other by simple fect for yourself using Demonstration Box 6-2.
 The Structure of the Ear 157

bins, 1980; van Bergeijk, 1967). All vertebrate ears

seem to have evolved from the sense of touch.
Complex
wave Whether primitive or advanced, they seem to be
specializations of groups of cells with protruding
hairs, much like those found on the skin of your
arm. In fact, we can use the skin to demonstrate a
number of phenomena associated with human hear¬
ing (Bekesy, 1960).
One of the first steps in the evolution of the
modem mammalian ear is the lateral line. The lat¬
eral line is a linear array of nerve endings in the
Fourier skin of fish and some amphibians from which pro¬
components
trude jellylike masses in which are embedded sen¬
sory hairs. Often these hairs are pigmented and
form a horizontal line that runs the length of the
animal’s body. As the water moves or vibrates be¬
cause of sounds made by prey or predators or other
FIGURE 6-3 Fourier components (sine waves) of stimuli these sensory hairs bend. In a sense the fish
a complex sound wave. does not so much “hear” sound as “feel” it. The
lateral line system is especially useful under con¬
ditions of low light levels, such as in the Arctic or
THE STRUCTURE OF THE EAR
Antarctic during the winter. For example, the lat¬
eral line system in one Antarctic fish is tuned very
Evolution and Anatomy of the Ear
precisely to the sounds made by the plankton it
The human ear is a complex piece of biological feeds on (Montgomery & MacDonald, 1987).
engineering, yet biologists have traced its origins In addition to the lateral line system, some
to simple organs in quite primitive animals (Steb- types of fish have primitive internal ears that work

DEMONSTRATION BOX 6-2. Ohm’s Acoustical Law

This demonstration is done with a piano or a guitar, fact that the chord formed by these notes is quite a
but if neither is available use three glasses filled with complex sound pattern. Notice that the individual
water to different heights so that they produce a fairly sounds do not lose their identities, and can be dis¬
high note, a middle note, and a low note when tapped criminated from the others in the complex sound.
with a butter knife. Now have some friends strike the With enough practice a person can learn to separate
high, middle, and low notes simultaneously a few as many as six or seven different components of a
times. Without telling you, have them drop out one complex chord or “clang.” The separation of sound
note, sounding only two a few times, then put it back components by the auditory system is known as
in. Notice that it is quite easy to determine which of Ohm’s acoustical law.
the three notes was added or subtracted, despite the
158 Chapter 6/The Auditory System

on much the same principles as do human ears. It yrinth, rather than the single bone found in these
is believed that these internal ears evolved from a other species. Bekesy (1960), in a series of detailed
specialized, deeply sunken part of the lateral line studies, established that all these various types of
system. In such fish, vibrations in the water cause ears function in a similar manner. He was able to
similar vibrations in an air bladder in the fish’s link many of the performance differences with dif¬
body cavity. From there the vibrations are passed ferences in the physical properties of the ears (such
on to a series of small bones, derived from some as in the length of the cochlea). Thus, the human
of the vertebrae and ribs, and then to a complex ear, with which we will be concerned in the re¬
organ called the labyrinth, whose looping passages mainder of this chapter, is a part of a large family
are filled with fluid. This fluid contains hairs that of roughly equivalent organs. This fact makes it
are sensitive to the movements of the liquid in the possible to extend the results of studies of other
labyrinth caused by the vibrations picked up by the mammalian ears to the human auditory system.
air bladder. These hairs send the auditory informa¬
tion to the fish’s brain by way of sensory nerves.
Physiology of the Human Ear
In many of its elements, this system is quite similar
to that found in humans, even to the composition We shall now follow a sound wave through the
of the fluid in the labyrinth. Many other animals structure of the human ear, and trace the neural
have ears that are considered to be developments pathways to the cortex of the brain. The ear can be
from these primitive versions. An interesting tran¬ divided into three major parts, the outer, middle,
sition form is that of the frog. A tadpole has a hear¬ and inner ears. Figure 6-4 is a schematic represen¬
ing apparatus similar to those of fishes. Sound tation of the human ear. The outer ear consists of
waves are picked up by the (not fully functional) the pinna, which is the fleshy part of the ear visible
lung and transmitted to the labyrinth. During meta¬ from the outside. Only mammals have pinnae, and
morphosis into a frog, however, the ear changes. It they are thought to function mainly to channel the
develops an external membrane (the eardrum) that sound waves into the auditory canal, although
takes the place of the lung in the tadpole hearing they may also be involved in the localization of
system. sound (see Chapter 7). Some mammals, such as
Another major component, which is common bats and dogs, have highly mobile pinnae that al¬
to all mammals and birds and also appears in the low them partially to select the direction from
crocodilian reptiles, is the cochlea. The cochlea is which sounds are received. The sound waves that
a specialized development of the labyrinth that con¬ enter the auditory canal are channeled along it until
tains a long membrane covered with sensory cells they encounter the eardrum (or tympanum). The
from which (of course) hairs protrude. Its name eardrum vibrates in resonance with incoming sound
(which means “shell”) is derived from its coiled waves, moving back and forth at a high rate for
seashell appearance in mammals. high-frequency sounds, and more slowly for low-
All mammalian ears have the same basic frequency sounds. As we mentioned when discuss¬
parts, although they differ somewhat in proportions ing sound, these vibratory movements are quite
(with the elephant of course having one of the larg¬ small. For frequencies in the middle of the audible
est). There are also differences in sensitivity. Bats, range (1000-6000 Hz), the shape of the auditory
dolphins, and dogs have extraordinarily keen hear¬ canal helps to concentrate the sound and to increase
ing over a very wide range of frequencies. The ears its force against the eardrum.
of mammals differ from those of birds, reptiles, The middle ear consists of a set of three tiny
and fish in that mammalian ears typically have bones (ossicles): the malleus (“hammer”), the in¬
three small bones to transmit vibrations to the lab¬ cus (“anvil”), and the stapes (“stirrup”). These
 The Structure of the Ear 159

bones transmit the vibrations of the eardrum to the drum. In total, the middle ear amplifies sounds by
transducer mechanism located in the inner ear. a factor of about 30. This amplification is needed
Some interesting and important subtleties of oper¬ because the tiny motions of the air molecules
ation of this part of the ear may help explain why caused by the sound wave ultimately must cause
the complex system of ossicles evolved. The mid¬ the entire body of cochlear fluid to vibrate. Without
dle ear amplifies sound waves in two ways as it this amplification, only very intense sounds would
transmits them to the cochlea. First, the energy of be heard.
a sound wave may be approximately doubled by its A second important aspect of the operation of
transmission through the ossicles since they operate the middle ear is the ability of the ossicle system
mechanically as a system of levers. Second, the to selectively decrease the amplification it provides
area of the oval window is only about 1/15 that of in order to protect the ear from damage by intense
the vibrating area of the eardrum. Elementary sounds. This is accomplished by means of a change
physics tells us that a given force applied to two in its relative orientation against the oval window.
surfaces of different areas yields a greater force per Sounds in the normal range of intensity cause the
unit area on the smaller surface; thus, the forces stapes to push directly on the fluid in the cochlea.
applied to the oval window by the stapes are an For very intense sounds, however, the angle at
additional 15 times greater than those at the ear¬ which the stapes moves changes and the effect of
160 Chapter 6/The Auditory System

the stapes on the cochlear fluid is greatly reduced. process equalizes the air pressure on both sides of
In addition, a muscle attached to the stapes con¬ the eardrums. Sometimes, for example when we
tracts via a neuromuscular reflex when intense have a head cold, the eustachian tubes become
low-frequency sounds strike the ear, while another blocked, and the pressure in the middle ears cannot
muscle attached to the eardrum contracts to stiffen adjust to that of the outside air. Also, when we
the eardrum. These mechanisms make the eardrum climb to cruising altitude in a commercial jetliner,
less able to vibrate in sympathy with sounds in the the cabin pressure may become considerably lower
canal, and, taken together, all these mechanisms than the pressure within our middle ears at the time
tend to protect the ear from long-lasting excessive of takeoff. Such inequalities in internal versus ex¬
stimulation by loud sounds. ternal air pressure can cause you to experience tem¬
The bones of the middle ear are surrounded by porary hearing loss and even some pain.
air. The air pressure is kept approximately equal to We have already noted that the footplate of the
that of the surrounding atmosphere by means of the stapes rests on a membrane called the oval win¬
eustachian tube, which opens into the back of the dow. This is the only part of the inner ear directly
throat. The equalization of pressure on either side receiving mechanical vibrations. The oval window
of the eardrum is important since a pressure differ¬ is at one end of one of the three canals that run the
ential would cause the membrane to bulge and length of the cochlea, as can be seen in Figure 6-
stiffen, resulting in less responsiveness of the ear¬ 5. Two of these, the vestibular and the tympanic
drum to the sound striking it. If the eustachian tube canals, are connected at the apex of the cochlea by
were not present, the pressure on the inner side an opening called the helicotrema. They are filled
would gradually drop because of absorption of the with a fluid resembling salt water. Since this fluid
air by the surrounding tissue. However, the eusta¬ is relatively noncompressible, a point is needed
chian tubes (one for each ear) open briefly every where the pressure from the vibration of the stapes
time we swallow, allowing air to flow into the mid¬ at the oval window is released. This release point
dle ear cavities from the mouth and lungs. This is provided by the round window, which is a

Sound

FIGURE 6-5 Movements of the eardrum in response to sound are transmitted by the ossicles to the
fluid in the canals of the coiled cochlea.
 The Structure of the Ear 161

membrane at the base of the tympanic canal that with a different, more viscous fluid. Reissner’s
opens onto the middle ear. When a movement of membrane is very thin (only two cells thick) and
the stapes causes the fluid to move away from the has no function other than to form one wall of the
oval window, the round window bulges. This in¬ cochlear duct. The basilar membrane is the func¬
dicates that some fluid has been displaced, and the tionally important one. In humans it is about 3 cm
resulting pressure change has been transmitted long and varies in width from about 0.08 mm near
through the helicotrema into the tympanic canal the base (where the windows are) to about 0.5 mm
and to the round window at its base. Of course, this at the apex (where the helicotrema is), and is about
system allows any vibration of the stapes to set up 100 times stiffer at the base than at the apex. A
similar vibrations in the internal fluid in the coch¬ third membrane within the cochlear duct is also im¬
lea. This is why so much amplification of the sound portant. The tectorial membrane extends into the
energy in the middle ear is needed. The amplifica¬ cochlear duct from Reissner’s membrane, and
tion is necessary to allow small and relatively weak touching it are some of the hairs of the organ of
displacements of the eardrum to cause the entire Corti (Figure 6-7). This is the part of the cochlear
body of cochlear fluid to vibrate in the same fash¬ duct that accomplishes the final transduction of
ion as the eardrum (Schubert, 1978). the mechanical energy of a sound wave into elec¬
The third canal of the cochlea, the cochlear trochemical energy interpretable by the nervous
duct (or scala media), is relatively self-contained. system.
It neither opens to the middle ear nor joins the ves¬ The organ of Corti rests on the basilar mem¬
tibular or tympanic canals. It is formed by two brane along its entire length. It is composed of
membranes that run the length of the cochlea: Reiss- about 23,500 cells, which resemble the cells of the
ner’s membrane and the basilar membrane. To¬ skin in that hairs protrude from them. A single row
gether they form a rough triangle with the wall of of about 3,000 inner hair cells is found on the
the cochlea (Figure 6-6). The cochlear duct is filled inner side (left side in Figure 6-7) of the tunnel of

FIGURE 6-6 Cross section of the cochlea reveals its three canals and the organ of Corti, the auditory
receptor.
162 Chapter 6/The Auditory System

Vestibular canal

Reissner’s membrane

Tectorial membrane
Hairs

Inner hair cell

Auditory nerve fibers

FIGURE 6-7 The detailed structure of the organ of Corti. (Based on Gulick, 1971)

Corti, and three or four rows of outer hair cells per inner hair cell). The other 5 percent of the fi¬
are located on the outer side. Inner hair cells have bers from the spiral ganglion, called outer spiral
about 40-60 hairs each, which extend into the vis¬ fibers, each make synaptic contact with about 10
cous fluid that fills the cochlear duct under the tec¬ outer hair cells. No outer hair cell receives more
torial membrane but which apparently do not touch than about four such contacts (Spoendlin, 1978).
the membrane (Lim, 1980). Each outer hair cell Figure 6-8 shows a general picture of how the fi¬
may have as many as 100-120 very tiny hairs pro¬ bers of the spiral ganglion cells innervate the coch¬
truding from it. The tallest of these hairs are firmly lea. The radial fibers have a large diameter and are
embedded in the tectorial membrane; the shorter probably myelinated, whereas the outer spiral fibers
hairs apparently do not touch the membrane. In are smaller in diameter and probably unmyelinated,
both kinds of hair cells, a graded set of hairs forms hence they have slower neural conduction speeds.
a hair bundle. Each hair bundle contains hairs that In addition, the two types of fibers come from gan¬
are connected to a shorter hair on one side and a glion cells that have a noticeably different shape
longer hair on the other. All of the hairs in a hair (Kiang, Rho, Northrop, Liberman & Ryugo,
bundle tend to move, or bend, as a unit. 1982). Given so many structural differences, it is
About 30,000 nerve fibers, whose cell bodies likely (though as yet undetermined) that these two
are located in the spiral ganglion, form synaptic different sets of fibers carry different types of au¬
connections on the bases of the hair cells. About ditory information. The axons of these spiral gan¬
95 percent of them, called radial fibers, make sin¬ glion cells make up the auditory nerve, which is
gle connections with the inner hair cells (about 10 the neural pathway to the brain.
 The Structure of the Ear 163

pects of the transduction of mechanical energy into

electrochemical energy accomplished here. The
first is the movements of the basilar membrane that
result in the stimulation of the sensory cells, and
the second is the actual transduction mechanism it¬
self. We discuss each of these in turn.
The movements of the cochlear fluid cause
mechanical waves to travel down the basilar mem¬
brane from the base (near the oval window) to the
apex. Such a wave is really a traveling bend or
kink that moves down the length of the membrane,
much like the motion when you crack a whip. Fig¬
ure 6-9 is a schematic drawing of such a wave. The
existence of these traveling waves was demon¬
Scala media
strated by Georg von Bekesy (e.g., 1960), who re¬
(all bony parts of cochlea removed)
ceived the Nobel Prize for his work on the
FIGURE 6-8 A view of the cochlea with the bone mechanics of the ear. The waves themselves travel
and other covering removed leaving only the soft very quickly, going from the base to the apex in
membrane and neural tissue.
about 3 msec (Kitzes, Gibson, Rose & Hind,
1978). Each wave begins in the stiffer, narrower
part of the basilar membrane and travels toward the
Wave Motion on the Basilar Membrane
looser, broader part.
Sound waves cause the bones of the middle ear to The variations in elasticity and width of the
vibrate, and the vibration of the last of these bones basilar membrane are responsible for the direction
(the stapes) causes the oval window, and thus the and the speed of the traveling wave. They are also
cochlear fluid, to move back and forth at the same responsible for differences in the size or amplitude
frequency. This movement causes pressure differ¬ of the wave. Bekesy was able to demonstrate that
ences across the cochlear duct and movements in the basilar membrane reacts differently to sound
the basilar membrane, stimulating the sensory cells stimuli of different frequencies. Although the entire
of the organ of Corti. There are two important as¬ membrane vibrates for any given stimulus, the

Oval window

FIGURE 6-9 A traveling wave on the basilar membrane.

164 Chapter 6/The Auditory System

FIGURE 6-10 A graph of the relative sizes of traveling waves along the basilar membrane for three
different frequencies of tone. Notice that as the frequency increases, the waves reach their maxima
nearer the oval window and stapes (base).

traveling wave reaches a maximum at a different the amplitudes of traveling waves at different
place along the basilar membrane for each different places along the basilar membrane for pure tones
sound frequency. Traveling waves caused by low- of several different frequencies. This characteristic
frequency sounds grow steadily in size as they action of the basilar membrane in response to
travel toward the apex of the membrane, and do not sound provides a basis for the frequency analysis
reach a maximum until they are near the apex. of complex tones, as well as for the more basic
High-frequency sounds, however, cause traveling ability of distinguishing between pure tone stimuli
waves that reach their maximum near the base of of different frequencies. Demonstration Box 6-3
the basilar membrane and then quickly dissipate, shows you how to demonstrate the differences in
causing little deformation of the membrane near the the ability of low and high frequencies to travel
apex. This is shown in Figure 6-10, which displays down a membrane.

DEMONSTRATION BOX 6-3. The Skin as a Model for the Basilar Membrane

The basilar membrane is set into vibration by incom¬ teeth. Now make a loud, low sound (try to imitate the
ing sound stimuli. How the membrane vibrates, how¬ low sound of a foghorn), and notice that your entire
ever, depends on the frequency of the sound input. finger seems to vibrate, perhaps all the way down to
Low frequencies tend to cause vibrations of signifi¬ the knuckle at its base. Next make a high pitched
cant magnitude along the entire length of the mem¬ sound (try to imitate the whistling of a teakettle or the
brane; high frequencies cause vibrations that are sig¬ test tone on a TV station that has ended the day’s
nificant only near the base. You can easily broadcasting). Notice that the feeling of vibration
demonstrate the frequency-specific nature of the vi¬ covers only a tiny region, perhaps your fingertip, or
bration by using your finger as a model of the basilar down to the first joint. In a similar fashion, lower
membrane, since skin has about the same resiliency sound frequencies induce waves that extend over the
and elasticity. Place your finger in your mouth, rest* length of the basilar membrane, whereas higher fre¬
ing your finger tip firmly against the front of your quency waves are restricted spatially in their effects.
 The Structure of the Ear 165

Mechanisms of Transduction Close-up view

of stereocilia
We have been discussing the ear as if it were a
purely mechanical device. The organ of Corti,
which rests on the basilar membrane, is the site of
the transduction of sound energy from its mechan¬
ical form into the electrochemical energy that is the
communication medium of the nervous system (see
Appendix). The movements of the basilar mem¬
brane described above cause the basilar membrane
and the tectorial membrane to move laterally with
respect to each other. Since the outer hair cells are
attached at their base to structures connected to the
basilar membrane, and the longest of their hairs are
embedded in the tectorial membrane, a shearing
force is applied to these hairs, causing them to
bend. The inner hair cells may be bent in a some¬
what different way. Since they appear not to make
contact with the tectorial membrane, they are not
subject to the shearing force applied to the longest
hairs of the outer hair cells. It is possible that they
do make contact with the tectorial membrane as a
result of the movements of the basilar membrane
and are bent when they rub against it (Crane,
1982). However, it is also possible that they are
bent by the flow of the viscous fluid in the cochlear
duct caused by these same movements (Dallos,
FIGURE 6-11 Model of transduction in hair cells.
1978). Although this seems to be a less efficient (A) A typical inner hair cell. (B) and (C) Illustration
way of producing bending, there appear to be no of operation of a typical transduction channel in
differences in the absolute sensitivities of the inner the hairs. (Based on Hudspeth, 1985)
and outer hair cells (Dallos, Santos-Sacchi &
Flock, 1982).
The mechanism by which the bending of hairs about 20 percent of the time. Potassium ions,
is transduced into electrical changes in the hair which have a positive charge, flow into the cell
cells is still not fully understood (see Dallos, whenever these pores are open. The inward flow of
1981). However, recent work suggests that flexing positive charge is just balanced by the outward flow
the hairs causes parts of the outer membrane of the of positive charge caused by other “pumps” else¬
hair to function like little “trap doors.” Bending where in the cell, maintaining the cell’s resting
causes these to open and allow potassium ions to membrane potential at about -60 millivolts (see
flow into the interior of the hair cells (see Hud¬ Appendix). When the hair is bent in the direction
speth, 1985). Figure 6-11 shows this graphically. of the tallest hairs, however, the filament attached
In the model suggested by Hudspeth (1985) each to the trap door pulls on it and keeps it open more
hair has such a trap door, which is attached to the of the time, allowing more positive charge to flow
next longer hair by a very thin filament. Ordinarily into the cell. This causes a depolarization (making
the trap door “rattles around,” and is open only the potential more positive) of up to 20 mV, which
166 Chapter 6/The Auditory System

in turn is thought to release transmitter substances ulus or specific stimulus “features.” When we
from the synaptic cleft at the bottom of the hair record the responses of single axons in the auditory
cell. When these transmitter substances are re¬ nerve, we find that some respond only to complex
ceived in large enough quantities by the spiral gan¬ noises and are completely unresponsive to pure
glion cells making the synapses with the hair cells, tones. Others respond only to brief transient sounds
the spiral cells generate action potentials that then such as clicks. Still others respond to a wide vari¬
travel up the auditory pathways. ety of different frequencies of pure tones. Appar¬
This model is very successful in accounting ently, auditory encoding is very complex even at
for many properties of the auditory system. For ex¬ this peripheral level. Thus, in much the same way
ample, because the trap doors are constantly open¬ that we found selective analysis taking place early
ing and closing, response to a sound can be very in the visual system, similar selectivity seems to be
rapid, as data indicate it is (less than 1 msec). characteristic of audition.
Moreover, because the process is probabilistic (hair Although neurons possessing a great many
bending simply increases the probability that the different types of response characteristics are pres¬
door will be open at a given instant), it is capable ent in the auditory nerve, one type of neuron is
of responding to very low levels of sound (of the extremely common. Such a neuron is often called
exceedingly small vibration distances mentioned a tuned neuron, in analogy to the way we “tune
above) if averaged over a sufficiently long time. in” a radio station to its broadcast frequency. Fig¬
All in all, this seems a very good candidate for the ure 6-12 shows the absolute thresholds (minimum
correct model of transduction. intensity of sound that stimulates a neuron to fire
above its resting rate) of some typical neurons of
this type. The results are called threshold response
ELECTRICAL ACTIVITY OF THE curves, and as can be seen from the figure each
AUDITORY NERVE neuron has a best, or characteristic, frequency for

We have followed sound energy to the point where

it is converted into patterns of electrical activity in
the auditory nerves. These signals are spike poten¬
tials, hence we can use the same techniques of
electrophysiological recording that we used to in¬
vestigate the visual system to study the neural pro¬
cessing of auditory information. If you have
forgotten what these procedures are specifically,
now would be a good time to glance back at the
Appendix, which provides this information. Basi¬
cally, the technique involves inserting electrodes
into cells in the auditory pathways and recording
the electrical activity of individual neurons in re¬
sponse to a variety of sounds.
As a first guess, we might suppose that the
auditory system would be responsive mainly to
simple physical aspects of a sound, namely fre¬ Frequency (Hz)
quency and intensity. Actually, some cells seem to
FIGURE 6-12 Threshold response curves for
mimic visual processing in that they are quite se¬ auditory nerve fibers in the cat. (Based on
lective, responding to complex aspects of the stim¬ Whitfield, 1968)
 Electrical Activity of the Auditory Nerve 167

which its absolute threshold is lowest. The sensitiv¬ ment of the basilar membrane at that location,
ity of tuned neurons decreases (the threshold is which in turn results in less vigorous bending of
higher) as we move away from the best frequency the hair of the hair cell, and finally less vigorous
in either direction. Another way of looking at the responding of the neuron.
tuning of such an auditory neuron is simply to pres¬ This passive mechanical tuning of the basilar
ent a tone varying in frequency but fixed in inten¬ membrane may not be the entire answer to the
sity. A graph of the response rate of a tuned neuron question of how neural tuning occurs. Most mea¬
treated in this way is often called a tuning curve, surements of the mechanical tuning curves of the
shown in Figure 6-13. In the auditory nerve there basilar membrane (similar to those of Figure 6-10)
are neurons tuned to frequencies over the entire give curves that are less sharp (change less steeply
range of hearing. with frequency) than the tuning curves measured in
The tuned nature of these cells probably auditory nerve fibers (see Dallos, 1981). One ex¬
comes about because of the mechanical properties planation for this difference might be that the outer
of the basilar membrane (Khanna & Leonard, hair cells help to sharpen the auditory nerve tuning
1982). Most of the tuned auditory nerve fibers that curves, because destruction of such cells results in
have been studied are connected to inner hair cells broader tuning curves (Lynn & Sayers, 1970;
(Kiang et al., 1982; Liberman, M.C., 1982), and Schmiedt, Zwislocki & Hamemik, 1980). These
there are more spiral ganglion cells synapsing with outer hair cells might actually act mechanically on
each hair cell the closer the hair cell is to the base the basilar membrane, dampening activity every¬
of the cochlea (Keithley & Schreiber, 1987). Re¬ where but at the maximum of the traveling wave
member that different parts of the basilar mem¬ and thus sharpening the mechanical tuning (see
brane vibrate maximally to different frequencies of Hudspeth, 1985). This active cochlear process is
sound, and it is this vibration that causes the hair being studied intensively, and several plausible
cells to bend and in turn initiate the neural re¬ models of how it could operate have been sug¬
sponse. A hair cell will respond most strongly gested (Jen & Steele, 1987; LePage, 1987a, 1987b;
when a tone of a certain frequency creates a trav¬ Zwicker, 1986). Such an active process has been
eling wave on the basilar membrane that has its used to explain why complex vibrations are found
maximum amplitude near the location of that hair in the cochlea in the absence of any sound inputs
cell. The frequency of this tone will be the fre¬ at all (Brownell, Bader, Bertrand & de Ribau-
quency to which the neuron that synapses with that pierre, 1985; Hudspeth, 1985).
hair cell is “tuned.” Other frequencies of sound of Further evidence for interaction between inner
this same intensity will cause less vigorous move- and outer hair cells comes from a finding called
two-tone inhibition (Rose, Galambos & Hughes,
1959; Sachs & Kiang, 1968). This phenomenon is
seen when recording from an auditory nerve fiber
that is stimulated by its characteristic frequency and
is responding vigorously. If a second tone of a dif¬
ferent frequency, which is moderately close to the
tuned frequency, is now briefly presented, the re¬
sponse rate in the tuned neuron drops. Such an in¬
hibitory response may result from the action of
outer hair cells (Javel, 1981). This is supported by
the fact that two-tone inhibition disappears when
FIGURE 6-13 Tuning curve of a typical auditoiy outer hair cells have been selectively damaged by a
nerve fiber. (Based on Lindsay & Norman, 1977) drug (Schmiedt et al., 1980). Such inhibition
168 Chapter 6/The Auditory System

provides an alternate means to sharpen tuning said to be saturated. These facts imply that for a
curves, since it involves the suppression of re¬ sound of a given intensity and frequency there will
sponses to frequencies other than the tuned one. be a population of neurons all firing at different
What is the nature of the information about response rates. This situation is represented hypo¬
intensity and frequency carried from the ear to the thetically in Figure 6-14A, for two levels of inten¬
brain? One suggestion is that low frequencies may sity. Stimuli of different frequencies tend to cause
be encoded directly in the number of spike poten¬ different populations of neurons to fire above their
tials in the neural response (Johnson, 1980; Rose, background rates, as is shown in 6-14B. As the
Brugge, Anderson & Hind, 1967). Thus, if the fre¬ intensity of the stimulus increases, other neurons,
quency of the stimulus is 100 Hz, the firing in the tuned to nearby frequencies, may also be recruited,
auditory nerve will tend to be approximately 100 resulting in an increase in the number of neurons
Hz. This does not mean that any given individual responding, as well as an increase in their response
neuron fires at this frequency. Rather, individual rates. Thus, the entire pattern of auditory activity
neurons seem to fire at fixed points in the cycle of changes with changes in the stimulus. Frequency
the sound wave. For example, one neuron may fire seems to be indicated by which neurons are firing,
at every second peak of the wave (and thus at 50 whereas intensity seems to be roughly indicated by
Hz), whereas another may fire at every fifth peak how many are firing, and to a minor extent how fast
(and thus at 20 Hz). This is called phase-locking. they are firing (Whitfield, 1978).
For a great many neurons firing out of phase with Our increasing understanding of how the au¬
one another there will tend to be a spike (or sev¬ ditory nerve encodes frequency and intensity has
eral) occurring at every peak of the wave and thus allowed some people with hearing disabilities to re¬
a composite response rate of 100 Hz could be cover some auditory ability. If the hair cells do not
achieved. This ability of the auditory nerve to fol¬ function, either because they were damaged or con¬
low the frequency of the stimulating sound wave genitally malformed, a person is deaf. Recently,
(up to about 4000 Hz) has been a central compo¬ otologists have been able to implant in the ear de¬
nent of several theories of pitch perception (e.g., vices (known as cochlear implants) that stimulate
Wever, 1979) that will be discussed in Chapter 7. the auditory nerve electrically in response to exter¬
The important factor to be noted is that the pattern nal sounds (see Schindler & Merzenich, 1985).
of response of the whole auditory nerve rather than Modem cochlear implants first analyze the fre¬
the responses of individual neurons may be signif¬ quency composition of the sound into ranges of
icant in conveying frequency information to the stimulus frequencies. The implant itself involves a
brain. series of electrodes that have been implanted at dif¬
The importance of pattern of response in the ferent points along the basilar membrane. These
auditory nerve cannot be overestimated. Even different regions of the basilar membrane are then
though many cells in the auditory nerve may be stimulated, based on the initial analysis of the fre¬
tuned to the same characteristic frequency, they quencies, in an attempt to mimic the way the trav¬
may still vary in terms of their threshold intensities eling wave on the basilar membrane stimulates the
over a range of about 20 dB (Evans, 1975). Once hair cells. Deaf patients with such devices can dis¬
the intensity of a sound has exceeded a cell’s criminate the frequencies of different sounds
threshold, further increases in intensity up to a (Townshend, Cotter, Van Compemolle & White,
level 30-50 dB above the threshold value will in¬ 1987) and can even recognize speech sounds quite
crease its rate of response. Intensity increases be¬ well, especially when a speech preprocessor is at¬
yond this point (30-50 dB above threshold) will tached to the cochlear implant to isolate certain
not cause the neuron to increase its rate of re¬ speech-relevant frequency changes (Blarney, Dow¬
sponse; the neuron is firing as fast as it can and is ell, Brown, Clark & Seligman, 1987). There is
 The Auditory Pathways 169

Response to
<D
03 strong stimulus
v_
0)
(/) Response to
c weak stimulus
o
Q_
cn
CD
cr
ilfc
Population of auditory nerve fibers
A

FIGURE 6-14 Hypothetical distributions of response rates for the population of auditory nerve fibers
firing in response to (A) weak versus strong stimuli of the same frequency and (B) stimuli of the same
strength but of different frequencies.

some promise that, in future, such devices will al¬ sual pathways. First, the axons of the spiral gan¬
low nearly a full range of hearing experience for glion cells make up the auditory nerve, which
people with this form of deafness. projects to the cochlear nucleus, located in the
lower back part of the brain. As shown in the fig¬
ure, the auditory nerve fibers enter the ventral
THE AUDITORY PATHWAYS (front) cochlear nucleus, where each divides into
at least two branches. One branch synapses with
Figure 6-15 diagrams the principal pathways taken cells in the ventral cochlear nucleus, and the other
by auditory information in the brain. The auditory proceeds to the dorsal (back) cochlear nucleus.
pathways are somewhat more complex than the vi¬ The cells of the ventral cochlear nucleus send about
170 Chapter 6/The Auditory System

half their axons to the superior olive on the oppo¬ Most cells in the inferior colliculi send axons to
site side of the brain and half to the one on the the medial geniculate, although a few go to the su¬
same side. The cells of the dorsal cochlear nucleus perior colliculus as well. Since the superior collicular
send all their axons to the opposite side of the pathway has been implicated in visual localization,
brain, eventually to terminate in the inferior colli¬ perhaps the auditory fibers that go there also carry in¬
culus. Thus, most of the information from the right formation about location. This sound localization in¬
ear is sent to the left side of the brain, and vice formation could then be related to visual data to yield
versa. The cells of the two superior olives send a more complete “picture” of space (see Chapter 7).
most of their fibers to the inferior colliculi (which From the medial geniculate, fibers project to a part of
are located just below the superior colliculi, dis¬ the temporal cortex that is often called the primary
cussed in Chapter 3). At the level of the inferior auditory projection area, or Brodmann’s Area 41.
colliculus, considerable fiber crossing takes place The adjacent Brodmann’s Area 22 also receives ax¬
from one side of the brain to the other, so that each ons directly from the medial geniculate, although
inferior colliculus has full information about what fewer of them (these regions of the brain were pic¬
is going on in the other. tured in Figure 3-13).

Dorsal
cochlear Auditory cortex
nucleus
Medial geniculate

Inferior colliculus

Auditory nerve
from right ear
Cochlear nucleus
Superior olive
Ventral
cochlear
nucleus

FIGURE 6-15 The major auditoiy pathways in the brain. (Based on Lindsay & Norman, 1977)
 The Auditory Pathways 171

Electrical Activity of the Lower the visual system seems to exist. These cells ac¬
Auditory Centers tually reduce their response rate below their spon¬
taneous activity level at the onset of a tone, and
We have already mentioned that many of the re¬ then give a burst of activity at its offset. An inter¬
sponse patterns found in the fibers of the auditory esting variation of this is the presence of tuned cells
nerve suggest that there is some form of auditory that reduce their activity level when a stimulus dif¬
feature analysis going on. Although some of these ferent from the “best” one is present. Of particular
neurons respond only to clicks or other stimuli that interest is a set of cells that are reminiscent of the
are not pure tones, the majority of the fibers are on-center/off-surround cells observed in the visual
axons from “tuned” neurons. Tuned neurons re¬ system (Chapter 3). Rather than having a receptive
spond to a range of frequencies of pure-tone field that consists of a region in space, these cells
sounds, but each has a “preferred” band of fre¬ have a receptive field consisting of a band of fre¬
quencies to which it is most sensitive and to which quencies. For instance, there are cells in the medial
it produces its maximum response. Similar tuned geniculate with “W-shaped” receptive field pat¬
neurons are found in the cochlear nucleus, superior terns. These neurons respond above their back¬
olive, inferior colliculus, and medial geniculate. ground rate of firing to a particular best frequency,
Also, as described above for the auditory nerve, and below the background rate for frequencies
there is a region of nearby frequencies that pro¬ close to that frequency on either side. The firing
duces an inhibition of response (as in two-tone in¬ rate gradually returns to background level for fre¬
hibition) in these same cells. quencies progressively more removed from the best
In addition to the frequency tuning of cells in frequency and its surrounding “worst” frequencies
the more central nuclei of the auditory pathway, a (see Webster & Atkin, 1975).
number of more complex neural response patterns This type of correspondence between proper¬
appear. When Pfeiffer (1966) recorded from single ties of neurons in different sensory systems is very
neurons in the cochlear nucleus of adult cats, he common, although each sensory system also dis¬
found a variety of different types of neural re¬ plays coding principles idiosyncratic to its own
sponses to a simple tone. On neurons gave a burst stimulus modality. We seem to be dealing with ba¬
of responses immediately after the onset of the tone sically the same kinds of neural units at the electro-
burst and then ceased responding, no matter how physiological level, regardless of the sensory
long the tone persisted. Pauser neurons exhibited system considered. It is as though some master
a similar burst of firing at the onset of a tone, but technician found a useful biological coding scheme
this was followed by a pause and then a weaker for stimuli, and simply modified it to fit the various
sustained response until the tone was turned off. requirements of different sensory systems.
Choppers gave repeated bursts of firing followed Because different points along the basilar
by short pauses, with the vigor of successive bursts membrane vibrate most strongly for different fre¬
decreasing. Primarylike neurons gave an initial quencies of sounds, we refer to the response of the
vigorous burst of firing when the tone was turned basilar membrane as tonotopic (from the Greek
on, then the firing rate decayed to a lower level that tono for “tone” and topus for “place”). This
was sustained for the duration of the tone. These means we are actually representing sound frequen¬
primarylike neurons have been shown to be capable cies with a sort of spatial code. This spatial encod¬
of encoding some critical aspects of speech sounds ing also seems to be preserved at other points in
(Palmer, Winter & Darwin, 1986). the auditory system. In the cochlear nucleus of the
In addition to these response patterns, a re¬ cat, Rose, Galambos, and Hughes (1960) recorded
sponse analogous to the off responses observed in the responses of single cells to pure tones as a
172 Chapter 6/The Auditory System

microelectrode was moved through the tissue. As tory centers of humans varies in depth as the fre¬
the electrode was pushed along, the frequency of quency varies. Responses to high frequencies are
stimulus that gave the best response (highest rate of deep, whereas those to low frequencies lie near the
firing) changed systematically from high near one surface of the brain. A similar result has been ob¬
edge of the nucleus to low near the other. This in¬ tained using the technique of positron emission
dicated that the cells were arranged in an orderly tomography, which allows researchers to visualize
spatial layout, with cells tuned to similar frequen¬ areas of the brain that are most active during var¬
cies lying closer together in the nucleus than those ious activities, such as listening to sounds of dif¬
tuned to different frequencies. Similar tonotopic ar¬ ferent frequencies (Lauter, Herscovitch, Formby &
rangements of cells have been reported for the au¬ Raichle, 1985). These results are consistent with
ditory nerve and other noncortical nuclei (Gulick, the idea that the frequency of sound waves is coded
1971; Webster & Atkin, 1975). In general, the ton¬ mainly by place (which neurons are firing), both on
otopic organization of the basilar membrane is pre¬ the basilar membrane and in the central auditory
served even up to the cortex, where we find cells system.
that respond to high frequencies clustered in some
areas, and cells that respond to low frequencies
clustered in other areas. You can see this clustering THE AUDITORY CORTEX
for the cat auditory cortex in Figure 6-16 (where
the high- and low-frequency areas are marked, and Studies of nonhuman animals have provided us
the intermediate frequencies fall in between). with most of the information we now possess about
This result has been confirmed in humans by the physiology of sensory systems. For the lower
some ingenious measurements of the magnetic field levels of analysis we can be fairly confident in gen¬
created in the brain by its electrical response to eralizing the concepts to humans. When we begin
sounds (Romani, Williamson & Kaufman, 1982). to discuss the cortex, however, we are on shakier
The maximum brain activity observed in the audi- ground. The human cortex is more complex than
that of the most common experimental animals
(such as the cat, the preferred subject of such stud¬
Secondary ies), so that generalization of findings to humans is
visual area
more tenuous. Nonetheless, animal studies have
yielded a significant amount of useful information
on the activity of the auditory cortex. Figure 6-16
shows the wide variety of areas on the cat cortex
that have been shown to be responsive to sound
stimuli. In this chapter we have been concerned
mainly with the primary auditory areas, but a large
number of nonauditory areas also respond to sound
stimuli, especially association and visual areas.
Presumably, similar responses occur in the human
cortex, especially in the speech and general asso¬
ciation areas.
Cells in the auditory cortex exhibit a variety
I 1 Primary auditory areas (l) Low-frequency locations
of complex responses to sound stimuli. In the ap¬
BESBHH Secondary areas (h) High-frequency locations proximately 60 percent of the cells that respond to
pure tones, there occur on responses, off responses,
FIGURE 6-16 Auditory areas of the cat cortex. on-off responses, and more general excitatory and
 The Auditory Cortex 173

inhibitory responses (see Figure 6-17). These re¬ have obvious utility for the cat, whose “war cries’’
sponses, of course, resemble the response patterns and “love calls” consist of just these types of
of cells in the visual system. The other 40 percent sounds.
of the cells seem to respond selectively to more These cortical neurons respond to sound pat¬
complex sounds, including noise bursts, clangs, or terns in much the same way that visual cortical
clicks. neurons respond selectively to light patterns. There
Another complex type of neuron found in the are even auditory analogues to the highly specific
auditory cortex of the cat is the frequency sweep visual cells (face and paw detectors) observed in
detector (Whitfield & Evans, 1965). These cells the temporal cortex (see Chapter 3). Swarbrick and
respond only to sounds that change frequency in a Whitfield (1972) found cells in the auditory cortex
specific direction and range. Some cells respond of the squirrel monkey that are sensitive to the vo¬
only to increases but not to decreases in frequency calizations of other squirrel monkeys. Some of
in the same range. Others respond to decreases in these cells are unresponsive to the presentation of
frequency but not to increases. A third type re¬ simple tones, although they respond vigorously to
sponds only to increases in frequency for low-fre¬ the presentation of extremely complex vocaliza¬
quency tones. Since these types of stimuli are often tions (Funkenstein, Nelson, Winter, Wolberg &
encountered in speech and music, such detectors, Newman, 1971). Watanabe and Katsuki (1974)
if present in humans, could have an important role found cells in the cat auditory cortex that in addi¬
in speech and music perception. Certainly they tion to being tuned to a specific band of frequencies

Response ill Hi
An on response
Tone t- I—r :
On On

Response
An off response
Tone

Response HU
An on-off
response
Tone

Response Hill 11'HI 1111 II

An excitatory
response
Tone \...m
On

Response .HI Ml M_LLUliil_ill

An inhibitory
response
Tone
On

FIGURE 6-17 Different types of response to pure tones recorded from neurons in the auditory cortex
of the cat. (Based on Whitfield, 1967)
174 Chapter 6/The Auditory System

also responded with a unique pattern of activity to Cycles per second The unit to measure frequency of
recordings of cat vocalizations. Interestingly, these sound waves, now usually referred to as hertz.

cortical cells did not respond with the same pattern Decibel (dB) The unit used to measure sound inten¬
sity; one-tenth of a bel.
of activity to any of the individual components of
Dorsal cochlear nucleus The back half of the nu¬
the cat vocalization. This indicated that the cells
cleus in the lower back part of the brain where the audi¬
were integrating the outputs of cells from lower tory nerve fibers end.
levels of the auditory pathway that do respond to Eardrum The membrane at the end of the auditory
simpler components of the sound pattern. We could canal that vibrates in resonance with incoming sound
say that the lower level cells were detecting the waves.
various features of the vocalization and the cortical Eustachian tube A channel from the back of the
cell was responding only to the combination of all throat to the middle ear; when we swallow it opens and
allows the air pressure in the middle ear to equalize with
of the features (Whitfield, 1980).
the outside.
Although these physiological findings are
Fourier components Simple sine waves that add to¬
quite intriguing and suggestive, the ultimate test of gether to form a complex waveform.
any hypotheses about the significance of neural en¬ Frequency The number of cycles a sound wave
coding or analysis of auditory patterns rests on data completes in one second.
about an individual’s actual auditory percepts. In Frequency sweep detector A neuron that responds
Chapter 7 we consider what is heard and try to in¬ only to sounds that change frequency in a specific direc¬
tegrate it with our present knowledge of how it is tion and range.
heard. Hair bundle A graded set of hairs protruding from a
hair cell in the cochlea and tending to move, or bend, as
a unit.
Helicotrema An opening, between the vestibular
and tympanic canals, at the apex of the cochlea.
GLOSSARY
Hertz (Hz) The unit (cycles per second) used to
measure frequency of sound waves.
The following definitions are specific to this book.
In phase When the peaks and valleys of different
Auditory canal The canal conducting sound waves sound waves coincide over time.
to the eardrum. Incus One of the three middle bones involved in
Basilar membrane The membrane within the coch¬ sound conduction to the cochlea. Also known as the an¬
lea of the ear on which the organ of Corti lies. vil.
Bel The basic unit used to measure the relative inten¬ Inferior colliculi Auditory processing centers in the
sity of a sound wave. midbrain that are the termini for cells of the superior
Chopper A neuron that gives repeated bursts of im¬ olives.
pulses followed by short pauses, with the vigor of suc¬ Inner ear The part of the ear containing the cochlea.
cessive bursts decreasing, in response to presentation of Inner hair cells Cells found on the inner side of the
a pure tone. tunnel of Corti.
Cochlea A snail-shaped part of the labyrinth of the Labyrinth A structure of fluid-filled canals and
ear that contains the auditory receptors. chambers in the head that contains organs of hearing and
Cochlear duct One of the three canals in the coch¬ the vestibular senses.
lea; also called the scala media. Malleus The inner ear bone that is attached to the
Cochlear implant A device implanted in the cochlea eardrum; also called the hammer.
that electrically stimulates the auditory nerve similarly to Medial geniculate The brain structure that receives
how it is stimulated by hair-cell activity. inputs from the inferior colliculus and sends axons to the
Cochlear nucleus A structure in the lower back part auditory cortex.
of the brain that receives input from the auditory nerve. Middle ear The part of the ear consisting of the os¬
Cycle In sound, the completion of a full sequence of sicles (malleus, incus, and stapes) that transmit the ear¬
air compression and rarefaction. drum vibrations to the inner ear.
 Glossary 175

Off responses Neural responses that commence with Saturated The state in which a neuron cannot fire
the termination of stimuli. any faster, even if stimulus intensity is increased.
Ohm’s acoustical law The auditory system separates Sound pressure level (SPL) The log-relative pres¬
complex sounds into simple (Fourier) components. sure amplitude of a sound wave, measured in decibels.
On neurons Neurons that fire immediately and ex¬ Spiral ganglion The cells whose axons form the au¬
clusively after the onset of a tone. ditory nerve.
Organ of Corti The part of the cochlear duct that Stapes The bone in the chain of middle ear ossicles
transduces mechanical sound wave energy into electro¬ that makes contact with the oval window; also called the
chemical energy interpretable by the nervous system. stirrup.
Out of phase When the peaks and valleys of differ¬ Superior olives The brain termini for axons leading
ent sound waves do not coincide over time. from the ventral cochlear nuclei.
Outer ear The pinna, the auditory canal, and the Tectorial membrane Within the cochlear duct, the
eardrum. membrane, extending from Reissner’s membrane, in
Outer hair cells Cells found on the outer side of the which some of the hairs of the organ of Corti are embed¬
tunnel of Corti. ded.
Outer spiral fibers Fibers extending from the spiral Threshold response curve A graph of neural abso¬
ganglion to the outer hair cells. lute threshold as a function of sound frequency.
Oval window A membrane in the cochlea that re¬ Timbre A sound attribute associated with the com¬
ceives sound vibrations from the stapes. ponents of a complex sound wave.
Pauser neurons Neurons, similar to on-response Tonotopic Characteristic of the place of response to
neurons, that exhibit an initial response to stimuli, fol¬ a sound depending on the frequency of the sound.
lowed by a pause, and then a weaker, sustained response Tuned neuron A neuron that responds optimally to
until the stimulus stops. tones of a characteristic frequency.
Phase The particular point in the compression-rare¬ Tuning curve A graph showing the rate of firing of
faction cycle of a sound wave at one instant of time. an auditory neuron for different tone frequencies; it usu¬
Phase angle The degree to which one sound wave is ally has a single peak.
out of phase with another, considering a complete cycle Tunnel of Corti A structure in the cochlea.
as 360 degrees. Two-tone inhibition Inhibition of neural response to
Phase-locking The tendency of individual neurons to a characteristic frequency of sound that occurs when a
fire at fixed points in the cycle of a sound wave. second tone of a different frequency is presented; inhi¬
Pinna The fleshy visible part of the outer ear. bition occurs during and for a brief period after the pre¬
sentation of the second tone.
Pressure amplitude A measure of the degree of
compression or rarefaction at the peaks or valleys of a Tympanic canal One of three canals running
sound wave. through the cochlea.
Primary auditory projection area The area of the Tympanum See Eardrum.
temporal cortex that receives most of the fibers from the Ventral cochlear nucleus The front half of the nu¬
medial geniculate; also known as Brodmann’s Area 41. cleus in the lower back part of the brain where the audi¬
Primarylike neuron A neuron that gives an initial tory nerve fibers end.
burst of firing in response to a stimulus, and then contin¬ Vestibular canal One of three canals running
ues firing at a lower level until the stimulus stops. through the cochlea.
Radial fibers Fibers extending from the spiral gan¬ Wave The pattern of air molecule motion that char¬
glion to the inner hair cells. acterizes sound.
Reissner’s membrane One of two membranes mak¬ Wavelength The distance from one peak to the next
ing up the cochlear duct. of a sound wave.
Round window The membrane at the base of the
cochlea facing the middle ear.
 ' CHAPTER
7
Hearing
DETECTION OF SOUNDS
Temporal, Frequency, and Binaural
Interactions
Auditory Masking
Sound Discrimination
Sound Localization
Direction Cues: Simple Tones
• • Direction and Distance Cues: Complex
Sounds
Physiological Mechanisms
SUBJECTIVE DIMENSIONS OF SOUND
Loudness
Pitch
• • Theories of Pitch Perception

177
178 Chapter 7/Hearing

T he moon was yellow and full in the night sky.

Tamara and Scott crouched nervously under
mining an observer’s absolute threshold for sound
presented through earphones we measure the sound
the large old oak tree and stared at the ancient man¬ pressure level (see Chapter 6) associated with the
sion. Giving each other a reassuring glance, they threshold stimulus very close to the eardrum. This
advanced to the edge of the porch. They weren’t value is called the minimum audible pressure.
sure what to expect, but the sound that greeted their We might argue that this is a rather artificial situa¬
arrival was utterly chilling: it sounded exactly like tion, since the sound wave at the eardrum has al¬
someone scraping metal claws across a piece of ready been somewhat amplified and distorted
slate. The grating shriek shattered their determina¬ during its travel through the ear canal. Another,
tion and sent them racing toward home in a panic. perhaps more natural, procedure would be to deter¬
Inside the mansion, a timid, but very reclusive her¬ mine the absolute auditory threshold for an ob¬
mit gave thanks that he had paid attention when his server sitting in an open space that is free of echoes
introduction psychology instructor had given the and other distortions. In this situation, sounds are
lecture about the sounds most feared by people. presented by a speaker placed at various angles to
Perhaps he had bought himself another few years the observer’s ear and the intensity of the threshold
of seclusion. He ardently hoped so. stimulus is measured at the location of the observ¬
No one knows why the sound of metal (or fin¬ er’s head. This measurement of threshold is called
gernails) on slate is so aversive to so many people; the minimum audible field (indicating that the in¬
it isn’t the high frequencies present because remov¬ tensity of the threshold stimulus was measured in a
ing them makes little difference (Halpem, Blake & free field rather than at the eardrum).
Hillenbrand, 1986). Perhaps it is an ancient mem¬ In a classic study at Bell Telephone Labora¬
ory of sounds made by predators or fellow primates tories, Sivian and White (1933) systematically var¬
in distress. Whatever the reason, the phenomenon ied the frequency of a pure-tone stimulus as they
demonstrates the powerful effect sound can have on took a series of threshold measurements under
behavior and the richness of the psychological ex¬ carefully controlled conditions. Their results are
perience of sound. In this chapter we investigate shown in Figure 7-1. Notice that the absolute
some of this richness, concentrating on the more threshold is different for different frequencies of
basic sensations associated with our perception of sound. The ear appears to be most sensitive to
sound (in Chapter 12 we deal with the more com¬ sounds with frequencies between 1000 and 5000
plex phenomena of speech and music perception).
Every sound gives rise to a number of perceived
qualities including location, duration, loudness, Dynamic
range , \ Threshold
pitch, timbre, volume, and density. Although some 140
of pain
of these sensory qualities seem to be associated 120
S'
100
with particular aspects of the physical stimulus, as Minimum
is often the case in many kinds of perception, our audible
pressure
sensation from a sound is often not directly pre¬
dictable merely from knowledge of the physical na¬
ture of the stimulus. Minimum
-20 audible field

100 500 1000

DETECTION OF SOUNDS 5000 10000

Frequency (Hz)
Clearly, the minimum auditory experience is the
FIGURE 7-1 The dynamic range of hearing,
detection of the presence of a sound. What is the from minimum audible intensities to the
minimum sound intensity we can hear? In deter¬ threshold of pain. (Based on Sivian & White, 1933)
 Detection of Sounds 179

Hz, being about 100 times less sensitive to a sound females, do sometimes have peaks in the 3000-Hz
at 100 Hz than to a sound at 3000 Hz. Notice that range. You can demonstrate for yourself the effect
the minimum audible field measurements are con¬ of frequency on your ability to detect sounds using
siderably lower than the minimum audible pres¬ Demonstration Box 7-1.
sures. The reasons for this are rather complex. The lower limit of sensitivity for the ear seems
Probably the most important is that the free-field to be determined by the sound of blood rushing
situation allows resonances and amplifications from through the tiny vessels in the middle and inner ear
the shape of the pinna (the outer cup of the ear) (or perhaps the random noise generated by motion
and the ear canal to come into play. This is sup¬ of the air molecules; see Hudspeth, 1985), whereas
ported by the fact that we are most sensitive to the upper limit is determined by the stimulus inten¬
sounds of 3000-4000 Hz in free-field presentation, sity that produces pain. The difference between the
which is also in the range of the natural resonance absolute threshold and the pain threshold for a par¬
frequency of the external ear canal. From an adap¬ ticular frequency of sound waves defines the dy¬
tive point of view, the benefit that humans derive namic range of the ear for that frequency (see
from increased sensitivity in this particular range of Figure 7-1). For stimuli with frequencies between
frequencies may at first seem obscure, although about 1000-5000 Hz the ear has a dynamic range
some important frequencies found in speech are in of up to 150 dB, which is equivalent to a 7.5-mil¬
this range. Another, somewhat macabre possibility lionfold increase in sound pressure from the weak¬
is, as Milne and Milne (1967) speculate, that we est sound detectable to the most intense sound
leave a “channel open—as though reserved for tolerable. Few stereo systems can approach the dy¬
emergencies—for any high-pitched scream.” In namic range with which you were born.
fact, screams of agony or terror, especially those of The dynamic range of the ear is distinct from

DEMONSTRATION BOX 7-1. Sound Frequency and Threshold

Many people are aware of the problems associated radio, both of which lack loudness compensation cir¬
with replaying recorded music so that it sounds as it cuits, would be perfect. Find a station (or a record)
did when it was recorded. Recording techniques re¬ where a full orchestra is playing. Turn down the
produce the frequencies produced by musical instru¬ sound and listen to the instruments you can hear.
ments, but the replay is often at a lower intensity. Now, gradually turn up the sound. As you do this,
Most of the sounds of musical instruments lie in fre¬ you will find that you become more aware of the bass
quency ranges where the absolute threshold is most violin and cello, the larger brass pieces, such as the
affected by changes in frequency. Thus, unless you tuba, and some of the lower notes of the harp or bas¬
listen to recordings of an orchestra at reasonable in¬ soon, as well as some of the higher tones from the
tensity levels, you will not hear many of the frequen¬ violins, flutes, and piccolos. When the volume has
cies produced by the instruments. Many high-quality been considerably increased so that you can hear the
audio amplifiers have been modified to include cir¬ entire orchestra and many of the pieces (placing your
cuits that compensate for such psychological mecha¬ ear close to the speaker helps), gradually turn down
nisms. These circuits are set to emphasize very low the volume again. Now, many of the lower and
and very high frequency sounds. higher frequency instruments seem to disappear as
For this demonstration you will need a radio or an¬ certain frequencies they produce drop below thresh¬
other sound source that produces orchestral music. A old. The middle frequencies of the orchestra, how¬
cheaper unit, such as a portable radio or your car ever, are still quite audible.
180 Chapter 7/Hearing

the frequency range over which our ears respond to sounds must be more intense than longer sounds in
sound. Young adults can hear sounds between order to be detected with the same (threshold) like¬
about 20 and 20000 Hz, and some young children lihood. This relationship is a good approximation
can hear sounds with frequencies up to 27000 Hz. for threshold sounds up to a duration of about 200
Unfortunately, with age a progressive loss in sen¬ msec. Beyond 200 msec, increasing stimulus du¬
sitivity occurs, particularly for higher frequencies, ration does not seem to improve our ability to de¬
so that this range gradually decreases as we grow tect sound. Notice the strong similarity to Bloch’s
older (see Chapter 16). Demonstration Box 7-2 law, which shows a similar relationship between
provides a simple test for the upper range of your time and intensity in the visual system, as we dis¬
own hearing. cussed in Chapter 4.
We may also increase the likelihood that a
sound will be heard by increasing the number of
Temporal, Frequency, and Binaural
different tones, or frequencies, that are presented
Interactions
together. Suppose we present an observer with two
A number of factors other than frequency and in¬ tones, neither of which would reach threshold by
tensity determine our ability to detect sounds. One itself. Even if each tone is only about half the in¬
of these factors is the duration of the stimulus. The tensity needed for threshold, a sound will still be
auditory system seems to act as if a fixed amount heard. It seems the nervous system adds the neural
of sound energy is necessary to stimulate the ear responses of different tones, producing a composite
sufficiently so that we hear a sound. It doesn’t response based on the sum of the intensities of the
seem to matter if this energy comes at a high inten¬ various single stimuli. The tones should not differ
sity over a short time interval or at a lower intensity in frequency by too much, however, or their ener¬
over a longer time interval. We can describe this gies will not sum, and the threshold intensities will
relationship algebraically as be the same as if we presented each tone alone.
Just as there was a critical duration beyond which
T = 1 x D temporal summation did not occur, there is a criti¬
cal band of frequencies beyond which adding tones
where I is the intensity of the sound, D is its du¬ does not facilitate detection (Scharf, 1975). This
ration, and T is a constant value necessary to reach critical band is not the same width for all frequen¬
threshold. This simple formula says that brief cies, being much narrower for low frequencies than

DEMONSTRATION BOX 7-2. High-Frequency Hearing Limits

You can make a simple test of your own high- much younger. You should find that the older individual
frequency hearing using your television set. Turn it cannot hear this sound, whereas the younger one can.
on and then lower the sound completely. Now lean You might also try moving away from the set (if pos¬
over the back of your set and listen for a soft, high- sible) until you can just hear the sound. This is your
pitched whine. If you can hear it, this means that you threshold distance. Now have your other observers do
can detect frequencies on the order of 16000 Hz. the same and determine their threshold distances. The
Now, try this test on someone who is considerably greater your threshold distance, the more sensitive
older than you are and then with someone who is your ear is to these high-frequency sounds.
 Detection of Sounds 181

it is for high frequencies. Thus, if we start with a stimuli do not have to occur simultaneously in the
400-Hz tone, adding a tone between 350 and 450 two ears. If the tones are presented to the ears one
Hz will improve our ability to detect the sound, but at a time, and the total stimulus duration of the
adding a tone beyond these limits will not. If we combined input is less than 200 msec, the pair of
started with a 5000-Hz tone, however, any added tones will be detected even if each individual tone
tone between about 4500 and 5500 Hz would im¬ is only about one-half of the intensity needed to
prove our ability to detect the sound. The band¬ reach threshold when presented monaurally (Schen-
width has increased from 100 Hz in the lower kel, 1967).
frequencies to 1000 Hz at the higher. Figure 7-2 In addition to lowering the absolute threshold,
demonstrates how the critical bandwidth varies presentation of the same stimulus to the two ears
with frequency. causes the subjective impressions of loudness from
An additional factor affects our ability to de¬ each ear to add together (see e.g., Marks, 1979b).
tect the presence of sound stimuli. When sounds Thus, a binaural presentation will sound about
are presented to both ears, as opposed to only one, twice as loud as a monaural presentation of the
absolute thresholds are lowered. Presentations to same tone. If you have a sound source nearby such
one ear are called monaural (from the roots mon as a radio or a television, you can demonstrate this
for “one” and aural for “ear”); presentations to for yourself by assessing the loudness when you
two ears are called binaural (from the root bi for hear the source with two ears, then covering one
“two”). At first it was believed that this was only ear and noting how the apparent loudness dimin¬
because one of the ears was more sensitive than the ishes.
other, and the most sensitive ear determined the ab¬
solute threshold (Sivian & White, 1933). However,
Auditory Masking
later work demonstrated that the threshold for two-
ear stimulation is about one-half of that for one-ear We have all been in a noisy meeting, convention,
stimulation (Chocolle, 1962). An interesting aspect or theater and found that we could not hear or un¬
of the interaction between the ears is that the two derstand a speaker very well. When the crowd
quiets down, however, we find that the speaker’s
voice is audible immediately. This illustrates that
whether a particular sound can be heard or not de¬
pends not only on its own intensity but also on the
presence of other sounds in the environment. We
just discussed how sounds can interact to facilitate
detection; in the present situation the effects are re¬
versed. Now we present an observer with a sound,
which is audible by itself, and then add another
sound only to find that the target tone can no longer
be heard. We usually say that the second tone (the
masker) is masking the first (the target). When
target and masker are presented at the same time,
we have simultaneous masking. A masking sound
Frequency of target tone (Hz) does not simply make all other tones more difficult
to hear. Masking sounds act rather selectively.
FIGURE 7-2 The relation between critical
A set of experiments demonstrating this was
bandwidth, within which added tones will facilitate
detection, and frequency of target tone. done by Zwicker (1958), who masked target tones
182 Chapter 7/Hearing

using a narrow band of noise with a middle fre¬ back to Figure 6-10 in the previous chapter, which
quency of 1200 Hz. He measured the threshold in¬ shows how the vibration pattern of the basilar
tensity for a listener to detect a target tone with and membrane varies with the frequency of a pure tone.
without the masking stimulus. When he measured Notice that tones of low frequencies produce a very
the amount of masking for several different fre¬ broad vibration pattern, extending over much of the
quencies of target tone, he obtained the results membrane, whereas tones of higher frequencies
shown in Figure 7-3. As you can see from the fig¬ produce vibration patterns nearer to the oval win¬
ure, as the intensity of the masking stimulus is in¬ dow and not extending as far along the membrane.
creased, the intensity of the test tone also must be Now look at Figure 7-4. Notice that when we have
increased for it to be audible. The most striking a weak test tone, and the masking noise is of a
aspect of these data, however, is the asymmetry of higher frequency than the target, the pattern of vi¬
the masking effect. Of course, the greatest masking brations set up in the basilar membrane by the
is found for tones that have frequencies that are masking noise only extends part way up the mem¬
very similar to the masker itself (the thresholds are brane. Because the lower frequency target tone vi¬
highest for tones of frequencies around 1200 Hz). brates more of the membrane, the target’s pattern
But there is still a great deal of masking of tones extends beyond the flank of the vibration pattern
higher in frequency than the masking sound (the produced by the masker. Thus, the target is detect¬
threshold curves fall relatively slowly to the un¬ able. However, the vibration pattern produced by
masked threshold curve on the right side of Figure the target tone when the noise is of a lower fre¬
7-3), whereas tones of a lower frequency are rela¬ quency than the target is completely covered by the
tively unaffected (the threshold curves fall quickly masker’s vibration pattern, and it is thus not de¬
to the unmasked curve on the left of Figure 7-3). tectable as a separate tone. The intensity of the
Why does the added noise most effectively higher frequency test tone in the presence of low-
mask tones higher in frequency than itself? The an¬ frequency noise must be increased (to the “in¬
swer may lie in the physiology of the ear. Turn tense” level in Figure 7-4) before its own vibration
pattern at last extends beyond that of the masker
and can be detected as a separate tone. You can
experience some aspects of the frequency-specific
effect of a masker by performing Demonstration
Box 7-3.
Although sound-masking effects seem to be
largely explained by the interaction of the patterns
of vibration on the basilar membrane, this expla¬
nation is not adequate for all masking phenomena.
For example, consider what happens if target
and masker are not presented at the same time. If
the masker is presented first, followed after some
interstimulus interval by a brief target, the de¬
creased ability to hear the target is called for¬
ward masking. Many studies (see Zwislocki, 1978)
FIGURE 7-3 Thresholds for a pure-tone target in
have found, as you might expect, that the forward
the presence of a narrow band of masking noise
centered at 1200 Hz. The higher the curve, the masking effect increases as the intensity of the
higher the threshold, hence the more effective the masking sound increases, and decreases as the in¬
masking. (Based on Zwicker, 1958) terstimulus interval increases. For interstimulus in-
 Detection of Sounds 183

Weak tone

Distance along basilar membrane

Medium tone

frequency than noise frequency than noise

Masking noise Target tone

vibration pattern vibration pattern

FIGURE 7-4 The interactions of patterns of vibration of the basilar membrane resulting from a target
and a noise stimulus. (Based on Scharf, 1964)

tervals longer than 300 msec there is no measurable higher than the masker. It is unlikely, however,
forward masking. Also, in general, the lower the that we can explain these effects by interaction of
frequencies of both the target and mask pair, the excitation patterns on the basilar membrane, since
more masking takes place (Jesteadt, Bacon & Leh¬ the excitation of the masker is not present when the
man, 1982). In addition, the same asymmetry we target is presented at a later time. Some sort of in¬
discussed above for simultaneous masking shows teraction of more central neural processes must be
up in forward masking: there is little masking of involved. For the situation discussed above, it is
target tones with frequencies lower than the mas¬ likely that the masker is lowering the sensitivity of
ker, but a great deal for target tones of frequencies the hair cells, or their synapses with auditory nerve
184 Chapter 7/Hearing

DEMONSTRATION BOX 7-3. Auditory Masking

To experience several different masking phenomena the car engine. Turn up the volume until the high and
you need two major sources of sound, one for a low frequencies (which should now be masked) are
masking sound, and one for the target sound that will just barely audible again and take notice of the differ¬
be masked. Good sources are the noise of a car en¬ ence between the volume settings before and after the
gine for a masking sound and the car radio for a noise was introduced. You could map out a masking
source of target sounds. If you have a car with a ra¬ curve for particular frequencies in a piece of music by
dio, get into it and turn on the radio without starting varying the revolutions per minute of the motor to
the engine. Find some music with a good range of vary the intensity of the noise and by varying the fre¬
frequencies. Classical music is best, but any music quency of the sounds whose audibility you are using
will do. Modern music with a lot of steel guitar as a criterion for radio volume adjustment. Note that
(country) or electrically amplified guitar (rock) is also even with intense masking noise, you can still hear
good. Take particular note of the high and the low the middle frequencies, where most of the singing is,
frequencies. Turn the volume knob on the radio to an while the higher and lower frequencies are masked.
intensity where you can just barely hear these fre¬ This is a reflection of the superior sensitivity of the
quencies. Now start the car motor. Press on the ac¬ ear to these frequencies. You also experience speech
celerator (with the car out of gear!) to make the en¬ masking in your car. When the masking noise is of
gine turn over at high revolutions per minute. This sufficient intensity (be careful not to damage your en¬
creates a source of intense broad-band masking noise. gine), even the middle frequencies (where most
Now listen for the high and the low frequencies that speech sounds occur) are masked, and you cannot un¬
were clearly audible in the music before you started derstand the singer or the radio announcer.

fibers, to stimulation by the target tone, thus raising masker could build up faster than excitation caused
the threshold for that tone. by the target tone, thus overlapping with it in time
What happens when the target tone precedes and cancelling it to some extent, even when the
the masker? Nothing, you might think; how could target occurs appreciably earlier than the masker.
a tone that follows another affect our perception of In addition to separating target and masker in
the first tone, which has already been processed? time, we can separate them by presenting a target
Yet backward masking does occur, albeit some¬ sound to one ear and a masking sound to the other.
what differently from forward masking. For in¬ This is called central masking, since again there
stance, your ability to hear a click may be reduced can be no interaction of the sounds on the basilar
if another click follows it by as long as 25 msec membrane and the masking is therefore assumed to
(for loud masking clicks). Backward masking is take place in more central brain areas. In this situ¬
more difficult to measure for tones than for clicks, ation the masker must be about 50 dB more intense
because tones must extend longer in time, but it than when a masking sound is presented to the
does occur. When a tone is masked with noise, the same ear as the target tone. Under these conditions
masker may have some effect on the threshold of a it can be shown that the effect of the mask is usu¬
tone that begins up to 400 msec before the masker ally much more symmetrical and does not spread
is turned on (Wright, 1964). The explanation of so widely as we vary the frequency of the test tone
these backward masking effects is still not clear. (Zwislocki, Damianopoulos, Buining & Glantz,
One possibility is that inhibition caused by the 1967). Only when the frequency of the masking
 Detection of Sounds 185

sound is quite low (less than 200 Hz) is there ap¬ tion in respect to the perception of sound: “How
preciable asymmetry of masking (Billings & Sto- different must two sounds be in order for the dif¬
kinger, 1977). ference to be detected reliably?” To answer this
Interaction between the two ears can give rise question precisely, we must separate two of the
to some strange and interesting phenomena. For physical dimensions along which a sound stimulus
example, Hirsh (1948) presented a pure tone plus a may differ, namely, intensity and frequency.
broad-band noise (that is, one containing many fre¬ Let us begin by considering our sensitivity to
quencies) to the same ear. He adjusted the target so intensity changes. A good deal of care must be
that it could just be heard above the background taken when studying such abilities, since turning a
noise. Next he presented some additional noise to tone on or off, or changing its frequency or inten¬
the other ear, so that the two noises were in phase sity abruptly, can cause the perception of a
(meaning that the peaks and valleys of sound pres¬ “click.” In a threshold situation an observer might
sure coincided, as we discussed in Chapter 6). Un¬ respond to this click rather than to the actual inten¬
der these circumstances the target tone, instead of sity or frequency change in which we are inter¬
being more difficult or perhaps impossible to hear, ested. Because of this, Riesz (1928) at the Bell
became more clearly audible. To reach the thresh¬ Telephone Laboratories resorted to a rather elabo¬
old value again, it was necessary to lower the tar¬ rate technique based on a phenomenon known as
get’s intensity. What seemed to be happening was beats. When we simultaneously listen to two tones
that the two masking tones were masking each that are similar in intensity but slightly different in
other. Again, since the two noise masks were com¬ frequency, we may perceive the occurrence of
ing into different ears and did not share the same beats—a throbbing quality or an alternating rising
basilar membrane, the interactions must have been and falling in loudness of the sound, much like the
occurring more centrally, that is, in the brain. vibrato of a singer. The frequency with which the
Such central interactions become very impor¬ loudness fluctuates is the frequency difference be¬
tant when we consider the processing of more com¬ tween the two tones that are combined. The air
plex and meaningful sounds, such as speech (see compressions and rarefactions in the sound waves
Chapter 12). Interestingly, it is sometimes impos¬ will add when the maximum of one wave is occur¬
sible to ignore components of complex sounds that ring at the same time as the maximum of the other,
are far from the target in frequency, resulting in and subtract when the maximum of one wave co¬
what is called information masking. When a mask¬ incides with the minimum of the other. For exam¬
ing sound made up of several different frequencies ple, when two tones differ by 3 Hz, the maxima of
chosen at random is presented simultaneously with the two sound waves will coincide (add) three
a target, the target is more difficult to detect even times each second. The sound will thus seem to
if none of the frequencies is particularly close to wax and wane at 3 Hz (three times per second). As
the target frequency (Neff & Green, 1987). the frequency difference becomes larger, the discri-
minability of the beats decreases. At a large enough
difference between the two tones, the sound begins
Sound Discrimination
to take on a harsh or rough and grating quality.
In some respects the problem of masking is really Figure 7-5 shows how two sound signals can com¬
a discrimination problem, of much the same sort as bine to form a separate beat frequency.
that discussed in Chapter 2. Basically, the observ¬ Riesz (1928) used the perception of beats to
er’s task is to discriminate the target sound from determine the limits of intensity discrimination in
the masking sound. We may simplify this problem an observer. He first presented a tone that was
somewhat by asking the basic discrimination ques¬ clearly audible by itself (say, at a frequency of
186 Chapter 7/Hearing

Time-
3 msec

A1000 Hz
sine wave

A 1100 Hz
sine wave

The two signals

above, combined

FIGURE 7-5 A 1000-Hz sine wave is added to a 1100-Hz sine wave to give a beat frequency of 100 Hz.
This means that the beat pattern repeats every 10 msec as the overall envelope of sound pressures varies.
(From Lindsay & Norman, 1977)

1000 Hz). He next presented another tone that was of our ability to discriminate sounds from one an¬
close enough in frequency (say, 1003 Hz) so that it other. Figure 7-6 shows Riesz’s (1928) results for
would cause the perception of beats if it were in¬ how the Weber fraction varies as we vary the inten¬
tense enough. He gradually increased the intensity sity of the standard (/) stimulus. Notice that we
of the added sound until the listener first detected have plotted four different curves for four different
the fluctuation in loudness caused by the beating. frequencies. As you can see, the size of the Weber
On the basis of this he could compute the differ¬ fraction is smallest (discrimination is best) for stim¬
ence threshold, without contaminating the sound uli in the middle range of frequencies. Increasing
with the “click” caused by sudden changes in in¬ or decreasing the frequency results in a decrease in
tensity at onset or offset of the stimulus. our ability to discriminate intensity changes, al¬
You will probably recall from Chapter 2 that though such variation in discrimination with
one measure of our ability to discriminate between changes in frequency are not always found to be as
two stimuli is given by the Weber fraction. We de¬ large as those shown here (see e.g., Jesteadt, Wier
fined the Weber fraction as AHI, where A/ equals & Green, 1977), nor is the Weber fraction always
the intensity change necessary to be just noticed, found to drop off so smoothly with intensity (Long
and I is the standard stimulus from which the & Cullen, 1985). For moderate stimulus intensities
change is taken. This fraction represents the pro¬ and frequencies, however, the Weber fraction is
portion by which a stimulus must be changed in rather constant. It would, of course, be perfectly
order for us to detect that change. Thus, a Weber constant if Weber’s law were completely true, as
fraction of 0.5 means we must increase (or de¬ we discussed in Chapter 2. More modem measure¬
crease) the intensity of a stimulus by 50 percent in ments (using pure-tone stimuli that were turned on
order to discriminate the change. and off gradually in order to avoid the spurious
We can use the Weber fraction as a measure click we mentioned above) seem to indicate even
 Detection of Sounds 187

FIGURE 7-6 Intensity discrimination measured in terms of the Weber fraction for various intensities
and frequencies of standard stimuli. (Based on Riesz, 1928)

less of a violation of Weber’s law, with much less gle chance available when monaural presentation is
of a rise in the Weber fraction at lower intensities used. Similarly, the intensity difference threshold
(Green, Nachmias, Kearny & Jeffress, 1979; is smaller the longer the duration of the stimuli,
Hanna, von Gierke & Green, 1986). Figure 7-6 over a range of 2 msec to 2 seconds (Florentine,
shows that the auditory system is capable of de¬ 1986). The longer stimulus durations give more in¬
tecting an approximately 10 to 20 percent change formation about the stimulus difference and thus
in stimulus intensity (5% under optimal conditions) more opportunity to detect it.
across a broad range of frequencies and intensities, A similar, but special, case of intensity dis¬
covering the stimulus range where most of our every¬ crimination arises in what has been called profile
day hearing takes place. analysis (Green, 1987). In this situation, a listener
We have discussed only studies of sound dis¬ is presented with a complex sound made up of
crimination where the stimuli were presented to a many different frequencies (often more than 20 of
single ear (monaural presentation). You might ex¬ them) and is asked to discriminate a slight change
pect that, just as is the case with detection, it would in the intensity of one of the component sounds.
be easier to discriminate intensities of sounds pre¬ For example, say a complex sound (a standard
sented to both ears (binaural presentation) than “flat” profile) consists of pure tones with frequen¬
those presented only to a single ear. In fact, inten¬ cies of 250, 500, 1000, 2000, and 4000 Hz, all
sity difference thresholds are about 33 percent with amplitudes of 60 dB. A comparison sound
smaller when the stimuli are presented simulta¬ could consist of tones with the same frequencies
neously to both ears than when they are presented and amplitudes except, for instance, the 1000-Hz
only to one or the other ear alone (Jesteadt & Wier, tone, which has been raised to an amplitude of 65
1977). This is probably because the binaural pre¬ dB. A listener would be asked to say whether the
sentation gives the observer two chances to hear the standard or comparison tone complex has occurred
difference (one in each ear) rather than just the sin¬ at a particular time. Under these conditions
188 Chapter 7/Hearing

intensity thresholds can be much smaller than for the telephone company would have an interest in
single tones. Thresholds actually decrease as the discovering the limits of our ability to discriminate
number of component frequencies is increased from sounds). The basic experiment involves presenting
3 to 21 (see Green, 1987). The auditory system an observer with a tone of a given frequency and
seems to carry out a “profile analysis,” which is a intensity and then varying (modulating) the fre¬
comparison of the relative intensities of compo¬ quency of the tone by larger and larger amounts
nents in a complex sound, and seems better able to until the observer is just able to detect a change in
detect “bumps” in the profile (changes in a single pitch. Again, we may measure the limits of dis¬
component) when there are more component fre¬ crimination using the Weber fraction. In this case,
quencies present. It is probably not surprising that however, the fraction consists of Af/f, where /rep¬
listeners must practice for many trials before they resents the frequency of the standard tone, and A/
can detect these profile differences. Once they can, represents the change in frequency necessary to be
however, the thresholds so obtained are robust and just noticed as different from the standard. Figure
consistent with thresholds obtained with older 7-7 shows Shower and Biddulph’s measurements
methods. For example, profile discrimination is of the Weber fraction for frequency discrimination
best when the intensity increment is added to com¬ for a number of different intensity levels and a
ponents of intermediate frequencies, as in the ex¬ broad range of frequencies. Notice that above 1000
ample above (Bernstein & Green, 1987). Hz the Weber fraction is constant, and quite small
Thus far, we have dealt with the question of for most intensity levels (around 0.005). This
discrimination of differences in intensity of sound means that if we presented a listener with tones of
stimuli. We may also ask, “By how much must 1000 Hz and 1005 Hz, this small difference in fre¬
two tones differ in frequency for this difference to quency (1/2%) would be detectable. At lower in¬
be noticed?” Again, the classic study was done at tensity levels our discrimination of frequency
the Bell Telephone Laboratories, this time by differences is not quite this good.
Shower and Biddulph (1931) (you might guess that As was the case for intensity discrimination.

FIGURE 7-7 Frequency discrimination measured in terms of the Weber fraction for various intensities
and frequencies of standard stimuli. (Based on Shower & Biddulph, 1931)
 Detection of Sounds 189

more recent studies of the effect of intensity and the sound intensity at one ear is less than the inten¬
frequency on the frequency difference threshold sity at the other ear.
have used as stimuli tones that are turned on and Measurements have been made of intensity
off gradually and that are only on for a brief dura¬ differences between the ears as both the azimuth
tion (500 msec or so). The method of constant of the sound source and the frequency of the emit¬
stimuli (see Chapter 2) applied to these tones then ted sound are varied. These show that the intensity
yields a measurement of the difference threshold. difference between the ears increases as a sound
The most comprehensive of the modem studies was source is moved toward one side. In addition, al¬
done by Wier, Jesteadt, and Green (1977). Their though low-frequency sound waves (those less than
results were similar to those of Shower and Bid- 3000 Hz) bend around the head very readily, high-
dulph (1931), that is, the Weber fraction for fre¬ frequency tones tend to rush right past the hidden
quency depended on both frequency and intensity ear unless deflected into it. This exaggerates the
in a way similar to that shown in Figure 7-7. intensity differences caused by the presence of a
Finally, as was the case for intensity discrimina¬ sound shadow for higher frequency sounds. The
tion, binaural frequency difference thresholds are changes in intensity difference as the angle of the
about 33 percent smaller than are monaural ones sound source changes can serve as a cue to direc¬
(Jesteadt & Wier, 1977). tion. A large intensity difference between the two
ears indicates that the source of the sound is posi¬
Sound Localization tioned to one side; the greatest intensity difference
occurs when the sound source is located at 90 de¬
Sounds are usually perceived as having a location grees azimuth. The ear receiving the loudest input
in space, as emanating from sources to the right or is perceived as closest to the sound source.
left of, in front or behind of, above or below our When a sound source is at an angle, sound
bodies. Some sounds appear to come from close must travel different distances to reach the two
by, others from a distance. Our auditory systems ears. This is always the case unless the sound
use a variety of aspects of sound to construct a sort source is positioned at either 0 or 180 degrees,
of auditory space, with our bodies at the center, when the ears are at equal distances from the
within which sounds can be localized and their source of the sound. Because sound takes time to
sources approached (“Hey Jill, nice to see you!”) travel through space, there is a time difference in
or avoided (“Grrrrooowwwlll”). the arrival of the sound at the two ears. For exam¬
ple, for a sound at 0 degrees azimuth, there is no
Direction Cues: Simple Tones
time difference between the stimulation of the right
When a sound comes from some distance away and and the left ears since they are at equal distances
from a particular angle to the listener, a number of from the sound source. For a sound at 90 degrees
cues indicate the direction to the right or left, the azimuth in either direction, however, the ear closer
azimuth, of the sound source. Figure 7-8 shows a to the sound is stimulated approximately 0.8 msec
typical situation when a sound is coming from a earlier than the hidden ear. Intermediate azimuths
source positioned at about 45 degrees left azimuth. result in intermediate values for this time difference
Notice that one ear receives the sound directly from (see Figure 7-8). Such a time difference may be a
the source while the other ear is in what could be cue to the location of the sound source and may
called a sound shadow. The shadowed ear receives result in the experience of an apparent direction for
only those sounds from the source that are bent it. You may demonstrate the effects of this time
around the head, or diffracted by the edge of the difference on direction perception for yourself us¬
head. The presence of a sound shadow means that ing Demonstration Box 7-4.
190 Chapter 7/Hearing

180° azimuth

FIGURE 7-8 The path of sound to the two ears for a sound source at 45 degrees left azimuth. (Based on
Lindsay & Norman, 1977)

Under certain circumstances, the time differ¬ at one ear (as it would if the sound source were
ence between the stimulation of the two ears results positioned at about 62 degrees azimuth), it would
in a phase difference. If a sound is arriving earlier always be 0.5/1 or 1/2 cycle ahead of the sound
at one ear it will be in a different portion of its arriving at the opposite ear.
cycle of compression and rarefaction of the air Although phase difference could be a cue to
molecules than the sound arriving at the other ear sound direction, it provides ambiguous information
(this aspect of sound is discussed fully in Chapter when we consider the full range of sound frequen¬
6). This is especially true for low-frequency cies. For instance, with a tone of 10000 Hz at 62
sounds, where the time taken to complete one cycle deg azimuth, the time difference between the ar¬
is more than the maximum time difference of the rival of the sound at the two ears would once again
arrival of sound at the two ears. For example, it be 0.5 msec. However, a 10000-Hz tone takes only
takes a 1000-Hz tone exactly 1 msec to complete 0.1 msec to complete one cycle. This implies a
one cycle. If such a tone arrived 0.5 msec earlier phase difference of 0.5/0.1 or 5 cycles. Thus the
 Detection of Sounds 191

DEMONSTRATION BOX 7-4. Time Differences and Auditoiy Direction

For this demonstration you will need a length of rub¬ section of tube on one side, the sound must travel
ber hose or flexible plastic tube. Hold one end up to farther before reaching one of your ears. This delay
each ear as shown in the figure. Now, have a friend is perceived as a shift in direction of the sound. No¬
tap the tube using a pencil. At the point where she tice how the sound seems to change direction as dif¬
taps, a sound wave starts moving in both directions ferent parts of the tube are tapped, causing different
down the tube. If she taps so that there is a longer patterns of sound delays.

sound at the ear closest to the source is 5 cycles quencies by different amounts. Such differential
ahead of the sound arriving at the more distant ear. delays and amplifications apparently provide cues
However, every cycle is identical. Therefore, how to the location of complex sound sources, espe¬
can the observer tell just what the phase difference cially their elevation (Oldfield & Parker, 1984),
may be? It could be 5, 4, 3, 2, or 1 cycles. Even since a positive relation exists between the apparent
at the lower frequencies, where potentially the location of sounds and the amount of delay or am¬
phase difference cue could be more useful, the plification provided by the pinnae (see also Oldfield
same phase difference is characteristic of sounds & Parker, 1986).
positioned directly opposite to one another (in ref¬ In 1907, Lord Raleigh proposed a dual, or
erence to a line drawn through the head in any di¬ two-process, theory of sound localization. He sug¬
rection). gested that we localize low-frequency sounds by
Finally, it has also been argued that the pinnae using time or phase differences, or both, at the two
(the fleshy parts of the ears outside of the head) ears, and that we localize high-frequency sounds by
delay (Batteau, 1967) or amplify (Butler, 1987; using the intensity differences at the two ears
Flannery & Butler, 1981) sounds of different fre¬ caused by the sound shadow and differences in
192 Chapter 7/Hearing

their distance from the sound source. This notion Direction and Distance Cues: Complex
has been confirmed by later research. For example, Sounds
Stevens and Newman (1934) had observers with
When we are in an ordinary room, the sound from
their eyes closed make judgments as to the direc¬
any source may go bouncing around the room, re¬
tion of a sound source. They played sounds of dif¬
flecting from the walls, ceiling, and floor many
ferent frequencies from a variety of azimuths and
times before it reaches our ears. Figure 7-10 illus¬
recorded the listener’s errors of localization for
trates this phenomenon. Why do we not experience
each sound. Their data are shown in Figure 7-9.
an overwhelming auditory confusion as these
The solid line in this graph represents a summary
sounds ricochet around us? Typically, we respond
of the data they collected, with errors averaged
only to the first of the many replicas of a particular
over all the locations at a particular frequency. As
complex sound in echo-producing surroundings.
you can see, most errors occur in the region of
We do not respond to the echoes that arrive several
1500-3000 Hz. Fewer errors occur above or below
milliseconds later. In fact, we do not even experi¬
this frequency range. We can interpret this as in¬
ence echoes until the reflecting surface, which hurls
dicating the efficient use of at least one cue in the
the sound back at us, is far enough away so that
low- and high-frequency ranges. Performance is
the echoes take a substantial time to reach us (more
worst in the midrange, however, where neither cue
than 35 msec or so). Groups of sounds that arrive
to localization is particularly useful. This interpre¬
at interstimulus intervals of less than 35 msec are
tation has been confirmed by Mills’s (1958) work
fused together into one sound. The first arrival ap¬
on the minimum audible angle, which is the
pears to be the major determinant of where in space
smallest amount of movement of a sound source
we perceive the sound source to be. This phenom¬
that can just be detected. The minimum audible an¬
enon is called the precedence effect and has been
gle also varies as a function of frequency and lo¬
cation of a sound source; the variations are
consistent with those observed in experiments like
those of Stevens and Newman (Mills, 1960).

FIGURE 7-9 Relative cue effectiveness in

arbitrary units for interaural intensity and phase
differences (dashed lines) as a function of
frequency. The solid line shows mean localization FIGURE 7-10 Some of the echoes produced by
errors as a function of frequency. (From Gulick, 1971. sound reflecting from the walls of a room. Unless
Copyright 1971 by Oxford University Press, Inc. Reprinted by the walls are quite far away, the echoes are not
permission. Data from Stevens &. Newman, 1934.) perceived. (From Lindsay & Norman, 1977)
 Detection of Sounds 193

extensively studied (see Rakerd & Hartmann, 1985; ments, which in turn may help to build up a rep¬
Wallach, Newman & Rosenzweig, 1949; Zurek, resentation of the immediate environment (Kobler,
1980). The experiments of Wallach et al. (1949) Isbey & Casseday, 1987). We might call this rep¬
indicated that the earlier of a pair of fused sounds resentation “auditory space” to distinguish it from
(separated by 2 msec) was 6-10 times more impor¬ our more familiar perception of space based on vi¬
tant than the later of the pair in determining the sion.
perceived direction of the sound source. Wallach et There are a number of other cues to the spatial
al. (1949) also pointed out that the precedence ef¬ location of complex sounds. As we suggested
fect is an important part of our ability to listen se¬ above, head movements are important in building
lectively to one source of sound out of a larger up our representation of auditory space. They serve
group of competing sounds (see the discussion of to resolve ambiguities of location, such as whether
the cocktail party problem in Chapter 15). You can the sound is in front of or in back of the listener
experience the effects of precedence on the locali¬ (Noble & Gates, 1985; Wallach, 1939). They also
zation of sound by using Demonstration Box 7-5. provide information indicating that the sound is
Under appropriate conditions, echoes can be really “out there” as opposed to one generated in¬
important to the judgment of the location of side the head, such as the ringing in the ears called
sounds. For example, blind individuals apparently tinnitus that some people experience if they take
use echoes to help them locate obstacles and thus too much aspirin or have a bad head cold.
avoid them (Supra, Cotzin & Dallenbach, 1944; Distance information is also carried by com¬
Worchel & Dallenbach, 1947). Animals such as plex sounds. One major source of this information
bats and whales have highly developed echoloca- is the relative intensity of a sound, with nearer
tion systems, similar to sonar, which they can use sources being more intense. Changes in sound dis¬
to locate objects with the same facility with which tance are thus reliably coded by changes in sound
we use our eyes (see Neuweiler, Bruns & Schuller, intensity (Mershon & King, 1975). Of course, it is
1980 for a review). They may even have special always possible that the more distant source is ac¬
auditory brain pathways analogous to the tectopul- tually emitting a stronger intensity signal,-making
vinar visual pathway (serving visual localization, this cue unreliable in determining the absolute dis¬
see Chapter 3) to control scanning head move¬ tance of a sound source unless the sound is a

DEMONSTRATION BOX 7-5. Precedence and the One-Speaker Stereo Illusion

For this demonstration you will need a radio, phono¬ speaker nearest you. You no longer get any sensa¬
graph, or tape recorder that has stereo speakers lo¬ tion of sound coming from the more distant speaker
cated about 2 m apart. Turn on some music and stand (although it still affects sound quality as you can
about midway between the two speakers, facing a demonstrate by turning it off). A few steps to the
point between them. You will notice that the sound other side will reverse this effect, making it appear as
seems to envelop you. It comes from both sides, and though all the sound is coming from the other
you can clearly identify sounds coming from one speaker. As you move toward a speaker, you alter the
speaker or the other. Take a few steps (you need not time that it takes for the sound to reach your ears.
go very far) toward one side where a speaker is lo¬ The precedence process then takes the sound arriving
cated. After only a step or two you will suddenly find first and emphasizes it, giving you the impression that
that all the sound seems to be coming from the all the sound emanates from that source.
194 Chapter 7/Hearing

familiar one. Through experience we build up object that could be the source. Thus, the ventrilo¬
memories of what a phone bell or a car engine quist’s dummy seems to be talking because its
sounds like when these sounds are made at differ¬ mouth moves and the ventriloquist’s does not (if
ent distances from us. In later encounters we can the ventriloquist is a good one). Of course, echoes
use this knowledge to judge how far away a sound and reverberation do not play a role in this effect
source may be based on the remembered loudness (Mershon, Desaulniers & Amerson, 1980). In ad¬
of other similar sound sources. dition, the illusion that a sound is coming from a
Another important source of distance infor¬ likely visual object can be so compelling that it can
mation is the relative amount of reverberation in affect the perceived loudness of the sound. If the
the impinging sound. As we stated above, sound sound seems to emanate from far away, it sounds
reaches our ears both directly from a source and louder than if it seems to emanate from close by
after being reflected from (reverberating from) var¬ (Mershon, Desaulniers, Kiefer & Amerson, 1981).
ious surfaces such as walls (see Figure 7-10). In Observers seem to correct for the fact that actual
general, as a sound source gets further away from sound intensity diminishes rapidly as the distance
an observer, the amount of sound that directly from the sound source increases, a phenomenon
reaches the ears decreases more rapidly than the termed loudness constancy (see Chapter 14 for a
amount reaching the ears after reverberation. Thus, discussion of the constancies).
the relative amount of “reverberation sound”
(which has a distinct quality, like an echo) is a cue
Physiological Mechanisms
to the distance of a sound source from an observer.
Bekesy was one of the first to investigate this cue The auditory system seems to contain neural units
systematically. In 1938 (Bekesy, 1960), he showed that respond to both time and intensity differences
that altering the proportion of reverberant sound al¬ between the two ears, which may, in turn, signal
ters judgments of perceived distances of sounds. the location of a sound source. For instance, some
More recent work (Butler, Levy & Neff, 1980; neurons in the superior olives, inferior colliculi
Mershon & Bowers, 1979; Mershon & King, 1975) (Semple & Kitzes, 1987), and auditory cortex of
has confirmed and extended this earlier work. various birds and mammals respond best to bi¬
Another cue to distance that seems to be as naural stimuli that reach the two ears at slightly
compelling as the amount of reverberation is the different times or intensities (see Erulkar, 1972 and
frequency makeup, or spectrum, of a complex Phillips & Brugge, 1985 for reviews). Different
sound. Sounds composed mostly of high frequen¬ neurons have different “best” interaural time dif¬
cies seem to come from quite nearby, and the more ferences, or different “best” interaural intensity
the sound is dominated by low-frequency compo¬ differences. In other words, different neurons are
nents the farther away its source seems to be. “tuned to different time or intensity differences
Butler et al. (1980) suggested that this is because between the two ears. Since these differences are
more distant sounds typically are more dominated cues to the location of sounds, we could say that
by low-frequency components, perhaps because the these tuned neurons encode sound location much as
high-frequency components are so easily blocked neurons tuned to sounds of different frequencies
by intervening obstructions. Again, our previous encode frequency. It is possible that such neurons
experiences may play a role in determining how far constitute a kind of map of auditory space, with
away we judge a sound to be based on the fre¬ each neuron having a region of auditory space to
quency spectrum. which it responds best, a sort of “auditory recep¬
A final important cue to the distance of a tive field” much like the visual receptive fields dis¬
sound source is the presence of a compelling visual cussed in Chapter 3.
 Subjective Dimensions of Sound 195

There are problems with this idea, however. there would be a direct correspondence between the
The major one is that the tuning of the neurons is experienced qualities of the sensation and proper¬
too gross to account for the accuracy with which ties of the physical stimulus. For a long time it was
animals, including humans, can localize sounds. In taken for granted that every qualitatively different
other words, the auditory “receptive fields” of psychological variable would reflect almost per¬
these neurons are too large to account for the de¬ fectly some corresponding quantifiable physical
gree of accuracy shown in actual behavioral data. variable. For example, it was firmly believed that
In some species, such as the bam owl, much the subjective dimension of loudness was a direct
smaller, more intricately organized auditory recep¬ reflection of the physical dimension of amplitude of
tive fields have been found using electrophysiolog- the sound-wave stimulus. In similar fashion it was
ical recording techniques (Knudsen & Konishi, believed that the subjective dimension of pitch
1978a). An interesting nuance in the bam owl is (whether a sound appears to be high or low in tone)
that the receptive fields of these neurons have a was simply the psychological experience of the fre¬
center-surround organization (Knudsen & Konishi, quency of the sound wave. This sort of mechanistic
1978b). That is, not only do these neurons fire viewpoint has been opposed by many investigators,
above their background rate in response to stimuli who have pointed out that we should separate con¬
in their best areas of space but also they are inhib¬ cepts and expressions that describe our conscious
ited in their response by sounds in areas outside or phenomenal experience from those that describe
their best areas, thus resembling, in many ways, the physical stimulus. The subjective qualities of
the center-surround organization of neurons at var¬ loudness and pitch are complex perceptions that de¬
ious levels of the visual system (see Chapter 3) and pend on the interaction of several physical charac¬
other parts of the auditory system (see Chapter 6). teristics of the stimulus, as well as the physical and
So far there has been no direct evidence that such psychological state of the observer.
center-surround neurons exist in the auditory sys¬ The deeply rooted older view maintained that
tems of mammals, but it is possible that the time at best the observer could be expected to distin¬
and intensity difference detectors are preliminary guish only two phenomenal dimensions (loudness
stages leading to such neurons. Interaction of time and pitch) because there are two predominant phys¬
and intensity difference detectors that have a ical dimensions (intensity and frequency). Ac¬
center-surround organization could give rise to tually, we can differentiate many qualitatively
higher-level neurons that have relatively restricted different experiences arising from sound stimuli.
receptive fields and might allow a fairly accurate These include not only pitch and loudness but also
mapping of auditory space. the perceived location of a sound (where it seems
to come from), its perceived duration (how ex¬
tended in time it appears), its timbre (that complex
quality that allows us to distinguish a note played
SUBJECTIVE DIMENSIONS OF SOUND on a clarinet from the same note played on a vio¬
lin), its volume (the sense in which it fills space
Up to now we have concentrated on describing our and seems large or small), and its density (a com¬
ability to detect the presence of a sound, or to dis¬ plex feeling of the compactness or hardness of the
criminate one sound from another. However, such sound), as well as consonance or dissonance (how
analyses do not deal directly with the subjective two tones seem to “go together” or “clash”). Our
quality of a sound as we experience it. Early in the auditory experience is composed of these and other
history of the psychological investigation of audi¬ sensory qualities—not simply the registration of the
tion, experimenters were inclined to believe that frequency and the intensity of the stimulus. We
196 Chapter 7/Hearing

have already discussed sound localization above; power function varies with stimulus frequency,
we discuss some of the other subjective qualities in being somewhat larger for frequencies lower than
more detail in the sections that follow. 1000 Hz (Scharf, 1978).
On the basis of his and others’ work, Stevens
suggested a new unit by which to measure loud¬
Loudness ness. He called this unit the sone. One sone is de¬
We cannot predict a subjective experience totally fined as the loudness of a 1000-Hz stimulus at an
from a single physical dimension. However, the intensity level of 40 dB. For most of the stimulus
experienced loudness of a sound is affected greatly range a linear relationship exists between the loud¬
by stimulus intensity. If everything else is held ness measured by the logarithm of the number of
constant, it is fair to say that as we increase the sones and the intensity measured in decibels. To
amplitude of the sound stimulus, we increase its double the loudness (for instance, from 1 to 2
apparent loudness. The experience of loudness, sones) we have to increase the intensity of the
however, is not identical to stimulus intensity, and sound by about 10 dB. For very weak sounds (be¬
decibels are not measures of phenomenal loudness. low 30 dB), however, the change in apparent loud¬
In order to study loudness, we must use psy¬ ness is much more rapid with increases in intensity
chophysical scaling procedures such as the magni¬ (see e.g., Canevet, Heilman & Scharf, 1986). This
tude estimation techniques we discussed in Chapter relationship is shown in Figure 7-11, which also
2, or the matching of one stimulus to another on a indicates the loudness in sones of some typical
different sensory continuum (cross-modality match¬ sounds.
ing). Stevens (1956) did a classic study of this type Sound intensity does not provide a full de¬
using magnitude estimation. He gave an observer a scription of loudness. For example, our perception
standard stimulus tone and a set of tones that varied of the loudness of a tone is also affected by its
in intensity but had the same frequency (1000 Hz) frequency. A typical procedure to measure the re-
as the standard. The standard tone was assigned a
value of 100 units of loudness. The observer sim¬
ply assigned numbers to the variable tone on the
basis of its perceived loudness. Thus, a tone that
sounded twice as loud as the standard would be
called 200 units, and a tone that sounded half as
loud would be called 50 units. Stevens found that
the perception of loudness varied according to a
simple equation:

L = al06

where L is the apparent loudness, I is the physical

intensity of the sound (in units of pressure ampli¬
tude), and a is a constant. The loudness of the
stimulus was increasing as approximately the 0.6
power of the physical sound pressure. Other expo¬
nents have also been found depending on the spe¬
FIGURE 7-11 The relationship between loudness
cific stimuli used and the test conditions employed (measured in sones) and stimulus intensity
(Marks, 1974). For example, the exponent of the (measured in decibels).
 Subjective Dimensions of Sound 197

lationship between frequency and loudness involves A series of such equal loudness contours is
presenting an observer with a standard tone of a shown in Figure 7-12. Each curve represents a dif¬
given frequency and intensity. She is then asked to ferent sound intensity for the standard tone in deci¬
adjust the intensity of another tone (differing in fre¬ bels. Notice that the lines are not flat. If sounds of
quency) until it matches the loudness of the first. various frequencies sounded equally loud when
When this is done for a number of comparison they were the same intensity, all of the curves
tones, we can plot a curve that describes the inten¬ would be straight lines. The fact that the contours
sity at which tones of varying frequencies appear rise and fall with frequency (much as the graph of
to be equally as loud as the standard tone. Such a absolute threshold varies with frequency) means
curve is called an equal loudness contour. that tones of equal intensity but of different fre-

FIGURE 7-12 Equal loudness contours. (From Lindsay & Norman, 1977. Data from Robinson & Dadson, 1956.)
198 Chapter 7/Hearing

quencies appear to differ in loudness. Tones of less with comparison tones of different durations. When
than 1000 Hz, or greater than 6000 Hz, must be we do this we get a curve similar to that shown in
considerably more intense to match the loudness of Figure 7-13. Thus a 2-msec burst of sound must be
tones between 1000 and 6000 Hz. Thus, tones in 15 dB in order to sound as loud as a 90-msec burst
the middle range of frequencies sound considerably at 5 dB. This sort of finding suggests that the au¬
louder than equally intense tones outside this range. ditory system may sum all the inputs coming in
What if we wished to compare the loudness of over a 200-msec window of time (Gulick, 1971).
tones of frequencies other than 1000 Hz, say an Other sounds occurring at the same time, or
8000-Hz, 40-dB tone and a 400-Hz, 30-dB tone? just before, a sound to be judged can also affect
Figure 7-12 shows that both tones fall nearly on the apparent loudness. For example, if a continuous
30-dB contour. This means they are both approxi¬ tone is played to one ear and an intermittent tone
mately equal in loudness to a 1000-Hz, 30-dB to the other, the loudness of the continuous tone
tone, and thus they also match each other in loud¬ appears to diminish with time (Botte, Canevet &
ness. We could be more precise and specify their Scharf, 1982). The reduction in apparent intensity
loudness in sones by finding the loudness in sones for a continuously presented stimulus is called au¬
of the 1000-Hz, 30-dB tone. By graphical (see Fig¬ ditory adaptation. Adaptation is weak for a con¬
ure 7-11) or computational methods we find this to tinuous tone alone, but can be quite dramatic when
be about 0.46 sone. Thus, both the 8000-Hz, different tones are presented to the two ears. The
40-dB tone and the 400-Hz, 30-dB tone have a apparent loudness of the continuous tone actually
loudness of 0.46 sone. Table 7-1 summarizes es¬ diminishes to zero if the tone in the other ear is
sential aspects of sones and other audiometric units close to it in frequency and is played for 40 sec
discussed in this chapter. (Botte, Baruch & Scharf, 1986).
Duration also influences the apparent loudness A related phenomenon, called auditory fa¬
of a tone. For tones briefer than about 200 msec, tigue, is caused by exposing the ear to very intense
we must increase intensity to match the loudness of sounds. The resultant reduction of loudness of
a longer tone. We could create an equal loudness other stimuli presented after the intense sound
contour, of a type similar to that for variations with ceases may persist for a considerable period of
frequency, by having a standard of fixed duration time, depending on the intensity and duration of the
and requiring an observer to match its loudness fatiguing stimulus. For instance. Postman and Egan

Table 7-1. Audiometric Units

Audiometric term Unit What is measured How measured

Pressure amplitude Dyne/cm2 Variation of sound pressure Measure peak compressive

from atmospheric force per 1 cm2 area
Sound pressure level Decibel (dB) Ratio of pressure amplitudes 20 log(PAP0)
of two sounds
Frequency Hertz (Hz) Number of cycles of Count cycles per second
compression/rarefaction
Loudness Sone Subjective impression of 1 sone = loudness of 1000-
sound intensity Hz tone at 40 dB
Pitch Mel Subjective impression of Pitch of 1000-Hz tone at 40
sound frequency dB is 1,000 mels
 Subjective Dimensions of Sound 199

Another factor that influences our perception

of loudness is the complexity of the stimulus. Most
of the sounds we hear in our everyday environment
are made up of mixtures of a large number of dif¬
ferent frequencies of sound. We can create a dif¬
ferent kind of equal loudness contour by asking
listeners to adjust the intensity of a pure tone of
1000 Hz until its loudness matches that of some
FIGURE 7-13 Equal loudness contour showing
the changes in intensity needed to maintain a complex sound. Suppose we take a complex sound
constant loudness as the duration of the standard composed of a group of frequencies centered
is varied. (From Gulick, 1971. Copyright 1971 bv Oxford around 1000 Hz. We can systematically increase or
University Press, Inc. Reprinted by permission.!
decrease the range of frequencies included. We
usually refer to the range of frequencies as the
(1949) exposed observers to an intense sound (115 bandwidth. As the bandwidth of the sound is in¬
dB) for 20 min. They then measured the sensitivity creased, the intensity of each of the component fre¬
of their observers over a period of several days. quencies is decreased in order to keep the overall
The results are shown in Figure 7-14. The horizon¬ intensity of the sound the same. Figure 7-15 dis¬
tal line represents pre-exposure sensitivity, and the plays an equal loudness contour for such a complex
other curves represent the hearing loss, which sound. Notice that when only a small band of fre¬
could be interpreted as a reduction in loudness, for quencies makes up the complex sound, increasing
varying periods of time following the exposure to the bandwidth does not affect our perception of the
the stimulus. As you can see, the largest hearing loudness of the stimulus. This is reasonable, since
loss immediately follows the exposure to the in¬ the overall intensity of the sound is not changing
tense noise; however, it persists to measurable ex¬ but only the number of different frequencies in¬
tent over a period of 24 hr. You can experience an cluded in it. Notice, however, what happens when
interesting analog to this experiment using Dem¬ the frequencies reach a critical bandwidth of about
onstration Box 7-6. 160 Hz. From this bandwidth onward, loudness

FIGURE 7-14 Prolonged reduction of loudness following exposure to an intense (115-dB) sound for 20
min. (Based on Postman & Egan, 1949)
200 Chapter 7/Hearing

DEMONSTRATION BOX 7-6. Auditory Fatigue

During an average day you are exposed to many loud. During the night your ears have recovered from
noises and sounds, from individuals who talk with the auditory fatigue caused by exposure to the sounds
you, from stereos, televisions, radios, and numerous you heard during the previous day. The quiet of the
other sources. Set a radio or a stereo to an intensity night has given you a chance to recover your sensitiv¬
level where the sound seems comfortable for listening ity, hence all sounds now seem louder. This may ex¬
in the evening before you go to bed. At the day’s end, plain why an alarm clock, whose bell seems low and
your auditory system has become fatigued by the on¬ pleasant when bought one evening in a department
going, persistent noise of the day. When you awaken store, will seem so jarring and loud the following
in the morning, however, you may find that the radio, morning.
set to the same sound level, will appear to be too

begins to increase as we include a greater number lus—the high notes on a piano have higher
of frequencies, although the overall intensity of the frequencies than the low notes. For instance, the
sound is unchanged (Cacace & Margolis, 1985; dominant frequency of A4 (the 49th key on the ex¬
Gulick, 1971; Scharf, 1978). tended piano keyboard pictured in Figure 7-12,
counting from the left to the right) is 440 Hz; the
dominant frequency of the A note one octave
Pitch
higher (A5 or the 61st key from the left) is 880 Hz.
Every time you sing or play a musical scale, you Perhaps the first demonstration of the relation¬
are varying the subjective experience of pitch. ship between frequency and pitch was performed
Your do, re, and mi differ in this tonal quality. The by Robert Hooke in 1681. Hooke placed a card
most important physical determinant of our percep¬ against a wheel that had teeth notched in it, and
tion of pitch is the frequency of the sound stimu- then spun the wheel. The spinning teeth hit the card
and the resultant vibrations sent out a sound
wave—a sort of rough, buzzing musical note.
When the speed of rotation was increased the fre¬
quency of vibration of the card increased, and so
did the pitch of the note. The relationship between
frequency and pitch had been established. For cen¬
turies thereafter, the terms pitch and frequency
were used interchangeably on the assumption that
pitch rises and falls in exact step with frequency.
The most commonly used measure of the pitch
of a sound is probably the musical scale, which is
basically logarithmic in nature. Any note one oc¬
tave higher than another note of the same name has
FIGURE 7-15 The effect on loudness of exactly twice the frequency of the lower note.
increasing the bandwidth of frequencies in a Thus, the note A3 (the 39th key from the left in
complex tone. (Based on Gulick, 1971) Figure 7-12) has a frequency of 220 Hz, whereas
 Subjective Dimensions of Sound 201

A4 (the 49th key from the left) is one octave higher dB has been assigned a pitch of 1,000 mels (see
and has a frequency of 440 Hz. The musical scale Table 7-1 for a summary of measures of frequency
has undergone very little change over the years, al¬ and pitch). This frequency lies between the notes
though some attempts have been made to adjust the B5 and C6 on the extended piano keyboard in Fig¬
spacing between the notes in an attempt to repre¬ ure 7-12 (the 63d and the 64th keys from the left).
sent more accurately the pitches of different musi¬ When we compare the mel scale with the musical
cal notes. For example, there is a version of the scale, we find several large discrepancies. For in¬
musical scale called the equal temperament scale stance, the one-octave difference between C3 (Key
(W. D. Ward, 1970). Here each octave is divided 28) and C4 (Key 40) is 167 mels, whereas the one-
into 12 standard intervals (representing equal loga¬ octave difference between C6 (Key 64) and C7
rithmic steps) between the musical notes. These in¬ (Key 76) is 508 mels. Such measurements confirm
tervals are called semitones and each semitone can the feelings, often expressed by musicians, that the
be further divided into 100 cents. Thus, an octave higher musical octaves sound “larger” than the
consists of 1,200 cents, and the pitch of any tone lower ones. It is as if there is more “psychological
can be precisely described in terms of what octave distance” between the keys at the high end of the
it is in and how many cents it lies above the lowest piano than at the low end. The relationship between
tone in that octave. This scale has proved to be the musical scale and sound frequency is shown in
quite useful for musicians, although there are still Figure 7-16A, and is roughly linear, whereas the
arguments about the spacing of the standard inter¬ relationship between the perceived pitch (in mels)
vals within an octave. The reason for such argu¬ and sound frequency, shown in Figure 7-16B, is
ments seems to lie in the fact that the musical scale clearly quite different.
is not a direct representation of the psychological Just as a variety of factors other than sound
scale for pitch. intensity affect the perceived loudness of a sound,
The most useful psychological scale for pitch factors other than frequency affect its perceived
to date is the mel scale proposed by Stevens, Volk- 'pitch. The very existence of the mel scale and a
man, and Newman (1937). Like the sone scale of glance at Figure 7-16B demonstrate that pitch is
loudness, the mel scale can be created by various not identical to frequency. This is verified by the
psychophysical scaling techniques. For instance, in fact that when asked to find a sound that is half the
one experiment the researchers created a sort of pitch of a standard sound, a listener does not pro¬
electronic piano with 20 keys and 20 corresponding duce a sound that is half the frequency of the stand¬
knobs set above the keyboard. Turning a knob var¬ ard. Perhaps the major physical factor, other than
ied the tone produced by the corresponding key frequency, that affects the perceived pitch of a pure
through a wide range of frequencies. Listeners tone is its intensity . There is a classic demonstra¬
were asked to sit before the keyboard and to tune tion, first performed over a quarter of a century
the “piano” to produce pitch intervals that ap¬ ago, in which an investigator struck a tuning fork
peared to be equally wide. The results were some¬ tuned to middle C (C4 or 262 Hz) a few feet from
what surprising. Subjects did not tune the piano to the ear of a trained singer. She was asked to sing
equal steps on the frequency scale, nor did they the note she heard, and she reproduced the sound
tune them to equal steps on a logarithmic scale of with reasonable accuracy. Next, the investigator
musical intervals. held the same tuning fork a few inches from her
As in other psychophysical scaling techniques, ear. This increased the intensity of the sound reach¬
a standard must be designated against which all ing her ear, but it left the frequency unchanged
other items will be scaled. By definition, a sound since a tuning fork (when properly struck) produces
with a frequency of 1000 Hz and an intensity of 40 sounds of only a single frequency. Nonetheless, the
202 Chapter 7/Hearing

0)
03
O
CO

CO
o
CO
3

_1_____1___i
100 1000 10000 10 100 1000 10000

Frequency (Hz) Frequency(Hz)

FIGURE 7-16 (A) The relationship between frequency and the musical scale and (B) the relationship
between mels and frequency. (From Lindsay & Norman, 1977)

pitch the singer heard did change. She now sang a pitch. For higher-frequency tones the pitch tends
note that was considerably lower in pitch than mid¬ to rise as intensity increases, whereas for lower-
dle C, clearly demonstrating that sound intensity frequency tones an increase in intensity tends to
affects perceived pitch. lower the pitch.
Using an experimental technique similar to Another factor that affects our perception of
that used in producing equal loudness contours, we pitch is the duration of the stimulus. A pure tone
can produce equal pitch contours. The listener can that lasts for only a few milliseconds is always
be asked either to adjust the intensity of one of two heard as a click, regardless of the frequency. Be¬
tones that differ in frequency until the two tones fore a tone is perceived to have the quality of pitch,
match in pitch (Stevens, 1935) or to adjust the fre¬ one of two conditions must be met. For high-
quency of one of two tones that differ in intensity frequency tones (greater than 1000 Hz), the mini¬
until it matches the other in pitch (Gulick, 1971). mum length of time the stimulus must be sounded
Figure 7-17 shows the results from one listener is around 10 msec. For low-frequency tones (less
measured by Stevens (1935). The graph shows the than 1000 Hz), at least 6-9 cycles of the sound
percentage change in the frequency necessary to wave must reach the ear before it is perceived to
keep the pitch constant as intensity is changed. The have pitch, meaning that most lower-frequency
ordinate was chosen so that lines curving upward tones must last for considerably longer than 10
mean that the pitch is increasing (sounds higher) msec before they have pitch (Gulick, 1971). Even
and lines curving downward mean that the pitch is for tones that exceed the minimum duration for
decreasing (sounds lower). As the figure shows, number of cycles, the tonal quality continues to im¬
varying the intensity of the tone alters its perceived prove as the duration is increased up to about 250
 Subjective Dimensions of Sound 203

FIGURE 7-17 The relationship between apparent pitch and stimulus intensity for one listener. (Based
on Stevens, 1935)

msec. Listeners are better able to discriminate be¬ chord containing two distinct pitches—you do not
tween tones of different frequencies when their du¬ hear a single, unitary sound (as you discovered in
ration is longer. You may recall that the loudness Demonstration Box 6-2).
of a tone also increases as we increase stimulus du¬ When we are considering a complex stimulus
ration up to around 200-250 msec. Perhaps a composed of several sound-wave frequencies, the
quarter of a second represents some sort of funda¬ lowest (usually most intense) frequency sound
mental time period for sensory systems such as the wave is called the fundamental. Musical instru¬
ear. Our phenomenal impressions of the world in ments tend to produce complex sounds in which
many modalities seem to be based on averages or harmonics (frequencies higher than the fundamen¬
sums of energy changes taken over this small win¬ tal) occur at whole number multiples of the funda¬
dow of time. mental frequency. For example, we have already
noted that a sound wave of 220 Hz corresponds to
a musical note that we call A (it would be A3, or
Theories of Pitch Perception
the 37th key from the left in Figure 7-12). When a
The realm of pitch perception has proved to be a musical instrument sounds this note, the complex
testing ground for the major theories of auditory waveform produced will also contain some sound
perception. Several tests have been based on some energy at frequencies of 440 Hz, 880 Hz, 1320 Hz,
interesting characteristics of our subjective impres¬ and so on. These would be called high even har¬
sions of sound. Suppose we created a complex monics (because they represent frequencies of 2, 4,
sound by adding together several different sinusoi¬ and 6 times the fundamental frequency). The char¬
dal sound waves. The auditory system conducts its acteristic sound of a particular instrument depends
own waveform analysis of the complex sound, and on the specific harmonics it produces, a quality
we are able to perceive the pitch of the various called its timbre. Different instruments emphasize
components separately. For instance, if two tones different higher harmonics (sometimes called over¬
are played at the same time, you perceive a musical tones). The number of higher harmonics, or over-
204 Chapter 7/Hearing

tones, and their strength determine the complex An interesting and puzzling problem comes
phenomenal perception that allows us to distinguish from a particular illusion associated with the miss¬
between a note played on a piano and the same ing fundamental. Suppose you are presented with
note played by plucking a violin or guitar string. two complex sounds. One contains the fundamental
Thus, the pitch of a sound is greatly determined by and the higher harmonics and the other contains
the frequency of the fundamental, whereas the only the higher harmonics. The illusion lies in the
timbre is determined by the harmonics. Helm¬ fact that the pitch of both sounds appears to be that
holtz’s (1863/1930) summary of the various subjec¬ of the fundamental frequency, even though the fun¬
tive experiences pertaining to the composition of a damental frequency is not physically present in the
complex tone is shown in Table 7-2. waveform of the second sound (it is a missing fun¬
Because the fundamental frequency is the damental). This unusual phenomenon of hearing
greatest common denominator of all the harmonics plays an important role in a test of the two major
present in a complex sound, we could correctly de¬ theories of pitch. The first is based on the place
termine the fundamental from our knowledge of the principle and the second is based on the frequency
way harmonics work. For example, if we had har¬ principle.
monics of 600, 900, and 1200 Hz, the fundamental More than one hundred years ago, Helmholtz
frequency would be 300 Hz. As just noted, a com¬ became intrigued by the fact that the ear could sep¬
plex waveform normally contains both a fundamen¬ arate a complex sound stimulus into its component
tal frequency and several higher harmonics. For simple frequencies. He observed that since the bas¬
experimental purposes it is possible (through the ilar membrane consists of many fibers stretched
use of special electronic filters) to remove the fun¬ across its triangular shape, we could think of it as
damental frequency without changing the higher a harp with strings. He suggested that the longer
harmonic structure. Alternatively, we could present strings resonate to (that is, vibrate in sympathy
together a set of pure tones that had a particular with) lower frequency tones, whereas the shorter
fundamental without presenting the fundamental strings resonate to tones of higher frequency. Thus,
(for example, 600-, 900-, and 1200-Hz tones with¬ if we sounded a complex tone it would be auto¬
out the 300-Hz fundamental). Such an artificial matically decomposed into its component frequen¬
sound complex, from which the fundamental fre¬ cies on the basilar membrane. Each different tone
quency is missing, is called a stimulus with a miss¬ would cause a different place on the membrane to
ing fundamental. vibrate. This is the place principle.

Table 7-2. Sound Composition and Timbre (based on Helmholtz, 1863/1930)

Subjective
Makeup of complex tone impression

Fundamental alone Soft

Fundamental plus first harmonic Mellow
Fundamental plus several harmonics Broad or full
Fundamental plus high harmonics Sharp
Fundamental intense, harmonics less intense Full
Harmonics intense, fundamental less intense Hollow
Odd harmonics (for example, 1, 3, 5) dominating Nasal
Frequency ratios of 16:15, 9:8, 15:8, 7:5, or 7:6 Rough or screeching
 Subjective Dimensions of Sound 205

This basic idea was later supported and modi¬ wave, we would expect that when the noise is suf¬
fied by Bekesy in a series of precise experiments ficiently intense it would be very effective in mask¬
that ultimately won the Nobel Prize for him (see ing the fundamental tone, since it is stimulating
Bekesy, 1960). His basic procedure was to cut tiny approximately the place on the basilar membrane
holes in the cochleas of guinea pigs and to observe that is said to be vibrating by the place theory.
the basilar membrane with a microscope as the ear Nevertheless, despite the presence of this noise, the
was being stimulated by tones of different frequen¬ pitch of the complex wave is still perceived to be
cies. He discovered that high-frequency tones max¬ that of the fundamental frequency. Since the miss¬
imally displace the narrow end of the basilar ing fundamental phenomenon cannot be explained
membrane near the oval window, and tones of by distortions in the ear, or the local response of
lower frequencies cause displacement further to¬ the basilar membrane, it presents a problem for a
ward the other (wider) end of the basilar mem¬ place theory.
brane. The action of the basilar membrane was not The second major class of theory is based on
quite as simple as Helmholtz’s resonance notion, the. frequency principle. It also has a long history,
however, since low-frequency tones activated the having been championed first by August Seeback
entire membrane (see Chapter 6). This fact, to¬ in the 1840s (Green, 1976) and then carried for¬
gether with the result that tones of intermediate fre¬ ward to the present by Wever (1970) and in a mod¬
quency also displace a fairly broad area of the ified form by Goldstein (1973). This theory argues
membrane, made it unlikely that differential dis¬ that the vibrations of the basilar membrane repro¬
placement or vibration of the basilar membrane is duce at least partially the vibrations of the incom¬
sufficient to fully explain our ability to discriminate ing sounds. The frequency of the sound is
pitch (at least for lower to intermediate frequen¬ transmitted by the pattern of neural excitation re¬
cies). sulting from this vibration. This situation is analo¬
The place theory also has difficulty explaining gous to the microphone end of a telephone
the phenomenon of the missing fundamental. transducing the pattern of vibrations into variations
Helmholtz attempted to deal with this by suggest¬ of electrical signals as it vibrates in unison with
ing that the transduction process in the middle ear your voice. According to this theory, pitch is de¬
distorts the sound waves before they affect the termined by the frequency of impulses traveling up
cochlea. This distortion creates the fundamental the auditory nerve. The greater the frequency, the
frequency, so that the fundamental is present inside higher the pitch. Some studies have shown that for
the cochlea even though it is missing in the stimu¬ tones of up to about 4000 Hz, the electrical re¬
lus that contacts the outer ear. Bekesy (1960) mod¬ sponse of the auditory nerve tracks the frequency
ified this notion somewhat so that the distortion of the tone (see Chapter 6). A tone of 500 Hz pro¬
became part of the response of the basilar mem¬ duces a pattern of response that contains some 500
brane, which was said to respond “as if” the fun¬ bursts of electrical responses per second in the
damental were also physically present. nerve, and a tone of 1000 Hz produces twice as
Unfortunately, several experimental results many responses.
throw doubt on this distortion hypothesis for the Such a theoretical position could explain the
missing fundamental. The basic form of such ex¬ missing fundamental. Since there are many har¬
periments (Patterson, 1969) involves the presenta¬ monics but only one fundamental frequency, mask¬
tion of pairs of tones, such as 2000 Hz and 2400 ing the region of the fundamental should not
Hz, which would produce a missing fundamental appreciably change the overall pattern of sound ex¬
of 400 Hz. If a low-frequency band of noise, cen¬ citation. The low frequency of the fundamental
tered around 400 Hz, is now added to the complex may actually be signaled by the neurons that
206 Chapter 7/Hearing

respond to the higher harmonics, since it is these to be explained in terms of a volley principle
neurons that convey most of the information about (Wever, 1970), which describes cooperation be¬
the pattern of excitation. This may seem somewhat tween neural fibers so that they fire in groups or
topsy-turvy in that we are saying that the funda¬ squads. While one neuron is “reloading’’ (actually
mental is not fundamental, yet consider the exam¬ resting between impulses), its neighbor can dis¬
ple we used earlier in this discussion. Given a charge. The overall effect is that the pattern of
sound wave with harmonics of 600, 900, and 1200 neural firing is in direct correspondence to the fre¬
Hz, the fundamental is inferred to be 300 Hz. In quency of the stimulus, since if we count the total
much the same way that we infer the fundamental number of discharges, or volleys, per unit of time,
from knowledge of the harmonic structure, a higher they correspond to the frequency of the stimulus.
auditory center could infer the fundamental from An example of how this can work is shown in Fig¬
the pattern of excitation reported by neurons that ure 7-18.
respond to higher frequencies (see Goldstein, 1973; A major problem with a volley or frequency
Javel, 1981; Srulovicz & Goldstein, 1983). This principle of pitch perception is that we are requir¬
can be verified to a certain extent by the following ing the frequency of neural firing to encode both
experiment. We again present an individual with a the intensity and the frequency of the sound. Al¬
pair of tones, such as 2000- and 2400-Hz, tones to though at first this seems impossible, one way to
produce a missing fundamental of 400 Hz. If we resolve the problem is to differentiate a concept of
now introduce a high-frequency band of noise, cen¬ overall density of neural activity from one of a
tered at about 2200 Hz and extending for several number of volleys (or bursts of firing) per unit time.
hundred Hertz on either side of it (which should The density of neural firing is the total number of
mask the higher harmonics), the missing funda¬ neurons responding in each volley, or the number
mental is no longer heard (Patterson, 1969). of “shots” being fired in each volley down the au¬
Additional experiments give a similar picture. ditory nerve. An increase in the intensity of the
If we present one component of a complex tone, sound, although not changing the volley frequency,
say 600 Hz, to one ear, and another, say 800 Hz, could increase the number of neurons joining in the
to the other, a missing fundamental corresponding firing, or cause the rate of each individual neuron
to 200 Hz is perceived (Houtsma & Goldstein, to increase somewhat. If all the neurons were con¬
1972). Here there could be no activity on either nected to some higher center that ascertained pitch
basilar membrane corresponding to that created by the frequency of volleys and loudness by the
when a 200 Hz pure tone stimulates it, since each number of responses per volley, the problem would
membrane was stimulated only by a single tone far be solved. An example of how this could work is
from 200 Hz. Missing fundamental pitches seem to also shown in Figure 7-18.
be perceived using a different mechanism than that Since both the place principle and the fre¬
used for the pitch of pure tones. quency principle seem to be supported by some
A number of problems still exist with a fre¬ data, it seems likely, as Wever (1970) has sug¬
quency theory for pitch perception. One difficulty gested, that the ultimate explanation of pitch will
is that an individual neuron cannot fire at high include some aspects of both theories. Wever has
enough rates to account for the perception of high- proposed that in humans pitch is coded by the fre¬
frequency signals. Actually, an individual neuron quency principle for frequencies lower than about
can conduct only about 1,000 impulses per second. 4000 Hz (the theoretical upper limit for volleying).
Thus, the ability of the auditory nerve to track fre¬ The use of this volley mechanism for the percep¬
quencies above this point (up to about 4000 Hz; see tion of lower pitches may be supported by an indi¬
the discussion of phase-locking in Chapter 6) has vidual’s learning and experience (Hall & Peters,
 Subjective Dimensions of Sound 207

Intense
stimulus input

Neuron A

C
D
E

Composite response
in auditory nerve

FIGURE 7-18 The volley principle. Note that the composite neural response follows the frequency of
the stimulus. However, for the weaker stimulus (A) fewer neurons are firing in each volley than for the
stronger stimulus (B).

1982; Terheardt, 1974). For frequencies from 500 Goldstein, 1983). In one of the most intriguing
to 20000 Hz, the place principle seems capable of studies, Simmons, Epley, Lummis, Guttman,
explaining pitch perception. However, below 500 Frishkopf, Harmon, and Zwicker (1965) placed
Hz the vibration pattern on the basilar membrane electrodes in the auditory nerve corresponding to
seems too broad to explain our excellent pitch per¬ different parts of the basilar membrane of a sub¬
ception. Notice that for frequencies between 500 ject’s deaf ear. They found that electrical stimula¬
and 4000 Hz, both principles are operating. This tion at different locations produced perception of
could explain the superior performance of the ear different pitches. Similarly, modem cochlear im¬
for sounds in this range as compared to sounds of plants (see Chapter 6) produce perceptions of pitch
higher or lower frequencies. For frequencies out¬ differences by electrically stimulating different
side this range we must rely on only one mech¬ places in the cochlea (Townshend et al., 1987).
anism, hence performance is poorer. This, of course, supports the place principle. When
The place-frequency compromise is supported Summons et al. (1965) varied the frequency of the
by a good deal of research (see e.g., Srulovicz & electrical stimulus from about 20 to 300 Hz,
208 Chapter 7/Hearing

however, the appropriate changes in pitch percep¬ Equal temperament scale A version of the musical
tion were produced, regardless of the place where scale in which each octave is divided into 12 intervals,
called semitones, that are further divided into 100 cents.
the stimulating electrode was located. This supports
Forward masking Masking of a sound by another
the frequency principle. Clearly both mechanisms
sound presented before it in time. See Masking.
are needed to explain such data.
Frequency principle Asserts that sound frequency is
encoded by the overall frequency of firing in the auditory
nerve.
GLOSSARY Fundamental The lowest, usually most intense, fre¬
quency sound wave in a complex stimulus.
The following definitions are specific to this book. Harmonics Frequencies that are whole-number mul¬
tiples of the fundamental frequency in complex sounds.
Auditory adaptation Transient reduction in auditory
sensitivity due to prior or concurrent exposure to sound Intensity difference A difference in sound intensity
stimuli. at the two ears caused by the presence of a sound
shadow; cue to localization of higher-frequency sounds.
Auditory fatigue Prolonged reduction in auditory
sensitivity following exposure to very intense sounds. Interstimulus interval The time span between the
end of one stimulus and the beginning of the next.
Azimuth The direction of a sound source indicated
as degrees right or left around a horizontal circle, with 0 Loudness The subjective experience of the magni¬
deg straight ahead of and 180 deg directly behind an ob¬ tude or intensity of sound.
server. Masker A sound that, when presented, makes per¬
Backward masking Masking of a sound by another ception of another sound more difficult.
sound presented later in time. See Masking. Masking When a usually audible sound can no
Bandwidth The range of frequencies (frequency longer be heard because of the presentation of another
band) making up a complex sound. sound close to it in time.
Beats Fluctuations in loudness of a complex sound Mel A scale used to measure apparent pitch; the unit
caused by interactions of the simple sound waves com¬ of that scale—a 1000-Hz tone at 40 dB has a pitch of
posing it. 1,000 mels.
Binaural Sound presentation to both ears simulta¬ Minimum audible angle The smallest amount of
neously. movement of a sound source that can be detected.
Central masking Masking resulting from presenta¬ Minimum audible field The threshold intensity of a
tion of target sound and masking sound to different ears. sound stimulus presented and measured in a free field.
Consonance The quality of two tones blending or Minimum audible pressure The threshold stimulus
“going together.” intensity for sound stimuli presented through earphones
Density The quality of compactness or hardness of a and measured at the eardrum.
sound. Missing fundamental When a complex sound has a
Dissonance The quality of two sounds being discor¬ fundamental frequency that is not actually present in the
dant or “clashing.” sound.

Dynamic range For a particular frequency, the dif¬ Monaural Sound presentation to one ear.
ference between the absolute threshold and the pain Perceived duration The length of time a sound ap¬
threshold, measured in decibels. pears to last.
Echolocation system System used by some animals, Perceived location Where in space a sound seems to
such as bats and whales, to locate objects by analyzing come from.
self-emitted sound waves reflected from them. Phase difference The difference in the phase of a
Equal loudness contour A curve describing the in¬ sound wave between the two ears caused by the different
tensities at which tones of varying frequencies appear to distances the sound wave has to travel to reach each ear;
be equally loud. cue to localization of lower-frequency sounds.
Equal pitch contour A curve describing the fre¬ Pitch The psychological attribute of sound most
quencies at which tones of varying intensities appear to closely associated with sound frequency, described by
have the same pitch. the words high or low.
 Glossary 209

Place principle Asserts that sound frequency is en¬ Sound shadow An area in which only sounds dif¬
coded by what place on the basilar membrane vibrates fracted by the edge of the head are received by the ear,
most to each frequency. resulting in lower intensity especially for sounds of
Precedence effect The first of a group of sounds (for higher frequencies.
example, a sound and its echoes) to arrive at the ear is Target The sound to be detected in a masking situ¬
the major determinant of where in space the sound source ation.
is perceived to be. Timbre A sound attribute associated with the har¬
Profile analysis Detection of an intensity increment monics or overtones of a complex sound.
(“bump”) or other difference between complex sounds Time difference The difference in the time taken by
that differ in their profiles (amplitudes at the various fre¬ a sound wave to travel to the two ears when starting from
quencies making them up). an azimuth other than 0 or 180 deg; a cue to the direction
Reverberation A cue in determining the distance of of a low-frequency sound source.
a sound source; reverberant sound is sound that reaches Volley principle The theory that neural fibers fire in
the ears after having bounced off some surface. groups, one group of neurons firing while another group
Simultaneous masking When a sound is masked by “recharges.”
another sound presented at the same time. Volume The sound quality associated with the de¬
Sone A scale used to measure the apparent loudness gree to which a sound fills space and seems large or
of a sound; the unit of that scale—a tone of 1000 Hz at small.
40 dB has a loudness of 1 sone.
CHAPTER
8
Taste and
Smell
THE GUSTATORY (TASTE) SENSE
Taste Stimuli
Taste Receptors
Neural Responses in Taste
Taste Thresholds
Taste Adaptation
Taste Intensity and Qualities
THE OLFACTORY (SMELL) SENSE
Smell Stimuli and Receptors
Neural Responses in Smell
Smell Thresholds
Smell Adaptation
Smell Intensity and Qualities
Sensitivity to Pheromones
212 Chapter 8/Taste and Smell

T he jungle man stopped and looked at the fresh

gash in the tree. He touched his tongue to it.
THE GUSTATORY (TASTE) SENSE

The lingering taste of metal verified that humans Presumably, life began in the giant bowl of chem¬
were near. He turned and sniffed the wind. There ical soup that we call the sea. Various substances
it was—the faint scent of two men. He could tell suspended or dissolved in water were important to
by the scent that they were city people. He could the survival of primitive living things. Some sub¬
also tell by the message in the wind that she was stances provided food, some gave warning, some
with them. caused destruction. The most primitive, one-celled
Do you know anyone who tastes or smells this organisms clearly could not use anything like visual
keenly? Probably not. Such accounts are common or auditory sensory systems, which require large
only in fiction. We humans are quite inferior to many numbers of specialized cells. They relied on chem¬
other animals in our abilities to taste and smell; we ical or mechanical interactions with their environ¬
rely more heavily on sight and hearing. Nonetheless, ment mediated by the cell’s outer membrane. This
chemical sensitivities are important in many ways. was the first primitive “sensory system” used by
The flavor of your favorite dessert, the acrid taste of a living things. As life evolved, multicelled animals
broken aspirin on your tongue, the smell of fresh could afford a “division of labor” among the many
coffee, cut grass, or roses, all evoke intense feel¬ cells composing their bodies. Specialized cells
ings and guide our behaviors in many ways. were grouped together to pick up chemical
More important than aesthetics, chemical senses information from the surroundings. For example,
serve a survival function. A reasonable rule of fish have pits lined with cells responsive to a vari¬
thumb, at least for natural substances, is that things ety of chemical and mechanical stimuli. Insects and
that taste bad are likely to be harmful, indigestible, other invertebrates have such cells located on their
or poisonous, whereas things that taste good are apt antennae.
to be digestible and contain substances the body Although two anatomically separate systems
can metabolize. When you eat a hot pepper, you developed, in the sea there was little differentiation
usually experience a “burning” sensation in addi¬ between taste and smell. All important chemical
tion to (or masking completely) the flavor of the stimuli were dissolved or suspended in the same
pepper. Some cultures, such as Mexican, Indian, substance, water. When life moved on to land, the
or Szechwan, use this effect as a major flavor in¬ two existing chemical receptor systems came to
gredient, and most people, whatever their native serve different functions. The taste system became
culture, appreciate “hot” foods. In this, humans a “close-up” sense, which provided the last check
are unique in the animal kingdom in ignoring the on the acceptability of food. Smell turned out to be
warnings (the burning sensation) of the common useful as a distance sense, although it also retained
chemical sense and continuing to ingest, even find¬ an important function in dealing with food (see Ro¬
ing enjoyable, foods that burn (Rozin, Ebert & zin, 1982; and later in this chapter).
Schull, 1982).
Smell has similar survival functions, since
Taste Stimuli
foul odors often signify danger in the sense of
putrefied or spoiled substances that are no longer The physical stimuli for the taste system are sub¬
safely edible. Smell also helps the species survive stances that can be dissolved in water, although the
by conveying sexual and social information. Al¬ extent to which they can be dissolved in lipids
though often we are not consciously aware that we (fats) may be more important to how they taste
are responding to smells, even in humans some (Gardner, 1979). As is usual for physical stimuli,
odors serve to identify individuals and to convey the amount of a chemical substance present is re¬
information between members of our species. lated to the intensity of the taste we experience.
 The Gustatory (Taste) Sense 213

However, which property (or collection of proper¬ table salt is composed of two atoms, one sodium
ties) gives rise to the various different taste quali¬ (Na) and one chlorine (Cl). When dissolved in wa¬
ties is still unknown in detail. There are several ter, the atoms break apart. The sodium atom is now
possibilities, such as the size of the individual mol¬ a positively charged ion, and the chlorine atom is a
ecules of the substance, how the molecule breaks negatively charged ion. The ratio of the weights of
apart when it is dissolved in water, or how the mol¬ these positively and negatively charged ions may
ecule interacts with cell membranes. Unfortunately, be related to how salty a substance tastes. When
just which aspects of a stimulating substance cause the ratio is relatively low, substances taste salty. A
its characteristic taste will be known only when we high ratio is associated with substances that taste
agree on the basic dimensions of taste (cf. Erick¬ bitter (see Wybum, Pickford & Hurst, 1964). In
son, 1985). very low concentrations, salts tend to taste sweet.
There is general agreement that there are at Sour substances also break into two parts
least four primary taste qualities: sweet, salty, sour, when in solution, but they are usually acids (such
and bitter. These taste qualities are associated with as hydrochloric, sulfuric, acetic, and nitric) rather
some general types of molecules. A sweet taste is than salts. In all these substances, hydrogen is the
generally associated with so-called organic mole¬ positively charged ion. The behavior of the hydro¬
cules, which are made up mostly of carbon, hydro¬ gen seems to be directly related to the sourness of
gen, and oxygen in different combinations. These such acids, but other properties must also be im¬
organic molecules are commonly called sugars, al¬ portant because most acids taste sweet or bitter in¬
cohols, and so forth. Other sweet substances, like stead of sour. Thus, although we can relate some
saccharin, are also organic chemicals, but they are aspects of chemical stimuli to the tastes they pro¬
quite different from “natural” sweeteners, such as duce, this relationship is not simple. Any given
sugars, in their molecular structure. Many sweet¬ taste is probably the result of the interaction of sev¬
tasting substances have a particular structure in eral different properties of the physical stimulus
common, termed the AB,H system, that consists with the properties of the receptor cells (see Fau-
of two negatively charged atoms (represented by rion, Saito & MacLeod, 1980).
the letters A and B) and a positively charged hydro¬
gen atom (H) arranged in a special way. It is
Taste Receptors
thought that this molecular structure selectively in¬
teracts with special parts of some taste receptors to The tongue is covered with little bumps called pa¬
cause them to respond (see Bartoshuk, 1979). pillae. The major receptors for taste are groups of
Bitter taste is closely related to sweet taste. cells called taste buds that are found in three types
Many substances that taste sweet in small amounts of papillae (see Figure 8-1). Fungiform papillae are
taste bitter in large amounts (e.g., saccharin). Also, shaped like little mushrooms (fungi) and are found
a number of chemicals containing nitrogen (such as at the tip and the sides of the tongue. Foliate pa¬
strychnine, caffeine, quinine, and nicotine) taste pillae make up a series of folds (folia) along the
bitter. The relationship between sweet and bitter is sides of the rear portion of the tongue. Cirumval-
further strengthened by the fact that bitter-tasting late papillae are shaped like a flattened hill with a
molecules also often contain an AB,H system. In circular trench or valley surrounding it and are lo¬
bitter substances, however, the components have cated at the back of the tongue. The filiform papil¬
a different spatial arrangement than they do in lae are shaped like rough, tapered arrowheads or

sugars. blades of grass. Filiform papillae contain no taste

A salty taste is elicited by molecules that, buds, but serve the purpose of abrading the food
when dissolved in water, break into two electrically into smaller bits that will dissolve more easily.
charged parts called ions. For example, common There are also some taste receptors scattered over
214 Chapter 8/Taste and Smell

10,000 taste buds in your mouth when you are

young, although their number decreases with age.
Within each taste bud, the individual cells are
continually developing. Each cell has a life span of
only a few days, so that the composition of the
taste bud is always changing, with some immature
cells (around the outside), some mature cells (near
the inside), and some dying cells always present
(Beidler & Smallman, 1965). No one knows why
these cells have such a short life span; perhaps it is
because they are somehow damaged when they re¬
spond to substances in the mouth. Each taste cell
in the taste bud is leaflike and resembles a skin cell
FIGURE 8-1 A diagram of the human tongue (from which it probably evolved). A slender pro¬
showing the locations of the different types of jection from the top end of each cell lies near an
papillae.
opening onto the surface of the tongue called a
taste pore. It is thought that the actual reception
mechanism for taste is located in these slender pro¬
parts of the mouth other than the tongue, such as cesses (called microvillae).
on the soft palate (which is the back portion of the A number of theories have been proposed for
roof of your mouth). Figure 8-2A shows how the how the receptors interact with the stimulating mol¬
taste buds are distributed within a circumvallate pa¬ ecules to generate the electrical current that ulti¬
pilla. Each taste bud consists of several receptor mately results in action potentials in the taste
cells (perhaps up to 30) arranged like the closed nerves (see Teeter & Brand, 1987, for a review).
petals of a flower (Figure 8-2B). There are about There are probably several transduction mecha-

Enlarged view

Gustatory
cell

FIGURE 8-2 (A) A typical papilla with taste buds (one is circled); (B) enlarged picture of a single taste
bud. (From Wyburn et at., 19641
 The Gustatory (Taste) Sense 215

nisms, perhaps different ones for different classes somatosensory cortex (near where information from
of taste stimuli. In one proposed mechanism, salt touch for the face is projected) and the third is the
ions, such as sodium or lithium, directly penetrate anterior-insular cortex, which is a part of the
the cell membrane of the microvillae, causing elec¬ frontal cortex under the front end of the temporal
trical current to flow across the membrane. Another cortex.
theory suggests that reversible chemical bonds form As in vision and audition, most of our knowl¬
between specific parts of the taste stimulus mole¬ edge of the electrical activity of the taste system
cule (e.g., the AB,H system) and receptor mole¬ has come from studies of nonhuman animals. A va¬
cules in the cell membrane. These bonds, either riety of studies has shown that taste fibers respond
directly or through causing chemical reactions in¬ to increasing intensity (concentration) of the stim¬
side the receptor cell, change the flow of ions ulus by increasing their overall rate of firing. One
across the cell membrane, thus generating an elec¬ of the few studies that actually used human subjects
trical current. Finally, it is possible that some taste for direct recording (Diamant, Funakoshi, Strom &
stimulus molecules alter the electrical properties of Zotterman, 1963) capitalized on the fact that taste
the cell membrane itself, causing an electrical cur¬ pathways from the front of the tongue must be cut
rent. In all of these mechanisms, the final result is during a certain kind of ear operation. Electrical
the release of a neurotransmitter across the synapse activity in response to taste stimuli was recorded
of the receptor cell with the taste nerve cell, caus¬ from this nerve during the operation. The data from
ing spike potentials to travel up the taste nerves these patients show that the amount of neural re¬
(see the Appendix for more about spike potentials). sponse grows as the logarithm of the intensity of
the stimulus (in this case, table salt), much like the
responses in vision and audition to increasing stim¬
Neural Responses in Taste
ulus intensity (see Chapters 3 and 6). As in most
Three large nerves (the chorda tympani, vagus, and modalities, the neural code for intensity seems to
glossopharyngeal) carry fibers from the taste buds. be the overall amount of firing of all the sensory
They run from the tongue to the several nuclei in fibers.
the solitary tract, which is located in the medulla How is taste quality encoded? At first it was
(the place where the spinal chord widens to form thought that there would prove to be “sweet recep¬
the brainstem). In addition, there is information tors” and “salt receptors,” and so forth, one for
carried from the common chemical sense. In hu¬ each taste quality, perhaps even corresponding to
mans this consists mostly of the trigeminal nerve the different categories of papillae. However, most
of the head and its various nerve endings in the receptor cells seem to respond to all of the four
mouth and nasal cavity. The common chemical basic kinds of taste stimuli although at different
sense responds to a wide variety of different stim¬ rates (Arvidson & Friberg, 1980; Kimura & Bei-
uli, including and of particular interest to humans dler, 1961). This seems appropriate given the dif¬
several popular spices such as peppers and ginger ferent transducer mechanisms described above. The
(see Silver, 1987). same sort of responsiveness to most stimulus types
From the nuclei in the solitary tract, taste in¬ has been found in the solitary nucleus mentioned
formation is carried via a set of pathways called the above and in the thalamus as well (Doetsch, Gan-
medial lemniscus to the taste center of the thala¬ chrow, Nelson & Erickson, 1969; Scott & Erick¬
mus, which is situated at the top, rear, central por¬ son, 1971).
tion (ventral posterior nuclei) of the thalamus. The Actually, there is no theoretical need for spe¬
thalamic taste area projects to three areas in the cialized taste receptors as long as the various neural
brain. Two are regions at the base of the primary units have different stimulus-specific response
216 Chapter 8/Taste and Smell

rates. If this condition is met, then the code for qualities, different gustatory nerve fibers do seem
taste quality could be an across-fiber pattern of to be “tuned” to certain taste stimuli, much as au¬
neural activity (Erickson, 1985; Erickson & Schiff- ditory nerve fibers are tuned to certain frequencies
man, 1975; Pfaffman, 1955). Figure 8-3 shows of sound (see Chapter 6). Such fibers respond most
how this might work. Notice that although all of vigorously to their “best” substances and less vig¬
the fibers respond to all taste inputs to some extent, orously to others. Eventually, it may be possible to
the pattern of firing across the four diagramed fi¬ classify such fibers into a few classes, correspond¬
bers is different for each quality. Thus for a sugar ing to the basic taste qualities (Frank, 1975, 1985;
stimulus (S), we find fiber A responding vigor¬ Pfaffman, Frank & Norgren, 1979). The response
ously, B moderately, and C and D only weakly. patterns shown in Figure 8-3 could thus also be
For salt (NaCl), A and D respond weakly, whereas interpreted as those characteristic of such tuned fi¬
B responds strongly and C nearly as vigorously. ber types. On the basis of such data, Pfaffman
Erickson (1963) was able to show such distinct (1974) proposed the labeled-line theory of taste
across-fiber pattern differences. These patterns be¬ quality encoding. The basic idea is that each taste
come somewhat less distinct in the thalamus fiber encodes the intensity of a single basic taste
(Doetsch et al., 1969; Scott & Erickson, 1971). quality, that associated with its best stimulus. To
Although there do not seem to be taste recep¬ the extent that a stimulus activates the “sweet” fi¬
tors that respond only to one of the four basic taste bers, it tastes sweet; to the extent it activates the

CD
Fiber (type) A:
CO
c
Sucrose best
o “sweet”
Ql
CO
CD

>
03
DCD /p
11 '

s NaCl HCI Q
Stimulus

mm
Fiber (type) C: * 11 -i Fiber (type) D:
<D
<0 HCI best til Quinine best
c
o “sour” “bitter”
CL
CO
■
CD r-

ZD
CD
^4ft

NaCl HCI Q NaCl HCI Q

Stimulus

FIGURE 8-3 Using the across-fiber pattern theory, consider each graph to represent the response of a
unique taste fiber to the various stimuli. Using the labeled-line theory, consider each graph to represent
the average response of a group of more-or-less equivalent taste fibers.
 The Gustatory (Taste) Sense 217

“bitter” fibers, it tastes bitter. This means that if the molecular weight of the substance (in grams)
“simple” stimuli such as NaCl could have a com¬ is added to enough water to make one liter of so¬
plex taste if they activated several types of fibers, lution. Different solutions with the same molar
and this seems to happen. For example, salts tend concentrations have the same number of stimulus
to taste both salty and sour (Bartoshuk, 1978). This molecules in a given volume of liquid.
theory is compatible with the across-fiber pattern Once we have decided how to specify the
approach, except that here the code for taste quality stimulus concentrations, we can begin to measure
is a profile across a few fiber types rather than a thresholds. We must be careful to note, however,
pattern across many thousands of unique fibers (see that the region of the mouth or tongue stimulated
also Scott, 1987; Smith, 1985). also affects our sensitivity. For many years it was
Although it is unknown whether labeled-lines thought that the back of the tongue was especially
exist along the entire taste pathway, cortical neu¬ sensitive to bitter taste, the front more sensitive to
rons most responsive to the four basic tastes do sweet and salty stimuli, and the sides of the rear
seem to be localized in different parts of the taste most sensitive to sour stimuli. Work by Codings
cortex (Yamamoto, Yayama & Kawamura, 1981). (1974) altered this picture. She used as a stimulus
Furthermore, it is likely that some recoding of the a tiny piece of filter paper that had been soaked in
taste information takes place in the cortex. There a stimulus solution. This gave precise control over
may be specific cortical cells that give an “on” the location of the stimulus on the tongue. Figure
response to some taste stimuli and an “off” re¬ 8-4 shows a summary of her results for four differ-
sponse to others, similar to the feature-specific cells
in the visual cortex discussed in Chapter 3 (Funa-
koshi, Kasahara, Yamamoto & Kawamura, 1972).

Taste Thresholds

What are the limits of a human’s sensitivity to

taste? It is difficult to study the thresholds for taste
stimuli, since there are so many different stimuli to
consider. Also, we find that thresholds vary with
the viscosity of the mixture to be tasted (Paulus &
Haas, 1980) or with its temperature (Paulus &
Reisch, 1980). The study is further complicated by
the fact that the various parts of the tongue and
mouth are not equally sensitive to different stimuli.
Let us begin by seeing how we can measure
physically the amount of a taste stimulus present at
any moment. Although there is some debate on the
appropriate measurement procedures for taste stim¬ O-oS <=> "D -2
rt ch_ <2 <= M-
uli, one useful measure is the molar concentration LL a) O Li- CC O CD CO O CQ cc O

of a substance (Pfaffman, Bartoshuk & McBurney, Location of stimulation

1971). Molar concentration is based on the weight

FIGURE 8-4 Average absolute thresholds for four
of a substance dissolved in a given amount of a different taste stimuli at four locations on the
solvent (usually water when we deal with taste). A tongue and at a location on the soft palate. (Based
solution is said to have a concentration of 1 mole on Collings, 1974)
218 Chapter 8/Taste and Smell

ent parts of the tongue and the soft palate. For the Vanillin

bitter substance the lowest threshold on the tongue iw

T3
is at the front. For this substance, however, an O
_C
even lower threshold occurs on the palate. The tip C/D
CD Tablesalt
and back of the tongue are most sensitive to sweet,
whereas the front and sides are most sensitive to
salt. There is also differential sensitivity of parts of
the mouth to “hot” tastes, such as those associated
with chili peppers. Demonstration Box 8-1 helps
you explore these differences.
Individuals often differ in marked ways in
their sensitivity to certain tastes. This was shown
in a study by Blakeslee and Salmon (1935), who
tested the taste thresholds for 47 people using 17
different substances. Figure 8-5 summarizes some
of their results. Most substances, such as table salt 0)
-Q Hydrochloric acid
E
or saccharin, have a narrow range of thresholds for =5
_
different people. For others, however, such as va¬ Phenylthiocarbamide
nillin or PTC (phenylthiocarbamide), large individ¬ mm _ _*■
-———s—__m__
ual differences in sensitivity exist. These latter two Low High
threshold threshold
substances are interesting because some people are
Threshold values for each modality
apparently “taste-blind” to them. That is, at ordi¬
nary concentrations some people cannot taste these FIGURE 8-5 Frequency distributions of absolute
substances at all. PTC produces a bitter taste for thresholds of 47 observers to various taste stimuli.
those who are sensitive to it, but many can’t taste Phenylthiocarbamide (PTC) and vanillin have
particularly wide ranges of thresholds, and PTC
it even at high concentrations. If the stimulus is
clearly shows two modes—tasters with low
intense enough, however, even the taste-blind can thresholds and non tasters with high thresholds.
taste PTC, as Figure 8-5 shows. One similarity be¬ (Based on Blakeslee & Salmon, 1935)
tween taste blindness for PTC and color blindness
(discussed in Chapter 5) is that both appear to have
a genetic component and tend to run in families.

DEMONSTRATION BOX 8-1. Variations in Taste Sensitivity

Although you have all probably experienced spicy hot several drops of Tabasco sauce or Louisiana style hot
foods on numerous occasions, you may not know that sauce (both of which contain hot red peppers) on the
sensitivity to hot spices varies over the mouth. For tip of a cotton swab or a bit of paper napkin twirled
example, the tip of the tongue is most sensitive to red around a pencil or toothpick. Touch this “taste stim¬
and black peppers, and the anterior (hard) palate and ulator” to different parts of your mouth and tongue,
cheek are least sensitive (Lawless & Stevens, 1988). and notice how the sensations differ in strength for
You could demonstrate this for yourself by placing different locations.
 The Gustatory (Taste) Sense 219

You might be surprised to learn that another

substance associated with taste blindness is caf¬
feine, the stimulant found in coffee (Hall, Barto-
shuk, Cain & Stevens, 1975). For caffeine, the
taster and nontaster groups are not as distinct as
they are for PTC, but the range of thresholds is still
quite large and two groups clearly appear. Further¬
more, tasters of PTC tend to be tasters of caffeine
and vice versa, indicating that a common mecha¬
nism may be responsible for this threshold varia¬
tion. This is not simply a general “bitter” taste
mechanism, for in the same study Hall et al. (1975)
0 10 20 30 10 20 30
found that thresholds to two other bitter-tasting Time (sec)
substances were unrelated to thresholds for PTC
and caffeine. Similar results have been found for FIGURE 8-6 Adaptation and recovery from
adaptation to continued stimulation of the tongue
the bitter and sweet tastes of saccharin (Bartoshuk,
with table salt (NaCl). The three curves represent
1979). In general, the attempt to explain such cases three different adapting concentrations. The time
of taste blindness by attributing them to deficien¬ axis represents the amount of exposure to the
cies in some particular taste system (as is often adapting concentration, or recovery time, before
possible for color blindness) has not succeeded determination of absolute threshold. Note the
resemblance to dark adaptation curves (see
very well.
Chapter 4). (Based on Hahn, 1934)

Taste Adaptation

Taste thresholds can also be affected by stimuli that resalting your food if you eat something that is not
have reached the tongue prior to the threshold test. salty between bites of salty food (see Halpem &
The taste system adapts very readily to continued Meiselman, 1980). As Figure 8-6 shows, recovery
stimulation of the same type, and this adaptation from adaptation is virtually complete in about 10
temporarily raises the absolute threshold for the sec, no matter how much salt was eaten previously.
particular substance to which it has been adapted. Adaptation to one substance can also have an
Figure 8-6 shows an example of the effects of pre¬ effect on the threshold for (and the subsequent taste
vious stimulation on absolute threshold. Here both of) different substances. This is called cross¬
the adapting stimulus and the test stimulus were ta¬ adaptation. For example, adaptation to one salt
ble salt (NaCl). As you can see from the figure, the will raise the threshold to other salts. Similarly, ex¬
absolute threshold varied both with how long the posure to a sour substance will raise the threshold
tongue had been exposed to the adapting stimulus to other sour stimuli. In some cases, exposure of
and the strength of the stimulus (Hahn, 1934; the tongue to one stimulus may actually lower the
O’Mahony, 1979). Similar results can be demon¬ threshold to another taste stimulus (or make its
strated for the effects of adaptation on sensory in¬ taste more intense). This special case of cross¬
tensity (Gent, 1979). This sort of adaptation adaptation is called potentiation. Thus, adaptation
explains why at dinner some people resalt their to an acid, while reducing the sourness of another
food again and again. As they eat, they adapt to acid may increase the sweetness of a sugar. Adap¬
the salty taste and come to need more salt to ex¬ tation to urea (the bitter substance contained in hu¬
perience the taste at the same level. You can avoid man urine) will increase the intensity of salty
220 Chapter 8/Taste and Smell

sensations (McBumey, 1969). Perhaps the most increase the stimulus intensity, a number of inves¬
striking phenomenon associated with cross¬ tigators have used the magnitude estimation tech¬
adaptation is the variability of the taste of ordinary nique discussed in Chapter 2. In general, they have
tap water with adaptation to various substances. found that the sensation increase is described by a
In fact, all four of the basic tastes can be induced typical power function of the form S = aln, where
in water by previous adaptation to a suitable taste S is the sensation, / is the intensity of the stimulus,
stimulus (McBumey & Bartoshuk, 1972). Interest¬ a is a constant, and n indicates the rate at which
ingly, “off” responses to pure water have been the sensation increases. The exponent n is usually
recorded in the human chorda tympani nerve (Oak¬ approximately 1 (i.e., for table salt it is 0.91, for
ley, 1985); such neural responses may partly me¬ quinine hydrochloride it is 0.85, for hydrochloric
diate our experience of the taste of water. You can acid it is 0.99, and for sucrose it is 0.93) (Meisel-
experience these tastes for yourself by trying Dem¬ man, Bose & Nykvist, 1972; Norwich, 1984). An
onstration Box 8-2. exponent of 1 means that sensation grows linearly
with stimulus intensity, that is, changes in stimulus
intensity of the same size cause changes in sensa¬
Taste Intensity and Qualities
tion intensity of the same size no matter how in¬
Our ability to discriminate intensity differences in tense the stimulus. Contrast this with vision and
taste, regardless of the stimulus tested, is really hearing, where the corresponding exponents are
quite poor. The Weber fraction (which measures less than 1 and changes in weaker stimuli are per¬
the fractional amount by which two stimuli must ceived as greater than changes in more intense
differ for them to be discriminated as being differ¬ stimuli. An important caution here is that we must
ent, as we discussed in Chapter 2) ranges from a be careful to specify exactly the conditions under
relatively poor 1/10 to an awful 1/1, making taste which such exponents are measured, for they are
the least sensitive of the senses by this criterion greatly affected by adaptation. For example,
(Pfaffman et al., 1971). O’Mahony and Heintz (1981) found an exponent
To see how the sensation of taste grows as we for NaCl of only about 0.7 under conditions some-

DEMONSTRATION BOX 8-2. The Taste of Water

Water has a distinctive taste, especially when you your tongue and palate while eating it. Water tasted
have been eating or drinking some other substance after eating the artichoke usually tastes sweet. An¬
before you taste the water. In this demonstration, you other way to obtain the sweet taste is to swish a
will be able to make water taste sweet, bitter, sour, mouthful of strong (caffeinated) coffee on the tongue
or salty. Although you may not be able to (or want for 30 sec before tasting tap water. To make water
to) try all of the demonstrations, be assured that all taste bitter or sour, take some very salty water and
of them work under the carefully controlled condi¬ swish it around in your mouth for 30 sec and then
tions of the laboratory (Bartoshuk, 1974). spit it out. Afterward, taste some tap water. Some¬
The most pleasant of the demonstrations requires thing that has been tried in the laboratory but that you
that you eat a few cooked artichokes of any variety may not want to try is to swish some urea (a major
(canned, fresh, or frozen) and then taste a sip of wa¬ component of urine) on the tongue for 30 sec. Tap
ter. Be sure to thoroughly mash the artichoke onto water tasted after this treatment tastes salty.
 The Gustatory (Taste) Sense 221

what different from those used by other investiga¬ in much the same way we specified colors in Chap¬
tors (see also Meiselman et al., 1972). ter 5. He proposed that this space would be in the
Several studies have compared the changes in form of a pyramid, with the primary tastes at the
the rate of neural response with changes in stimulus comers. Intermediate tastes would be represented
intensity to the psychophysical data. They have by points on the surface of (or within) this space,
shown that the two types of response vary with which is shown in Figure 8-7A. Schiffman and Er¬
stimulus intensity in a similar fashion (Borg, Dia- ickson (1971) used a sophisticated mathematical
mant, Oakley, Strom & Zotterman, 1967; Diamant technique to test whether this notion actually de¬
& Zotterman, 1969). It seems likely that our sen¬ scribes the way people perceive taste stimuli. The
sation of the intensity of a taste is directly related technique is based on the judged similarity between
to the overall amount of neural activity evoked by stimuli. Using a procedure called multidimensional
the stimulus, which in turn depends on the intensity scaling, observers’ similarity judgments can be
(molar concentration) of the stimulus (see Norwich, converted into a set of distances between points in
1984, for a theoretical discussion of this relation¬ a geometric space with the smallest possible num¬
ship). However, the situation is rather complex ber of dimensions. Figure 8-7B shows their results.
since our sensation of the intensity of any one taste Notice that Henning’s taste pyramid fits reasonably
stimulus may be affected by the presence of other well into this space. This has been verified in an¬
taste stimuli. Demonstration Box 8-3 provides one other study using a larger set of stimuli (Schiffman
way you can explore this complex interaction be¬ & Dackis, 1975).
tween tastes. Such techniques have also been used to ex¬
Modem psychophysical scaling techniques have plore the relationships between the tastes of more
also been used to look at the relationships among similar substances, such as sodium salts (Schiff¬
taste qualities. In 1916, Henning proposed that we man, McElroy & Erickson, 1980) and sweeteners
could specify the qualities of all tastes using a spe¬ (Schiffman, Reilly & Clark, 1979), and to try to
cially shaped three-dimensional geometrical space, discover the physical and psychological properties

DEMONSTRATION BOX 8-3. Putting Out the Fire

You may have had the experience of putting too much quinine, or possibly beer) don’t seem to help any
pepper or other hot spice in your mouth and finding more than simply waiting for the burning to cease,
the burning too much to bear. Or you may have been and salty substances are intermediate in effect. The
prevented from eating some tasty dish because it was cool temperature explains some but not all of the
“too hot.” These are circumstances in which you cooling effect, since different tastants at the same
want to reduce the intensity of sensations arising from temperature have different effects. If you are daring,
stimulation of your common chemical sense. Surpris¬ you might try swishing some diluted Tabasco sauce
ingly, the effectiveness of swishing various liquids in around in your mouth until it begins to burn and then
your mouth to put out the fire depends to some extent experimenting with different quenching substances to
on how they taste (Stevens & Lawless, 1986). The see whether you can confirm these results. Remember
most effective liquid has a sweet (e.g., soda pop) or to leave plenty of time between trials for the burning
sour (e.g., lemon or other citrus juice) taste, and is at sensation to fade completely.
a cool temperature. Bitter-tasting substances (e.g.,
222 Chapter 8/Taste and Smell

"Sour" ‘Salty-bitter”
‘Salty”
HAc
?—
HCI
-V- hno3
Sour

\A 0-— NaOH “Alkaline’

'Sweet” /V NCI2CO3
"Bitter”
QHCI
Q2SO4

Sweet Bitter

FIGURE 8-7 (A) Henning’s taste pyramid. (B) Multidimensional scaling results for similarity judgments
of taste stimuli. Henning’s taste pyramid is superimposed on the three-dimensional space to show the
correspondence. (From Schiffman and Erickson, 1971)

that underlie the similarity judgments. In general, out who had been there recently, or exploring a
multidimensional scaling does provide results that new room by sniffing the furniture, we tend to
are consistent with the existence of the four pri¬ think that smell is unimportant; it is often referred
maries—sweet, salty, sour, and bitter, as shown in to as a “minor” sense. Although much of our per¬
Figure 8-7—although there is some controversy ceptual processing of odors is unconscious, our
about other relevant facts (see Erickson, 1985; Er¬ sense of smell plays a role that is far from minor.
ickson & Covey, 1980; McBurney & Gent, 1979). For example, it is very difficult to recall smells, or
It seems likely that investigators will continue to to name them, but the experience of a particular
refer to four basic taste qualities because they pro¬ smell at a particular moment can stimulate a flood
vide a useful summary of most of our research of memories of episodes in which that smell was
findings in taste. However, their interpretation may present (Engen, 1982, 1987). These memories of
change. For example, it has been suggested that our past are often rich in emotional tones. Thus,
each of the basic tastes should be considered to be the scent of cinnamon might evoke feelings of joy
a separate sensory modality, much as the skin sen¬ associated with your mother baking apple pies, or
ses can be separated into touch, warmth, cold, and the scent of ether might evoke the memory and
pain (McBurney & Gent, 1979). fears associated with a childhood visit to the hos¬
pital.
In many common situations, smell works in
THE OLFACTORY (SMELL) SENSE conjunction with taste. When we have a bad head
cold food seems flavorless, yet our nasal passages
All of us are familiar with the sight of a dog or cat are most affected by the cold, not our mouths
sniffing at some object, clearly processing infor¬ where the taste receptors are located. A major part
mation provided by their sense of smell. Since we of the experienced flavor of food and drink is
never see a human being sniffing the ground to find caused by the odor of the substance (Brillat-
 The Olfactory (Smell) Sense 223

Savarin, 1825/1971; Hyman et al., 1979; Murphy & Actually, smell acts as if it has two separate
Cain, 1980). When our nasal passages are clogged modes of action that may result in different percep¬
with mucus our olfactory (smell) receptors cannot tual experiences, and different forms of information
function properly. This affects our ability both to extraction (e.g., Rozin, 1978). One mode, which
smell and to experience flavor, since much of the we have already mentioned, is associated with the
richness and subtlety of our experiences with food experience of the flavors of food. The second is as
and drink come from their odors. When we cannot a distance sense, responding to molecules that float
smell, our ability to identify foods by taste alone is about in the air carrying information about the ob¬
significantly inferior, unfortunately for some of the jects or organisms from which they have become
most preferred foods, as Figure 8-8 shows. And detached. For instance, we leave molecules of
this happens often. In an informal survey con¬ ourselves on the ground and in the air near the
ducted by National Geographic magazine, it was ground whenever we take a walk. Insects and some
found that over two-thirds of the analyzed returns higher animals secrete volatile chemicals (phero¬
reported occasional loss of the sense of smell, mones) whose molecules waft through the air to
many of them quite often (e.g., every time a head other members of the species. The specific mole¬
cold occurred) (Gilbert & Wysocki, 1987). Many cules secreted can carry messages about fear and
people over 70 years of age also experience major sexual availability, among other things. Species
loss of the smell sense (Gilbert & Wysocki, 1987; that possess sensory systems that respond to these
Rabin & Cain, 1986) with similiar consequences low concentrations of molecules in the air can take
for the experience of flavor (see Chapter 16). Dem¬ advantage of this information, hence improving
onstration Box 8-4 allows you to experience this their ability to survive (see Gibbons, 1986, for a
for yourself in a controlled way (i.e., you don’t popular review). It is interesting to note that there
have to wait for a head cold). are times when the “distant smell” and “food

□ With smell

FIGURE 8-8 Identification of some common foods with and without smell. (Based on Mozel et al., 1969)
224 Chapter 8/Taste and Smell

DEMONSTRATION BOX 8-4. Flavor without Smell

The simplest way to experience flavor without smell ply need to get a friend to help you. First prepare
is to pinch your nostrils closed before coming close several different substances to be identified—the ones
to the substance to be tasted, and then put some of listed in Figure 8-8 will do. They should be in a liq¬
that substance into your mouth and swish it around uid state (i.e., mash up the garlic and mix it with
while paying attention to its flavor. Then release your water). Then seat your friend at some distance from
nostrils, open your mouth slightly, and breathe in the solutions and blindfold him or her. For each sub¬
gently through both mouth and nose. You should ex¬ stance, first ask him to pinch his nose and when he
perience a significant change in the flavor when you has done that, put some of the solution in his mouth
do this. Try it for the various substances listed in Fig¬ and ask for its name. After he has tried that, tell him
ure 8-8 and any others you can think of. Your expe¬ to release his nostrils and again attempt to name the
riences with nose pinched approximate those of el¬ substance. Repeat this for each substance, and several
derly people who have experienced large deficits in friends if you can, and see how closely the results
their sense of smell. match those of Figure 8-8.
It is rather easy to show that food identification is
impaired when the sense of smell is absent. You sim¬

smell” functions give different perceptual experi¬ ant and is used in making some of our most expen¬
ences. For instance, Limburger cheese has a long¬ sive perfumes.
distance smell that is quite strong and, most people In general, any molecule may be described as
think, offensive, but whose “smell” when in the having a specific size, weight, shape, and vibration
mouth contributes in a positive way to the flavor of frequency. This last property has to do with the fact
the cheese, which many people find quite pleasing. that atoms in a given molecule often are not held
firmly in place but move around in a characteristic
Smell Stimuli and Receptors pattern, at predictable speeds that are different for
different substances (Wright, 1977, 1982). In ad¬
Which aspects of a molecule give it the quality of dition, the particular atoms that make up a mole¬
evoking the sensation of smell? First, it must come cule, and the number of electrons available for
from a volatile substance (one that has a gaseous chemical bonding with other molecules on the
state at ordinary temperatures—in other words, smell receptors, are probably important compo¬
something that can evaporate), since air currents nents of smell stimuli. As yet there is no consensus
carry the molecules to the smell receptors in the as to which of these factors is critical; given the
nose. However, the most volatile substances do not large number of molecules that can be smelled, it
necessarily smell the strongest. Water, which has a is likely that several of them are important.
high volatility, has no smell at all (if it is pure). In The receptor cells that interact with the smell
fact, the extent to which a smell stimulus separates stimuli are located in a relatively small area in the
itself chemically from water (hydrophobicity) is upper nasal passages (see Figure 8-9) called the ol¬
highly correlated with the perceived intensity of factory epithelium, which translates to “smell
that stimulus (Greenberg, 1981). Conversely, musk skin.” The actual receptor cells are oval in shape.
(a secretion obtained from some deer and beavers) Each sends a long extension (called the olfactory
has a low volatility, yet it is a very powerful odor¬ rod) toward the surface of the olfactory epithelium.
 The Olfactory (Smell) Sense 225

and in addition sends its axons toward the brain. sensitivity. For example, a dog has about 200 mil¬
Thus, the receptors are actually specialized neu¬ lion olfactory receptors as opposed to only about
rons. Remarkably, the receptor cells only function 10 million in humans. Also a dog has 100-150
for about four to eight weeks before deteriorating; cilia per receptor as opposed to a paltry 6-8 cilia
new cells are continually being produced by the in humans (Brown, 1975).
basal cells (see Figure 8-9 and Costanzo & Gra- As is the case with the other senses, the mech¬
ziadei 1987). These are one of the few types of anism by which the stimulus molecules cause an
neuron in adult mammals that can regenerate. electrical response in the receptors of the olfactory
The olfactory rod of the receptor cell reaches epithelium is still something of a mystery. There
up toward the surface of the epithelium, where it are probably at least two classes of transduction
expands to form a knob. From this knob protrude a mechanisms (Gesteland, 1986; see also Getchell &
number of olfactory cilia, which are hairlike struc¬ Getchell, 1987). One is made up of highly selective
tures embedded in mucus secreted by a set of spe¬ processes in which specific receptor-cell proteins
cial glands found nowhere else in the nasal form reversible chemical bonds with specific parts
passages. These cilia are probably the receptor ele¬ of odorant molecules. These then trigger generator
ments that actually make contact with the smell potentials either directly or through other biochem¬
stimulus (Cagen & Rhein, 1980). Both the number ical processes inside the cell. The generator poten¬
of cilia per receptor and the total number of recep¬ tials of course produce the action potentials that
tors seem to be correlated with olfactory sensitiv¬ travel up the axons of the receptor cells. Notice that
ity. Animals that have more receptors and more in this theoretical formulation the specific receptors
cilia per receptor have keener senses of smell. Hu¬ are specialists for a specific odorant. Such selective
mans are at the lower end of the scale of smell processes may be distributed unevenly over the ep-
226 Chapter 8/Taste and Smell

ithelium as well as being located on specific parts temporal lobe of the cortex. Another type of cell
of the cilia. The second class of mechanism con¬ sends axons to a variety of lower brain centers, es¬
sists of less selective processes, in which various pecially the limbic system (which is involved in our
chemicals directly affect the receptor cell mem¬ experience of emotion and memory), as well as to
brane anywhere they contact it, causing ions to leak the smell cortex. The number of fibers leaving the
in or out, thus producing generator potentials. olfactory bulb is much smaller (about a thousand
These processes are the same for all receptor cells times smaller) than the number entering it, so pre¬
and may constitute what may be called generalist sumably many receptor cells contribute to the ac¬
smell receptors. Some insects seem to possess both tivity of each of the cells in the olfactory bulb and
specialist and generalist receptors in their olfactory later centers (Allison, 1953). The major route of
system (D. Schneider, 1969), and it is likely that information from the olfactory bulb to the smell
mammals do too. cortex is called the lateral olfactory tract. After
Two particular theories of the more selective the primary smell cortex the neural pathways be¬
mechanisms have been popular among psycholo¬ come extremely complex, including projections to
gists. The lock-and-key theory maintains that var¬ the thalamus and several other cortical areas (see
iously shaped molecules fit into special sites on the Price, 1987).
receptor membrane like a key into a lock. When a The study of the electrophysiology of the
molecule fits into a receptor site, a change in the smell system is still in its infancy. As we men¬
structure of the cell membrane results, allowing tioned earlier, however, a number of studies have
ions into or out of the cell and generating an elec¬ found that the intensity of the neural response var¬
trical current (Amoore, 1970). The vibration the¬ ies directly with the intensity of the stimulus. Most
ory maintains that the stimulus molecule ruptures contemporary investigators have focused on the
certain chemical bonds in the cell membrane, caus¬ more subtle problem of how different smell quali¬
ing the release of stored-up energy, which in turn ties are signaled to the brain. They have tried to
generates an electrical current. Which bonds are find evidence of specific types of receptors for dif¬
ruptured in which cells depends on the unique vi¬ ferent types of stimuli. One major early study was
bration frequency of each stimulus molecule that of Ottoson (1956), who measured the electrical
(Wright, 1977, 1982). response of the entire olfactory epithelium to var¬
ious stimuli. He discovered that passing a puff of
Neural Responses in Smell odor-laden air across the epithelium resulted in a
unique type of electrical response, a slow change
The olfactory receptor cells send their axons in the electrical charge of the receptor cells. This
through tiny holes in a bone at the top of the nasal change is thought to generate spike potentials in the
cavity to form the olfactory nerve. The nerve goes axons of these cells.
straight to the olfactory bulb, which is located in With improved experimental techniques, Ges-
front of and below the main mass of the brain (see teland, Lettvin, Pitts, and Rojas (1963) were able
Figure 8-9). The axons of the receptors and the to record the responses of single receptors in the
dendrites of cells from the olfactory bulb form olfactory epithelium. They recorded both the slow
complex clusters of connections in the bulb. These potential response to stimuli and the spike poten¬
clusters may be grouped according to the type of tials generated in the axons of the receptors by the
receptor or type of stimulus molecule involved same stimuli. These two types of electrical re¬
(Kauer, 1980, 1987). One type of olfactory bulb sponses are shown one on top of the other in Figure
cell seems to send axons directly to the primary 8-10. As you can see in the figure, this particular
sensory cortex for smell, which is located in the receptor responded vigorously to a musky odor.
 The Olfactory (Smell) Sense 227

than for taste. This is due to the fact that there is

no small list of primary smells for olfaction as you
will see below.

Smell Thresholds
A dog’s sense of smell can be amazingly acute.
Droscher (1971) tells the story of a dog trainer who
had worked with dogs for years. This trainer
brought his dog to a university professor for test¬
ing, claiming that the dog’s ability was “supernat¬
LLi
-' Pyridine
ural” since it could track people by scent even if
they were wearing rubber boots! Actually, the dog
FIGURE 8-10 Slow potential and spike potential was simply using its acute sense of smell to detect
responses of olfactory receptor cells to four
the millions of sweat molecules that leaked through
different smell stimuli. (Based on Gesteland et al.,
1963) the rubber boots. Why can’t we humans smell such
things? The sensitivity of each smell receptor is in
fact about the same for humans and dogs. The
less to nitrobenzene, hardly at all to benzonitrite, really important difference is that a dog has many
and not at all to pyridine. Gesteland et al. (1963) more smell receptors than a human, with more
thought these responses indicated the existence of hairlike cilia projecting from each, as we men¬
the sought-after receptor types, although they were tioned earlier.
cautious in making this interpretation. Such caution In the early 1960s, Stuiver studied the abso¬
seems to have been well founded, since later re¬ lute sensitivity of human smell receptors by making
cordings from single cells in the epithelium, olfac¬ a model of the nasal passages around the olfactory
tory bulb, and in cortical and lower brain centers epithelium. He used this model to calculate just
that receive olfactory information have found that how much of an olfactory stimulus actually arrived
each neural unit responds to a broad range of stim¬ at the surface of the epithelium (see Figure 8-11).
uli (Cain & Bindra, 1972; Giachetti & MacLeod, When threshold stimuli were considered, he found
1975; Kauer, 1987; MacLeod, 1971; O’Connell & that it takes 8 molecules at most to stimulate a sin¬
Mozell, 1969). gle receptor in the human. Considering all aspects
The same sort of across-fiber patterns that are of the manner in which molecules of odor stimuli
found in taste seem to be present in the olfactory are distributed in the nose, it can be argued that a
system, and it is likely that the “code” for smell single receptor cell needs to be contacted by one
qualities will be found in these patterns (Erickson stimulus molecule in order to respond (De Vries &
& Schiffman, 1975; Kauer, 1987). At present there Stuiver, 1961). Quite clearly this is the greatest
is no evidence of olfactory fibers falling into groups sensitivity that any single olfactory receptor cell
as the taste fibers seem to, although some investi¬ could have, hence a dog’s (or any other animal’s)
gators have proposed that receptors sensitive to the receptors can’t be more sensitive than a human’s.
same stimuli send their axons to the same part of Because the dog has more receptor sites, the like¬
the olfactory bulb (see Kauer, 1987). Furthermore, lihood that a very weak stimulus will actually stim¬
a labeled-line theory, such as the one we discussed ulate the olfactory receptors and produce a
for taste, would be somewhat cumbersome for ol¬ noticeable sensation is greater for the dog (see
faction, requiring many more types of labeled lines Marshall & Moulton, 1981).
228 Chapter 8/Taste and Smell

smell a strong stimulus of n-butyl mercaptan,

which has a foul, putrid odor. Amoore was able to
divide these 76 anosmias into about 31 classes,
each of which might represent a primary odor qual¬
ity. Such specific anosmias are consistent with the
idea that humans have two classes of olfactory re¬
ceptors, the specialists and the generalists that we
mentioned above (Wright, 1978a).

|--100 Smell Adaptation

FIGURE 8-11 A model of the nasal passage used As in most sensory systems, adaptation (from pro¬
by Stuiver to calculate the proportion of longed exposure to a particular stimulus) affects
molecules in a smell stimulus that reaches the smell thresholds and the perceived intensity of an
olfactory epithelium (O). Air follows the path odor. It even affects the pleasantness of odors
indicated by the arrows (measurements of
(Cain & Johnson, 1978). One of the great disap¬
distances are in mm). Stimulus molecules reach O
via eddy currents (like the eddies in a stream). pointments of wine tasting is that the aroma and the
(From De Vries & Stuiver, 1961. Copyright 1961 by the MIT bouquet of the wine seem to last only for a few
Press.)
sniffs. The rich complexity of a great wine soon
fades into a bland, featureless odor, even for
More traditional attempts to determine the ab¬ the most experienced “smeller,” unless frequent
solute thresholds for various odors encounter prob¬ breaks of about 15 sec are taken. Luckily, the
lems similar to those found when measuring taste odors of sweaty bodies and rotten eggs, and the
sensitivity. Thresholds vary across methods. They sulfurous smell of air pollution, also soon fade
seem to depend on the purity of the odorant, the away if you continue to sniff.
way it is delivered to the olfactory epithelium, and An important early study of adaptation of the
how the stimulus intensity is measured. Different smell sense was that of Moncrieff (1956). He stud¬
substances also have different thresholds. ied the effects of previous exposure to an odorant
In much the same way that we found individ¬ on the threshold for that odorant (self-adaptation).
ual differences in taste sensitivity, there are also He also studied how exposure to one odorant af¬
individual differences in smell sensitivity (see Ra¬ fected the threshold for different odorants (cross-
bin & Cain, 1986), with some individuals relatively adaptation). As you might expect, the largest loss
“odor-blind” to certain substances (see Engen, of sensitivity was found for conditions of self¬
1982). Amoore (1969, 1975; Amoore, Pelosi & adaptation. Cross-adaptation effects varied with the
Forrester, 1977) has reported the results of an ex¬ similarity of the smells of the two stimuli. Stimuli
tensive search for instances of odor blindnesses as with similar smells gave large cross-adaptation ef¬
a part of his plan to identify the primary odors. By fects; those that differed in smell gave smaller ef¬
1975 he was able to report 76 different anosmias fects. Surprisingly, all the adapting odors had some
(odor blindnesses) ranging from the smell of skunk effect on observers’ sensitivity for the others. This
to the smell of vanilla. Some of these are quite means that it was not possible to classify the var¬
common; for example, about 1 out of 3 individuals ious stimuli into a small number of primary classes
can not smell a strong stimulus of 1,8 cineole, based on adaptation data, which is similar to
which produces a camphorous odor (Pelosi & Pis- Amoore’s conclusions based on anosmias. You can
anelli, 1981). Conversely, some anosmias are quite experience self- and cross-adaptation for yourself
rare, as when only 4 people out of 4,030 couldn’t by trying Demonstration Box 8-5.
Color Plate 1

Color Plate 2

Color Plate 3
Color Plate 4

• m 9 •
9 9 • « • •
O
LO
O

O
O

30° 20° 10° 0°

Color Plate 5
 Color Plate 7

A Read through this list of color names as quickly as possible.

Read from right to left across each line.

RED YELLOW BLUE GREEN

RED GREEN YELLOW BLUE

YELLOW GREEN BLUE RED

BLUE RED GREEN YELLOW

RED GREEN BLUE YELLOW

B Name each of these color patches as quickly as possible.

Name from left to right across each line.

C Name the color of ink in which each word is printed as quickly as possible.
Name from left to right across each line.

RED BLUE GREEN YELLOW

YELLOW BLUE RED GREEN

BLUE YELLOW GREEN RED

GREEN BLUE YELLOW RED

BLUE YELLOW RED GREEN

Color Plate 9
 The Olfactory (Smell) Sense 229

DEMONSTRATION BOX 8-5. Smell Adaptation

You experience self-adaptation of odorants every day. tematic. You should use your own judgment and ex¬
The next time you notice a strong odor, take several plore a range of odorants using the general method
deep sniffs and then take a more usual sniff and pay described here. When two odorants you wish to test
close attention to the intensity of the odor as com¬ have been obtained, first step into another room,
pared to what you at first experienced. You should where you can’t smell them, and take several deep
notice a significant decrease in sensation intensity. sniffs. Then approach the odorants with pinched nose.
Alternatively, prepare yourself a cup of coffee or ar¬ Release your nostrils near the first (test) odorant, take
omatic tea, keeping your nostrils pinched while you a gentle sniff, noting the intensity of the odor. Then
do. When the steaming cup is in front of you, release take several deep sniffs of the other (adapting) odor¬
your nostrils and take a gentle sniff, noting the inten¬ ant, and return to the test odorant and take another
sity of the odor. Then take several deep sniffs fol¬ gentle sniff. Compare the odor intensity on this sniff
lowed by another gentle sniff and compare the odor to the first one. If it is less intense, you have experi¬
intensity during the final gentle sniff to that during the enced cross-adaptation, if more intense you have ex¬
first gentle sniff. perienced facilitation, as sometimes happens with bi¬
It is a bit more difficult to demonstrate cross¬ ologically significant odors (see Engen, 1982).
adaptation, since the effects are weaker and not sys¬

Cain and Engen (1969) looked at the effects stimulus intensity. Cain reported that the size of the
of adapting stimuli on the perceived intensity of exponent is directly related to the degree of water
odorants. They used a magnitude estimation pro¬ solubility of the odorant, with exponents for com¬
cedure (such as described in Chapter 2), and found pletely water-soluble odorants about 2.5 times as
that the higher the concentration of the adapting large as those for non-water-soluble odorants. More
stimulus the greater the reduction in the apparent recent work by Wright (1978b, 1978c) showed that
intensity of the test stimuli presented afterward. the exponents for various odorants can be predicted
This relationship does not hold for extreme test from specific ways in which the odorant molecules
stimulus values, however, since very intense test interact with receptor cell membranes.
stimuli all appeared to arouse about the same sen¬ In absolute terms the sense of smell is remark¬
sory response, regardless of the state of adaptation. ably acute, but for a long time it was thought that
humans could not discriminate between different
odor intensities very well. Most recent measure¬
Smell Intensity and Qualities
ments, however, indicate that the olfactory system
When we consider how the intensity of an odor is actually more sensitive than the taste system in
varies as stimulus intensity is varied, we again find this regard. In fact, the olfactory system may be as
it useful to refer to the exponent of the psycho¬ sensitive as the visual or auditory system in dis¬
physical power function to describe this relation¬ criminating changes in intensity, with Weber frac¬
ship. The exponents of power functions fitted to tions as low as 0.05, meaning that a change in
magnitude estimations of odor intensity differ intensity of only 5 percent can be detected about
across the various odors scaled. Cain (1969) found half the time (Cain, 1977).
that exponents ranged from about 0.7 to a low of As in all of the other sensory modalities, at¬
about 0.15. Remember that the higher the exponent tempts have been made to isolate a small set of
the faster the sensation increases as we increase “primary” smell qualities. If such could be found
230 Chapter 8/Taste and Smell

then all olfactory sensations could be predicted as Amoore’s (1975) suggestion that there may be as
the result of a combination of these primary re¬ many as 31 primary odors. Some of these particular
sponses. As we have already seen, the data from primaries seem to be associated with receptor sys¬
adaptation and anosmias do not suggest such a tems that respond to odors produced by the human
small cluster of primaries for smell. The classical body. These are the same types of odors emitted by
attempt to describe smell primaries was that of most animals in various situations relevant to the
Henning (1915). Figure 8-12 shows his “smell survival of the species, such as danger and sexual
prism,” which had six primary qualities arranged contact, and humans seem to respond to this class
at its comers. of stimuli as well.
Although it was a standard representation of
the “smell primaries” for quite some time, Hen¬
Sensitivity to Pheromones
ning’s prism apparently does a poor job of describ¬
ing the perceived relationships between odorants. Some smells have been said to have a special bio¬
When multidimensional scaling techniques are used logical significance for humans. Around the turn of
to provide a geometrical representation of odor- the century Ellis (1905) noticed that both men and
similarity judgments, in much the same way as for women often emit strong odors during sexual ex¬
taste judgments, neither Henning’s smell prism nor citement, and some researchers contend that some
any other readily identifiable classification scheme human behavior may be under the control of these
emerges. Figure 8-13 displays some representative olfactory stimuli (e.g., Comfort, 1971). Chemicals
results from this kind of study. This type of result secreted by animals that transmit information to
has led some investigators (e.g., Erickson & other animals (usually of the same species) are
Schiffman, 1975; Southwick & Schiffman, 1980) called pheromones. Although it has long been rec¬
to speculate that only a complex set of physio- ognized that such olfactory signals are important
chemical considerations could account for the inter¬ for many animal species, the possibility that scents
relationships shown in the qualities of smell affect human behavior has been entertained only
stimuli. Such results seem consistent with recently. As you might expect, manufacturers of
colognes and perfumes, ever searching for ways to
enhance the sales of their products, have immedi¬
Foul ately responded to this suggestion. In fact, a num¬
ber of products are now on the market that contain
alpha androstenol, which is known to be an effec¬
tive sexual pheromone for pigs and also occurs in
some human secretions.
Before we discuss the data available on hu¬
mans, it will be useful to put the problem in con¬
text by describing some of the work done on other
animals. Pheromones were first studied, indeed dis¬
covered and named, in insects (see Wilson, 1971).
There are two major types: releasers, which on re¬
ception by an animal “release,” or automatically
FIGURE 8-12 Henning’s smell prism. Six
trigger, a specific behavioral response; and prim¬
“primary” odors are at the corners; the surfaces ers, which trigger glandular and other physiologi¬
represent stimuli that resemble more than one cal activities in the recipient. Most of the aspects
primary. of the lives of insects, particularly social insects
 The Olfactory (Smell) Sense 231

Clove

Camphor

Eucalyptus
Feces
• •

Cinnamon Menthol

Sweat

Vanilla

Lemon Burnt

Citrous

Rotten eggs
Turpentine

FIGURE 8-13 A composite replotting of selected odors from multidimensional scaling of similarity
judgments of odor pairs (Schiffman, 1974). Dimension / is thought to be “pleasantness.” This mapping
bears little resemblance to Henning’s smell prism (Figure 8-12).

like ants, bees, and termites, are regulated via cies, including rodents, dogs, and monkeys (see
communication by pheromones. Insects attract their Brown & MacDonald, 1985a, 1985b, for a re¬
mates, recruit fellows for food gathering or fight¬ view). Mammalian releaserlike effects include the
ing, and recognize each other and their own species mutual recognition of mother and young offspring.
via releasers. Queen ants, bees, and termites con¬ Male and female sexual attraction is also affected
trol swarming, new-queen production, proportion by such scents, as is easily demonstrated by ob¬
of types of workers, and so on, using primers. The serving the effect of the odor of a female dog in
important thing to remember about insect phero¬ heat on the male (Goodwin, Gooding & Regnier,
mones is that usually a specific chemical, produced 1979). Other pheromones elicit aggression in
by a specific gland, and detected and recognized by males, inhibit aggression in females, and are used
a specific (or specialist) receptor is involved in for signaling purposes, such as trail marking or de¬
such communication. Insect behavior is directly fining territorial boundaries. Primerlike effects in¬
under the control of these pheromones. clude effects of scents on the estrous cycle of
Mammals, of course, are more complex ani¬ females, the age at which puberty is attained, and
mals than insects, and the effects of such olfactory even the likelihood of becoming pregnant.
stimuli are more subtle. Many pheromone-related Some effects of pheromones have been ob¬
effects have been found, however, in several spe¬ served in species rather close to humans on the
232 Chapter 8/Taste and Smell

evolutionary scale, namely monkeys. However, that people can reliably detect their own body odor
here the story becomes even more complex. Some from among a set of similar stimuli contributed by
early research studied the effect of vaginal secre¬ other people. The actual stimuli were collected by
tions from female monkeys in estrus (that is, those having people wear T-shirts for several days. When
that are sexually receptive) on the copulating be¬ asked to rate the pleasantness of the body odors and
havior of males. Early work seemed to indicate that to describe the individual whose odor they were
a powerful pheromone was involved. Normally, smelling, there was a remarkable degree of consis¬
male monkeys seldom attempt to engage nones- tency across people. In general, the shirts having
trous females sexually. Such behavior could be unpleasant odors were usually described as if the
elicited, however, if the females’ sexual skins had individuals also would be likely to have socially
been rubbed with the secretions of the sexual skins undesirable traits. They were supposedly dumb,
of females who were in estrus. These secretions, ugly, fat, and unhealthy. The more pleasant odors
presumably containing pheromones, were dubbed were described as coming from people with more
copulins (Michael & Keveme, 1968; Michael, desirable traits (McBurney et al., 1977).
Keveme & Bonsall, 1971). Later work indicated People seem to be able to discriminate reliably
that the suspected releaser pheromone was not as the sex of an odor donor. In one study, the odors
powerful as had been thought, and that monkey of males were characterized as “musky” whereas
sexual behavior was under the control of many in¬ female odors were described as “sweet” (Russell,
teracting factors, only one of which was the smell 1976); other studies have found that the stronger
of the partner’s sexual skin (Goldfoot, 1981; Gold- (and more unpleasant) odors are more likely to be
foot, Kravetz, Goy & Freeman, 1976; Keveme, attributed (correctly) to males (Doty, 1985; Doty,
1978). In fact, odor cues are not necessary at all, Green, Ram & Yankell, 1982; Doty, Kligman,
for anosmic male monkeys show normal mating Leyden & Omdorff, 1978). Similar results are
behavior (Goldfoot, Essock-Vitale, Asa, Thornton found in different cultures (Schleidt et al., 1981).
& Leshner, 1978). In all of the pheromonelike ef¬ Males and females can also be discriminated on the
fects studied with mammals, communications de¬ basis of hand odor (Wallace, 1977) and breath odor
livered by chemical means, although often very (Doty, et al., 1982). Females do better at this
important, have not been found to be as imperative (Doty et al., 1982; Wallace, 1977) and in fact do
as they are with insects. Mammals have more com¬ better than males at odor identification in general
plex brains and their behavior is influenced by (Doty, Applebaum, Zusho & Settle, 1985). You
many different environmental factors, only one of can try your nose at some of these tasks using
which is odorants (see MacDonald & Brown, Demonstration Box 8-6.
1985). Human sensitivity to human odors seems to be
When we consider humans, the behavioral ef¬ present from an early age. For instance, Russell
fects of pheromones seem to be most subtle, since (1976) found that a small sample of 2-week-old ba¬
social and learning factors influence our behavior bies reliably responded to the odor of their own
even more than they do that of other mammals. mother’s breast and not to that of a strange mother;
Some very interesting results, however, indicate this ability may exist as early as 6 days after birth
that smells may play an important, although prob¬ (MacFarlane, 1975). Breast-feeding babies (but not
ably subordinate, role in many aspects of human bottle-feeding babies) of the same age can also rec¬
social behavior. First, several studies (McBurney, ognize their mothers’ armpit odor (Cemoch & Por¬
Levine & Cavanaugh, 1977; Russell, 1976; ter, 1985). Actually, female (but not male)
Schleidt, Hold & Attili, 1981) have demonstrated neonates (less than 2 days old) develop a prefer-
 The Olfactory (Smell) Sense 233

DEMONSTRATION BOX 8-6. Social Significance of Human Odor

This demonstration is a bit complex, but is worth the group of friends put on scent-free T-shirts and wear
effort. It is designed to see whether you or your them for at least 24 hours. At the end of the time
friends can discriminate male from female body period, put the T-shirts (with paper labels under them
odors. The most important factor here is to eliminate to indicate the sex and identity of the donor) into
the effects of perfumes and deodorants, which is dif¬ clean plastic bags. You now have your odor stimuli.
ficult since almost all commercial products now con¬ Have each friend sniff at the opening of each bag and
tain scents. Ivory soap is one product that does not, indicate whether the sample came from a male or a
so it is advisable to have odor-donors wash them¬ female. You might also want to have them try to
selves with this soap before contributing the odor, or guess whose T-shirt it is, and which one is their own,
to simply soak for a period of time in clear water. and you might even want to ask for a rank order in
The same should apply for the clothing mentioned be¬ terms of pleasantness. When all have done this, tab¬
low. The simplest experiment involves having a ulate the results and see how well everyone did.

ence for an artificial odorant simply by being ex¬ region) sweat, it may play a role in human sexual
posed to it (Balogh & Porter, 1986). attraction, especially since humans have been
One of the necessary conditions for most odor shown to reliably discriminate among axillary (un¬
identification, at least in adults, seems to be a lot derarm) odors. A few published studies have found
of experience with the odor and its source (Cain, positive effects. For example, androstenol affects
1979; Rabin & Cain, 1984). Certainly the relation¬ ratings of various social characteristics of hypothet¬
ship between parents and children contains much ical applicants for a job, and affects male and
such experience. We might thus expect that parents female ratings differently (Cowley, Johnson &
could recognize the odors of their own offspring, Brooksbank, 1977). Similarly, women in photo¬
and that perhaps siblings could recognize each oth¬ graphs were rated as more sexually attractive by
ers’ odors as distinct from those of other children. both men and women wearing surgical masks im¬
This seems to be the case, raising the possibility pregnated with androstenol than by those wearing
that smell plays an important role in interactions control masks (Kirk-Smith, Booth, Caroll & Dav¬
among humans from the same family (Porter, Bal¬ ies, 1978). Finally, more women and fewer men
ogh, Cemoch & Franchi, 1986; Porter & Moore, (than in a control condition) used a seat in a den¬
1981). tist’s waiting room that had been sprayed with an-
A second line of research with humans has drostenone (Kirk-Smith & Booth, 1980).
been aimed at finding a sex attractant based on re¬ Such results are exciting to perfume manufac¬
leaserlike pheromone action. The most promising turers (and, perhaps, to the sexually deprived) but
candidate for such a substance is the substance al¬ they may not be easy to interpret. Rogel (1978)
pha androstenol mentioned above, which causes a critically reviewed a number of such studies along
sow to become immobile, and thus receptive to a with those of monkeys, and concluded that al¬
boar’s sexual advances, when the boar secretes it though it is possible that olfaction does influence
in his saliva. Since alpha androstenol is also pres¬ many aspects of social behavior, it would be wrong
ent in human apocrine (a gland in the underarm to believe that human behavior could be controlled,
234 Chapter 8/Taste and Smell

to the extent seen in lower mammals and insects, Lateral olfactory tract The main route, composed
by such pheromones. This means that human sex¬ of axons, from the olfactory bulb to the smell cortex.
ual choice, contrary to the claims of some perfume Lock-and-key theory A theory of smell mechanism
in which variously shaped molecules fit into holes in the
manufacturers, is apt to be more a matter of higher
walls of olfactory receptor cells like a key into a lock,
mental processes than of primitive responses to causing an electrochemical change that in turn causes an
sexual odors. action potential in the axon.
The most recent work on pheromonelike ef¬ Medial lemniscus The pathway that conveys taste
fects in humans has been more promising but less information from the solitary nucleus to the thalamus.
dramatic. It has been found that regular (weekly) Molar concentration A measure of the amount of a
intimate contact with a male, or even with male tastant present; 1 mole equals the molecular weight of a
substance in grams added to enough water to make 1 liter
underarm secretions, causes the female menstrual
of solution.
cycle to become more regular in length (Cutler,
Olfactory bulb A complex brain nucleus where the
Preti, Krieger, Huggins, Garcia & Lawley, 1986). axons of olfactory receptor cells terminate; sends axons
Regular contact with other females, or with their to various brain centers including the olfactory cortex
underarm secretions, causes female menstrual cy¬ and limbic system.
cles to become more synchronized (Preti, Cutler, Olfactory cilia Hairlike projections extending from
Garcia, Huggins & Lawley, 1986; see also Mc- the knoblike end of the olfactory rod, protruding through
Clintock, 1971). These are primerlike effects, since the surface of the olfactory epithelium.

they involve physiological changes and not specific Olfactory epithelium The small area of oval-shaped
cells in the upper nasal passages that respond to smell
behaviors. They can be important—for example,
stimuli.
menstrual regularity is associated with healthy re¬
Olfactory nerve The bundle of axons of smell re¬
productive functioning and fertility—but they are ceptor cells that passes through the top of the nasal cavity
not direct and powerful influences on human be¬ and terminates in the olfactory bulb.
havior. Olfactory rod A long extension from smell receptor
cells toward the surface of the olfactory epithelium.
Papillae Small bumps on the tongue in which taste
GLOSSARY buds are located.
Pheromones Chemicals secreted by animals that
transmit information to other animals, usually of the
The following definitions are specific to this book.
same species.
AB,H system A chemical structure consisting of two
Potentiation A case of cross-adaptation in which ex¬
negatively charged atoms (A,B) and a positively charged posure to one taste stimulus lowers the threshold to an¬
hydrogen atom (H) arranged in a special way; involved other taste stimulus.
in sweet and possibly bitter tastes.
Primers Pheromones that trigger glandular and other
Across-fiber pattern A pattern of neural activity in physiological responses.
which various neural units have different stimulus-spe¬
cific response rates. PTC Phenylthiocarbamide, a substance that shows
large variations in absolute threshold across different in¬
Anosmia Relative insensitivity to an odor.
dividuals; some people are “taste-blind” to it.
Anterior-insular cortex The cortical center for taste
Releasers Pheromones that trigger specific behav¬
information.
ioral responses.
Cross-adaptation A phenomenon in which exposure
Self-adaptation A phenomenon in which exposure
to one tastant (or odorant) affects the absolute threshold
to a tastant (or odorant) raises the absolute threshold or
or sensation intensity of other tastants (or odorants).
decreases the sensory intensity of the same tastant (or
Labeled-line theory A theory of taste in which each odorant).
taste fiber encodes the intensity of a single basic taste
Solitary tract The region of the brain stem that re¬
quality.
ceives information from cranial nerves about taste.
 Glossary 235

Taste buds The group of cells in which the major Vibration theory A theory of smell mechanism in
receptors for taste are located, on the tongue and parts of which a stimulus molecule ruptures chemical bonds in
the mouth. the cell membrane of olfactory receptor cells, causing a
Taste pore An opening in the surface of the tongue release of stored energy that generates an electrical cur¬
leading to the taste cells extending from the taste bud. rent and action potentials.
Thalamus The region of the lower brain that relays
impulses to the cortex.
CHAPTER
9
Touch and
Pain
THE SKIN SENSES
Skin Stimuli and Receptors
Neural Pathways and Responses
TOUCH
Touch Thresholds
Touch Adaptation
Touch Intensity
Tactile Pattern Perception
WARMTH AND COLD
Neural Coding of Temperature
Thermal Thresholds and Adaptation
Thermal Intensity and Qualities
KINESTHESIS
Kinesthetic Stimuli and Receptors
Neural Responses in Kinesthesis
Perception of Weight and Force
PAIN
Pain Stimuli and Receptors
Neural Responses to Pain Stimuli
Pain Thresholds, Intensity, and
Adaptation
Analgesia and Endogenous Opiates
238 Chapter 9/Touch and Pain

( ( Tguessed vaguely from my mother’s signs and in intimate contact with the outside world. In the
A from the hurrying to and fro in the house that most primitive one-celled organisms, the skin is the
something unusual was about to happen, so I went cell membrane, and it is responsible for all the or¬
to the door and waited on the steps. The afternoon ganism’s contacts with its environment. These con¬
sun penetrated the mass of honeysuckle that cov¬ tacts include such functions as taking in food,
ered the porch, and fell on my upturned face. My excreting waste, isolating the inside of the cell
fingers lingered almost unconsciously on the famil¬ from damaging outside substances, and responding
iar leaves and blossoms which had just come forth to all sorts of external stimuli. Although more ad¬
to greet the sweet southern spring. I did not know vanced organisms, such as mammals (including hu¬
what the future held of marvel or surprise for me. mans), have more specialized sensory organs to
. . . We walked down the path to the well-house, handle these tasks, their skins are also important
attracted by the fragrance of the honeysuckle with and complex organs. The skin plays a role in res¬
which it was covered. Someone was drawing water piration, temperature regulation, and protection. It
and my teacher placed my hand under the spout. also has a wide variety of sensory functions, and
As the cold stream gushed over one hand . . .” produces the sensations of touch, warmth or cold,
(Keller, 1931, pp. 21-22). and pain.
Did it occur to you, as you read the passage
above, that its author was blind and deaf? Helen
Skin Stimuli and Receptors
Keller’s perceptual world was restricted to smells,
tastes, touches, and feelings of warmth and cold The skin responds to a variety of physical stimuli.
and pain. Her teacher taught her to communicate When we press an object against the skin, it de¬
by finger taps on each others’ palms, demonstrating forms the surface and we experience the sensation
that as long as we have normal skin we are not cut of touch, or pressure. When an object makes con¬
off from the richness of language. Although we are tact with a hair, causing it to bend, we also expe¬
amazed and heartened by the accomplishments of rience touch. The temperature of the object with
someone like Keller, it is true that many species of which we touch the skin also elicits a sensation.
animals rely almost exclusively on the so-called Whether it is warmth or cold depends both on the
minor senses for information about the world. We temperature of the stimulus and on the temperature
have already described the senses of taste and smell of the skin. Finally, the skin responds to electrical
in Chapter 8. Here we deal with the skin senses, stimulation. For mild electrical stimuli, a type of
kinesthesis (which is closely related to touch), and touch sensation is usually felt, although tempera¬
pain, which can be felt anywhere but is most ture can also be experienced. When electrical stim¬
clearly understood in relationship to the skin. ulation becomes intense, the sensation usually
becomes painful.
In humans the skin has a very complex struc¬
THE SKIN SENSES ture. Figure 9-1 is a diagram of the most important
structures in hairy skin, which covers most of the
All living things have a “skin” of some kind. human body. A different kind of skin, found on the
Probably the most important function of the skin is palms of the hands, soles of the feet, parts of fin¬
to define the organism, that is, to set boundaries in gers and toes, and other places, has no hairs pro¬
space, inside of which exists the organism and out¬ truding from it and is called glabrous skin.
side of which exists the environment. The skin is Although there is a thick outer layer of dead cells
an interface (a place where two systems meet) be¬ in glabrous skin, there are also many free nerve
tween the organism and the environment, and it is endings embedded in this layer. This makes such
 The Skin Senses 239

skin effective protection but also extremely sensi¬ acteristics (e.g., receptors that respond selectively
tive to stimulation. All skin consists of two basic to mechanical, thermal, or painful stimuli). In par¬
layers. The outer layer, called the epidermis, con¬ ticular, most “corpuscular” endings—nerve end¬
sists of several layers of tough dead cells on top of ings with small bodies or swellings on the
a single layer of living cells. The living layer di¬ dendrites, including the Pacinian corpuscles, Meiss¬
vides constantly to generate the dead protective ner corpuscles, Merkle disks, and Ruffini end¬
layers above. The inner layer, called the dermis, ings—seem to be associated with various types of
contains most of the nerve endings in the skin. Un¬ fibers that are particularly responsive to touch stim¬
der these two layers is usually a layer of fat cells. uli, whereas “noncorpuscular,” or so-called free
In addition to these layers, the skin contains a va¬ nerve endings in subcutaneous fat are associated
riety of hairs, muscles, glands, arteries, veins, and with pain fibers (see Vierck, 1978). Free nerve
capillaries. Some of these are also shown in Figure endings projecting into the epidermis may be asso¬
9-1. ciated with cold fibers (Hensel, 1981) or pain fibers
Figure 9-1 also shows some of the most com¬ (Perl, 1984). Modem electrophysiological and his¬
mon nerve endings in the skin. It has been difficult tological techniques may eventually succeed in
to specify exactly which ending is responsible for identifying the nerve endings associated with the
which of the skin sensations. In fact, all the end¬ several types of nerve fibers that exist further along
ings seem to respond somewhat to all the different in the neural pathways (e.g., Torebjork, Ochoa &
types of stimulation. Recently, some progress has Schady, 1983). Much progress has already been
been made in associating various receptor types made with regard to Merkle disks and Pacinian cor¬
with nerve fibers that have different response char¬ puscles (Gottschaldt & Vahle-Hinz, 1981).

Merkle discs (touch)

Meissner's Epidermis
corpuscle
(touch) Free nerve
endings (cold)
Sebaceous gland Dermis

Duct of
End bulbs sweat gland
of Krause
Ruffini ending
(touch)
Nerve ending around
hair (basket ending) Free nerve
ending (pain)

Fat globules Pacinian

corpuscle
(touch)

FIGURE 9-1 A piece of hairy skin in cross section. (Based on YVoolard, Weddell & Harp man, 1940)
240 Chapter 9/Touch and Pain

As an example of how a skin receptor re¬

sponds to stimulation, consider the Pacinian cor¬
puscle (see Figures 9-1 and 9-2). It has been well
studied because it is large, easily accessible, and
occurs in nearly all animals that have complex ner¬
vous systems. The elegant work of Loewenstein
and his colleagues (see Loewenstein, 1960) in¬
volved peeling away the surrounding layers of the
cell (much as we would peel an onion) to allow the
FIGURE 9-2 A Pacinian corpuscle.
researchers to touch the axon itself. Thus they were
able to show that a mechanical stimulus operates
directly on the axon of the nerve by deforming its dix for information about ion flow and spike poten¬
membrane. This mechanical action causes numer¬ tials). The surrounding layers of material in the
ous tiny holes in the membrane to open, allowing Pacinian corpuscle seem to be present simply to
electrically charged particles (ions) to flow from make the effect of the stimulus less intense. Tem¬
one side of the membrane to the other, which in perature could act in a similar way, perhaps by
turn causes an electrical current to flow between the controlling chemical reactions that would affect the
point of stimulation and another point on the axon. ability of electrically charged particles to cross the
Finally, this electrical current generates a spike po¬ cell’s membrane. Electrical stimuli probably trigger
tential that jumps along the axon and carries the spike potentials directly. We do not know whether
message of stimulation to the brain (see the Appen- all the cutaneous nerve endings operate similarly,

DEMONSTRATION BOX 9-1. Inhibitory Interactions on the Skin

In this demonstration you will see how skin sensa¬ like two points or one on your skin. If you are push¬
tions interact. You will need two fairly sharply ing hard enough and paying close attention to your
pointed objects such as two toothpicks or two bristles sensations, at just about the separation where the two
from a hairbrush. The demonstration will work better points begin to feel like two distinct points on the
if you ask a friend to control the stimuli. Do not use skin, you should have a surprising experience. The
anything like a knife, for you will be pushing the magnitude of the sensation from the two points should
point quite strongly against your skin. First, try press¬ diminish greatly, perhaps vanish altogether for a short
ing one point against the skin of your palm. Notice time. The sensation should be very faint, even though
the spread of sensation around the stimulated point. two toothpicks (or brush bristles) are pushing with
Now put the two points as close together as you pos¬ some force against the skin. As you then move the
sibly can. Push them together on the same place on points even farther apart, you will perceive two dis¬
your palm. Notice that you feel only one point, al¬ tinct, full strength sensations, appropriately separate
though two are present. Now move the two stimulat¬ on the skin. This phenomenon is explained by the
ing points slightly apart. You should still feel only overlapping of regions of excitation and inhibition in
one point. Repeat this procedure several times, mov¬ adjacent receptive fields of the skin, as shown in the
ing the points apart by a little more each time and accompanying figure.
paying careful attention to whether the sensation feels
 The Skin Senses 241

but it is reasonable to suppose that they do. It is temperature, or noxious), (2) according to the way
also possible that each nerve ending may utilize they respond to those stimuli (slow- or fast-adapt¬
more than one of the several possible transduction ing), and (3) according to whether they have large,
mechanisms, as seems to be the case in taste (see ill-defined receptive fields or small, well-defined
Chapter 8). ones. By receptive field here, we mean much the
same thing that we did for vision (see Chapter 3),
only here it refers to that region of the skin that,
Neural Pathways and Responses
when stimulated, causes responses in a particular
Because there are so many different types of nerve neural fiber. These receptive fields can be quite
endings in the skin, you might think that the nerve complex. They are organized in such a way that
pathways to the brain would be hopelessly com¬ sensory activity in one region can interact with sen¬
plex. They are complex, but there is an organiza¬ sory activity in nearby regions. If the distance be¬
tional plan we can understand. The plan depends tween two stimuli is appropriate, a reduction in the
on two major principles associated with the type of perceived magnitude of the sensation may even oc¬
nerve fiber and the place of termination of the path¬ cur. This seems to indicate that the receptive fields
way in the cortex. in the skin have the same sort of excitatory-center,
The type of nerve fiber is important because inhibitory-surround organization that we observed
different types of nerve fibers carry different types in the visual system (Bekesy, 1967; Gardner,
of information to the brain. Fibers can be classified 1983). Demonstration Box 9-1 shows how you can
in at least three ways: (1) according to the type of demonstrate this organization for yourself with
stimulus that most easily excites them (mechanical. touch stimuli.
242 Chapter 9/Touch and Pain

Using the three classification criteria described though temperature fibers have also been found
above, humans have been shown to have at least (Hensel, 1981).
four different types of fibers that respond to me¬ The second major pathway is called the spi¬
chanical deformation of glabrous skin: (1) rapidly nothalamic pathway. This pathway is rather slow,
adapting fibers with small, well-defined receptive and is made up of many short fibers instead of a
fields, (2) slowly adapting fibers with similar re¬ few long axons. At the brain stem, it divides into
ceptive fields, (3) slowly adapting fibers with large, two branches, the paleospinothalamic (paleo
ill-defined receptive fields, and (4) rapidly adapt¬ means “old”) and the neospinothalamic (neo
ing fibers with similar receptive fields (Vallbo, means “new”). The paleospinothalamic pathway is
1981). Such fibers are reminiscent of the sustained evolutionarily older and seems specialized for sig¬
and transient pathways associated with the X and naling dull or burning pain—it probably receives
Y fibers that leave the retina of the eye (see Chap¬ most of its input from the small, unmyelinated,
ter 3). slow-conducting, C fibers that terminate in free
The second major principle is that where on nerve endings in the skin. The neospinothalamic
the skin a particular nerve ending is found deter¬ pathway seems specialized for signaling sharp or
mines where its information goes in the brain, re¬ pricking pain and probably receives most of its in¬
gardless of the type of fiber it represents. All the put from the small, myelinated, slow-conducting
sensory information from the skin is passed on to A8 fibers that terminate in free nerve endings in
the spinal cord through 31 pairs of nerves (1 mem¬ the skin. It also receives input from the large, fast¬
ber of each pair for each side of the body). These conducting A(3 fibers. (A(3, A5, and C fibers are
nerve paths enter into the back portion of the spinal described in more detail in the “Pain” section of
cord, hence are called the dorsal roots. There are this chapter.) The two spinothalamic pathways as¬
also 4 cranial nerves that collect cutaneous infor¬ cend on the opposite side of the spinal cord from
mation from the head region. These inputs are where their input fibers terminate in the skin and
gathered into 2 main pathways to the brain, each of then innervate several areas of the brain, the most
which seems to carry different types of informa¬ important being the thalamus and the limbic sys¬
tion. Figure 9-3 shows various aspects of these tem (responsible for emotion and memory). Fibers
pathways. from these areas then go to the somatosensory cor¬
The first pathway is called the medial lemnis¬ tex. This pathway seems to carry information about
cus. The nerve fibers that comprise it are large, temperature and touch as well as pain.
conduct information quickly, and mostly receive The somatosensory cortex is divided into two
inputs from the large, myelinated, fast-conducting main parts, labeled SI and SII, and SI has several
A(3 fibers terminating in corpuscular endings in the identifiable layers, labeled 1, 2, 3a, and 3b (see
skin. The pathway ascends the spinal cord on the Kaas, 1983). Thalamic neurons project mainly to
same side of the body until it reaches the brain one or more layers in SI, depending on where they
stem, where most of the nerve’s fibers cross over came from. SI neurons then project to SII (Pons,
to the other side. The pathway then continues to Garraghty, Friedman & Mishkin, 1987). This or¬
the thalamus and finally arrives at the somatosen¬ ganization is very similar to that of the visual cor¬
sory cortex, which is located in the parietal region tex (see Chapter 3).
of the brain (the upper central region, shown in The relationship between where a stimulus is
Figure 9-4). Thus, the pathway terminates in the applied to the skin and where neural activity occurs
somatosensory cortex on the opposite side of the in the somatosensory cortex is quite regular. One
body from where it started. This system has fibers classic “map” of this relationship was created by
that respond mostly to touch and movement, al¬ Penfield & Rasmussen (1950). These investigators
 The Skin Senses 243

FIGURE 9-3 Schematic drawing of some of the important neural pathways from the skin to the brain.
The sections of the spinal cord and brain stem are horizontal; that of the brain is vertical.
244 Chapter 9/Touch and Pain

Somatosensory cortex

FIGURE 9-4 Penfield and Rasmussen’s (1950) topographic map of projections of “touch” nerve fibers on
the somatosensory cortex. The length of the line next to the drawing of each body part is proportional to
the area of somatosensory cortex subserving that body part. (From W. Penfield and T. Rasmussen, The Cerebral Cortex
of Man. Copyright 1950 by Macmillan Publishing Co., Inc., renewed 1978 by Theodore Rasmussen.)
 Touch 245

electrically stimulated the somatosensory cor¬ TOUCH

tex of patients during brain operations. As various
points on the cortex were stimulated, the patients Given the fact that any point on the body’s surface
reported the sensations that they felt, which might can evoke the sensation of touch, it is surprising
be the tingling of one leg and so forth. The result¬ how often this modality is ignored. For instance,
ing somatosensory map of the body is shown in we seldom consider that major components of most
Figure 9-4. Notice that the spatial location of stim¬ sexual experiences are touch sensations. The very
ulation on the skin is preserved in the spatial loca¬ act of touching another person, in Western society,
tion of activity in the cortex. More recently, details is considered to be an act of considerable intimacy,
of how the cells in the somatosensory cortex en¬ whether it is the gentle touch of a friend or lover
code information about touch stimuli have become or the violent punch of an aggressor.
clearer. For example, these cells respond to move¬
ment of ridges across the skin in much the same
Touch Thresholds
way that cells in the visual cortex respond to mov¬
ing edges in the visual field (Darian-Smith, Sugi- One of the most striking aspects of our sense of
tani, Heywood. Karita & Goodwin, 1982; Gardner, touch is how our sensitivity varies from one region
1983; Leinonen, 1983; Whitsel, Dreyer, Hollins & of the body to another. Figure 9-5 shows the ab¬
Young, 1979). You can experience one of the con¬ solute touch thresholds for several different regions
sequences of a cortical map that encodes the stim¬ of the body (excluding most of the erogenous
ulation of adjacent regions of the skin into adjacent zones, which probably contain some of the lowest
regions of cortical activity by trying Demonstration thresholds). Such thresholds are obtained by apply¬
Box 9-2. ing a small rod or hair (pig bristles were used in

FIGURE 9-5 Average absolute thresholds for different regions of the female skin. Relative values for
males are similar but higher overall. (From S. Weinstein, in D. R. Kenshalo, ed., The Skin Senses, 1968. Courtesy of Charles C.
Thomas, Publisher, Springfield, Illinois.)
246 Chapter 9/Touch and Pain

DEMONSTRATION BOX 9-2. Aristotle’s Illusion

The famous Greek philosopher-scientist Aristotle no¬ the outsides of the two fingers because of your finger
ticed an interesting illusion of touch that is quite easy contortions, the information is being sent to two sep¬
to demonstrate. Hold your fingers as shown in Figure arate areas of the touch cortex, allowing you to ex¬
A and touch the point between them with a pencil, as perience two distinct touches. Such a cortical map¬
shown. Notice that you feel one item touching you ping is quite reasonable, since commonly a single
and the sensation of one single touch. Now cross your object between two fingers would be expected to
fingers as shown in Figure B, touching yourself again stimulate adjacent skin surfaces, hence should be en¬
with a pencil in the place indicated between the fin¬ coded as a single touch source. It is normally not pos¬
gers. Notice that you feel two distinct touches. The sible, however, for a single object to stimulate the
effect may be stronger if you close your eyes during outsides of two different fingers, and so two different
the touches. The simplest and most plausible expla¬ touches should be experienced in these circumstances.
nation of the illusion is that when the pencil is stim¬ The cortical mapping reflects these common situ¬
ulating the inside of the two fingers (Figure A), the ations. It seems that whether the fingers are actually
touch information is being sent to overlapping or ad¬ crossed or not, all processing makes the assumption
jacent areas of the touch cortex, resulting in the sen¬ that they are uncrossed (Benedetti, 1985).
sation of one touch. When the pencil is stimulating
 Touch 247

some of the earliest work) to the surface of the skin cies up to 20000 Hz (Verrillo, 1975). Absolute
with differing amounts of force. When an investi¬ thresholds for vibration also depend on skin tem¬
gator continues to use the same hair, these thresh¬ perature; they are usually lower the higher the tem¬
olds can be expressed in terms of the amount of perature. This effect is restricted to vibration
force applied to the hair, as they are in Figure 9-5. frequencies over 100 Hz for glabrous skin, but oc¬
Thus, the higher the bar, the greater the force curs at all frequencies for hairy skin, for example,
needed, and the lower the sensitivity. It is probably that of the forearm (Verrillo & Bolanowski, 1986).
best, however, to express absolute thresholds in Cooling also increases absolute threshold on the
units of force per unit area, since the actual stimu¬ tongue, at least for intermediate frequencies of vi¬
lus is the change in the amount of tension in the bration (Green, 1987).
skin and not the force itself (Frey & Kiesow, There seems to be an absolute minimum
1899). At any rate, it is apparent in Figure 9-5 that threshold measurable for touch, similarly to vision.
absolute thresholds vary considerably over the body Vallbo (1983) reports experiments that determined
surface. that 1 nerve impulse in 1 rapidly-adapting fiber
Even more dramatic variations of threshold from the hand could be detected (he was recording
exist within a relatively small area of skin, say, the the nerve impulse from a subcutaneous electrode
surface of the arm. If you explore a 2 x 2 cm area inserted into the nerve of an awake human volun¬
on your forearm with a toothpick or hairbrush bris¬ teer). This single impulse was stimulated in this
tle, pressing with the same light pressure every case by a 10-fim movement of a tiny probe placed
time you touch the skin (just enough to make the on the skin. This is a very small movement indeed.
bristle bend slightly), you will discover a number Certainly, we can expect no greater sensitivity at
of spots that respond to this stimulus with a distinct the neural level than the detection of a single spike
sensation of touch. You will also find a large num¬ potential as a touch sensation.
ber of places that will give only a faint sensation, One aspect of all tactual stimuli is that each
or none at all. touch sensation seems to be located at a particular
The sensitivity of the skin is often tested using place on the skin. Our ability to localize a sensation
a vibrating stimulus, which alternately applies and accurately varies across different regions of the
releases a force to the same small surface region at skin, but seems to be directly related to the amount
frequencies ranging from 20 to 20000 Hz. Gener¬ of neural representation each area has in the touch
ally, vibrating the touch stimulus results in a lower cortex. In general, the greater the representation of
absolute threshold; that is, when the stimulation is a particular area, the smaller are the errors of lo¬
intermittent (on again, off again) the skin seems to calization for that area (the relative cortical repre¬
be more sensitive. This is reasonable, since the sentation of areas of the body was shown in Figure
change in the tension in the skin is the stimulus for 9-4). One way of measuring our sensitivity for lo¬
touch sensation and vibrations cause these changes calization is to introduce a second stimulus and to
repeatedly. The threshold for touch sensation de¬ measure the two-point threshold. This refers to a
pends on the frequency of vibration of a vibrotac- fact discovered by Weber in the 1830s, who found
tile stimulator on the skin much as the threshold of that two-touch stimuli (such as the points of a
hearing depends on the frequency of a sound wave. drawing compass) will be felt as a single touch if
Also, similar to responses of the ear, the skin is they are close enough together. The two-point
sensitive only to a limited range of vibration fre¬ threshold is a measure of how far apart the stimuli
quencies. The range most investigators agree on is must be before they are felt as two separately lo-
from about 40 to about 2500 Hz. However, some calizable touches. It was during investigations of
researchers have claimed that under special condi¬ this kind that Bekesy (1967) discovered inhibitory
tions sensations can result from stimuli of frequen¬ interaction in the skin (see Demonstration Box
248 Chapter 9/Touch and Pain

9-1). A comprehensive determination of two-point tion from it. Zigler (1932) measured touch adapta¬
thresholds was provided by Weinstein (1968), and tion using several different areas of the body. He
a summary of his results is shown in Figure 9-6. reported that the heavier the stimulus, the longer it
Notice the remarkable differential sensitivity of the took for the sensation to disappear, but that the
lower face and the hands, and even the feet. Pre¬ larger the area covered by the stimulus, the less time
sumably this reflects the use of these areas in the it took for the sensation to disappear. You can dem¬
manipulation of objects. The high sensitivity of the onstrate this result using Demonstration Box 9-4.
feet may be a leftover from our primate ancestors, Another technique for measuring adaptation is
who could manipulate objects with their feet! An¬ to present a stimulus for some period of time and
other interesting aspect of touch localization is then introduce a second stimulus. The observer is
studied in Demonstration Box 9-3. asked to adjust the intensity of the second stimulus
until the sensation associated with it matches that
associated with the first. Since adaptation has re¬
Touch Adaptation
duced the sensation intensity of the first stimulus,
Touch sensations adapt, as do all other sensations. the difference between the magnitudes of the two
This can be shown by simply applying a stimulus to stimuli is a measure of the amount of adaptation
the skin and observing the gradual disappearance of that has taken place. This technique assumes that
the sensation. For example, when you first get the two stimuli would be equal in sensation mag¬
dressed in the morning you may be (uncomfortably) nitude if they were equal in intensity. Using this
aware of your belt or waistband, but after a while technique, Frey and Goldman (1915) determined
you no longer are consciously aware of the sensa¬ that adaptation to touch stimuli is similar to that for

FIGURE 9-6 Average two-point thresholds for different regions of the male skin. Relative values for
females are similar but lower overall. (From S. Weinstein, in D. R. Kenshalo, ed., The Skin Senses, 1968. Courtesy of Charles
C. Thomas, Publisher, Springfield, Illinois.)
 Touch 249

DEMONSTRATION BOX 9-3. Touch Localization

Bekesy has studied localization in all the major sen¬ stimulated approximately equally. Now use the same
sory modalities and has discovered similar phenom¬ finger to touch, with the same light, quick touches,
ena in all of them. One of the most striking is the your little toe or your ankle. Most people now report
discovery that a touch can be localized as being out¬ that the sensation seems to be located mostly in the
side of the body under the appropriate circumstances finger, rather than in the toe or the ankle, even though
(Bekesy, 1967). We will demonstrate a somewhat both are being equally stimulated. As it turns out, it
simpler phenomenon that apparently depends on the takes somewhat more than 1 msec longer for the
difference in arrival times of neural impulses from nerve impulses to travel from the fingertip to the brain
different parts of the skin surface to the primary sen¬ than for them to travel from the lip to the brain. Sim¬
sory areas of the brain. ilarly, it takes more than 1 msec longer for impulses
Touch your two index fingers together. Try to con¬ to travel from the foot to the brain than from the fin¬
centrate on experiencing where the sensations of ger to the brain. The impulses that arrive at the brain
touch are felt, that is, on which of the two fingers. first (providing the difference is more than 1 msec)
Most people report sensations of about equal intensity seem to dictate where the sensation will be experi¬
from both fingertips. Now touch your fingertip (either enced, even though the two places on the skin are
one) repeatedly to your lower lip with light, quick being stimulated equally. The various other parts of
touches. When asked to say where the sensation is, the body fall in between these extremes, but in all
most people report that they feel it mostly on the lip cases the localization depends on the relative lengths
and little or not at all on the fingertip, even though of the pathways from the touching parts to the brain.
both are of about equal sensitivity and are being

other modalities. The adaptation is very rapid for tion for static stimuli. Again, different parts of the
the first second or so and then gradually slows body respond differently. On the lip adaptation is
down. After 3 sec, the sensation level has de¬ complete after about 20 sec; for the forearm, how¬
creased to about 1/4 of the beginning value. ever, loss of sensation is more gradual, and adap¬
Bekesy (1959) used this same technique to tation is not completed even after 60 sec. These
measure the time course of adaptation for vibratory longer adaptation times are consistent with the
stimuli, which generally takes longer than adapta¬ greater effectiveness of the vibratory stimulus.

DEMONSTRATION BOX 9-4. Touch Adaptation

For this demonstration, you will need a watch with a peat this with the other piece of cardboard. Try the
sweep second hand, two pieces of cardboard (cut into experiment again, only this time press gently on each
small circles with diameters of about 1 cm and about cardboard. Notice that the lighter touches and the
4 cm), and a friend. Lay one piece of cardboard on larger surface area stimulations disappear faster from
the skin of your friend’s back and record the amount consciousness. Thus, they show faster adaptation.
of time before the sensation of touch disappears. Re¬
250 Chapter 9/Touch and Pain

Touch Intensity
When measuring the subjective intensity of touch
stimuli, vibrating stimuli are often used because, as
we just noted above, they do not adapt as quickly.
A study by Stevens (1959) demonstrated that mag¬
nitude estimations of the intensity of a 60-Hz vibra¬
tory stimulus on the fingertip followed the standard
psychophysical power function. The apparent in¬
tensity increased with increasing physical intensity
with an exponent of about 0.95, suggesting that
there was almost a one-to-one relationship between
the sensation magnitude and the magnitude of the
stimulus. More recent measurements using single
mechanical pulses applied to the skin of the hand
found that magnitude estimates produce similar re¬
sults for both hairy and glabrous skin, but that ex¬ FIGURE 9-7 Equal-sensation contours for a
ponents were somewhat lower for glabrous (about vibrating stimulus. These contours are quite
0.70) than for hairy (about 1.05) skin (Hamalainen similar to equal-loudness contours over the same
range of frequencies (see Figure 7-12). (From verrillo,
& Jarvilehto, 1981).
Fraioli & Smith, 1969)
We mentioned earlier that the action of a vi¬
bratory stimulus on the skin is quite similar to that
of sound on the ear in that sensitivity is greatest for
tends to respond best to different aspects of stimu¬
certain stimulus frequencies (see also e.g., Ges-
lus patterns than, for example, the visual system
cheider & Verrillo, 1982; Marks, 1979a). This
does (Klatzky, Lederman & Reed, 1987). Most of
means that for a given physical pressure, some vi¬
you have probably heard about Louis Braille’s tac¬
bration frequencies give a more intense touch sen¬
tile pattern alphabet, by which blind people can
sation than others. This can be seen in Figure 9-7,
read any suitably translated book or paper. In this
which shows a set of equal-sensation curves for a
alphabet, patterns of raised dots on paper play the
vibrating stimulus on the skin that are highly simi¬
role of the patterns of ink on paper that sighted
lar to the equal-loudness contours presented in Fig¬
people use for letters. The speed with which an ex¬
ure 7-12. Here, maximum sensitivity seems to be
perienced blind person can read with this alphabet
in the region of 200-400 Hz, with sensitivity de¬
is a testimony not only to long hours of practice (as
creasing dramatically as frequency declines. Notice
is any form of reading) but also to the remarkable
that stimulus intensity in Figure 9-7 is measured in
sensitivity of the touch system. Of course, the final
decibels (compared to a displacement of the vibra¬
interpretation of these patterns of touch stimuli in¬
tor surface by one millionth of a meter, which then
volves a number of complicated cognitive pro¬
represents 0 dB). This makes the vertical scale a
cesses (Krueger, 1982).
logarithmic scale similar to the logarithmic decibel
Braille is not the only method by which pat¬
scale used in hearing (see Chapter 6).
tern information is used to convey information
through touch. Alternate methods have been in¬
Tactile Pattern Perception
vented by those who wanted to more directly
The sense of touch has the ability to discriminate substitute patterns of touch for those of vision.
and recognize complex stimulus patterns (see e.g., White, Saunders, Scadden, Bach-y-Rita, and Col¬
Klatzky, Lederman & Metzger, 1985), although it lins (1970) developed a vision substitution system
 Touch 251

in which a television camera is used to scan a vi¬ difficult to do with newspapers and magazines,
sual pattern. The information gathered by the tele¬ which are published in large numbers. The Opta-
vision camera is then converted into a pattern of con (Bliss, Katcher, Rogers & Shepard, 1970)
vibrating points on the skin of the back of an ob¬ works similarly to the vision substitution system
server (see Figure 9-8). The observer can move the except that the size of the visual field scanned is
camera to view different parts of the visual scene. about 1 printed letter, and the pattern of vibrations
When visual stimuli are presented tactually in this corresponding to each letter is formed on the fin¬
way, observers can identify a wide variety of dif¬ gertip instead of the back. After approximately 50
ferent stimulus patterns (up to 25). Remarkably, hours of training, blind users can read untreated
the relative distance of several visual objects in a material at about 20 words per minute. Experienced
scene can be perceived from the tactile pattern, and users attain rates as high as 60 words per minute.
even visual illusions can be experienced. These In 1977, nearly 2,000 blind people were using the
types of findings raise questions similar to those Optacon for reading (Craig, 1977).
raised by visual and auditory pattern perception; for The Optacon and similar devices are now
example, we might ask if there are also feature de¬ being used for research into the mechanisms of tac¬
tectors for touch. The findings also point to the im¬ tile pattern perception (see e.g., Craig, 1981,
portance of higher-level cognitive processes in even 1983b; Loomis, 1981; Schneider, Hughes, Epstein
the simplest types of perceptual experiences. & Bach-y-Rita, 1986). One particularly active area
Reading in Braille requires that the paper or involves the interaction between successively pre¬
book be translated into Braille. This is especially sented patterns. If the presentation of one pattern

FIGURE 9-8 The vision substitution system. (From White, Saunders, Scadden, Bach-y-Rita & Collins, 1970)
252 Chapter 9/Touch and Pain

on the skin makes it more difficult to identify an¬ duction of the overheated sweat to the surface of
other pattern we refer to this as masking (see Craig, the skin where it can radiate more efficiently and
1978). For example, if a letter pattern is presented by the cooling of the skin surface through evapo¬
on the Optacon and then followed immediately by ration. When we are too cold, the blood vessels in
a rectangular pattern, observers have a harder time the skin contract, thereby slowing heat loss, and we
identifying the letter than if no masking pattern is begin to shiver, which generates more heat from
presented. This is called backward masking, since our muscles. The thermal sensitivity of the skin
the masking pattern seems to act backward in time plays a major part in this complex, mostly reflex-
to interfere with the perception of the earlier target. operated temperature-regulating system that keeps
Forward masking also occurs with tactile patterns; our internal body temperatures around 37 deg C
here, the masking pattern is presented first, fol¬ (98.6 deg F) (see Hensel, 1981). Usually the sys¬
lowed by the target. As in audition and vision (see tem functions so well that we do not notice any
Chapters 7 and 11), masking studies often reveal temperature sensations, at least in temperate envi¬
basic mechanisms of tactile perception. For exam¬ ronments. The ability of this system to regulate
ple, there is usually more backward masking when body temperature is limited, however, and when
the time interval between target and mask is short the limits are exceeded the body needs to take more
and more forward masking when the time interval dramatic steps, such as changing clothing or start¬
is longer (Craig, 1983a). This is consistent with the ing a fire. Such actions bring the temperature of the
idea that the perceptual representations of tactile skin’s environment within its safe limits again.
features persist for about 1,200 msec (Craig & Such necessities are signaled by the conscious sen¬
Evans, 1987; Evans & Craig, 1986), and reveals sations of warmth and cold.
something about how those representations are in¬
tegrated over time (Evans, 1987). These masking
Neural Coding of Temperature
effects do not depend very much on where on the
body the patterns are presented, although pattern Temperature seems to be sensed in the skin by
discrimination and identification in the absence of nerve endings that are quite similar in appearance
masking does vary with location on the body (Cho- to those sensitive to mechanical contacts. Two
lewiak & Craig, 1984). types of nerve fibers respond when the skin is
cooled or warmed. Cold fibers respond to cooling
of the skin with an increase in firing relative to
WARMTH AND COLD their resting rate, and to warming with a firing de¬
crease. Warm fibers respond to warming with an
Are you cold right now? Are you warm? Probably increase in firing rate and to cooling with decreased
you are feeling rather neutral, that is, comfortably firing (see Hensel, 1981). Also, after an initial
unaware of any temperature sensations. Our bodies change in firing rate in the direction appropriate to
contain a remarkable system of thermal sensors that the type of stimulus and the type of receptor, both
trigger reflexes regulating the flow of blood in the types of fibers gradually adopt a steady rate of fir¬
blood vessels in our skin, the activity of the sweat ing. This “resting” rate is related to the absolute
glands, and the tiny muscles located around the temperature of the skin (and thus of the receptor).
roots of hairs in the skin. When our internal body Cold and warm fibers have different patterns
temperature is too high (above 37 deg C), the blood of response over a broad range of skin tempera¬
vessels of the skin dilate, allowing more blood to tures. This is shown in Figure 9-9, which is from
flow and thus radiating more heat into the air. We the work of Zotterman (1959). As you can see in
also begin to sweat, losing heat both by the con¬ the figure, the cold fibers respond in the range from
 Warmth and Cold 253

FIGURE 9-9 Steady-state discharge of cold and warm fibers in the cat as it varies with skin
temperature. (Based on Zotterman, 1959)

about 13 to 35 deg C (55 to 95 deg F) and again means a stimulus has caused the skin temperature
from about 46 to 50 deg C (113 to 122 deg F). to drop below physiological zero. This reference
Above this limit the receptors become damaged and temperature is a floating neutral point, based on the
the response ceases. Notice that the cold fibers temperature to which the thermal receptors in the
have two response peaks, one at about 25 deg C skin have adapted. There is a neutral zone around
(77 F) and another at about 50 deg C (122 deg F). physiological zero, within which no sensation will
Warm fibers respond from about 23 to 46 deg C be felt if a stimulus within that temperature range
(73 to 117 deg F) with a peak at about 38 deg C is applied to the skin. This zone is seldom more
(100 deg F—just above body temperature). These than a couple of degrees on either side of physio¬
steady-state responses provide us with information logical zero, but it varies in width depending on
about the absolute temperature of the skin. This in¬ what the physiological zero is, where on the body
formation is quite important, since internal body the change occurs, and what kind of stimulus is
temperature must be maintained within a narrow applied. A good average value for physiological
range of values around 37 deg C (98.6 deg F) (see zero in a temperate environment (room tempera¬
Hensel, 1981). Demonstration Box 9-5 allows you ture) would be around 33 deg C (91 deg F). At
to experience the consequences of having both cold room temperature such a skin temperature main¬
and warm fibers stimulated optimally, something tains the normal internal body temperature of about
that wouldn’t happen under normal conditions. 37 deg C.
The measurement of absolute thresholds for
warmth and cold sensations is complicated by the
Thermal Thresholds and Adaptation
fact that relatively complete adaptation to thermal
Because the skin works to maintain a constant in¬ stimuli takes place over a range of temperatures,
ternal body temperature, sensations of warmth or and that thermal sensations are relative to the tem¬
cold are generally caused by departures from a par¬ perature to which the skin has become adapted.
ticular reference skin temperature called physiolog¬ Several techniques have been used to measure ad¬
ical zero. Thus, to talk about feeling “cold” aptation. Typically, an area of the skin is exposed
254 Chapter 9/Touch and Pain

DEMONSTRATION BOX 9-5. The Heat Grill

For this demonstration you will need two pipe clean¬ you will feel no coolness. Observers usually get a
ers bent as shown in Figure A. Be careful to bend the sensation of an intense stinging heat. Some people
pipe cleaners so that they fit together closely when may find the heat sensation sufficiently intense to
both are laid on a table, as shown in Figure B. Place cause them to withdraw their arms. The temperatures
one pipe cleaner in a glass of cool water and the other of the pipe cleaners are such (if you followed direc¬
in a glass of very warm (not unpleasantly hot) water. tions faithfully) that receptors of both cold and warm
Take the pipe cleaner out of the glass of cool water, fibers in the same general area of the skin are being
and place it on a flat surface. Working quickly, take stimulated near their optimum. Usually this does not
the pipe cleaner out of the glass of warm water and happen. When it does, you mislead your brain by
arrange it to form the configuration shown as Figure making both the cold and the warm fibers fire near
B. As soon as this is done, place your forearm over their maxima. The brain processes this information as
the set of pipe cleaners and press down as shown in emanating from a single, very hot stimulus. (Note in
Figure C. The temperature sensation you receive will Figure 9-9 that for very hot stimuli, the cold fibers
probably be quite surprising. Although the stimulus would be firing vigorously.)
consists of alternately cool and warm surfaces,

Pipe cleaner 1
 Warmth and Cold 255

to very high or low temperatures for some period Overall, the sensation of warmth or cold
of time (usually by dipping it in water at the re¬ caused by stimuli with temperatures near 33 deg C
quired temperature), and then a threshold for is determined largely by the temperature to which
warmth or cold is determined. In a different tech¬ the skin has been adapted before the stimulus is
nique, observers are exposed to a particular thermal applied. You can experience this in Demonstration
stimulus and asked to report when they no longer Box 9-6. Absolute thresholds for these sensations
feel any thermal sensation at that temperature. For can be defined as the amount of temperature change
instance, Kenshalo and Scott (1966) had observers (from the adapting temperature) necessary to cause
change the temperature of a sophisticated thermal an experience or a report of warmth or cold. Ab¬
stimulator just enough to maintain a detectable sen¬ solute thresholds differ for upward and downward
sation, while they adapted the skin to a given ther¬ changes in temperature depending on the adapting
mal level. The stimulator started at the previously temperature (Kenshalo, Nafe & Brooks, 1961).
measured temperature of the observer’s skin. Ad¬ Sensitivity to lowering of temperature is greater
justments were made every minute at first, and then when skin temperature is colder, but it is greater
every 5 min for up to 40 min. Figure 9-10 shows for rises in temperature when skin temperature is
the results obtained for four observers. Using this warmer, as can be seen in Figure 9-11. Thus, at
technique, Kenshalo and Scott (1966) found that relatively high or low skin temperatures we are
complete adaptation occurred over a range of about more sensitive to fluctuations in temperature, es¬
4 to 8 deg C centered around the average skin pecially those changes that indicate greater devia¬
temperature. This then serves as an experimental tion from our normal body temperature. This
measurement of the neutral zone around the phys¬ makes sense, since very low or very high temper¬
iological zero set by adaptation. Notice that when atures can be dangerous for the body, hence the
adaptation has not fully occurred (early in the pe¬ detection of thermal changes is more important
riod), the neutral zone is actually quite narrow when the stimuli are more extreme. The minimum
(perhaps about 2 deg), but it widens as the adapta¬ threshold values obtained are about 0.1 deg C of
tion becomes more complete. temperature change from the adapting temperature.
As in other aspects of touch, the place on the
body surface to which the stimulus is applied also
affects our sensitivity to warmth and cold. The
head is the most sensitive to warm stimuli, with the
limbs least and the trunk intermediate. Conversely,
the trunk, particularly the back, is the most sensi¬
tive to cold stimuli, with the limbs intermediate and
the head least (J. C. Stevens, 1979). This is easy
to demonstrate for yourself using pieces of metal
that have been dipped in water of different temper¬
atures (pay attention to the intensity of the sensa¬
tions aroused by the stimuli on the various parts of
FIGURE 9-10 Adaptation to thermal stimuli. your body). The mouth seems to be more sensitive
Each curve represents the average amount of to increases in temperature than to decreases
adjustment by four observers to maintain a just (Green, 1986).
noticeable warm or cool sensation with a thermal
There are significant differences in sensitivity
stimulator. Thermal changes closer to the 0 line
than the curves are felt as having neutral to warmth and cold even over a small patch of
temperature. (Based on Kenshalo & Scott, 1966) skin. If the thermal stimulus has a small area
256 Chapter 9/Touch and Pain

DEMONSTRATION BOX 9-6. The Dependence of Thermal Sensation on Physiological

Zero

For this demonstration, you will need to create three (as it should have relative to your average skin tem¬
water baths of different temperatures. You can use a perature of about 33 deg C) and the warmer one felt
meat thermometer to measure the temperatures or try warm. Now prepare the 35 deg C bath (quickly, if
the following formulas. you are doing it by yourself), and plunge both hands
Assuming that your tap water has temperatures of into it at once. The hand that was in the 30 deg C
10 deg C (cold) and 60 deg C (hot), you can make a bath should now feel as if it is in warm water while
30 deg C bath by adding 3% cups of cold tap water the other hand (40 deg C) will feel as if it is in cool
and 21/3 cups of hot tap water to a bowl. A 35 deg C water. Since the water is actually the same for both
bath requires 3 cups each of hot and cold tap water. hands, but their physiological zero has been changed
For a 40 deg C bath, use 214 cups of cold water and by the previous 5 min adaptation period, the sensa¬
3% cups of hot water. It is best to have a friend help tions of warmth and cold must be caused by the re¬
with preparing the water baths and with keeping them lationship of the stimulus temperature to the current
at a nearly constant temperature (by adding a little hot temperature of the skin (physiological zero). You can
water every minute or so). Your friend can then try experiment with this phenomenon further by trying
the demonstration while you make the baths. out more extreme adapting temperatures and adapting
First prepare the 30 deg and the 40 deg C baths. to them for longer times. See if you can find points
Place one hand in the 30 deg C and one in the 40 deg beyond which all stimuli feel either warm or cold.
C bath. Keep the hands in the baths for about 5 min. Changes in temperature near these points only result
Under these conditions most people report that after 5 in changes in the intensity of that sensation, rather
min there is no longer any sensation of warmth or than resulting in a change in sensation quality (Ken-
cold, although the cooler bath initially felt cold shalo & Scott, 1966).

(perhaps the size of a pin head), it is possible to

find some spots that yield only sensations of
warmth and others that produce only sensations of
cold. There are also regions that produce no tem¬
perature sensations at all (although they may re¬
spond to touch). Figure 9-12 shows a set of maps
of such warm and cold spots from an area of 1 cm2
on the skin of the upper arm (Dallenbach, 1927).
The spots were mapped on four successive days so
that the permanence of the spots could be deter¬
mined. As you can see from the figure, the spots
tend to be in the same places from day to day.
Temperature to which skin is adapted
Since all these spots are innervated only by free
FIGURE 9-11 Sensitivity to increases and nerve endings, the difference between the cold and
decreases in temperature varies as a function of warm spots probably has to do with the type of
the adaptation temperature of the skin. nerve fiber generating these nerve endings. Cold fi-
 Warmth and Cold 257

Cold

•• •• • • • •
• • • •• • • •
• • •
•
•• • • •• • • • • • • •
•• • • • • •
• ••
••
•
• •
•• • • • • •
Day 1234

Warm

FIGURE 9-12 Maps of cold and warm spots on a 1 cm2 area of the skin of the upper arm of a single
observer on four successive days. Notice how the spots tend to be in the same places from day to day.
(From K. M. DaUenbach, American Journal of Psychology, 1927, 39. Copyright by the University of Illinois Press.)

bers are typically larger in diameter and myelinated lips and tongue, although the exponents differ un¬
whereas warm fibers are smaller and unmyelinated der different conditions and locations, especially
(Hensel, 1981). for cold (Green, 1984). In general, however, the
sensation of warmth grows somewhat more rapidly
with stimulus intensity than does that of cold.
Thermal Intensity and Qualities
Because of the close relationship between the
Because of the complexity of the situations under sensations of warmth and cold, it should not sur¬
which warmth and cold sensations occur, it would prise you to learn of some very interesting and
seem to be difficult to measure the apparent inten¬ somewhat paradoxical phenomena associated with
sity of thermal stimuli. We can, however, measure thermal stimuli. One of these, called paradoxical
these sensations using direct scaling techniques cold, was discovered by Max von Frey in 1895.
such as those discussed in Chapter 2. For instance, He found that if cold spots (such as those illustrated
Stevens and Stevens (1960) had observers make in Figure 9-12) are touched with a very warm stim¬
magnitude estimations of the thermal sensations ulus above 45 deg C (113 deg F), a sensation of
caused by the application of warm or cool pieces cold will result. The opposite phenomenon, para¬
of aluminum to the skin of the forearm. They used doxical warmth, has been much sought after but
stimuli above the average skin temperature (about never convincingly demonstrated. This asymmetry
33 deg C or 91 deg F) for warmth sensations, and is consistent with the tendency of nerve fibers that
stimuli below this temperature for cold ones. They respond to cooling (cold fibers) to respond both in
found that warmth and cold sensations also follow the range of temperatures between about 12 and 35
the psychophysical power law when measured in deg C (54-95 deg F) and in that between 46 and
this way, with exponents of the power functions for 50 deg C (113-122 deg F). Warm fibers respond
warmth around 1.6 and for cold about 1.0. Similar only in the range between about 25 and 46 deg C
power functions have been found for the skin of the (77-117 deg F) as shown in Figure 9-9. When the
258 Chapter 9/Touch and Pain

skin temperature rises above 46 deg C (117 deg F), formation is generated by our continual battle
only the cold fibers respond. Therefore, very high against gravity. Both movement and postural re¬
temperatures actually result in neural activity sponses involve tension, compression, or twisting
equivalent to that caused by very cold tempera¬ forces on the muscles, tendons, or joints of limbs.
tures, and both are felt subjectively as cold. These physical forces should be considered the
stimuli for kinesthesis. Any position of the body,
even supine and fully relaxed, results in a complex
KINESTHESIS pattern of muscular tensions and compressions, and
consequent mechanical forces acting on tendons
We have been talking about receptors that respond and joints. The relative intensities of the various
to information generated by the world around us forces, or changes in those intensities over time,
when contact is made with our skin. There is also signal body movement and posture.
a vast amount of mechanical information generated Just as in the surface layers of the skin, a great
within our own bodies. This information indicates many other touch and stretch receptors are scattered
whether we are moving or stationary (see also throughout the body. First, at least two types of
“Motion” section of Chapter 13), and informs us nerve endings exist in the deeper layers of tissue
of the position and movement of our body parts. beneath the skin: free nerve endings and Pacinian
The neural processing of this information, and the corpuscles. The free nerve endings are thought to
sensations we feel, called collectively kinesthesis, be responsible for pain sensations. The Pacinian
bear striking resemblances to touch. corpuscles provide our sense of deep pressure
One of the things distinguishing animal from (which can be felt even when the overlying skin
plant life is the ability to move about in the world. has been anesthetized). Although the visceral or¬
In higher organisms, specialized receptor systems gans themselves are rather insensitive to touch,
inform the brain about the position of the limbs or temperature, or pain stimuli (with the obvious ex¬
the orientation of the body. The bodies of such or¬ ception of stretching or twisting forces that cause,
ganisms (including humans) are literally enmeshed for example, gas pains), these organs are sur¬
in a web of sensory receptors, which accurately rounded by muscle. This muscle is supplied with a
monitor the positions of various parts so that appro¬ variety of nerve endings that are responsive to the
priate action can be initiated. In many cases the movements of the viscera. Finally, several types of
signals of these sensory systems are not con¬ nerve endings are located in and around our joints
sciously perceived, but rather used in controlling in the muscles that move our limbs.
reflex actions that maintain an upright posture. Matthews (1933) divided the receptors in the
When these signals are perceived, they give rise to muscles into three major types, two of which are
the sensations of force or weight, which are often shown in Figure 9-13. His first type is called the A
used to help guide our voluntary movements, as in endings, which have two subtypes. The Ax endings
sports or other skilled motor performances. are often called “flower-spray” endings because
they look like a bouquet of flowers against the
muscles where they synapse. Endings of the A2
Kinesthetic Stimuli and Receptors
type are wound around strands of muscle fibers
The overt physical stimulus to which the kines¬ called muscle spindles. Both subtypes respond to
thetic system responds is movement (the root kine stretching of the muscle and therefore have been
is from the Greek word for “movement”). Some called stretch receptors or spindle organs, since
information about position is available, however, they both attach to muscle spindles. This type of
even when no movement is taking place. This in¬ nerve ending is found also in great numbers in the
 Kinesthesis 259

Golgi tendon organs. They seem to respond to

both stretching and contracting of the muscle,
whereas the spindle organs respond only to stretch¬
ing. Matthews’s third type of receptor, which he
labeled C, is thought to be the Pacinian corpuscle
(Geldard, 1972). Although it is important in the cu¬
taneous senses, it probably plays little role in ki¬
nesthesis. There seem to be few such Pacinian
corpuscles in muscles or near joints. In addition to
all of the above, muscles and joints are well-
supplied with free nerve endings that may be re¬
sponsible for pain sensations that occur in these
areas.

Neural Responses in Kinesthesis

The various receptors send their messages to the

brain via the two major neural pathways described
previously for touch: the medial lemniscus and the
spinothalamic pathways. Most of the fibers follow
the medial lemniscus, except for the free nerve
endings, many of which follow a spinothalamic
pathway. There are also a great many branchings
and interactions of these two pathways with others
of less importance. The nerves that terminate on
muscles or tendons and in joints also project to the
somatosensory cortex. Thus, information about
stimuli touching the skin of the arm and about the
FIGURE 9-13 Nerve endings in muscle. (1) Free
position and movement of the arm are both pro¬
nerve endings around an artery (ART); (2) efferent
nerves that regulate the size of the artery; (3) jected onto the same general area of the cortex.
motor nerves (MN) that cause the muscle to However, the cutaneous information is kept sepa¬
contract; (4) more motor nerves; (5) Matthews type rate from the position and movement information
A2, or annulospiral (AS), nerve endings on a even at the cortical level (see Vierck, 1978).
muscle spindle; (6) Matthews type Ai; or flower-
Let us look at the signals generated by these
spray (FS), nerve endings on the same muscle
spindle; (7) Matthews type B, or Golgi tendon kinesthetic receptors. Figure 9-14 shows an elec-
organs (TO), nerve endings on a muscle tendon trophysiological recording from a nerve fiber that
(MT). Matthews type C endings, or Pacinian terminates in the knee joint of a cat. The limb was
corpuscles, are not shown. (From Creed, Denny-Brown, repeatedly bent and returned to its original position.
Eccles, Liddell &. Sherrington, 1932)
The movement of the limb is signaled by a sudden
change in neural response rate, the size of the
joints between limbs. Matthews’s second type, the change indicating the speed of movement. The
B endings, look much like the flower-spray endings static position of the limb is signaled by the resting
but are attached to the tendons that connect the level of neural response. These are slow-adapting
muscles to the bones. B endings are also called fibers because after a limb movement it takes sev-
260 Chapter 9/Touch and Pain

may serve to give us the sensation of strain when

lifting something, or of pain. Their major role
seems to be to provide information to control our
postural reflexes, automatically adjusting our mus¬
cle tension to the requirements of whatever load we
are carrying.

Perception of Weight and Force

The sensations of weight and force can be de¬

scribed psychophysically much as other sensations
have been (see Jones, 1986, for a review). In fact,
the discrimination of lifted weights was the first
and the most generally used technique in the early
days of psychophysics. Weber was the first to study
it experimentally, and Fechner did an enormous
FIGURE 9-14 The response of a single nerve number of experiments testing Weber’s law with
fiber that terminates in the knee joint of a cat. lifted weights (see Chapter 2). Weight discrimina¬
The dotted lines represent bendings of the knee tion is about as good as discrimination of sound
joint by 10 degrees. The slow-adapting fiber gives a intensity or light intensity, with a typical Weber
large initial burst of impulses to the bending,
fraction of about 0.08 (meaning that an 8% change
which then gradually declines to a stable rate of
firing above background (no bending). When the in stimulus magnitude can just be detected).
limb is “unbent,” there is an inhibition of Force and weight can be scaled by direct scal¬
responses, so that nerve firing is below ing methods. Judgments are affected by many fac¬
background rate, and then firing climbs back to tors, for example the size-weight illusion in which
the background rate. (From Boyd & Roberts, 1953)
a larger-sized object feels lighter than a smaller
object of the same weight. Physiological factors
eral seconds for them to reach a new resting rate of also play a role, such as fatigue, which makes
firing corresponding to a new limb position. Cells everything feel heavier. Because of this there is
with similar response characteristics are also found disagreement about the exact form of the psycho¬
higher up in the kinesthetic system at sites in the physical function. Sensations of force or weight
thalamus and cortex (Mountcastle, Poggio & Wer¬ seem to increase as a power function of stimulus
ner, 1963; Mountcastle & Powell, 1959). In addi¬ intensity, but differently for lifted weights with a
tion, more rapidly adapting fibers and “postural moderate range (exponents near 1.0) versus grip or
neurons” (which respond only when a joint is in a squeezing force generated on a dynamometer (in
particular position) have been found in monkey so¬ which changes are more easily sensed, with expo¬
matosensory cortex, along with neurons that act nents of 1.5-2.0).
much like feature detectors for the different limb Gravity also affects the perception of weight.
components of particular postures (Costanzo & During parabolic flight, which provides periods of
Gardner, 1981; Gardner & Costanzo, 1981). Vir¬ 0 gravity at the apex and about 1.8 gravity at the
tually all positional and movement information nadir, magnitude estimates of weights were lowest
comes from the receptors located in the joints during 0 gravity, next lowest during 1.0 (ordinary)
(Adams, 1977). The other kinesthetic receptors gravity, and highest during 1.8 gravity (Ross &
 Pain 261

Reschke, 1982). Interestingly, the value of the We¬ succumbing to an assault it was never meant to
ber fraction approximately doubled under condi¬ experience.
tions of prolonged weightlessness (or 0 gravity) in Pain is a complex experience. It is usually as¬
the U.S. Spacelab (Ross, Brodie & Benson, 1984). sociated with damage to the body of an animal,
Actually, of course, everything weighs nothing un¬ and, in humans, it is usually accompanied by myr¬
der such conditions, but objects still have mass iad emotions and thoughts. Some sensory psychol¬
and, therefore, momentum. The Weber fraction ob¬ ogists consider it to be a sensation in its own right;
tained in 0 gravity was really a Weber fraction for other psychologists argue that pain is not a sensa¬
mass, obtained by having the astronauts shake the tion at all, but rather an emotion, or even a bodily
stimulus objects in order to generate forces that state akin to hunger or thirst (e.g.. Wall, 1979).
could be perceived. Perhaps the best view is a compromise. There cer¬
You might try demonstrating some of these ef¬ tainly seem to be identifiable sensory characteristics
fects for yourself by repeating Demonstration Box to the experience of pain. Pain apparently has ab¬
2-2 while riding in an elevator. The starting and solute and differential thresholds, it adapts, it may
stopping of an elevator, especially one of the fast have definable and separate physiological path¬
ones now in use in tall buildings, also generates ways, and it can be measured in a variety of ways.
short periods of lessened and greater than normal Thus, pain acts much like a separate sensory sys¬
gravity. As you ride the elevator, also pay attention tem, and we will therefore treat it as a separate
to the magnitude of the sensation of weight of the sensory modality. We must be aware, however,
objects you are trying to discriminate. Do they feel that interactions with other sensations and with
heavier as the elevator comes to a stop after a drop more complex cognitive and emotional processes
from the 10th to the 1st floor? Did they feel lighter are important for a complete understanding of pain
as the drop began? phenomena.

Pain Stimuli and Receptors

PAIN
The evolutionary significance of pain may be two¬
The shrill squeal of the siren seemed to pierce his fold. First, it is essential that an animal be able to
head like a knife. The pain was exquisite. Then respond appropriately to environmental situations
came a brilliant flash of light; he blinked, trying to that could destroy it. It must respond by avoiding
somehow relieve the savage pain flooding in such situations in the future or terminating the on¬
through his eyes. He stumbled, barking his shin on going dangerous stimulus. Light, sound, touch, and
a log, and another bright, fierce pain penetrated his temperature, when they occur at very high intensi¬
consciousness. On top of all this, the old World ties or for prolonged durations, can destroy the re¬
War II shrapnel wound in his hip began to throb ceptors that are specialized to receive them. If such
with a dull, sickening ache. Finally, he found the potentially damaging intensities are not signaled
water, diving into a dark, cool world that promised quickly to the brain, the organism will be damaged
to soothe his battered body. But something was beyond repair. This is often the unfortunate fate of
wrong with the water; instead of cooling it was those humans who are bom without a well¬
burning. His head seemed to explode as the caustic functioning pain sense (see Stembach, 1963). Sec¬
liquid burned its way along his nasal passages and ond, pain seems to have the effect of requiring a
forced its way between his lips. Finally, and person to cope appropriately with an injury if it
almost gratefully, he lost consciousness, his body does happen. Many people who receive serious in-
262 Chapter 9/Touch and Pain

juries do not feel pain until quite a while, some¬ away from the stimulated site on the body. There
times several hours, after they occur (see Melzack, are at least three major classes of such nerve fibers.
Wall & Ty, 1982). Their pain seems to have the Large, myelinated fibers, called A)3 fibers, seem to
function of inducing them to be still in order that respond especially well to light touch stimuli.
healing may occur, or to seek treatment for the in¬ These fibers conduct nerve impulses at high speed
jury. When an injury first occurs, more important (as fast as any in the nervous system, at least 40
responses than pain may be appropriate, for exam¬ m/sec) and connect with both the medial lemniscus
ple escape or fighting for life. In this view, the bi¬ and the spinothalamic pathways. Smaller myeli¬
ological significance of pain lies in its ability to nated fibers (A6) and small, unmyelinated fibers
promote healing (see Wall, 1979). During the re¬ (C) are much slower conducting (about 5-20 m/sec
covery phase, the stimulus for pain is the injury for A5 and less than 2.5 m/sec for C), have higher
itself, and the function of the pain is not to warn thresholds, and respond only to noxious stimuli
but to promote recovery. such as pinches, pinpricks, or extreme temperatures
All pain does not arise from overstimulation (see Willis, 1985). The A8 fibers seem especially
or serious injuries. Certain kinds of painful experi¬ sensitive to noxious mechanical stimuli, whereas
ences come from only moderately intense stimula¬ the C fibers respond to all kinds of noxious stimuli
tion, such as a pinprick or salt touching an open (they are often called polymodal nociceptors).
wound. Conversely, overstimulation can sometimes These fibers typically terminate in free nerve end¬
occur without eliciting pain. For instance, pain is ings, either in the subcutaneous fat or epidermis of
not experienced as one increases the concentration the skin (as mentioned earlier) or deep in muscles
of sugar stimulating the tongue. At present it is dif¬ and joints, and the A(3 fibers typically terminate in
ficult to say exactly what stimulus properties are corpuscular endings. A 8 and C fibers connect
responsible for the experience of pain. Intense mainly with the spinothalamic pathways, and are
stimulation and tissue damage are certainly only now generally acknowledged to be “the” pain fi¬
part of the story. bers (see Figure 9-3). Apparently the fast fibers
The best candidates for pain receptors are the and pathways (e.g., A(3 fibers and the medial lem¬
free nerve endings with which the skin and the rest niscus) are specialized for highly discriminative,
of the body are particularly well supplied. As we complicated processing of information, whereas the
mentioned earlier in this chapter when discussing slower fibers and pathways (e.g., A8 and C fibers
the skin senses, free nerve endings in the subcuta¬ and spinothalamic pathways), which are also the
neous fat under the dermis of the skin have been more primitive ones, carry less complicated infor¬
found to be connected to nerve fibers associated mation such as pain, temperature, and rudimentary
with pain (see Vierck, 1978). The position of these touch, as Henry Head (1920) first suggested.
endings in the subcutaneous fat makes them re¬ This view is appealing on physiological
spond only to higher intensity stimuli, whether me¬ grounds, and is consistent with some interesting
chanical or temperature. Other pain fibers terminate psychological phenomena. Consider, for example,
in the epidermis; these endings are wrapped in a the phenomenon called double pain. This is the
Schwann cell sheath (see Appendix), that allows experience of two distinct peaks of pain, differing
them to retain a high threshold even in this more in quality and separated in time, arising from a sin¬
exposed location (Perl, 1984). Modem methods gle pain stimulus. It is now generally accepted that
promise to reveal even more about the nerve end¬ the first, sharp or pricking pain arises from the re¬
ings that respond to noxious stimuli. sponse of the somewhat faster A8 fibers to the nox¬
Another view puts less emphasis on the recep¬ ious stimulus, whereas the second, dull or burning
tors than on the nerve fibers that carry information pain arises from the slower-conducting C fibers
 Pain 263

(Cooper, Vierck & Yeomans, 1986; Torebjork & the case. Unfortunately, simply knowing that a par¬
Hallin, 1973; Willis, 1985). You may be able to ticular stimulus was present cannot guarantee that a
experience this for yourself if you try Demonstra¬ particular sensation was present. Thus, injured hu¬
tion Box 9-7. mans, particularly those engaged in some demand¬
ing activity such as war or athletics, often do not
feel pain although horribly wounded, presumably
Neural Responses to Pain Stimuli
because of conflicting and more urgent responses.
The electrophysiology of the pain system(s) has We do know, however, that certain nerve fibers ap¬
been studied by applying noxious stimuli (such as pear to fire only when their receptive fields are
electric shock, pinching, or pricking) to animals stimulated by noxious stimuli. For example, Pog-
and recording the responses of neurons at various gio and Mountcastle (1960) found such neurons in
levels of the nervous system. It is assumed that the cat’s thalamus, and Casey and Morrow (1983)
these stimuli cause pain for the animals, as they found them in the thalamus of awake monkeys.
usually do for humans, but this may not always be Such neurons also have been found in area SI of

DEMONSTRATION BOX 9-7. The Production of Double Pain

This demonstration uses the method of Sinclair and Immerse your entire hand in the 35 deg C bath for
Stokes (1964) to generate two pains for the price of about 10 minutes. When this time has elapsed, mix
one. Double pain is experienced only under certain the 57 deg C bath, and carefully insert your finger
conditions. When these conditions are met, people re¬ into it until the water comes up past the second joint
port a first, sharp stinging sensation, followed about of the finger. Count “one-thousand-one” to yourself,
1 second later by a more intense burning pain that and then withdraw your finger. Pay careful attention
may spread to a wider area and fades more gradually. to the sensations you experience. Notice that first you
Although most people, under the appropriate condi¬ feel a sharp stinging and then about a second later a
tions, experience this sequence without being told burning feeling. You may try the experiment again
what to expect, we are telling you now so that you and again without fear of any damage if you immerse
will have a good chance to experience it. For this your hand in the 35 deg C bath between trials, and
demonstration you will need to find a source of hot always limit your immersion in the 57 deg C bath to
water, something to measure its temperature, and two 1 second. If you wish, you can try varying the tem¬
medium-sized bowls to hold it in. You need to pro¬ peratures of the two baths to find the limits of the
duce two water baths: one at 35 deg C (95 deg F) and conditions under which the phenomenon will occur.
one at 57 deg C (135 deg F). If you have access to a Also, in calculating the formulas for the two baths,
thermometer (a meat thermometer is fine for this we assumed that the cold tap water in your area has
demonstration), this would obviously be the best way a temperature of about 10 deg C (50 deg F), and the
to measure the temperatures of the baths. If you don’t hot tap water a temperature of about 60 deg C (140
have a thermometer, simply mix 3Vi cups of very hot deg F ). If your water temperatures vary significantly
tap water with 3 Vi cups of cold tap water for the 35 from these, you will have to adjust the proportions of
deg C bath. To keep it at about this temperature, add each to make up the baths.
a little hot water every minute or so. To create the 57
deg C bath, combine 6% cups of hot tap water with
Vi cup of cold tap water.
264 Chapter 9/Touch and Pain

the somatosensory cortex of rats (Lamour, Wilier active, we experience pain. A normal stimulus, say
& Guilbaud, 1983) and monkeys (Kenshalo & Is- a touch, would mostly stimulate the fast fibers,
ensee, 1983), and may also occur in area SII (see which have lower thresholds. This would excite the
Willis, 1985). It is clear that both the thalamus and substantia gelatinosa neurons, thus causing them to
the somatosensory cortex play roles in pain percep¬ inhibit the T cells, keeping them below the activity
tion. Just what these roles are, however, is far from level that is sensed as pain. A noxious stimulus,
clear. Probably there are many brain areas that par¬ however, would stimulate the higher-threshold
ticipate in the pain experience (see Casey, 1978; slow fibers as well as the fast fibers. Because the
Willis, 1985). slow fibers inhibit the substantia gelatinosa cells,
Perhaps the most interesting electrophysiolog- canceling their excitation by the fast fibers, the
ical fact about pain is that the several types of substantia gelatinosa neurons no longer inhibit the
nerve fibers involved in pain and in other cutaneous T cells, which fire more vigorously, and pain is
and kinesthetic sensations interact, sometimes in experienced. The substantia gelatinosa is the
opposition to each other. Melzack and Wall (1965, “gate” for the activity of the T cells—the fast fi¬
1982) devised an ingenious conceptual model of bers close this gate and the slow fibers open it.
pain, called the gate-control theory, based on the Generally speaking, chemical anesthetics act
interaction of two of these fiber types. This theory to inhibit the slow fibers, but do not affect the fast
provides the foundation for most modem accounts fibers. This allows the substantia gelatinosa to in¬
of a variety of pain phenomena, so it is important hibit the T cells and close the pain gate. Melzack
to understand it thoroughly at this point. Figure 9- and Casey (1968) suggested that the central path¬
15 presents the theory diagrammatically. Let us go way to the gate can also be responsible for closing
through it step by step, since there are several as¬ it, so that other perceptions, cognitions, and emo¬
pects to keep in mind. tions could be responsible for changing the nature
First, notice in Figure 9-15 that both fast of a potentially painful experience. Certainly there
(e.g., A|3) and slow (e.g., AS and C) fibers have is good evidence of these descending pathways
connections with the substantia gelatinosa (a inhibiting responses of spinal cord neurons to
group of neurons in the spinal cord; see also Figure noxious stimuli (e.g., Dickhaus, Pauser & Zim¬
9-3) and with the first transmission cells (T cells). merman, 1985; Willis, 1983; Zimmerman, 1983).
These T cells are part of the set of slow fibers that
make up the spinothalamic pathways and send pain
Pain Thresholds, Intensity, and
information up the spinal cord to the brain. The fast
Adaptation
fibers also have a direct connection (the medial
lemniscus) to the brain (or “central control,” to To treat pain as a sensation it is first necessary to
use Melzack and Wall’s terminology), which can define a pain threshold. Usually this is taken to be
in turn send information back down the spinal cord the intensity of a stimulus that will just barely pro¬
to the gate-control system. Notice in Figure 9-15 duce a sensation of pain. Obviously, thresholds
that the connections of both the fast and the slow will vary across the different conditions under
fibers to the T cells are marked with a plus sign, which they are measured, since pain can be aroused
meaning that they increase neural activity in, or ex¬ in so many different ways. As in the case of touch
cite, those cells. The actions of these fibers on the and thermal sensitivity, there are specific tiny
substantia gelatinosa cells are different, however. points on the skin that respond selectively to pain.
The fast fibers excite the neurons in the substantia These points will give the sensation of pain for
gelatinosa ( + ), whereas the slow fibers inhibit stimuli that do not produce painful sensations when
their action ( —). When the T cells are sufficiently applied to places other than “pain points.” In gen-
 Pain 265

FIGURE 9-15 An illustration of the gate-control theory of pain.

eral, pain points seem to correspond to receptive conditions for measuring pain thresholds was made
fields of the small, slow-conducting pain fibers (see by Hardy, Wolff, and Goodell (1943). They used a
Willis, 1985). The distribution of such pain points device that focused an intense beam of light on the
over the body seems quite variable, as can be seen ink-blackened forehead of a subject in order to pro¬
from Table 9-1. Pain thresholds vary in the oral- duce a painful heat stimulus. Since the device used
facial regions as well, with the tongue and inside radiant heat as a stimulus, the stimulus could be-
of the lip being less sensitive than other parts to precisely controlled and measured. They called the
heat-induced pain (Green, 1985). device dolorimeter, from the latin dolor, “pain,”
One major advance in the standardization of and meter, “to measure.” This development
266 Chapter 9/Touch and Pain

Table 9-1. Distribution of Pain age. They even moved the site of the painful stim¬
Sensitivity ulation from the forehead to the forearm because
the latter was more easily cared for when blistered
Skin region Pain points/cm1 by the pain stimuli. These rather extreme measures
resulted in some very important results. Hardy et
Back of knee 232 al. (1947) found that the differential threshold
Neck region 228 could be measured for pain, and that it is reproduc¬
Bend of elbow 224
ible under constant conditions. Moreover, they also
Shoulder blade 212
found that the Weber fraction remains remarkably
Inside of forearm 203
constant (as Weber’s law would assert) at about
Back of hand 188
184 0.04 (a mere 4% stimulus change) over quite a
Forehead
Buttocks 180 large range of stimulus intensities. This indicates
Eyelid 172 that we are quite sensitive to variations in pain in¬
Scalp 144 tensity. Weber fractions begin to increase dramati¬
Ball of thumb 60 cally at only the highest stimulus intensities. At the
Sole of foot 48 extremes, however, the data were not very reliable,
Tip of nose 44 because the skin damage sustained made it difficult
for the observers (the authors themselves) to con¬
Based on Geldard, 1972 centrate on the pain intensities.
Hardy et al. (1947) also created the first scale
permitted the investigation of the various condi¬ of pain intensity. Since they had established the va¬
tions that affect the level of the pain threshold. The lidity of Weber’s law for pain, they merely added
exact thresholds measured in this way are of little up jnds, as Fechner had done (see Chapter 2), to
importance to us here, since the units of any pain create a scale of pain intensity based on the discri-
threshold stimulus vary with the pain-producing de¬ minability of painful stimuli. They appropriately
vice or stimulus modality. However, Hardy et al. called this scale the dol scale (again based on the
(1943) were able to show that pain thresholds acted latin dolor). Later scales of pain intensity were cre¬
very much like the thresholds for other sensations. ated by more direct methods (see Chapman, Casey,
Pain thresholds were shown to be relatively stable Dubner, Foley, Gracely & Reading, 1985, for a
as long as the conditions were stable, but they var¬ review of pain measurement). For example, Ste¬
ied systematically with changes in the neurological, vens (1961) obtained magnitude estimations of the
pharmacological (drugs), or psychological state of intensity of pain produced by electric shocks. He
the individual. These results have been replicated found that the magnitude estimations were a power
many times, and more recently even social situ¬ function of the stimulus intensity with an exponent
ations have been shown to affect pain thresholds of about 3.5, making pain produced by electric
(Craig, 1978). shock the sensory modality with the largest power
Whether two pains are the same or different in function exponent. This implies that the sensation
intensity can be discriminated, indicating that pain of pain grows faster with increasing stimulus inten¬
has a differential threshold, too. The first good sity than does any other sensory experience. The
measurement of the differential threshold was done actual size of the exponent varies as a function of
by Hardy, Wolff, and Goodell (1947) using a mod¬ a variety of stimulus and even social factors (Craig,
ification of the dolorimeter. Since the authors felt Best & Ward, 1975; Stembach & Tursky, 1964),
the knowledge was important, they served as their and more recent studies have typically obtained
own subjects and as a result experienced both a somewhat lower exponents, usually under 2.0
large amount of pain and considerable tissue dam¬ (e.g., Rollman & Harris, 1987).
 Pain 267

Being able to measure pain is important to un¬ stimuli, which were only mildly painful, and less
derstanding pain phenomena. For example, it has complete for the more painful stimuli. Adaptation
recently been observed that pain summates, that is, may not take place at all for extremely painful
different pains add together to produce a higher in¬ stimuli, although recently it has been shown that
tensity of experienced pain than either one alone. even dental pain adapts (Ernst, Lee, Dworkin &
This is true both within modalities and across mo¬ Zaretsky, 1986).
dalities. Thus, stimulating two teeth at the same
time lowers the pain threshold compared to stimu¬
Analgesia and Endogenous Opiates
lating a single tooth, and turns mild discomfort into
pain (Brown, Beeler, Kloka & Fields, 1985). Also, Because pain is unpleasant we seek to minimize it.
pain from shock and loud noise experienced to¬ After it has served its function of warning that the
gether is roughly the linear sum of the pains expe¬ body is about to be, or already has been, damaged,
rienced separately (Algom, Raphaeli & Cohen-Raz, the pain signal is no longer necessary or desired.
1986). Yet strong pain signals often persist, or indeed may
If we want to argue for a sensory basis for the occur for the first time, well after the damaging
experience of pain, we must ask if pain sensations stimulus is gone. This may be because the pain in¬
adapt as do all other sensations. Dallenbach (1939) duces us to remain relatively immobile, so that
demonstrated that pain caused by needles, heat, healing may proceed optimally (Wall, 1979). How¬
and cold does adapt. Heat-induced pain was studied ever, humans are not content to simply accept this
by Hardy, Stolwijk, and Hoffman (1968) by having immobilizing pain, nor do they desire to experience
observers judge the degree of experienced pain as the pain from surgery or illness. Thus, we have
they sat with their hands in hot water over a period assembled an impressive array of analgesics and
of time. As can be seen from Figure 9-16, adapta¬ anesthetics to rid ourselves of pain. The major
tion is complete for the lower-temperature pain focus of much pain research seems to be the

FIGURE 9-16 Average estimations of pain intensity from hot-water immersions of different
temperatures at different durations.
268 Chapter 9/Touch and Pain

discovery of new ways to induce analgesia. The sible for both the generation of endogenous opiates,
most potent and reliable method of pain alleviation and for their analgesic effect. For example, electri¬
involves chemicals that are eaten (such as aspirin) cal stimulation of certain parts of the thalamus can
or are injected into nerves or muscles (such as the produce strong analgesic effects. This analgesia is
novocaine used by dentists). We can even buy reversible by naloxone and less strong for individ¬
sprays or tubes of salve that contain substances that uals who have developed a tolerance or relative in¬
cause a temporary analgesia on cut or burned skin. sensitivity to morphine, thus suggesting that this
For more severe pain we resort to narcotic drugs part of the brain may be one site where endogenous
(such as morphine, an opium derivative) or opt for opiates are produced (see Akil & Watson, 1980).
unconsciousness (as with ether or chloroform). It is interesting to note that sufferers of chronic
While studying how some of the more pow¬ pain have lower than normal levels of some endog¬
erful opium-based drugs produce analgesia, re¬ enous opiates in their spinal fluid, and electrical
searchers made a discovery that helped advance our stimulation of the brains of such people produces
understanding of how our bodies control pain nat¬ both analgesia and dramatic increases in the levels
urally. Opiates interact with specific receptors in of endorphins in their spinal fluid (Akil & Watson,
the brain to produce their analgesic and intoxicating 1980; Terenius & Wahlstrom, 1975).
effects. Since the brain has receptors for this family Our conscious experience of pain intensity is
of chemicals already, it seemed likely that there affected not only by the magnitude of the pain
must be a class of chemicals naturally present in stimulus but also by these chemical regulators,
the body that also interacts with these receptors. generated internally and acting directly on specific
Presumably, these endogenous opiates (meaning areas of the central nervous system. Study of en¬
opiates generated from within) should exhibit an¬ dogenous opiate systems may also provide clues as
algesic properties similar to those of opium deriv¬ to the mechanisms involved in the nonchemical
atives. Several such substances were discovered by methods for the reduction of pain. For example,
biochemists in the early 1970s (see Kosterlitz & there are many instances where purely psychologi¬
McKnight, 1981; Snyder, 1977). At least two ma¬ cal factors seem to cause reduced sensitivity to
jor classes of internally generated chemicals, the pain. Wilier, Dehen, and Cambier (1981) found
enkephalins and the endorphins, have significant that the psychological stress caused by the antici¬
analgesic effects, and seem to react with the same pation of a painful shock actually resulted in anal¬
sites that opiates do (see Millan, 1986; Yaksh, gesic effects. Presumably the stress triggered the
1984). When bodily levels are artificially raised (by endogenous opiate system (see also Lewis, Ter-
administration of extra amounts of these sub¬ man, Shavit, Nelson & Liebeskind, 1984). Simi¬
stances), the endorphins seem to be the more potent larly, women during the last two weeks of
and longer lasting. The opiumlike action of these pregnancy experience significant increases in pain
endogenous substances is further demonstrated by thresholds, which reduces their discomfort (Cogan
the fact that their analgesic effect can be blocked & Spinnato, 1986). Since pregnant rats experience
by the administration of naloxone, a potent antag¬ the same threshold increases, and since the effect
onist of opiates such as morphine, which is often is much reduced when the rats are given naltrexone
administered to those who have taken overdoses. (another opiate antagonist) (Gintzler, 1980), it is
Administration of naloxone by itself makes people likely that the endogenous opiate system is in¬
who are under stress more sensitive to pain, pre¬ volved.
sumably because it blocks the effectiveness of en¬ Similar factors seem to be involved in some of
dogenous opiates released naturally under these the more “mysterious” reports of reduced pain
circumstances (Schull, Kaplan & O’Brien, 1981). sensitivity. For example, placebo effects are some¬
Specific sites in the brain seem to be respon¬ times reversible by naloxone, suggesting that some
 Pain 269

opiate system is involved. Perhaps even more mys¬ have now established that pain reduction achieved
terious is the traditional Chinese technique for al¬ through acupuncture is mediated by release of en¬
leviating pain called acupuncture (from the Latin dogenous opiates (Akil & Watson, 1980; He, 1987;
acus meaning “needle” and pungere meaning “to He, Lu, Zhuang, Zhang & Pan, 1985; Kosterlitz &
sting”). In this technique, long thin needles are McKnight, 1981).
inserted at various sites on the body (see Figure The brain-chemical interaction we have been
9-17). These needles may be twirled, heated, or discussing provides only an incomplete picture of
have electrical current passed through them. Al¬ the factors influencing our perception of pain. For
though Western doctors have been cautious about instance, many forms of pain reduction, such as
accepting acupuncture as a valid means of reducing that achieved via hypnosis, do not appear to be me¬
pain, most studies support its effectiveness (see diated by endogenous opiates (Akil & Watson,
P. E. Brown, 1972; Chapman, 1978; Cheng, 1973; 1980; Kosterlitz & McKnight, 1981). It is therefore
Clark & Yang, 1974). A large number of studies not the case that all forms of “mysterious” anal¬
gesia can be explained by endogenous opiates. The
most recent evidence suggests that humans have at
least two pain control systems, and that only one
of them involves endogenous opiates (Akil & Wat¬
son, 1980; Mayer & Watkins, 1984; Watkins &
Mayer, 1982).
Some of the most interesting analgesic proce¬
dures involve cognitive processes. These include
such techniques as suggestion, attitude, concentra¬
tion of attention, and social modeling (see Craig,
1978; Weisenberg, 1984; Wolff & Goodell, 1943).
The efficacy and interpretation of these techniques
vary, but there is no doubt that they are real—pain
thresholds can be affected dramatically. For in¬
stance, social modeling, where observers see an¬
other person’s reactions to painful stimuli before
judging the painfulness of the same stimuli for
themselves, has been reported to affect both d' and
physiological reactivity to painful electric shocks
(Craig & Coren, 1975; Craig & Prkachin, 1978).
Demonstration Box 9-8 provides an opportunity for
you to assess the effectiveness of one form of cog¬
nitive control of the perceived intensity of pain.
Quite clearly, the perception of pain is com¬
plex, involving a number of different levels of con¬
trol. You can see how these many levels may be
integrated into one system by considering the gate-
control theory of pain that we discussed earlier.
FIGURE 9-17 A typical acupuncture chart. The According to this theory, pain is experienced when
numbers indicate sites at which needles can be the T cells are firing at a high enough rate. The
inserted, and then either twisted, electrified, or
theory describes not only a spinal gate controlled
heated. An impressive analgesia results in many
cases.
by fast and slow conducting fibers, but also allows
270 Chapter 9/Touch and Pain

DEMONSTRATION BOX 9-8. Cognitive Effects on Pain: The Lamaze Technique

A cognitive technique to alleviate the pain of child¬ must count the breaths (1-2-3-4-5-blow) or say a
birth is taught in many places in North America and short poem or nonsense sentence over and over again
Europe (see Beck & Siegel, 1980). The basic idea to the rhythm of your breathing (“Am I a bird or
was that of a French medical doctor named Fernand plant?”). Third, you must concentrate your visual at¬
Lamaze. One demonstration of how this technique tention on (look intently at) some clearly visible ob¬
works requires a friend to assist you. ject during this entire period. Practice these behaviors
Have your friend grasp your leg just above the knee until you feel fairly confident of your ability to main¬
with a hand. Have your friend squeeze gently at first, tain them for a couple of minutes. Then have your
then with steadily increasing force until you can feel friend give you the gradually increasing pressure on
a fairly severe pain. This should convince you that the leg, while you do your breathing. Under these
the stimulus is actually painful. Now you have three conditions, if your concentration is really intense, you
things to practice simultaneously. First, you have to might not feel the pain (or the pressure) at all, or at
breathe in a particular way. To do this you must take least it will be of much less intensity. According to
five short panting breaths in a row, followed by a the Melzack-Wall-Casey (Melzack & Casey, 1968;
strong blow outward (pant-pant-pant-pant-pant-Wow). Melzack & Wall, 1965) approach to pain, what is
Repeat this pattern during the entire period during happening here is that your central control system is
which the painful stimulus might occur. Do not closing the pain gate. Thousands of mothers claim
breathe too quickly for you might hyperventilate and that this basic technique of concentrating on some¬
get dizzy. If you do get dizzy, stop for a moment and thing removed from the source of the pain effectively
then start up again at a slower place. Second, you alleviates the pain and distress of childbirth.

inputs from higher levels of the nervous system to 12th-century English doctor prescribed the wear¬
open or close the gate. Thus, cognitive factors, ing of a copper bracelet on the left hand to relieve
motivational states, attentional factors, or other a pain in the right hand, and a modem Chinese
stimulation such as high-intensity hissing noises, doctor twirls needles stuck through the skin of
electricity, or music could all be responsible for an appendectomy patient—all claiming successful
controlling the gate via the pathway descending analgesic results. Perhaps each was stimulating
from the brain to the spinal corchgate (Willis, 1983, one of the several pain control systems we pos¬
1985). These descending pathways seem to be sess, rather than simply engaging in “empty
strongly implicated in analgesia caused by release superstition.”
of endogenous opiates. Perhaps activation of the
descending pathways causes release of endogenous
GLOSSARY
opiates into the spinal cord, thus closing the pain
gate (Watkins & Mayer, 1982). However, the
The following definitions are specific to this book.
mechanism involving the substantia gelatinosa does
not seem to use endogenous opiates to produce its Backward masking When a masking pattern inter¬
feres with the perception of a target presented earlier in
effects.
time.
We do not understand fully the story of the
Cold fibers Neurons that respond to cooling by fir¬
perception of pain. However, we now have a hint ing more rapidly and to warming by firing more slowly.
of why the 2nd-century physician Galen prescribed Dermis The inner layer of skin, containing most of
the shock from an electric fish for a headache, a the nerve endings.
 Glossary 271

Dol scale A scale of pain intensity based on the dis- Optacon A system, similar to the vision substitution
criminability of painful stimuli. system, that converts printed letters into vibration pat¬
Dolorimeter A device that delivers precise quantities terns on the fingertip.
of radiant heat; used to measure pain thresholds. Pacinian corpuscle A corpuscular nerve ending
Double pain The phenomenon of two distinct peaks found in skin and joints, sensitive to mechanical defor¬
of pain, differing in quality and separated in time, from mation.
a single pain stimulus. Paleospinothalamic pathway A part of the spinal
Endogenous opiates Analgesia-inducing opiates pro¬ cord, evolutionarily older, that conducts information
duced naturally in the brain and other areas of the body. (representing dull, burning pain, temperature, and rudi¬
Endorphins One of the major groups of endogenous mentary touch) from the skin, muscles, and joints to the
opiates. brain.

Enkephalins One of the major groups of endoge¬ Paradoxical cold The phenomenon of cold spots in
nous opiates. the skin responding to warm stimuli with a sensation of
cold.
Epidermis The outer layer of skin.
Parietal cortex The upper central region of the brain
Forward masking When a masking pattern inter¬
housing the somatosensory cortex.
feres with the perception of a target presented later in
time. Physiological zero A neutral point in the perception
of heat and cold, usually taken to be the skin tempera¬
Free nerve endings Noncorpuscular, branching nerve
ture.
endings in skin, joints, etc., that may be receptors for
pain and temperature. Somatosensory cortex The part of the cerebral cor¬
tex, located in the parietal lobe, that receives input from
Gate-control theory A conceptual model of pain
the thalamus and other brain regions representing touch,
based on the interaction of slow, high-threshold nerve
temperature, pain, and kinesthesis.
fibers and fast, low-threshold nerve fibers via the sub¬
Spindle organs See Stretch receptors.
stantia gelatinosa (the “pain gate”).
Spinothalamic pathway A slow pathway of short fi¬
Glabrous skin A type of skin that has no hairs (e.g.,
bers in the spinal cord that conducts information (repre¬
on lips); it is highly sensitive to stimulation.
senting pain, temperature, and rudimentary touch) to the
Golgi tendon organs Nerve endings attached to ten¬ brain from the skin, muscles, tendons, and organs.
dons that respond to both stretching and contracting of
Stretch receptors Nerve endings attached to muscle
muscle.
spindles that respond to stretching of the muscle.
Hairy skin Covering of the human body from which
Substantia gelatinosa The part of the spinal cord
numerous hairs protrude; it is both a protective and stim¬
implicated in pain transmission through the gate-control
ulus-sensitive covering. See Glabrous skin.
theory.
Kinesthesis Sensations of force, weight, and limb
Thalamus The region of the lower brain that relays
position and movement.
nerve impulses to the somatosensory cortex.
Limbic system A part of the brain, evolutionarily
Transmission (T) cells In the gate-control theory of
old, involved in emotion and memory.
pain, these transmit pain impulses to the brain.
Medial lemniscus A part of the spinal cord com¬
Two-point threshold The minimum distance neces¬
posed of large, rapidly conducting nerve fibers that con¬
sary between two pointed touch stimuli (such as two
duct touch information from the skin to the brain.
toothpicks) so that they will be felt as two distinct sen¬
Neospinothalamic pathway A part of the spinal sations.
cord, evolutionarily more recent, that conducts informa¬
Vision substitution system An instrument that con¬
tion (representing sharp, pricking pain, temperature, and
verts a visual pattern from a television camera into a pat¬
rudimentary touch) from the skin, muscles, and joints to
tern of vibrating points on the skin of the back; used for
the brain.
the visually impaired.
Neutral zone A temperature range surrounding
Warm fibers Neurons that respond to warming by
physiological zero. Stimuli within this range feel neither
firing more rapidly and to cooling by firing more slowly.
warm nor cold.
CHAPTER
10
Space
TYPES OF DEPTH AND DISTANCE
PERCEPTION
PICTORIAL DEPTH CUES
Interposition or Overlay
Shading
Aerial Perspective
Retinal and Familiar Size
Linear Perspective
Texture Gradients
Height in the Plane
Motion and Motion Parallax
PHYSIOLOGICAL CUES FOR DEPTH
Accommodation
Convergence and Divergence
PHYSIOLOGICAL CUES FOR DIRECTION
Eye Movements and Direction
Eye Dominance and Perceived Direction
RINOCULAR DEPTH PERCEPTION
Cues for Stereopsis
The Process of Stereopsis
INTERACTION OF CUES
DEVELOPMENT OF SPACE PERCEPTION
Species Differences
Experience and Depth Perception
274 Chapter 10/Space

I n 1621 Robert Burton noted, “All places are

distant from heaven alike.” Perhaps for a
observer to make object-relative localizations,
which are estimates of the distances between ob¬
clergyman-philosopher such a description of spatial jects in the environment. The judgment of relative
relations was sufficient. Yet for you, a simple distance is also involved in the perception of
mortal trying to pick up a cup of coffee from the whether an object is flat (as in a two-dimensional
tabletop, much more precision is needed. You picture) or solid (three-dimensional), in that this re¬
must be able to judge how far the cup is from quires estimation of the spatial relationships be¬
your hand with a good deal of accuracy, lest you tween parts of an object.
end up with a messy puddle of hot fluid. Your The accomplishment involved in seeing ob¬
very life may depend on your precision in judging jects in depth is quite amazing considering that the
depth and distance, as when you sense that you basic information available to the nervous system
are near the edge of a cliff. You must know how is just a flat image on our retinas. The question
close you are lest you fall, and you must also know of how we convert this two-dimensional image
the direction of the edge from your body lest you into our three-dimensional conscious impression
step toward it rather than away from it. We cannot of the world has stimulated a number of different
get cut by a knife edge pictured in a flat photo¬ theoretical approaches. One is called direct per¬
graph, but the real blade extending toward us in ception (Michaels & Carello, 1981) and is charact¬
space can produce painful contact. Thus, our daily erized by the work of J. J. Gibson (e.g., 1979).
tasks and our safety depend on the accuracy of our Direct perception has had its major influence
spatial perception. by stating several viewpoints. The first is that all
the information you need to see three-dimension-
ally is present in the retinal image, or in relation¬
ships among parts of the retinal image. The second
TYPES OF DEPTH AND DISTANCE is the concept that the visual scene is analyzed in
PERCEPTION terms of objects, textures, and surfaces, rather than
in terms of specific stimulus components. Fin¬
Your perception of space consists of at least two ally, direct perception suggests that the impres¬
different aspects. One aspect involves the percep¬ sion of depth or distance arises immediately in
tion of the actual distance of an object from you, the observer, and needs no further computation or
such as how far away a pencil is on your desk. This any additional information based on inferences or
is an estimate of absolute distance, which involves experience.
a process called egocentric localization. Most of An alternative approach, which contains some
us are familiar with the word egocentric in its every¬ of the flavor of direct perception, was proposed by
day use—you are egocentric if you are concerned a number of theorists who were influenced by
only about your own activities and their effect on developments in artificial intelligence (a part of
yourself. In the context of space perception, ego¬ computer science that attempts to make machines
centric means that we have a good sense of where perform human tasks). These include efforts to de¬
our bodies are positioned relative to other objects velop computer programs that will allow machines
in the external environment. We encounter a differ¬ to interpret visual information in the same manner
ent aspect of space perception when we ask that a human observer might. Such theorists, in¬
whether the pencil is lying nearer to the book or to cluding Marr (1982), began with the general as¬
the coffee cup, which are also on the table. This is sumption made in direct perception, namely, that
the judgment of relative distance, requiring the all the information you need to derive three-dimen-
 Pictorial Depth Cues 275

sionality is present in the visual inputs, but added depth. These are stimulus characteristics of which
the suggestion that this interpretation might require we are often consciously unaware, but that function
the detection of fairly subtle aspects of the stimu¬ to shape our perceptual responses.
lus, and might also require a number of computa¬
tions and several stages of analysis. This added
feature, and the assertion that these computations PICTORIAL DEPTH CUES
can be described precisely mathematically, have re¬
sulted in the label of computational theories for When you look at a realistic painting or a photo¬
this approach. (See Chapter 11 for more on com¬ graph you find it quite easy to perceive the spatial
putational theories.) relationship among the various items portrayed.
The last major approach is based on the as¬ Your impression of the relative distances in such
sumption that our perceptual representation of the scenes is based on a set of cues, appropriately
world is much richer and more accurate than might called pictorial depth cues. These cues are also
be expected on the basis of the information con¬ called monocular cues, since they appear in pic¬
tained in the visual image alone. This approach, tures (which are two-dimensional) and only require
which might be called intelligent perception, orig¬ one eye to register (remember a camera has only
inated with Helmholtz in 1867 and is today best one eye), and in order to distinguish them from
exemplified by Rock (1983). It suggests that per¬ cues that require two eyes. To understand these
ception is like other logical processes in that, in depth cues we must first recognize that our visual
addition to the information available at the mo¬ experience usually depends on the transfer of light
ment, we can also use information based on our reflected from an object in the external world to the
previous experience, our expectations, and so eye of the observer. A number of depth cues de¬
forth. In other words, our visual perception of pend on characteristic ways in which light travels
space may involve other sources of information, to the eye, and on ways in which it interacts with
some nonvisual in nature, some based on our past the medium (usually air) through which it passes or
history and cognitive processing strategies. Be¬ interacts with other objects, and on some specific
cause of this combination of several sources of aspects of the geometry of images.
information, these theories have been called
constructive theories of perception.
Interposition or Overlay
Although it is quite likely that each of these
approaches is true for some aspects of the percep¬ The vast majority of objects in the world are not
tual process (cf. Coren & Girgus, 1978; Uttal, transparent. Since light reflected from distant ob¬
1981), theorists who favor particular approaches jects cannot pass through opaque objects that stand
will tend to try to isolate different factors when between them and the observer, a nearer object
they consider the perception of depth or distance. tends to block the view of a more distant one. This
Some will concentrate on isolating structures within depth cue is called interposition, or overlay. It is
the sensory systems themselves, others on the easy to see that the cat in Figure 10-1 is nearer to
properties of the physical world and the way our you than the man’s leg because your view of the
sensory systems interact with it, and still others on leg is partially covered by the image of the cat.
the cognitive interpretive processes called into play Notice that interposition is a cue for relative depth
when we interpret the three-dimensional nature of only, indicating that the cat is nearer than the
the world. All of these approaches, however, usu¬ man’s leg, but not how far away the cat itself is
ally begin by attempting to isolate the cues for from you.
276 Chapter 10/Space

FIGURE 10-2 Shading makes it clear that we are

looking at a muffin tin with the cups pushed in.
Inverting the figure reverses the shadow pattern
and now makes it appear that the cups are
pushed out toward you.

Figure 10-2 again. Now it should look like an in¬

verted muffin tin, with the bottoms of the cups
pointed toward you (cf. Berbaum, Bever & Chung,
1984). Clearly, for this shading cue to work con¬
sistently in this example, we must be assuming that
the light is coming from above. If the light were
FIGURE 10-1 Interposition as a depth cue is coming from another position the shading pattern
illustrated by the fact that the cat is seen as closer would be different. Observers do seem to use their
than the man’s leg since it partially blocks your knowledge or presumptions about the location of
view of the leg.
the light source to help them accurately perceive
the three-dimensional nature of objects using the
shading cue (Berbaum, Bever & Chung, 1983; Ra-
Shading
machandran, 1988).
The fact that light cannot pass through most objects
gives rise to the interposition cue. The fact that
Aerial Perspective
light usually travels in straight lines gives us
another cue for relative depth. This is based on the The partial or complete blockage of light gave us
fact that the lower part of an in-going “dent” will our first two pictorial cues for relative depth. A cue
catch some of the light from above, whereas the for absolute depth emerges from the fact that the
upper part will be in shadow. This is illustrated in air is filled with light-absorbing and light-scattering
the “muffin tin” in Figure 10-2, in which the cups molecules, even on the clearest of days. As light
are clearly pushed in relative to the surface. If the passes through the air, it tends to be absorbed and
shadow pattern is reversed, with bright on the top somewhat scattered by these minute particles of
and shadow on the bottom, we perceive the same dust and moisture. Large particles (such as dust)
pattern as an out-going “bump.” You can see this scatter the light uniformly, causing a uniform dis¬
by rotating the book 180 degrees and looking at tribution of light or a blurring of the image. For
 Pictorial Depth Cues 277

particles that are small in comparison to the various the particles in the air could account for the percep¬
wavelengths of light (such as minute bits of water tion of a diminished brightness with increasing ob¬
vapor), the degree of scatter depends on the spe¬ ject distance, even though the distances may not be
cific wavelength, with shorter wavelengths (blue) as great as those described in the context of the
undergoing more scatter than longer wavelengths usual aerial perspective cue (Uttal, 1981). Cer¬
(Uttal, 1981). The combined effect of these phe¬ tainly, in the absence of any other cues, you will
nomena produces the cue called aerial perspec¬ tend to see the brighter of two identical objects as
tive, in which the image of a very distant object, closer (Ittelson, 1960).
such as a distant mountain, will be slightly bluer in
hue and hazier in appearance than the images of
Retinal and Familiar Size
nearer objects. Such changes in appearance can
provide information about the absolute distance of As an object moves farther away from you, its ret¬
relatively faraway objects. In some geographic re¬ inal image size begins to diminish. One country
gions (such as the prairies of the United States and song captured this effect in its lyric, “If you see
Canada), this can lead to considerable errors in dis¬ me getting smaller I’m leaving.” The geometry of
tance judgments, since the clear dry air reduces ae¬ this situation is shown in Figure 10-3, where the
rial perspective. Thus, a plateau that appears to be more distant person is casting a smaller retinal im¬
only 1 or 2 miles away on a clear day, when look¬ age. We tend to use these relative differences in
ing across a dry sector of Wyoming, may actually retinal image size as a cue for relative distance, as
be 20 or 30 miles distant. Conversely, this explains in Figure 10-4, where we see a row of puppy dogs
why objects seen in the morning fog or a mist ap¬ that seem to recede in distance from us because of
pear to be farther away than when seen in bright their decreasing image size. Thus, the comparison
midday sun (Ross, 1975). of the sizes of objects in the visual field, relative to
There is an interesting variant of the aerial each other, is an important part of the process of
perspective cue that is usually referred to as rela¬ perceiving relative distance.
tive brightness. The luminance of a lighted surface Retinal image size is a cue used by both direct
or object does not decrease with distance, but the perception and computational theories of percep¬
light from more distant objects must travel through tion. There is, however, another size cue that is
the atmosphere for a greater distance. Therefore, important in constructive theories. This has nothing
increased absorption or scattering of the light by to do with image size but rather with your previous

Near

FIGURE 10-3 Objects of the same physical size produce smaller retinal angle sizes with increasing
distance from the observer. Thus, relatively speaking, smaller images are perceived to be more distant.
278 Chapter 10/Space

experience with the usual or familiar size of the distant. You may demonstrate the effect of familiar
object. For example, playing cards all tend to be size for yourself by following the instructions in
the same size. Ittleson (1951) presented to observ¬ Demonstration Box 10-1.
ers three playing cards, under reduction condi¬
tions. This is usually a darkened room with all
Linear Perspective
other depth cues removed. One of the playing cards
was normal in size, a second was twice normal There is a well-known pictorial depth cue that may
size, and a third was one-half normal size. He be seen as an extension of the retinal image size
found that observers tended to judge the double¬ cue to distance. This cue is linear perspective. For
sized playing card as being much closer to them example, look at Figure 10-5, which is adapted
and the half-sized card as being much more distant from a book on how to depict perspective in draw¬
than the normal-sized card. This is the same pro¬ ings authored by Jan Vredeman de Vries in 1604
cess that causes you to see the dogs in Figure 10-4 (de Vries, 1604/1968). In this schematic scene we
as receding into the distance. You presume that all notice that physically parallel lines, such as those
of these images are about the same “dog size” and defining the paving blocks making up the floor,
use this familiar size information in conjunction seem to converge as objects become more distant.
with the changing retinal size to gain the impres¬ So do the hypothetical lines that connect all the
sion of changes in relative distance. tops and all the bottoms of the pillars, all of which
As long as the objects are commonplace, and are actually the same size. This means that physi¬
the distances not too extreme, familiar size can cally parallel lines, such as railroad tracks, appear
give you absolute depth information, not merely to converge and objects appear to get smaller in a
relative depth information (Epstein & Baratz, 1964; systematic fashion as their distance from you in¬
Fitzpatrick, Pasnak & Tyer, 1982). Thus, if we see creases. Eventually they will reach a vanishing
a very tiny elephant, we can use our knowledge point, where all the perspective lines will converge
that elephants are relatively large creatures to de¬ and objects will diminish to invisibility. This point
duce that the elephant has not shrunk in size but is usually on the horizon, as shown in the figure.
rather has moved away from us and is now more This is a simple geometric effect that occurs in the

FIGURE 10-4 Relative size differences are interpreted as cues for relative distance. Thus, we see a row
of puppy dogs that seem to recede in distance because of their decreasing image size.
 Pictorial Depth Cues 279

DEMONSTRATION BOX 10-1. Familiar Size and Distance

Look at Figure 10-4. Notice that the row of puppy volleyball or a basketball. Which dog is the same dis¬
dogs seems to recede into the distance. Off to the tance as the basketball? Notice that the ball apparently
right is a ball. If we told you that it is a tennis ball or “moved backward” in depth when you assumed that
a baseball, you would have no difficulty in deciding it was a larger object. This shows how knowledge of
which dog is at the same distance away from you as the size of an object can affect our judgment of the
the ball. After you decide this, return to this box. distance of the object, giving us the familiar size cue
Now, suppose we told you that the ball is really a to distance.

Horizon

fgjgjg
■IIIIIIIIHIIIIIIIIIIiliW**^

FIGURE 10-5 An example of linear perspective, in which physically parallel lines seem to converge as
they grow more distant. Notice that the lines have been extrapolated to show a vanishing point on the
horizon.
280 Chapter 10/Space

real world, and when we project a three-dimen¬ information into one cue, which he referred to as
sional scene onto a two-dimensional surface. It texture gradient. A texture is any collection of ob¬
provides a powerful relative depth cue. Hence, it is jects (Caelli, 1982), and the gradient (continuous
easy to determine that pillar B is farther away from change) is the change in the relative size and com¬
you than pillar A by utilizing the perspective cue. pactness of these object elements. The more distant
parts of the texture have smaller elements that are
more densely packed together (Gibson, 1950). This
Texture Gradients
cue is sometimes called detail perspective. Figure
J. J. Gibson (1950) suggested an interesting way of 10-6A shows a texture of lines. Since the texture
combining both linear perspective and relative size is uniform, it shows little depth, and looks much

A B

FIGURE 10-6 Examples of texture gradients in B and C, which appear as surfaces receding in depth. In
y\ there are no decreases in element size or spacing and thus the perception is of a flat surface.
 Pictorial Depth Cues 281

like a flat wall or garage door. If we introduce a Height in the Plane

gradient, however, as is done in 10-6B, with the
There is another cue to distance that depends on the
lines becoming more compact as we move toward
relationships between objects as their images are
the top, we now get an impression of depth. An
projected onto our retinas. This cue is height in the
even stronger impression of depth appears if we al¬
plane, or relative height, and refers to where an
low the gradient to appear in the horizontal placing
object is relative to the horizon line. In Figure 10-
of elements as well as the vertical, as can be seen
9, post B seems farther away than post A because
in the texture of dots in 10-6C. One important type
its base is closer to the horizon line. Hence it is
of information contained in texture gradients
said to be “higher in the plane,” or “higher in the
emerges from the fact that sudden changes in tex¬
picture plane” if we consider this as a two-dimen¬
ture usually signal a change in the direction or dis¬
sional representation. The reverse holds for targets
tance of a surface. Thus, Figure 10-7A shows how
above the horizon. Bird C seems farther away than
the gradient changes when we shift from floor to
bird D because it is “lower in the picture plane.”
wall, and 10-7B shows how the gradient changes
In other words, proximity to the horizon line sig¬
at a cliff or step down. The perception of depth
nals the greater distance.
obtained from texture gradients can be quite strik¬
ing. Texture helps us to define the shapes of objects
(Todd & Akerstrom, 1987; see also Chapter 11) as
Motion and Motion Parallax
well as delicate variations in distance, as shown in
the undulating surface depicted by texture cues We have been considering only depth cues that
alone in Figure 10-8. could also be represented in a photograph or a

Change in texture

FIGURE 10-7 How texture changes at a corner (A) and an edge next to a sharp drop in depth (B).
282 Chapter 10/Space

FIGURE 10-8 This rippling or undulating surface is defined completely on the basis of variations in
texture density.

FIGURE 10-9 Height in the plane and proximity to the horizon will determine which elements in the
diagram are perceived as more distant. In this case* B and C are seen as being farther awav because they
are closer to the horizon line.
 Pictorial Depth Cues 283

painting, but when we add motion to the incoming can also be generated by swinging your head back
visual pattern we provide some additional opportu¬ and forth while your body is stationary. The mo¬
nities for depth cues to appear. One of these move¬ tion parallax from head movements also gives
ment cues concerns the motion pattern of an object very good information about the depth of objects
as you travel past it. Suppose you are traveling in if they are not too distant from you (Ono, Rivest &
a car or bus and looking at the scene in Figure 10- Ono, 1986).
10. Suppose also that your direction of movement A special form of motion parallax occurs
is from right to left and that you are gazing at the when an object moves or rotates. The relative pat¬
spot marked “Fixation point.’’ Under these condi¬ tern of movement of parts of the object can give us
tions, all the objects closer to you than the fixation information about its three-dimensional shape
point will appear to move in a direction opposite to (e.g.. Carpenter & Dugan, 1983; Doner, Lappin &
your movement, while objects that are farther away Perfetto, 1984). The fact that motion cues can give
will appear to move in the same direction you are us information about the relative depth of parts of
moving. Not only the direction but also the speed an object has been called the kinetic depth effect
of movement varies with the objects’ proximity to (Gibson, 1966; Kaufman, 1974; Rock, 1975).
you and to your point of fixation. This cue to dis¬ Demonstration Box 10-2 allows you to see this
tance is called motion parallax. Motion parallax phenomenon for yourself. In Chapter 13 you will

FIGURE 10-10 Motion parallax. When an observer moves, objects at varying distances from the observer
will move in different directions at differing speeds. These differences can serve as cues for the relative
distance of the objects.
284 Chapter 10/Space

DEMONSTRATION BOX 10-2. The Kinetic Depth Effect

To see how subtle motion parallax effects can create on a blank wall as shown in the figure. Notice that
the impression of a three-dimensional form in a two- when the shape is absolutely motionless, the shadow
dimensional pattern, you will need a candle and a is seen as a flat pattern of lines. Now if you rotate the
piece of stiff wire (a coat hanger or a long pipe shape with your hand, the shadow suddenly changes
cleaner will do). Bend the wire into a random three- perceptually, becoming a three-dimensional object
dimensional shape. Now light the candle and darken that cannot be seen as flat, despite the fact that you
the room. Place the bent wire so that it casts a shadow are viewing a two-dimensional shadow.

also learn how motion parallax can give you infor¬ to arise from muscular responses and adjustments
mation about the direction of your movements of the eye.
through space.

Accommodation
PHYSIOLOGICAL CUES FOR DEPTH When we discussed the physiology of the eye in
Chapter 3, we described how the crystalline lens
Up to now, we have only considered cues for depth responds to targets at different distances from us.
that can be found in the retinal image itself. There Basically, we noted that the lens must change
are other cues for distance that come about because shape (actually it changes its amount of curvature)
of the way the visual system responds to or inter¬ in order to keep the retinal image in clear focus
acts with the visual stimulus. These may be called (Dalziel & Egan, 1982). This process is called ac¬
structural or physiological cues, since they tend commodation. There is only one particular curva-
 Physiological Cues for Direction 285

ture that will clearly focus the retinal image of an when the image of an object is focused on the
object viewed at a particular distance from the eye. fovea, eye movements are executed to bring the
Relaxed accommodation, where the lens is rela¬ image to this region of the eye. If the eyes move
tively flattened, is necessary if distant objects are in different directions these are called vergence
to be clearly focused on the retina, whereas a movements. If an object is close to you, you must
strongly curved lens is needed to image closer ob¬ rotate your eyes inward (toward the nose) in order
jects on the retinal surface. As we change the ten¬ to focus its image on the fovea. Such movements
sion on the ciliary muscles, which control the lens are called convergence (the root con means “to¬
shape, feedback from these muscular changes can ward”). When a target is farther away, the eyes
provide us with some additional, non visual, infor¬ must move away from each other in an outward
mation about the distance of the object we are rotation (toward the temples), hence these move¬
looking at. ments are called divergence (from the root di
In addition to feedback from the act of accom¬ meaning “apart”). Different degrees of conver¬
modation, the presence or absence of blur due to gence are illustrated in Figure 10-11.
an object being out of focus can serve as a cue for Each target distance, up to about 6 m, is as¬
relative distance. It has been shown that in the ab¬ sociated with a unique angle between the eyes
sence of all other depth information observers can called the convergence angle, as indicated in Fig¬
judge that two spots of light presented in complete ure 10-11. To achieve each eye position, a unique
darkness are at different distances. This is probably pattern of muscular contractions must occur. Feed¬
because accommodation cannot be correct for two back from such vergence movements could be
stimuli at different distances at the same time; useful in determining the distances of objects,
hence, one of the lights will be slightly blurred and although there has been some controversy about
out of focus, suggesting that the targets are not how useful and reliable such information is as a
equidistant (Kaufman, 1979). depth cue (Gogel, Gregg & Wainwright, 1961;
Accommodation is rather limited in the range Hochberg, 1971). Recent evidence, however, sug¬
of observer-to-object distances over which it is use¬ gests that convergence and accommodation to¬
ful (Graham, 1965). At a distance of around 3 me¬ gether may provide quite accurate absolute depth
ters, the lens has fully relaxed accommodation and information, even when the only visible stimulus is
doesn’t flatten out any further, regardless of how a single point of light whose distance observers are
far away an object is. There is a similar limit for asked to judge (Morrison & Whiteside, 1984).
close objects. If a target is within 20 centimeters of
your face, your lens has reached its point of maxi¬
mum curvature. Within the range of 20 to 300 cm,
however, accommodation may provide a useful, if PHYSIOLOGICAL CUES FOR
not very precise, auxiliary cue for distance (Hoch- DIRECTION
berg, 1971).
Distance is only one aspect of our perception of
space. The perception of the location of an object
Convergence and Divergence
will also include its direction relative to our bodies.
Another potential distance cue comes from the fact There are actually two types of directional judg¬
that we have two eyes. Two-eyed perception is re¬ ments that we integrate in a complex fashion to
ferred to as binocular, from bi, meaning “two,” give us our sense of up, down, right, and left
and ocula, meaning “eye.” Since (as we learned (Howard, 1982). The first one we call bodycentric
in Chapter 3) the best visual acuity is obtained direction. It uses as a reference location the midline
286 Chapter 10/Space

FIGURE 10-11 Convergence angle changes as a function of fixation distance. This may provide some
information about target distance.

of the body, an imaginary vertical line parallel to around the midline of the head, regardless of eye
the spine passing vertically through the naval. This position. The visual egocenter is the position in the
can be contrasted to headcentric direction where head that serves as our reference point for the de¬
the midline of the head is used as another reference termination of headcentric straight ahead. In some
location for right and left. This is an imaginary ver¬ respects, this is a very complex judgment, since we
tical line running up and down, centered on the seem to ignore the directions that the eyes are
nose. Of course, bodycentric and headcentric direc¬ pointing, and to compute a straight ahead that
tions are potentially different since it is possible to seems to be located in front of the middle of the
rotate your head independently of your body. The head. Researchers often refer to this compromise
distinction between these two aspects of direction direction as the location pointed to by a hypotheti¬
is shown in Figure 10-12. cal Cyclopean eye, a name derived from the myth¬
Most of the research on direction has concen¬ ical Greek giant Cyclops who had a single eye in
trated on one aspect of headcentric perception we the middle of his forehead. Demonstration Box 10-
can refer to as the visual straight ahead. Generally 3 shows how you can experience for yourself the
speaking, we tend to define our notion of “straight referring of the visual direction of the two eyes to
ahead” as a direction in front of us, oriented this common egocenter.
 Physiological Cues for Direction 287

FIGURE 10-12 The distinction between bodycentric and headcentric directions. The point B is straight
ahead of the body midline, whereas H is straight ahead of the midline of the head. Notice, however, that
these two straight ahead directions can be two different points in visual space.

Eye Movements and Direction ments, in the form of convergence and divergence,
conveyed information about the distances of ob¬
A number of variables affect our sense of the direc¬ jects. It would also seem reasonable that feedback
tion of objects. Stimulus factors are, of course, im¬ from eye movements could help in determining the
portant; the more stimuli available, the more stable visual direction of an object. At least two sources
our directional judgments, which accounts for the of eye movement information could be used to
fact that our ability to judge direction is much less compute the direction of an object in space. The
stable in the dark. Also, specific visual configura¬ first arises from the movement commands sent to
tions influence the judgment of direction. For ex¬ the eye muscles (the efference copy); the second
ample, if you are looking at a square or rectangle (the afference copy) arises from the eye movement
that is not exactly centered in your visual field, that is planned or intended (although not necessar¬
there will be a tendency for you to judge the ily executed, e.g., Hershberger, 1987). There is a
straight ahead direction as the center of this dis¬ good deal of controversy about which of these
placed square or rectangle. It is as if the perceptual types of movement information is more important
system confuses what is straight ahead of the ob¬ (e.g., Matin, 1982; Shebilske, 1976; Stark &
server with what is centered with respect to the Bridgeman, 1983), but it seems clear that eye
other contents in the visual field (Roelofs, 1935). movement information does play a role in localiz¬
However, nonvisual factors also play a role in di¬ ing targets in space.
rectional localization. There are many examples demonstrating how
In the previous section we saw how eye move¬ eye movements affect our localization of targets.
288 Chapter 10/Space

DEMONSTRATION BOX 10-3. The Common Visual Direction of the Two Eyes

You can experience how the visual directions of the

two eyes are referred to one common direction in the
center of the head. First, take a sheet of stiff card¬
board (20 X 27 cm will do) and place it in front of
the eyes as shown in the figure. Put a dot in the mid¬
dle of the far end of the cardboard and stare at it
while a friend marks the exact center position of each
of your pupils on the end of the cardboard closest to
your face. Next, draw lines from these marked points
until they form an angle, or V (as pictured). Finally,
reposition the cardboard in front of your face at a
point slightly below your eyes. Now, stare at the far
point where the two drawn lines intersect, and you
should see, in addition to the two lines you have
drawn, an additional, somewhat more shadowy line
running between them. This “new” line is the fusion
of the views of the two eyes and should appear to
point directly at a spot close to the midline of the
head. This demonstrates that, although the direction
of each eye’s view is different (as shown by the spa¬
tial separation between the two drawn lines converg¬
ing on the far point on which you are fixating), the
visual direction of the combined binocular view is re¬
ferred to a common point between the eyes. This
point is called the egocenter, or the Cyclopean eye.

The information about target localization seems to Festinger & Easton, 1974; Honda, 1984; Mack &
be associated with where the eyes are pointing at Herman, 1972).
any given moment. Imagine that the fovea of the The role of eye movements in target localiza¬
eye serves as a reference point. In the absence of tion can be quite subtle. For example, there is a
other information you localize a target as being in suggestion that the eye movement does not actually
the direction that your fovea is pointing when you have to be made, but that the eye movement you
try to look at it. This means that if you accurately compute in order to move the eye at some later
image the target on your fovea you will accurately time may bias your perception of direction (e.g.,
perceive the direction of the target. If, on the con¬ Coren, 1986). Furthermore, information may be
trary, you inaccurately point your eyes, such as passed from one response system to another. Thus,
when the eyes lag behind a moving target that you Mather and Fisk (1985) were able to show that the
are trying to track (see Chapter 13), you should information you obtain from looking at a target can
inaccurately localize the target. There is a good aid you in accurately pointing to the target. Con¬
deal of evidence that this is exactly what happens versely, information obtained from pointing can as¬
(e.g., Coren, Bradley, Hoenig & Girgus, 1975; sist the eyes in accurately looking at other targets.
 Binocular Depth Perception 289

DEMONSTRATION BOX 10-4. Sighting Dominance and the Straight Ahead Direction

The visual straight ahead may depend on a single eye point on the wall whether one or both eyes are
(Porac & Coren, 1976, 1981; Walls, 1951). Try the opened. The eye that maintains the alignment is
following demonstration to see how this works. Stand called the sighting-dominant eye. You will notice that
in front of a wall at a distance of about 3 m. Pick a regardless of which hand you use to perform the
point on the wall that is directly in front of you (a alignment, you will tend to line up a near (your fin¬
small crack or bump will do). Now, with both eyes gertip) and a distant (the point on the wall) target in
open, quickly stretch out your arm and align your fin¬ terms of the same eye. The presence of a sighting-
gertip with the point on the distant wall. When the dominant eye, and our tendency to make a straight
alignment has been completed, alternately close each ahead alignment in terms of this eye, indicates that
eye. You will find that the point on the distant wall the locus of the egocentric straight ahead direction
will shift out of alignment for one of the eyes. How¬ may be shifted toward the side of the sighting-domi¬
ever, the other eye will seem to be aligned with the nant eye.

Eye Dominance and Perceived 1981). Rather, it means that the location of the
Direction egocenter, or Cyclopean eye, is biased toward the
side of the sighting-dominant eye (Barbeito, 1981).
We have considered eye movements of either eye For instance, Porac and Coren (1986) tested ob¬
to be interchangeable, but there is some data that servers in a totally darkened room and had them set
suggest that the two eyes are not used equivalently a point of light so that it appeared to be visually
in the computation of visual direction. Before we straight ahead. Whether observers used only one or
consider this evidence, it is important to understand both eyes, they tended to set the point so that it
that there are some tasks we habitually do with one was closer to the side of the dominant eye, rather
eye. In sighting tasks where only one eye can be than midway between the two eyes.
used at a time (such as in looking through a tele¬
scope), 65 percent of all observers consistently use
their right eye, whereas the remainder consistently BINOCULAR DEPTH PERCEPTION
use their left (Coren, Porac & Duncan, 1981). The
preferred eye for such tasks is usually called the Except for vergence movements, all the cues to
sighting-dominant eye (Porac & Coren, 1981; depth we have discussed work just as well with one
Ruggieri, Cei, Ceridono & Bergerone, 1980). eye as they do with two. Nevertheless, having two
Demonstration Box 10-4 shows how you can deter¬ eyes confers a great advantage in trying to estimate
mine which eye is your sighting-dominant eye. relative depth. For example, in many common
For the purpose of our discussion of the per¬ tasks involving judgments of relative depth, such
ception of direction, sighting dominance is impor¬ as threading a needle, inserting items into slots, or
tant because the visual direction associated with even placing cards behind alphabetic dividers in a
straight ahead is more strongly influenced by the box, people perform up to 30 percent faster and
dominant eye (Porac & Coren, 1976, 1981). This more accurately when using both eyes than they do
certainly does not mean that only one eye is used with one eye alone (Sheedy, Bailey, Burl & Bass,
to determine visual direction (Ono & Weber, 1986). Demonstration Box 10-5 allows you to see
290 Chapter 10/Space

DEMONSTRATION BOX 10-5. Binocular versus Monocular Depth Perception

For this demonstration, which was suggested by Sek- open. You will probably find that you were much
uler and Blake (1985), you will need two sharp pen¬ more accurate with both eyes open than with one eye
cils. Hold one in each hand at a relaxed arm’s length, alone. However, not all of you will show this im¬
so that their tips are pointed toward each other but provement with binocular viewing, since between 5
separated by about 10 cm. Now close one eye, and and 10 percent of the population do not have stereop-
slowly bring the two pencils together so that the two sis, which is the term used to describe the ability to
points touch each other. Now try this with both eyes see depth based on binocular disparity.

how much better your depth perception is under ent depending on whether the unfused image is in
binocular (two-eyed) conditions, as compared to front of or in back of the target you fixated. In the
monocular (one-eyed) viewing. demonstration we defined these patterns as
crossed versus uncrossed disparity. Objects more
distant than the point of fixation are seen in un¬
Cues for Stereopsis
crossed disparity, whereas closer objects are seen
The cues for binocular depth perception (stereop¬ with crossed disparity. Hence, we can use the type
sis) depend on the fact that, in humans and other of double image as a cue to relative distance.
animals, the two eyes are horizontally separated. In Only objects at about the same distance as the tar¬
humans, the distance between the two pupils may get we are fixating will be fused and seen singly.
be up to 6.5 cm. Because of this separation, each When we map out all the points where targets are
eye has a different direction of view and, hence, a at about the same convergence or fixation distance
different image of the world. We call the differ¬ in visual space, we trace out an imaginary curved
ences between the two eye’s images binocular dis¬ plane called the horopter. A narrow region on ei¬
parity. You can see how different the images may ther side of this hypothetical plane includes all
be by following the instructions in Demonstration points in visual space that are fused into single
Box 10-6. images. It is called Panum’s area. Figure 10-13
The process by which we merge these dispa¬ contains a diagram of the horopter and Panum’s
rate images into a single unified percept is called area. The size and shape of Panum’s area actually
fusion. As a process, however, fusion is fairly lim¬ changes a bit with varying fixation distances.
ited in its range of operation and many parts of the However, it always remains the zone in the visual
total visual image do not fuse. This failure of the field where the disparate images are seen as fused
two eyes’ views to merge completely gives rise to into a single object.
double vision, or diplopia. Under normal viewing The process of fusion has also been looked at
conditions you are usually not consciously aware of in terms of corresponding retinal points. These
this diplopia, however you can readily learn to see are areas on the retina that represent a common di¬
the double images in the unfused portion of the vi¬ rection in the visual areas of the brain. The foveas
sual field. Demonstration Box 10-7 shows how this of each eye are corresponding retinal points and,
is done. according to this conceptualization, the horopter rep¬
In Demonstration Box 10-7, you should have resents the zone in visual space that stimulates corre¬
noticed that the pattern of double images is differ¬ sponding retinal points for one fixation distance.
 Binocular Depth Perception 291

DEMONSTRATION BOX 10-6. Binocular Disparity

You can see the difference between the views of your right eye open, it appears angled toward the left; with
eyes by holding a pencil up near your nose, as shown. the left eye open, it appears angled toward the right.
The tip of the pencil should be toward you and angled With both eyes open, the fused view is of a pencil
slightly downward. Now alternately close each eye. straight ahead of your nose.
The pencil seems to swing back and forth. With your

The Process of Stereopsis ings or photographs. Thus, in Figure 10-14A, we

have two rods at different distances from the ob¬
In the 1830s two physicists, Charles Wheatstone server. If we drew the image each eye sees, we
and Sir David Brewster, independently created a would get something like Figure 10-14B. Notice
technique to recreate the impression of depth from that the images are disparate, since the rods are
flat pictures using only the binocular disparity cue more widely separated in the right eye’s image than
(see Wade, 1984). Basically, this involves recreat¬ in the image in the left eye. Now the resulting im¬
ing the disparate views each eye would see and ages are viewed in an optical instrument known as
representing them to the eyes in the form of draw¬ a stereoscope, which places different stimuli into
292 Chapter 10/Space

FIGURE 10-13 The horopter and Panum’s area for one fixation distance. The regions of fusion and
disparate images are shown. Crossed disparity is present at distances closer to the observer than the
fixation distance; uncrossed disparity is present beyond the fixation distance. The presence of disparate
images may provide a cue to distance.

the two eyes simultaneously, as shown in Figure Mayhew & Frisby, 1980). From a physiological
10-14C. When this is done, the disparate images viewpoint, however, the initial breakthroughs in
fuse and the objects are seen as if they were an understanding began in the late 1960s, when
actual three-dimensional scene. For a period of investigators began to find disparity-tuned detec¬
time, every Victorian living room had a stereo¬ tors in the visual cortex of the cat (see Petti¬
scope and a set of travel pictures of famous places grew, 1978a).
that had been taken using a two-lensed camera. These detectors are cells that are finely tuned
This produced the “visual magic” of depth from to small differences in the relative horizontal place¬
flat images. Unfortunately, the trend toward more ment of images in the two eyes (Bishop, 1981). For
prurient photographic subjects relegated the stereo¬ example, suppose there is no disparity in the im¬
scope to unsavory establishments, although it did ages of the two eyes and a particular cell responds
not end the interest in stereoscopic depth. maximally to this condition. This particular neuron
Understanding how stereopsis is achieved is would represent a spatial position that lies on the
more difficult than setting up the conditions that horopter, or the zone of fused images in external
allow us to see binocular depth. There have been space. In a like manner, other neurons may be
several computational approaches to this problem, tuned to particular disparities that represent loca¬
most of which involve selecting a particular loca¬ tions in space that lie in front of or behind the
tion in space, then comparing and computing the horopter. There is even evidence that there are
relative positions of parts of the images, and on the different cells that respond to crossed versus un¬
basis of this deriving the relative depth of the ob¬ crossed disparities (Mustillo, 1985). Each of these
jects being viewed (e.g., Marr & Poggio, 1979; neurons seems to be tuned to only one particular
 Binocular Depth Perception 293

FIGURE 10-14 Disparate retinal images. (A) The two retinal images of a scene are different because the
two eyes view the world from slightly different directions. (B) A stereogram is a flat representation that
mimics the differences between the two retinal images. (C) A stereogram is viewed in a stereoscope that
allows for the separate but simultaneous stimulation of the two eyes. The phenomenon is called
stereopsis.

disparity value. Some seem to be responsive only tions in Demonstration Box 10-8, you will see a
to stimuli that lie close to the horopter; others are dotted square floating in front of the background of
tuned to large disparities or to stimuli that lie far random dots. This perception comes about because
from the horopter (Aslin & Dumais, 1980; De- of disparity cues built into the dot patterns. Figure
Valois & DeValois, 1980). This means that rather 10-15 shows how this disparity, which consists of
large populations of cortical neurons are needed to a horizontal shift in a group of these random dots,
represent all the possible disparity values in the vi¬ is created.
sual scene. You might suppose that since there is disparity
It has been suggested that the existence of dis¬ built into the random-dot stimulus, neurons tuned
parity-tuned detectors is not enough to explain ster¬ for such information should be capable of detecting
eoscopic depth perception. The problem arises in depth from these arrays. The problem comes about
situations like that in Demonstration Box 10-8, because the stimulus is composed of identical dots,
which contains a random-dot stereoscopic display. rather than discrete and identifiable contours. If
These displays were first introduced by Julesz stereopsis is based on the action of disparity-tuned
(e.g., 1971), who used them to demonstrate the no¬ detectors, each of which responds to one disparity
tion of global stereopsis. If you follow the instruc¬ value in the array, any dot potentially could be
294 Chapter 10/Space

DEMONSTRATION BOX 10-7. Double Images and Disparity

Find a piece of transparent colored material like cel¬ present in the visual field; however, we are usually
lophane (any hue will work). Place it before your not aware of them unless forced to attend to them as
right eye. If you wear glasses you can affix it to the in this demonstration.
frame over the lens in front of your right eye; if Once you have become comfortable with this pro¬
not, use a piece of tape to hold it to your forehead. cedure, fixate the near target and then close your right
Now align two index fingers directly in front of your eye. You should notice that the image of the far target
nose with the closer finger about 10-20 cm from your (the uncolored image) appears to lie to the left of the
nose and the farther finger about 8 cm behind the nearer, fixated object. Now close the left eye and
closer one. open the right and you will notice the opposite. The
Now that you have arranged the appropriate situa¬ image of the far target (the colored image) now ap¬
tion, fixate your nearer finger. However, simulta¬ pears to lie to the right of the closer, fixated target.
neously try to pay attention to what the far finger The fact that the right eye is seeing the right disparate
looks like. This is a pretty difficult feat to accomplish image and the left eye is seeing the left disparate im¬
at first, but with practice you should be able to fixate age means that when both eyes are open the far target
one target while simultaneously paying attention to is seen in uncrossed disparity. The opposite will
what is going on beyond the fixated area. When you happen if you change your fixation to the far target.
fixate the near target, you will notice that two images Now the closer object appears as diplopic (double). If
of the far target will be seen. The fact that one eye is you once again alternately close each eye, you will
viewing the image through a colored filter should help notice that the right eye is now seeing the image that
make the presence of double images beyond the fixa¬ lies to the left of the fixated target (the colored im¬
tion point more apparent. If you switch your fixation age), while the left eye is viewing the image that
to the farther object, the closer of the two targets will lies to the right. In the case of double images that
appear as a double image. Targets that lie away from lie closer to us than the point of fixation, we have
the area surrounding the point of fixation are not a situation of crossed disparity. As the text
fused into a single image. They produce disparate explains, these differences in disparity may be a cue
retinal images. Disparate, unfused images are always to distance.

combined with any other dot, with each of the rons working together in this fashion, only one
many possible combinations producing a different depth solution would be common to this facilita-
depth. For this reason there must be a method of tory-inhibitory process and only one global stereo¬
eliminating or avoiding false combinations and se¬ scopic view would be seen (Julesz & Schumer,
lecting only correct disparity pairs. Marr (1982) has 1981). In any event, the solution used by the neural
suggested a very sophisticated computational system probably involves a good deal of computa¬
scheme that would allow one to program a com¬ tion, since it often takes a good deal of time for
puter to do exactly this. However, some investiga¬ observers to actually see the depth in random-
tors have looked for physiological solutions that dot stereograms. It has been suggested that ste¬
assume that detectors tuned to the same disparity reoscopic information may be integrated over
cooperate, whereas those tuned to different dispar¬ intervals of up to a half second in length (White
ities inhibit each other. With a population of neu¬ & Odom, 1985).
 Interaction of Cues 295

INTERACTION OF CUES to confirm this relationship. Chaotic and conflict¬

ing cues, such as those shown in William Hogarth’s
Although it is the case that each of the cues for 1754 engraving False Perspective (Figure 10-16),
depth we have discussed is sufficient, by itself, to virtually never occur in “real world” settings.
give the conscious impression of a three-dimen¬ How do cues for depth combine? Jameson and
sional arrangement in space, the accuracy of our Hurvich (1959) suggested that an observer’s sensi¬
perception of distance often depends on the interac¬ tivity to a difference in distance when several cues
tion of a number of cues. Under normal conditions, are available is approximately the arithmetic sum
if a cue such as interposition suggests that your of the sensitivities obtained with each cue alone.
friend Fred is standing closer to you than your friend This has been demonstrated experimentally by van
Maria, other cues, such as relative size, height der Meer (1979), who found that information from
in the plane, and binocular disparity, will tend binocular disparity and linear perspective added
DEMONSTRATION BOX 10-8. Random-Dot Stereograms and Global Stereopsis

You may demonstrate how depth cues can bring about overlap and the frames around the outside seem to be
the perception of a binocular form by using the ac¬ at the same distance. Viewing it in this way, you will
companying figure. You will need a pocket mirror, see a square form emerge, floating above the back¬
which should be placed on the dotted center line of ground, created completely by the depth cue of bin¬
Figure B while you hold your head as shown in Fig¬ ocular disparity. Notice that this square simply can’t
ure A. Adjust the images until the two views seem to be seen in either monocular view alone.
 Interaction of Cues 297

1 0 1 0 1 0 0 1 0 1 1 0 1 0 1 0 0 1 0 1

1 0 0 1 0 1 0 1 0 0 1 0 0 1 0 1 0 1 0 0

0 0 1 1 0 1 1 0 1 0 0 0 1 1 0 1 1 0 1 0

0 1 0 Y A A B B 0 1 0 1 0 A A B B X 0 1

1 1 1 X B A B A 0 1 1 1 1 B A B A Y 0 1

0 0 1 X A A B A 1 0 0 0 1 A A B A Y 1 0

1 1 1 Y B B A B
v
0 1 1 1
V' B B A B X 0 1

1 0 0 1 1 0 1 1 1 /0 1 1 0 1 1 0 1
\° °/
1 1 0 0 1 1 0 i
V V 0 0 1 1 0 1 1 1

0 1 0 0 0 1 1 i 1 cN /0 1 0 0 0 1 1 1 1 0

Displaced center square

FIGURE 10-15 Figure A is a random-dot stereogram. Figure B shows how A is constructed, and C
illustrates that a central square is seen floating above the background when the two views are combined
in a stereoscope. (From B. Julesz, Foundations of Cyclopean Perception. Copyright 1971 by the University of Chicago Press.)
298 Chapter 10/Space

FIGURE 10-16 Ambiguity of depth cues gives a confusing, difficult interpretation to a scene, as shown
in Hogarth’s 1754 engraving False Perspective. The more you study this figure, the more contradictory
depth cues you find.
 Development of Space Perception 299

together in the final judgment of perceived depth. depth perception, in which our ability to decipher
Although we cannot prove that there is a simple and to combine depth cues may be a learned ability
addition of information from various cues to dis¬ that becomes more efficient with experience. It is
tance, it is certainly the case that the more cues on the issue of whether there is a learned compo¬
available, and the more consistent they are, the nent in the perception of depth that direct and con¬
stronger the perception of depth (Berbaum, Tharp structive theories of perception have focused much
& Mroczek, 1983). of their argument. In the older literature this has
Another important aspect of the interaction of been referred to as the linativist versus empiricist”
depth cues is the fact that absolute and relative question (Hochberg, 1972). The nativists argue that
depth cues may help calibrate each other. For ex¬ these perceptual abilities and our capacity to use
ample, binocular disparity is most effective as a them are inborn and automatically invoked by stim¬
relative depth cue, indicating whether one target is ulation, whereas the empiricists maintain that inter¬
nearer or farther than another, rather than how action with the world, through which we learn
close they both are to an observer. The geometry about its properties and organization, is crucial to
of the situation, however, is complicated by the our spatial awareness and abilities. Let us now con¬
fact that if we keep the relative distance between sider this issue briefly.
two targets constant but move both targets farther
away from you, the amount of disparity in the im¬
DEVELOPMENT OF SPACE
age grows smaller. Nonetheless, our perception of
PERCEPTION
the relative difference in the depth of these targets
remains the same, a phenomenon sometimes called
One of the most common ways to assess whether
stereoscopic depth constancy (Ono & Comerford,
there is a constructive aspect to space perception is
1977). Apparently, this adjustment of the relative
to observe the behavior of young organisms when
depth obtained from binocular disparity comes
they are placed in situations that call on their abil¬
about as a consequence of using other cues for ab¬
ities to perceive distance or direction. Since infants
solute distance. In other words, we recalibrate our
and young animals have limited experience with
interpretation of the magnitude of depth difference
the world, their abilities to deal with such situa¬
signaled by a particular amount of binocular dis¬
tions should shed some light on the role of inborn
parity by using other information about absolute
versus learned components in the perception of
depth (Cormack, 1984).
visual space.
The phenomenon of stereoscopic depth con¬
stancy reveals a problem with depth perception
Species Differences
originally noticed by Brunswick (1952, 1956). He
argued that each separate distance and depth cue is The evidence is quite clear that in certain simpler
ambiguous under certain conditions, such as those animals the perception of direction and distance is
depicted in Figure 10-16, and that sometimes any inborn. For example, in salamanders it is possible
single cue may lead us to an incorrect interpretation surgically to rotate the eye 180 deg, thus inverting
of depth or distance within a scene. The importance the retina. When this is done, animals consistently
we assign to a particular cue will be determined by swim and snap in the opposite direction when pre¬
its reliability and our past experience with its ac¬ sented with a food lure (Sperry, 1943). Since the
curacy. Simply shifting our attention to a given same results occur when similar operations are per¬
depth cue seems to increase its power in determin¬ formed during the animals’ embryonic stage, it is
ing how we perceive three-dimensional space (Ka- clear that visual direction is related innately to the
wabata, 1986). If we accept such an argument, we location of retinal stimulation in this species
are, in effect, accepting a constructive theory of (Stone, 1960). Similarly, immediately after birth
300 Chapter 10/Space

chicks peck at small objects with reasonable accu¬ A simple and popular procedure for measuring
racy. When experimenters optically displaced the depth perception in young animals uses an appa¬
images of the targets to one side (using special len¬ ratus called the visual cliff (Walk & Gibson,
ses attached to hoods), the chicks proceeded to 1961). A diagram of a typical visual cliff arrange¬
peck systematically to one side. This pattern of in¬ ment is shown in Figure 10-17. Basically it
accuracy showed little improvement over time, consists of two sections, divided by a “start
suggesting that this response to the apparent direc¬ platform.’’ Each section provides a different depth
tion of stimuli was not changeable by experience impression. The “shallow” side is a piece of glass
(Hess, 1950). In higher animals, such as mammals, that lies directly over a patterned surface. The
experience may play a larger role. “deep” side has the same type of patterned sur¬
To allow more exact determination of which face, but looks like a sharp drop since the surface
factors may be influenced by experience, investi¬ is placed at some distance below the glass. For
gators frequently use controlled-rearing procedures, testing, a young animal is placed on the central
such as rearing an animal in total darkness from starting platform that separates the apparently shal¬
birth until testing. Such dark-rearing eliminates all low and deep surfaces. It is assumed that from this
externally-generated visual experience. If experi¬ position the animal can see that the shallow side is
ence with various visual depth cues is necessary for safe, whereas the deep side, with its simulated cliff¬
the development of normal depth perception, these like drop-off, would be perceived as being danger¬
dark-reared animals should have measurable defi¬ ous. Investigators make the presumption that if the
cits when required to respond to distance cues. If animal consistently chooses the shallow over the
depth perception simply matures as the animal deep side, then it can perceive the difference in ap¬
ages, then restricting the animal’s visual experience parent depth and is attempting to avoid a fall.
should not affect its behavior and the only impor¬ A number of different types of animals have
tant variable should be its chronological age. been tested on the visual cliff, including rats.

Deep pattern seen start board Shallow

FIGURE 10-17 The visual cliff.

 Development of Space Perception 301

chickens, turtles, goats, sheep, pigs, cats, dogs, Tees (1974) shows how deprivation of visual ex¬
monkeys, and humans (Walk & Gibson, 1961). In perience during a sensitive period can affect later
all cases, even when testing very young animals, depth perception. Dark-reared rats were compared
there was a preference for the shallow over the to light-reared rats on their preference for the deep
deep side of the cliff. There were some interesting versus the shallow side of the visual cliff. The
species differences, however, which seemed to be amount of time that the animals were dark-reared
related to the habitat features of the natural envi¬ was varied, and in addition, the strength of the
ronment for the various species (Sloane, Shea, depth information was varied by varying the dis¬
Procter & Dewsbury, 1978). For instance, aquatic tance to the bottom of the deep side of the visual
animals, such as certain turtles, did not show the cliff. In this study, the age of the animal, the
marked preference for the shallow side that the amount of distance information, and the amount of
other, more landbound species displayed. Perhaps visual experience all interacted. It was only among
the survival value of cliff avoidance may not be as animals that had been dark-reared for a compara¬
pronounced in animals that spend much of their tively long time (60 to 90 days) that the effects of
lives swimming, since changes in depth of water rearing conditions revealed themselves. For these
are not as perilous as sudden sharp drops on land. animals, although depth could be discriminated
when the drop-off was large, there was an insensi¬
tivity to weaker distance cues. These data indicate
Experience and Depth Perception
that there may be inborn components in the ability
Although the cross-species differences observed in of rats to discriminate depth on the visual cliff.
depth perception are of interest, the visual cliff ap¬ These are probably sharpened through experience
paratus has been used primarily to generate data with depth cues in the environment, a finding sup¬
concerning the development of depth perception. A ported by other research as well (e.g., Kaye,
combination of controlled rearing followed by ob¬ Mitchell & Cynader, 1982).
servations of behavior on the visual cliff has been The developmental time course and the effects
the experimental technique most commonly used in of experience seem to be different for the various
animal studies. In general, the findings have sug¬ depth cues. Binocular depth perception develops
gested that experience and innate factors interact to quite early, since evidence for the use of binocular
produce an animal’s ability to perceive depth. For disparity for a depth cue may be found in 3- or 4-
example, when cats or rats are initially reared in month-old infants (Birch, Shimojo & Held, 1985;
the dark they show little depth discrimination on Braddick, Atkinson, Julesz, Kropfl, Bodis-Wollner
the visual cliff when first tested. However, as they & Raab, 1980; Hutz & Bechtoldt, 1980; Petrig, Ju¬
receive more and more experience in a lighted lesz, Kropfl, Baumgartner & Anliker, 1981). In¬
world their depth discrimination rapidly improves fants at this age are too young to crawl and cannot
until they are indistinguishable from normally be tested on the visual cliff, so other techniques
reared animals (Walk & Gibson, 1961; Tees & must be used. For example, Fox, Aslin, Shea, and
Midgley, 1978). Dumais (1980) presented random-dot stereograms,
When the visual experience takes place also similar to those in Figure 10-15, to infants between
seems important. There seem to be sensitive peri¬ 2 and 5 months of age. When viewed with special
ods in an animal’s development, referring to the glasses, these infants saw a square floating in front
fact that there are particular ages and particular du¬ of a background only if they could combine the
rations of time when depriving an animal of a par¬ disparate information from the two eyes’ views.
ticular type of visual experience may produce the (You saw in Demonstration Box 10-8 that such
largest perceptual deficits (Aslin, 1981b; Mitchell, patterns are meaningless unless you can make use
1981; Timney, 1985). For instance, a study by of the disparity cues hidden in each monocular
302 Chapter 10/Space

view.) This square was made to move. When in¬ derstood unless we allow some components to be
fants could see the square in depth, they tended to explained by inborn factors, whereas others may
follow it with their eyes. An alternative technique require active experience to emerge. These studies
for measuring depth perception in young infants suggest that adhering strictly to either a direct per¬
capitalizes on the fact that infants tend to reach for ception or constructive perception viewpoint might
objects that appear to be near them (Granrud, be too limiting. Innate components of perception
Yonas & Pettersen, 1984). Results using this tech¬ must mature, and certain types of experience can
nique verified the finding that binocular depth per¬ help or hinder the achievement of a high level of
ception is present by 4 months of age (Granrud, perceptual functioning. In much the same way,
1986), suggesting either an innate or a rapidly some aspects of depth perception seem to be given
learned process. This may be contrasted to earlier directly, and others require memory and experience
animal work, which suggested that the use of mon¬ to allow them to function properly.
ocular cues for depth is much more dependent on
specific experience (Eichengreen, Coren & Nach-
mias, 1966).
The ability to use kinetic depth information GLOSSARY
seems to develop at about the same time that the
ability to use binocular depth information appears, The following definitions are specific to this book.
at roughly 3 to 5 months of age (Kellman, 1984; Absolute distance The distance of an object from
Owsley, 1983; Yonas & Granrud, 1985b). One bi¬ the observer.
ologically important aspect of depth perception, L^ccommodation The change in focus of the lens of
the eye, which may serve as a depth cue.
namely sensitivity to information about object mo¬
Aerial perspective A distance cue in which objects
tion toward the body (which may indicate an im¬
appear hazy, less distinct, and bluer the farther away
pending collision), seems to be present at an even they are, because of the interaction of light with dust and
earlier age. It has been measured at ages as young moisture particles in the air.
as 2 to 3 weeks (Ball & Vurpillot, 1976; Yonas, Afference copy Information about eye movement
1981). Even at this young age, infants will blink arising from an intended or planned eye movement.
their.eyes when presented with an object that seems Binocular Pertaining to two eyes.
to be moving closer and seems to be growing close "''''ffinocular disparity The difference in the monocular
enough to strike them in the face. views of the two eyes.
Studies that have looked at sensitivity to pic¬ Bodycentric Direction with the midline of the body
used as a reference point.
torial depth cues have shown a slower developmen¬
Computational theories Theories involving the
tal process, by about 3 to 4 months. For example,
presumption that certain perceived qualities must be
several studies have shown that 6- to 7-month-old
computed from stimulus information and that these com¬
infants respond to linear perspective information putations can be precisely described mathematically.
(Kaufman, Maland & Yonas, 1981; Yonas, Constructive theories Theories maintaining that
Cleaves & Pettersen, 1978) or to familiar size perception may involve the integration of several sources
(Granrud, Haake & Yonas, 1985). Use of more of information, and may be affected by cognitive factors
complex cues, such as the perception of relative and experience.

depth in a picture, based on the direction of shad¬ ^Convergence The inward rotation of both eyes to¬
ward the nose as a fixated object becomes closer.
ows cast in the pictorial representation, may not
•/Corresponding retinal points Areas in the two ret¬
appear until the age of 3 years (Yonas, Goldsmith
inas that share a common visual direction when the two
& Hallstrom, 1978). monocular inputs are processed in the brain.
Developmental studies indicate that the per¬ ‘''Crossed disparity A cue for relative distance,
ception of depth and distance cannot be fully un¬ where, when there are double images, the unfused image
 Glossary 303

in the right eye appears on the left and that in the left Object-relative localizations The estimation of the
eye on the right. relative positions of objects (other than the observer)
'-'Cues Features of visual stimuli that prompt the per¬ within the environment.
ception of depth or distance. Overlay See Interposition.
Cyclopean eye An imaginary point midway between Panum’s area The region around the horopter where
the eyes thought to be used as a reference point for the all images in space are fused.
straight ahead direction.
Physiological cues See Structural cues.
Diplopia Double vision.
-—Pictorial depth cues Cues for distance that can be
-—Direct perception The idea that all the information found in photographs and pictures.
needed for the final conscious percept is in the stimulus Reduction conditions An experimental procedure in
array and is apprehended directly. which an attempt is made to eliminate or reduce most
Divergence The outward rotation of both eyes away depth cues.
from the nose as a fixated object becomes more distant. Relative brightness A depth cue in which the
Efference copy Information about eye movement brighter of two otherwise identical objects will be seen
arising from the commands issued to the ocular muscu¬ as closer.
lature to rotate the eye. —Relative distance Distances of objects relative to
Egocenter The position in the head that serves as the one another.
reference point for the determination of headcentric Retinal image size A potential depth cue based on
straight ahead. the size of the image on the retina.
Egocentric localization The awareness of where our Sensitive period An age range during which the de¬
bodies are positioned relative to other objects in the ex¬ velopment of a perceptual ability may be strongly influ¬
ternal environment. enced by the presence or absence of relevant stimulation
Familiar size The known or remembered size of ob¬ or experience.
jects. Sighting-dominant eye The eye whose use is pre¬
Fusion The process by which disparate views are ferred in monocular tasks such as looking through a tele¬
synthesized into one percept. scope.
Global stereopsis Stereopsis that is not dependent Stereopsis The ability to see depth based solely on
on local contour elements (e.g., as in random-dot ster¬ the disparity of the two retinal images.
eograms) . Stereoscope An optical instrument enabling two dif¬
Headcentric Direction judged using the midline of ferent images to stimulate the two eyes simultaneously to
the head as a reference point. produce an effect of depth.
Height in the (picture) plane A cue for distance re¬ Structural cues Depth and distance cues arising
ferring to where an object is relative to the horizon. from adjustments of the eye(s) in interaction with the vi¬
-—Horopter An imaginary plane in external space used sual stimulus.
to describe the region of fused images. "fexture gradient Distance cue based on variations
Intelligent perception The presumption that cogni¬ in surface texture as a function of distance from the ob¬
tive processes and experience can affect perception. server.
—"Interposition The depth cue based on the blocking "'‘Uncrossed disparity A cue for relative distance in
of an object from view by another closer object. which, when there are double images, the unfused image
in the right eye appears on the right and that in the left
Kinetic depth effect The perception of the three-di¬
eye on the left.
mensional qualities of an object based on cues generated
by the object’s motion. Vanishing point In linear perspective a point on the
horizon at which converging parallel lines seem to meet.
c-—Linear perspective The apparent convergence of
physically parallel lines as they recede into the distance. Vergence movements Movements of the eyes in
which the two eyes move together but in opposite direc¬
-—■-Monocular cues Depth cues requiring only one eye
tions, either converging or diverging.
to be used.
Visual cliff A table with shallow and deep sides over¬
Motion parallax The apparent relative motion of
laid by a sheet of glass; used for measuring depth per¬
objects in the visual field as the observer moves his head
ception in young animals.
or body.
CHAPTER
11
Form
THE VISUAL FIELD
CONTOUR
Contour and Change
Contour Emergence
THE PERCEPTUAL OBJECT
Figure and Ground
Figural Organization
Information, Symmetry, and Good
Figures
Texture Segregation
Spatial Frequencies and Figural
Organization
Features
Feature Integration Theory
OBJECT RECOGNITION AND
IDENTIFICATION
Recognition versus Identification
Data-driven versus Conceptually-driven
Processing
Global versus Local Processing
Integral versus Separable Stimuli
Context and Identification
THEORIES OF OBJECT IDENTIFICATION
Pandemonium
Identification-by-Components
Computational Approaches
306 Chapter 11/Form

^lA/hen I lift my eyes from the paper on eyes at any given moment is called the visual field.
V V which I am writing I see the chairs and Figure 11-1 portrays the visual field as seen by the
tables and walls of my room. . . . I see, from my left eye of the physicist/psychologist Ernst Mach,
window, trees and meadows, and horses and oxen, as he lay on the couch in his study in Prague,
and distant hills. I see each of its proper size, of its Czechoslovakia, sometime around 1885. Reflected
proper form, and at its proper distance; and these light from the visual field forms a retinal image,
particulars appear as immediate transformations of consisting of a two-dimensional distribution of light
the eye. . . . How, then, is it that we receive ac¬ of various intensities and wavelengths on his retina.
curate information, by the eye, of size and shape This distribution carries information about objects
and distance?” (Mill, 1829, p. 97) in the environment, because light is altered
Look around you, as John Stuart Mill did. uniquely by each object it falls on. Some of the
You will be struck, as he was, with how filled the light may be absorbed before being reflected to the
perceptual world is with objects. When you enter eye; this will alter the intensity of the image on the
your room you see a desk, a chair, some books, retina, so that, for example, his shoes will look
and so forth, not patches of light, brown, gray, red, darker than the couch on which he rests. When
and green, which is the actual stimulus on your ret¬ light of certain wavelengths is selectively absorbed,
ina. You not only see these patches as objects but areas of the image appear colored (see Chapter 5);
also identify them as members of particular classes thus, if one of his books absorbs all but light of the
of objects, and can recall whether you have seen
them before. This is a common everyday form of
perceptual achievement that we each do automati¬
cally and easily, yet describing how we manage to
accomplish this has turned out to be very difficult.
In this chapter, we tell you something of what we
now know about the perception of form (Mill’s
other questions, about size and distance, are dealt
with in Chapters 10 and 14, respectively). We de¬
scribe first the information about the world that is
present in the distribution of light on our retinas,
and then we describe how the visual system trans¬
forms that into ‘‘perceptual objects.” Finally, we
discuss how we recognize and identify these per¬
ceptual objects. In several places in the chapter, the
traditional psychological approach to the perception
of form is contrasted with the newer “computa¬
tional” approach.

THE VISUAL FIELD

The eye receives information in the form of re¬

flected light from the environment. The total of all FIGURE 11-1 The visual field as seen by the left
the parts of the environment that stimulate your eye of Ernst Mach. (Based on Mach, 1959/1886)
 Contour 307

longer wavelengths, the portion of the retinal image Blank out can occur in natural environments
that corresponds to it will contain only red-appear¬ as well. For instance, snow blindness is a kind of
ing light. The process of getting from the flat natural blank out, caused by the lack of contour in
patches of light that make up the retinal image to the retinal image when the visual field contains a
the world of objects about us is the major problem lot of snow and ice. The snow and ice scatter much
in the perception of visual form. of the light in all directions, and together are often
very uniform in texture, which creates a kind of
natural Ganzfeld. You can experience some of
CONTOUR these sensations for yourself by trying Demonstra¬
tion Box 11-1.
At the most basic level, the visual system divides For vision to occur, there must be not only
the visual field into shapes, sometimes called local variations in the intensity of light, such as
“blobs,” separated by contours. A contour is a contours, but also variation in the pattern of illu¬
place in the retinal image where the light intensity mination on the retina over time. This can be dem¬
(or wavelength composition) changes abruptly. Ex¬ onstrated using a stabilized retinal image. In this
amples of contours include a black line on a sheet technique, the retinal image is made to stay in one
of white paper, the edge of a blackboard, and the place on the retina, in spite of where or how the
outline of the moon against the night sky. Generally eye moves. Normally, the retinal image is contin¬
speaking, shapes are regions in the retinal image ually in motion, because of saccadic eye move¬
that are surrounded by contours. Under appropriate ments made as we look around the environment,
circumstances we may interpret these blobs in the smooth tracking movements, and drifting movements.
retinal image as objects in the real world; however, There are also microsaccades (tiny, involuntary eye
since this is an interpretation, at times we may be movements) that keep our eyes shivering in their
wrong. sockets. These microsaccades occur many times a
second, and cause the retinal image to shift from
place to place on the retina even when we are
Contour and Change
trying to hold our eyes completely still. The effect
Contours are the basic building blocks of visual of the microsaccades, combined with other forms
perception; in their absence we actually lose our of eye movement, is to constantly move any con¬
ability to see. We can demonstrate this by looking tours across a number of different retinal receptors.
into a Ganzfeld (German for “whole field”). A This causes temporal variations in the stimulation
Ganzfeld is a visual field that contains no abrupt of any given receptor (in the form of “on” times
luminance changes and thus no contours. When ob¬ followed by “off’ times) as contours are swept
servers look into a Ganzfeld, they usually report across the retina.
seeing “a shapeless fog that goes on forever.” Any When we do manage to stabilize the retinal
hint of color soon fades to gray, even if the entire image, usually by having an observer wear a spe¬
field is illuminated with, say, green or blue or red cial contact lens that moves with the eye and has a
light (Cohen, 1958). Many observers even experi¬ mirror or a tiny projector mounted on it, the stabi¬
ence perceptual blank out, a feeling that they can't lized image soon fades from view (e.g., Pritchard,
see, after prolonged viewing. This feeling quickly Heron & Hebb, 1960; Riggs, Ratliff, Comsweet &
disappears the instant any kind of luminance Cornsweet, 1953; Yarbus, 1967). The contours dis¬
change is introduced into the visual field (Avant, appear in chunks, and the color also fades away. If
1965; Cohen, 1958). we now flicker the stabilized image on and off, it
 308 Chapter 11/Form

DEMONSTRATION BOX 11-1. The Ganzfeld

Although Ganzfeld situations have been produced view will be flooded with diffuse, contourless light.
with elaborate laboratory equipment, there are several Stay in this position for a few minutes and monitor
simple ways to produce a Ganzfeld that will allow any changes or alterations in your conscious percep¬
you to experience this contourless field for yourself. tual experience. If the light originally had a tint, you
You can take a table tennis ball and cut it in half, will soon notice that the color will fade into a gray.
placing one half over each eye, or you can use two After a while you will suddenly feel that you cannot
white plastic spoons (like those probably available in see. This feeling of blindness is called “blank out.”
any campus cafeteria) to produce a Ganzfeld by plac¬ It seems that in the absence of contours in the field,
ing the bowl of a spoon over each eye as shown in vision ceases. If a friend now casts a shadow over
the accompanying figure. Direct your gaze toward a part of the field (say, with a pencil across the spoons),
light source (a fluorescent lamp, say) prior to placing vision will immediately return with the introduction
the objects before your eyes, so that your field of of this contour.

will reappear and, if the flicker rate is high enough, terns when the retinal image is stabilized, and its
it will not fade (Comsweet, 1956). Flickering the reappearance when the retinal image is flickered,
image has re-introduced the temporal changes in supports the idea that constant temporal change in
light intensity usually caused by the eye’s move¬ the stimulation of the individual receptors is a nec¬
ments. The disappearance of our perception of pat¬ essary condition for vision. Without this constant
 Contour 309

change, receptors soon cease to respond differently in two opposite contours appearing in the same
from neighboring receptors, hence we lose the sig¬ place on the retina separated by different intervals
nal that should be present if a contour is in the of time. Figure 11-2 contains a diagram of this sit¬
image (cf. Norwich, 1983). In the absence of sig¬ uation and also summarizes the results of the ex¬
nals indicating that a contour is present, much like periment, which were quite striking.
our experience in the Ganzfeld, our perception of For very short interstimulus intervals, less
patterns will fade and disappear from consciousness. than 100 msec, observers tended to see only a large
You may experience the reappearance of a sta¬ black disk, a combination of the disk and the ring.
bilized image that has disappeared by trying Dem¬ It was as if the two had merged, or added together.
onstration Box 3-3 (see Chapter 3) again. In that For long interstimulus intervals, greater than 200
demonstration you mapped the pattern of blood msec, observers saw both of the stimuli clearly, the
vessels that lie above your retina. Ordinarily you disk before the ring. But for interstimulus intervals
don’t see these blood vessels because they create a between 100 and 200 msec, observers tended to see
stabilized image on your retina as light passes only the ring. It was as if the disk had not been
through them. They are always in the same place presented at all, although it stimulated the retina
so they create no temporal changes in stimulation exactly as it had at other interstimulus intervals.
of the receptors. By moving a flashlight placed at Wemer interpreted this masking of the disk by the
the comer of the eye, however, you cause their ring to mean that the later-appearing contour had
shadows to move across the retina, making them interfered with the processing of the earlier one,
visible. You will note that as soon as you stop preventing its appearance in consciousness. This
moving the flashlight (thus stopping the temporal kind of masking is called backward masking since
change), the blood vessels disappear, since their the mask appeared after the target and seemed to
image is now stabilized again. be acting backward in time. A more general term,
used to refer to any situation in which contours on
adjacent parts of the retina interfere with one an¬
Contour Emergence
other, is metacontrast.
Temporal factors are important to our perception of Studies of masking have helped shed light on
contours in another way. We are not immediately several aspects of form perception, such as how con¬
aware of a contour when it appears on our retinas. tours interact in the visual system (e.g., Kahneman,
It takes some time for the contour information to 1968; Lefton, 1973). One interesting suggestion is
be processed and the perceived contour to emerge that for each contour detected the visual system
into our consciousness. If anything interferes with produces both excitation, to indicate the existence
that processing before we see the contour, it may of the contour, and inhibition of contours detected
never become visible. The first demonstration of nearby. If two contours are detected closely enough
this was by Werner (1935) and it created an exper¬ in time and are spatially close to each another, then
imental technique that is now used to investigate the inhibition produced for one (e.g., the ring) may
many other phenomena. Wemer presented observ¬ cancel the excitation produced for the other (e.g., the
ers with a brief view of a black disk on a white disk; see Weisstein, 1968; Weisstein, Ozog & Szoc,
background (the target) and then, after a variable 1975). Another suggestion is that if the visual sys¬
amount of time (called the interstimulus interval), tem does indeed decompose the retinal image into
presented a brief view of a black ring (the mask) in spatial frequency components (as suggested in
the same place in the visual field. The inner con¬ Chapter 4), different components may be processed
tour of the ring was exactly the same size as the in different channels (see Graham, 1981). Because
outer contour of the disk, so the procedure resulted of this separation, contours composed of similar
310 Chapter 11/Form

STIMULUS SEQUENCE Interstimulus

interval
On

Target Mask
Off
Time
Mask
Target

WHAT THE OBSERVER SEES

Interstimulus interval Interstimulus interval Interstimulus interval
less than 100 msec between 100 and 200 msec greater than 200 msec

Only

FIGURE 11-2 A typical metacontrast experiment. Notice that when the interstimulus interval is of the
appropriate length, the target is often not seen at all.

spatial frequency components will mask one an¬ image. The miraculous thing is that the perceptual
other, whereas those composed of different com¬ objects so closely resemble what physics tells us is
ponents will not (Weisstein, Harris, Berbaum, the nature of the “real” objects “out there” in the
Tangney & Williams, 1977). world. An apple to our consciousness is very like
an apple “out there.” How does this come about?
The best guess at this time is that the perceptual
THE PERCEPTUAL OBJECT object is constructed in focal attention from a group
of features (or attributes) detected at a particular
The quotation from Mill beginning this chapter em¬ spatial location. In what follows we discuss how
phasizes that the visual world is filled with objects. this happens.
That is, under normal viewing conditions, our vi¬
sual systems operate to produce perceptual objects
Figure and Ground
from the array of blobs and contours contained in
the distribution of light on the retina. There are no The simplest perceptual object is a two-dimensional
perceptual objects on the retina, they exist only in figure on a two-dimensional ground. A figure is
our minds, and only as the result of many levels of simply an integrated group of contours. A shape
processing and interpretation applied to the retinal can be a figure, but shapes can also form part of
 The Perceptual Object 311

the background (or ground) from which the figure erally speaking, the smaller an area or a shape is,
emerges. Thus, a ball resting on a field of grass is the more likely it is to be seen as figure (see Weis-
a figure (here a round shape) resting on a ground stein & Wong, 1986). This is demonstrated in Fig¬
consisting of many elongated shapes (the blades of ure 11-3, where it is easier to see the vase when
grass). How can we tell the difference between a the white area is smaller (11-3A) and easier to see
figure and its ground? For example, how can you the faces when the black area is smaller (11-3C).
tell the book you are reading apart from the table Once a particular interpretation has been ar¬
on which it rests, or the printed words apart from rived at, for example the vase, figure and ground
the page? On your retina the blobs of contours that take on distinct properties. When the white area is
make up figures and grounds are all run together, seen as the vase, it appears to be “in front of” the
intersecting and overlapping, but somehow the vi¬ black area seen as the ground, and the contours in
sual system separates the book from the table and the pattern seem to “belong” to the vase. How¬
the words from the page. ever, when the interpretation changes, the contours
To begin with, this separation is a psycholog¬ are now seen to belong to the faces, and the faces
ical achievement, as can be seen in Figure 11-3B. seem closer than, and in front of, the white back¬
This is the famous Rubin (1915, 1921) face-vase ground. Furthermore, figures appear to be more
ambiguous figure. When you look at this figure you “thinglike” and appear to have a shape whereas
might see a pair of silhouette faces gazing at each the ground appears formless. Figures are seen as
other or you might see an ornate vase. The vase “richer” and more meaningful and are remem¬
appears white against a black ground, whereas the bered more easily. Figures also contrast more than
faces are black against a white ground. Notice that the ground, appearing brighter or darker than
as you look at 11-3B for a few moments the two equivalent patches of light that form part of the
pattern organizations alternate in consciousness, background (Coren, 1969). Finally, work based on
demonstrating that the organization into figure and the relative ability of figures and ground to mask
ground is in your mind, not in the stimulus. Notice other stimuli in a metacontrast-type of experiment
also that the faces and the vase never appear to¬ suggests that stimuli seen as figures are processed
gether. You “know” that both are possible but you or registered more quickly than stimuli seen as
can’t “see” both at the same time. It is impossible ground (Weisstein & Wong, 1986).
for a given part of a visual pattern to be simulta¬ Figure-ground organization affects many other
neously interpreted as both figure and ground. Gen¬ aspects of how we respond to the visual world, as

FIGURE 11-3 (B) A reversible figure-ground stimulus in which a pair of black faces or a white vase (or
perhaps a birdbath or goblet) are seen alternately. When the white area is smaller (A) the vase is easier to
see; when the black area is smaller (C) the faces are more easily seen.
312 Chapter 11/Form

has been demonstrated by a series of experiments that actually is not present (Coren, 1972). The con¬
by Naomi Weisstein and Eva Wong. Wong and tours bounding the rectangle are called subjective
Weisstein (1982) reported that the tilt of a short, contours or illusory contours because they exist
straight line was easier to discriminate when it was only in your mind and are not physically present on
flashed on a part of an outline version of Figure 11- the retina.
3B that was currently being seen as figure (either Although many factors can contribute to the
the face or the vase), than when flashed on a part formation of subjective contours (e.g., Halpem,
currently being seen as ground (e.g., the white part 1981; Ware, 1981), the presence of depth cues
when the faces are figure). This finding is consis¬ seems to provide a powerful impetus to organize
tent with the notion that figures are usually given parts of the field into simple figures, even when
some processing priority over ground. they are not defined by physical contours. This is
Not only can separation into figure and ground consistent with the fact that observers do report that
result in perceptual effects affecting other variables subjective figures created by subjective contours
but it appears that other forms of perceptual pro¬ appear to lie in front of their backgrounds, even
cessing can affect our perception of figure and when all other depth cues are carefully removed
ground. Figure 11-4A illustrates how form (and (Coren & Porac, 1983b). Figure 11-4B shows an¬
figure) can emerge from a two-dimensional array other subjectively contoured figure in the form of a
because of the existence of depth cues in the figure white triangle in which the triangle clearly appears
(see Chapter 10). In this case the cue is interposi¬ to stand out in front of the other elements in the
tion, where an object apparently blocking another figure. We saw another example of subjective contours
object is seen as closer than the occluded object. created by depth information in the random-dot
Despite the compelling percept of a white rectan¬ stereograms created in Demonstration Box 10-8.
gle, which is brighter than the white making up the Interestingly, subjective contours act very
other parts of the figure, there is in actuality no much like real contours, to the extent that they can
rectangle drawn in here. The implicit depth cues mask real contours (Lehmkuhle & Fox, 1980;
cause you to conclude that such a figure should be Weisstein, Matthews & Berbaum, 1974), improve
present, and you then reorganize the perception of judgments of the position of a dot (Pomerantz,
the array to perceive a figure (the white rectangle) Goldberg, Golder & Tetewsky, 1981), and cause

^TADI
Ul Vi ■
A

FIGURE 11-4 The white rectangle across the word STOP in A and the white triangle in B are bounded
by subjective contours. They actually do not exist in the stimulus. (From Coren, 1972. Copyright 1972 bv the
American Psychological Association. Reprinted by permission.)
 The Perceptual Object 313

motion aftereffects (Smith & Over, 1979). They School of psychology. Gestalt is a German word
can even meld with, or “capture,” real contours that can be translated to mean “form,” “whole,”
superimposed upon them (Ramachandran, 1986). or even “whole form.” They were interested in
Subjective contours, then, are one example of how processes that cause certain elements to seem to be
certain factors interact to segregate the perceptual part of the same figure or grouping and others to
field into figure and ground (see Petry & Meyer, seem to belong to other figures or groups. They
1987, for a full review of subjective contour formulated several laws of perceptual organization
research). that govern the emergence of a visual figure
(Wertheimer, 1923). Their basic observation was
that elements within a pattern do not seem to op¬
Figural Organization
erate independently. At the phenomenal level, there
Most forms or objects we see are composed of a appear to be attractive “forces” among the various
number of elements. We have already found that elements that cause them to form a meaningful and
the organization of elements into perceptual objects coherent figure, much as gravity organizes the
involves an active constructive process. We can see planets, sun, and moons of our solar system. They
the action of this “urge to organize” in Figure 11- described how certain regular properties of ele¬
5, where the many possible organizations of the ments within a pattern bring about the emergence
elements seem to alternate in a rapid unstable manner. of stable figures.
The general question of how all perception A Gestalt principle is illustrated in Figure 11 -
comes to be organized into patterns, shapes, and 6A, which is usually seen as two clusters of dots.
forms was central to a group of psychologists (Max Although the array actually contains 12 individual
Wertheimer, Kurt Koffka, and Wolfgang Kohler dots, perceptual processes organize them into an
were the most influential) who formed the Gestalt experience of two groups. This is an example of

<« * ■* •
A 4 m * s M • •
•• ••
»• :• *••
»..v . •*
••• • ••
• • .*•••••►•
•• •• • .
• *.• • * ••••
A * * > ;v;*v§4 4
4 ••• •* •*•• 4 •** • , ^ ••
p • • •
•• •* •• •• **•
^ 4 A ► ••••••• t
•••.••• •• •• ••
• •, .* • ••••
•• • • • • •► • •
V*-”*4 V
*• • •*
9 *• » **•
..• * « A T 4 **•••«
• •• " • . , # •
.• 4 jl 4 ... 4
•:•••>:: • X •
*• • V • •* •• I !• '# •
* ► •••••
• • *•,*^* ▼ 'k•••• • «
* •: • :• •••

k M • •• ^ ^ A ^ •
;r.v.

FIGURE 11-5 A stimulus revealing the “urge to organize.” There are many possible organizations of the
various small elements, and the visual pattern you see is continually changing as you shift from one
organization to another.
314 Chapter 11/Form

FIGURE 11-6 Examples of the Gestalt principles of figured organization: (A) grouping by proximity; (B)
and (C) grouping by similarity; (D) good continuation; (E) an extreme example of closure in which two
common figures are hidden; (F) closure; (G) and (H) the distortion of the distance between elements (the
distances marked by the arrows are the same) by the principle of proximity.

the law of proximity, which states that elements Another example of this law is seen in Figure 11-
close to one another tend to be perceived as a unit 6C where the two halves of the circular field appear
or figure. Figure 11-6B is normally seen as a tri¬ quite separate because of grouping by similarity
angle composed of black dots on a background of (for a more advanced analysis of such effects see
(or surrounded by a swarm of) X’s. This is an ex¬ Julesz, 1981). Figure 11-6D is viewed usually as a
ample of the law of similarity, which maintains spiral of dots with one standing outside. This is an
that similar objects tend to be grouped together. example of the law of good continuation, which
 The Perceptual Object 315

states that elements that appear to follow in the nanz means approximately “conveying the essence
same direction (as in a straight line or simple of something.” Because prevailing conditions are
curve) tend to be grouped together. Figure 11-6E sometimes not ideal, as in line drawings or on a
is an example of the law of closure, which states foggy night, the essence can be “better” than the
that when a space is enclosed by a contour it tends reality. Seeing complex patterns of contours as per¬
to be perceived as a figure. Most people see a dia¬ ceptual objects makes further processing of the vast
mond between two vertical lines here. Actually, array of information in the retinal image simpler
Figure 11-6E also could be seen as a letter W and faster. For example, it would take you less
stacked on a letter M, or a normal K and a mirror- time to count 12 dots if they were organized into
image K facing each other, were it not for the com¬ two triangles of the sort shown in the left half of
pelling nature of closure. Closure also allows us to 11-6A, than if the 12 dots were randomly grouped
complete broken contours as in Figure 11-6F, in a single cluster (Oyama, 1986). Demonstration
which is seen as a triangle rather than as three sep¬ Box 11-2 allows you to explore the concept of
arated acute angles. Pragnanz further.
The Gestalt principles are so powerful they
may even be responsible for visual illusions. For
Information, Symmetry, and Good
example, distances between parts of a pattern that
Figures
are organized into the same group, or figure, are
underestimated relative to the same distances when Although the notion of “figural goodness” seems
the same parts belong to different groups (Coren & intuitively reasonable, it has proved difficult to ob¬
Girgus, 1980). Figures 11-6G and 11-6H illustrate tain a precise definition of it for purposes of testing
this effect. The figures show two different group¬ the law of Pragnanz in various situations. Sugges¬
ings formed by operation of the law of proximity. tions have been made, but there still seems to be
The distance between the two lines pointed to by no consensus. One important suggestion was that
the arrows in each figure is identical, yet that dis¬ figural goodness can be defined in terms of the
tance seems larger in the lower figure, where the amount of information needed to describe a particular
lines are parts of two different groups. Such distor¬ organization (Hochberg & Brooks, 1960). Hoch-
tions support and enhance the operation of the Ge¬ berg and Brooks gave a formula to compute figural
stalt laws to form perceptual objects from discret^, complexity, based on the number and sizes and va¬
parts of the visual array. L' rieties of angles and line segments in the organized
All the Gestalt laws operate to create the most figure, where these aspects are considered to be the
stable, consistent, and simple forms possible within information needed to specify the organization.
a given visual array. The Gestalt psychologists They argued that the organization with the lowest
called this process the law of Pragnanz, which figural complexity would be the most likely one to
states that the organization of the visual array into be seen, and provided evidence that this is true.
perceptual objects will always be as “good” as the Another way of looking at the amount of in¬
prevailing conditions allow. Here the meaning of formation in a figure, and its consequences for per¬
good encompasses concepts such as regularity, ception , was offered by Attneave (1955). He proposed
simplicity, and symmetry. The law of Pragnanz is that quantifying patterns in terms of information
also a way of saying that the perceptual systems theory (see Chapter 2) would help in understanding
work to produce a perceptual world that conveys figural goodness. To see how information theory
the “essence” of the real world, that is, to assure applies to patterns, consider Figure 11-7, which is
that the information about the real world is cor¬ an image of a woman’s right eye as it might appear
rectly interpreted. In fact, the German word Prag¬ as an enlargement of a newspaper photo. Notice
316 Chapter 11/Form

DEMONSTRATION BOX 11-2. Pragnanz

Look at the accompanying figures for a moment and contains no right angles? that the “triangle” has two
(without looking back again) draw them on a separate rounded comers and an open one? that the “X” is
piece of paper. When you have finished, return to this actually made up of curved lines? Look back at your
box. reproductions. If you drew (or remembered) just a
Now carefully compare the figures you drew to good circle, square, triangle, and X, your percepts
the actual figures. Did you pick up the fact that the have been “cleaned up” by the action of Pragnanz.
“circle” is actually a tilted ellipse? that the “square”

OUAA
that the picture is made up of a number of dots, 8A. Suppose we asked you to guess the figure pres¬
each of which can be either black or white. We can ent, without actually seeing it, by simply guessing
analyze any figure into such an array. Consider whether each cell was black or white. Since each
Figure 11-8, where we have broken up square vi¬ guess deals with two alternatives, the answer to
sual fields into separate, smaller squares called each contains one bit of information, as we pointed
cells or pixels (we call the whole array a matrix). out in Chapter 2. If we filled in the pattern ran¬
Notice that we can construct a variety of patterns domly, in order to guess the complete pattern you
by simply filling in cells, as we have done in 11- would need 64 guesses (1 for each cell) or 64 bits
of information (1 for each guess needed). If we told
you that the left side was the mirror image of the
right side, called a vertically symmetrical pattern
since the mirror images are symmetrical around a
vertical line, you would only need to guess 32 cells
either on the right or on the left in order to guess
the pattern, thus reducing the amount of informa¬
tion to 32 bits. Therefore, a vertically symmetrical
figure, such as 11-8B, contains less information
than an asymmetrical figure, such as 11-8A. Figure
11-8C, which is symmetrical around both the ver¬
tical and horizontal axes, contains even less infor¬
mation than the other two patterns (16 bits), since
only one comer (16 cells) must be known before
FIGURE 11-7 An example of a figure made up of the entire pattern can be derived.
dots that can be either black or white. Since good figures are generally symmetrical
 The Perceptual Object 317

FIGURE 11-8 Examples of symmetry in patterns: (A) no symmetry; (B) symmetry around a vertical axis;
(C) symmetry around both horizontal and vertical axes.

and reguiar, we can now see that they also contain Clearly, Figure 11-9A is the “best” and 11-9C the
less information. This means that they should be “worst” under this definition. Since Figure 11-9A
easier to remember and easier to recognize. Att- is unique, it is the least informative about its set of
neave (1955) found that this was so. More recently, alternatives (which has no members); Figure 11-
Freyd and Tversky (1984) found that symmetrical 9C, in contrast, is one of eight patterns that can be
forms were often matched to even more symmetri¬ created by reflection and rotation. Seeing it indi¬
cal forms, indicating a perceptual bias towards cates that the other seven alternatives did not ap¬
symmetry in the representation of forms. They ar¬ pear, and thus conveys three bits of information
gued that detection of overall symmetry in a form (since, as you learned in Chapter 2, the amount of
led to the assumption that it is symmetric in its de¬ information is equal to log2 of the number of stim¬
tails as well, which might explain the results you ulus alternatives). Gamer and his colleagues (Gar¬
got when you tried Demonstration Box 11-2. Yo- ner & Clement, 1963; Handel & Gamer, 1965)
dogawa (1982) has given a mathematically rigorous found that the smaller the set of possible alterna¬
measure of pattern symmetry, based on information tives, the more likely an observer was to rate a pat¬
theory, that nicely predicts perceptions of pattern tern as “good” (the observers didn’t see the set of
symmetry and pattern complexity. alternatives). This approach, and the others de¬
A final suggestion about goodness that has scribed above, all support the conclusion that good
been influential was made by Gamer (1962, 1974), figures are simple and contain less information. In
whose definition of the amount of information in a this sense it doesn’t matter which approach we pre¬
pattern is somewhat different. It depends on the fer, the law of Pragnanz is the same in all of them.
number of possible alternatives that a figure could
be drawn from, much like the definition for the dif¬
Texture Segregation
ficulty of recognition that we used in Chapter 2
when we introduced the concept of information. Shapes and figures can be formed by changes in the
Garner argued that the smaller the set of possible stimulus pattern other than of intensity or wave¬
alternatives that could be created from a figure by length . For instance, object boundaries may be defined
rotation and reflection, the less the information and by areas of the visual array that differ in visual
the better the figure. Figure 11-9 shows the set of texture. Visual textures are collections of tiny con¬
alternative patterns for three different dot patterns. tour elements or shapes. For example, if you look
318 Chapter 11/Form

Set of all possible unique rotations and reflections

FIGURE 11-9 Relationships between figural goodness and the set of possible alternative stimuli that can
be created by rotations and reflections: (A) is the “best” figure and (C) is the “worst.” (Based on Handel &
Garner, 1965)

back at Figure 11-6C you will see two regions de¬ millions of tiny points of light as a result of the
fined by different textures—one a texture of dots, retina being made up of millions of tiny receptors
the other a texture of dashes. Notice that there is with gaps between them.
apparently a boundary, or contour, between these The ease with which you can make out shapes
regions. This contour is a form of subjective contour, defined only by textures depends on the nature of
since it is not actually present in the stimulation, the textural elements. It has often been suggested
and is generally referred to as a textural contour. that the segregation of parts of the field on the basis
Perhaps the most common shapes formed by such of textural elements is really an example of group¬
textures are the pictures in newspapers, which are ing by similarity, which we discussed above. Al¬
formed by arrays of thousands of tiny dots that dif¬ though the number of dimensions in which groups
fer in size forming textures that define the shapes of textural elements differ from one another is in¬
of people, buildings, and other objects in the pho¬ deed quite important, predictions based on a simple
tographs reproduced there, such as the one in Figure definition of grouping by similarity are often not
11-7. In a sense, all shapes are defined by textures, confirmed. A good example is shown in Figures
since the retinal image is effectively made up of 11-10A and 11-10B. Although people judge the
 The Perceptual Object 319

tttttttttt T T T T T T T T T T
tttttttttt T T T T T T T T T T
tttttttttt T T T T T T T T T T
TTTLLLLTTT T T T /* 7 7 7 T T T
TTTLLLLTTT T T T 7 7 7 7 T T T
TTTLLLLTTT T T T 7 7 7 7 T T T
TTTLLLLTTT T T T 7 7 7 7 T T T
TTTTTTTTTT T T T T T T T T T T
TTTTTTTTTT T T T T T T T T T T
TTTTTTTTTT T T T T T T T T T T
A B

AAAAAAAAAA □ A A □ A □ □ A □ A
AAAAAAAAAA A A A □ □ A □ A A □
AAAAAAAAAA A □ A □ □ A □ A □ A
AAADBDBAAA □ A □ A B B A □ A □
AAABDDBAAA A A A B B B A □ A □
AAADBBDAAA □ A □ A B A B A □ A
AAADBOBAAA □ A A B B A B A A □
AAAAAAAAAA A □ A □ A A □ A □ □
AAAAAAAAAA A □ A □ A □ A □ □ A
AAAAAAAAAA □ A □ A □ A A □ □ A

C D

□ A A □ A □ □ A □ A Z 7 Is (7 ts 7 Is 7 7 Is
A A A □ □ A □ A A □ A Is X) 7
V Is 7 k 7 /I
A □ A □ □ A □ A □ A E tsA A ts XI IS A A l\
□ A □ A B B A □ A □ 7 Z K 71 SI 71 xj A A
A A A B B fl A □ A □ XI 7 7 1A F 1A IS A
□ A □ A B A B A □ A ts XIIs 71 71 SI 7! XI A Is
□ A A B B A B A A □ 7 Z 7 K IZ XI IZ X] X] 7
A □ A □ A A □ A □ □ X| X]7 A Is 17 t\ Cs 7 A
A □ A □ A □ A □ □ A Z K 7 ts 17 XI A A XI xi
□ A □ A □ A A □ □ A Zl 7 IS ts A X] A X] 7 Is

E F

FIGURE 11-10 Examples of shapes defined by textural contours of varying strength.

tilted Fs to be more similar to the upright 7”s than blobs of a particular color, length, width, or ori¬
are the L’s, the square shape created by the tilted- entation, line ends (terminators), or line crossings
T/upright-T boundary is much more apparent than (intersections). For example, the square shape that
is the shape created by the LIT boundary (see Beck, appears in Figure 11-10F is defined by a boundary
1966, 1982). Apparently the orientation of small between elements that have no terminators and ele¬
contours is an important aspect of their contribution ments that have three terminators. Notice that all of
to texture. the textons are made up of the same parts, a slanted
Texture segregation is usually easy and auto¬ line in two different orientations and a “comer”
matic when there are differences in the number, shape in four different orientations. The local tex¬
density, or type of a few classes of local elements ture elements differ only in how these parts are put
that are generically called textons (Julesz, 1981; together.
Julesz & Bergen, 1983). These include elongated Although Julesz’ theory and demonstrations
320 Chapter 11/Form

are convincing, they must be modified somewhat. Spatial Frequencies and Figural
To begin with, all textons are not equally powerful Organization
in their ability to define textural contours. This is
demonstrated in Figure 11-IOC and D. In C the Several investigators have tried to combine certain
central square is formed by a shape difference, aspects of figural organization with suggestions that
whereas in D it is formed by a color difference. come out of a consideration of textural segregation
Notice that the color difference seems much more and of spatial frequency analyses (such as we dis¬
dramatic, and that color differences between ele¬ cussed in Chapter 4) to produce a multiprocess,
ments actually make the shape boundary harder to multilevel explanation of how the perceptual object
see in C. Color differences are clearly dominant is formed. In the context of the separation of figure
over shape differences in forming textures, and this and ground, for example, Julesz (1978) argued that
differential salience strongly affects texture segre¬ texture segregation and a kind of “early warning’’
gation (Callaghan, Lasaga & Gamer, 1986; Enns, to detect regions requiring finer analysis were ac¬
1986; Gumsey & Browse, 1987). Certain higher- complished by background processes, whereas the
level factors, such as closed versus open figures, finer, more detailed analysis of important shapes
can often act as a texton when the closed figures was accomplished by figure processes. This would
are very different from the background figures mean that figural analysis would involve higher
(Enns, 1986). The best suggestion seems to be that spatial frequencies (e.g., finer details) and would
texture segregation is determined by the degree to require more stability, but would be slower to
which textons unique to a particular region of the emerge. Ground processing, in contrast, would in¬
visual field are salient in the context of textons in volve lower spatial frequencies (e.g., a crude
other, surrounding, regions (Enns, 1986; cf. Olson “blob detection’’), but would require less stability
& Attneave, 1970; Beck, 1982). and would be faster.
Finally, the number of stimulus dimensions These speculations have been confirmed. For
involved is important. Treisman and Gelade (1980) example, Calis and Leeuwenberg (1981) found that
showed that although texture segregation was easy ground processing is faster than figure processing.
and automatic when either shape or color of ele¬ Weisstein and Wong (1986; Wong & Weisstein,
ments differed, forming a boundary, it was impos¬ 1987) reported several experiments that demon¬
sible when particular conjunctions (pairings) of strate that simply flickering different regions of a
shape and color differed. For example, look at 11- uniform field of dots can make the flickering area
10E. Is there a central square here? Actually there stand out as figure or ground, depending on the
is, but only conceptually, not perceptually. It is flicker rate. Usually, a rapid flicker rate induces a
formed where the outside elements are only empty perception of ground, whereas a slower flicker rate
squares or filled triangles and the inside elements causes the flickering area to be seen as figure.
are only empty triangles or filled squares. But be¬ Finally, Klymenko and Weisstein (1986) reported
cause the element boundary is defined only by dif¬ that when regions of a visual field were defined by
ferences in conjunctions of basic properties, the different spatial frequency sine wave gratings, the
textural contour does not emerge. To find the region containing the higher spatial frequency grat¬
boundary you must scrutinize the shapes one by ing was more often seen as figure and the one con¬
one. Apparently the form of grouping by similarity taining the lower frequency grating was more often
that results in textural contours depends on ele¬ seen as ground. This is demonstrated in Figure 11-
ments that differ in only a single dimension, and 11, where the faces are much easier to see as figure
doesn’t work when you require combinations of di¬ because they are filled with high spatial frequency
mensions or properties (see Treisman, 1986b). gratings. Thus, there is evidence of a low temporal/
 The Perceptual Object 321

in this approach is that the feature set is arbitrary

and it may change as the shapes we wish to differ¬
entiate change.
An exception to the above characterization of
features is when they can be traced to more funda¬
mental units, such as textons, or line or angle “de¬
tectors” in the visual cortex. One such approach,
dealt with in the previous section, involves speci¬
fying the spatial frequency components that make
up a shape as its “features.” As was pointed out
in Chapter 4, the two-dimensional distribution of
light on the retina can be completely described in
FIGURE 11-12 A reversible figure-ground terms of simple sine wave gratings (making some
stimulus in which the faces are easier to see than simplifying assumptions). One suggestion has been
the vase because they are filled with relatively high that these Fourier components are the features used
spatial frequency gratings. (Based on Klymenko &. to identify perceptual objects (e.g., Campbell &
Weisstein, 1986)
Robson, 1968; Pollen, Lee & Taylor, 1971). It has
now been shown that this approach cannot work if
high spatial frequency processing system for figure the Fourier analysis is applied over the entire visual
and a high temporal/low spatial frequency process¬ field (e.g., Caelli, 1984; Cavanaugh, 1984; De-
ing system for ground. Valois & DeValois, 1987), but it may be useful if
the analysis is done over smaller regions, around
the size of cortical receptive fields.
Features
Unfortunately, the spatial frequency approach
The shapes that emerge into our consciousness is really different from other forms of feature anal¬
from the retinal image can be said to possess fea¬ ysis in more ways than simply being describable in
tures that differentiate them from other shapes. For terms of mathematical equations. First, the spatial
example, letters consist of combinations of lines of frequency components are not apparent to con¬
various orientations and curvature. Sometimes sciousness, and thus are not useful in describing
these lines intersect, sometimes they touch without what we see. Second, they are not related to texture
crossing. A given letter may be printed in red ink, segregation, since regions of the visual array that
or may be an outline. The relevant features of the are clearly different in their Fourier components
letters are those that help to differentiate a given cannot be seen as a shape (Julesz, 1981; Julesz &
letter from other letters, or from numbers or other Bergen, 1983). However, this approach does have
similar shapes (Gamer, 1974). Some of these fea¬ the advantage that the set of possible Fourier com¬
tures may be textons, such as an intersection. Oth¬ ponents is not arbitrary and does not change with
ers may be unique to letters, such as a “bend” in the set of shapes, so it may ultimately be made
a particular line. Whatever they are, we will often useful through some modifications.
speak of shapes being different in their features, Some idea of the complexity of feature analy¬
especially when we look at how they are recog¬ ses of forms appears when we begin to re-analyze
nized and identified. The strength of using a set of the principles of figural organization. The early
features to describe a shape is that the features ap¬ Gestaltists thought of a whole figure as composed
pear in our consciousness; they can be seen directly of parts “glued” together by the attractive forces
and thus talking about them is easy. The weakness we described in our discussion of figural organiza-
322 Chapter 11/Form

tion. However, recent attempts to describe in more pulvinar systems, or registered in different “maps”
detail the way the Gestalt laws actually operate of the visual field in different places in the cortex;
have questioned this conclusion. For example, see Chapter 3). The second stage of the process
Pomerantz (e.g., 1986) has demonstrated that when consists of active focal attention (see Chapter 15)
parts configure into wholes an emergent feature is selecting a spatial location and integrating the fea¬
usually created. In effect, this suggests that some¬ tures registered at that location into a perceptual
thing like closure might be considered a feature object. This perceptual object, or object file in
rather than an organizing principle. Thus, if you Treisman and Kahneman’s terms (see Kahneman &
look at a figure like 11-6F, and a similar closed Treisman, 1984), consists of a temporary represen¬
triangle (without the gaps), they look different, be¬ tation of the current appearance of the object that
cause one has the feature of being closed and the is constantly updated as new information is ob¬
other does not. This kind of analysis suggests that tained. When attention shifts to another object, be¬
features may result from higher-level processes, not cause the current one disappears or in the process
simply relationships among contour elements. of scanning the environment or being distracted,
the old object file vanishes and is replaced by a
new one at a different location. This view explains
Feature Integration Theory
how objects in the “real world” can be seen as
In our discussion so far, although we have referred enduring and solid, even though the information
to the perceptual object being constructed from the about them we receive changes drastically from
information available in the retinal image, we have moment to moment, since each object is associated
perhaps given the impression that all the construc¬ with a particular location in space. This is some¬
tive activity happens automatically, without any ef¬ what reminiscent of earlier speculations by Ames
fort and for all parts of the visual field at once. This (1955) that the minimum properties of any per¬
is probably not the case. Quite a bit of evidence is ceived stimulus are its thereness, referring to its
accumulating that we can construct only one per¬ location in space, and its thatness, referring to its
ceptual object at a time (as noticed in a reversible object properties.
figure) and that usually we must focus our attention A lot of data have been accumulated to test
on a particular location in “visual space” in order these ideas. One of the most intriguing findings is
to synthesize a perceptual object from all the infor¬ the discovery of illusory conjunctions. Feature in¬
mation coming from that location. tegration theory predicts that if attention is over¬
A particularly attractive version of this view¬ loaded, so that it is difficult to integrate separable
point is Treisman’s feature integration theory features into a perceptual object, the features might
(Treisman, 1986a, 1986b, 1988). The theory as¬ combine incorrectly, giving rise to the perception
serts that there are two major stages to the con¬ of an object that is not really there (an illusory con¬
struction of a perceptual object. First, preattentive junction). For example, an observer may be ex¬
processes (those that happen without active atten¬ posed briefly to a red O and a blue X, but may see
tion playing a part) operate on the information in a red X, which wasn’t really there (Treisman &
the retinal image to produce shapes and register Schmidt, 1982). A simple analogy to this is if you
their features. A crucial assumption about this stage meet a man named Fred, who is an architect, and
is that features such as color, the various linear fea¬ another named George, who is an accountant, then
tures (e.g., line orientation, closure), and three- in recounting your experiences to a friend you
dimensional location are registered separately from might end up recalling Fred, the accountant. The
one another in special processing systems (e.g., name and the occupation are features, which you
carried separately in the geniculostriate and tecto- assign to the particular stimulus (the person) sepa-
 Object Recognition and Identification 323

DEMONSTRATION BOX 11-3. Figural Integration

It is rare that we see an object in its entirety. Either (including himself or herself) behind a narrow slit
we or the object is moving about, resulting in created by a door that is slightly ajar (Shimojo &
glimpses that are partially occluded by other objects Richards, 1986). Try varying the size of the slit and
(e.g., a dog running though some trees), yet we have the speed with which the object passes across it. You
no difficulty identifying the object. Parks (1965) stud¬ will find that over a surprisingly wide range of size
ied this phenomenon by moving a shape behind a nar¬ and speed conditions the object will be identifiable.
row window, or slit. His surprising result was that a Also, pay attention to the strength of the feeling you
wide variety of shapes could be easily recognized, will have that the entire object is present, even though
even though the shape was never seen except as a at any moment you are only receiving a fragmentary
series of fragments. Because of one of the shapes view of it. When the conditions are optimal, that
Parks used, the phenomenon of easy recognition of impression is very strong. This demonstrates the im¬
shapes presented by moving behind a slit has come to portance of the integration over time required to cre¬
be called “Parks’s camel.” ate the perceptual object out of a chaotic and con¬
In order to experience ‘‘Parks’s camel” for stantly changing retinal image.
yourself, arrange to have a friend pass various objects

rately. In the perceptual synthesis, however, the pronounceable letter strings. Again, the preatten-
objects don’t even have to be similar for their prop¬ tive analysis of the contours in the retinal image
erties to be exchanged in this way. For example, into shapes that may signify objects in the real
color is exchanged among dissimilar-shaped objects world seems to constrain feature integration. Since
(e.g., a triangle and a circle) just as easily as it is we learn through experience with the world which
among similar-shaped objects (e.g., two small out¬ features are likely to go together, feature integra¬
line triangles) (Treisman, 1986a). This is possible tion is less subject to error when we know which
because the individual features are processed sepa¬ objects to expect (Treisman, 1986a). One role
rately and then synthesized with reference to a played by the feature integration process seems to
given place in space via a different process. be to maintain the continuity and coherence of the
Just how the Gestalt principles interact with perceptual object in the face of the constantly
the feature integration process is an interesting, and changing array of visual information about it.
still controversial, question. Prinzmetal (1981) Demonstration Box 11-3 allows you to experience
found evidence that illusory conjunctions were less the coherence of the perceptual object in a some¬
likely when the components came from different what different way.
Gestalt groupings of visual contours than when
they came from the same grouping. He argued that
perceptual organization prevented the illusory con¬ OBJECT RECOGNITION AND
junction of features from groups of contours that IDENTIFICATION
were likely to be parts of different objects (see also
Butler & Kring, 1987). Also, Prinzmetal and Mil- Look at Figure 11-12 and study it for a minute or
lis-Wright (1984) found that illusory conjunctions so. You have never seen it before, because it was
were more likely to occur within letter strings that especially constructed for this book. It is clearly an
formed pronounceable words than in random, un¬ object, but what is it? Now look at it again. You
324 Chapter 11/Form

Recognition versus Identification

The first thing we must do is make our terminology

clear. In what follows, when we refer to object
recognition we mean the experience of “perceiv¬
ing something as previously known” (Mandler,
1980). Object identification means naming an ob¬
ject, correctly classifying it in some categorization
scheme, knowing in what context it is usually en¬
countered, knowing its relation to other concepts,
and so on—in short, remembering something more
about it than merely having seen it before. Accord¬
ing to Mandler (1980), the experience of familiarity
comes about because the more exposure you have
to a perceptual object, the more you have organized
the various processes that create and maintain that
object out of a particular combination of critical
FIGURE 11-12 An object you have never seen
before. (Based on Biederman, 1987) features. The process of feature integration leaves
memory traces, which are then experienced as fa¬
miliarity the next time that particular conjunction
recognize it because you have seen it before (just a of features is encountered. The more times an ob¬
minute or so ago), although you still are not able ject file is constructed for an object or event, the
to identify it (it is not a “real” object). You expe¬ more detailed that file can become, and the more
rience familiarity, but the object makes no sense. familiar it will seem when it is next encountered.
Now study it a little longer. Eventually you may Identification, however, clearly requires some
begin to be able to classify it into some object class sort of retrieval process. The representation of the
or other, perhaps it resembles a distorted version of perceptual object created for the moment, the ob¬
a streetside hot dog vendor’s cart, or perhaps a ject file, must be compared to other representations
child’s toy. The longer you look at it, the more in memory, along with the connections these other
associations it generates, although it still doesn’t representations have to other information stored in
have a name. If we told you that it was a “Horned memory. Most investigators agree on this much.
wheeler,” the name might suggest that it is an ap¬ What they don’t agree on is the composition of this
paratus for conveying or transporting things, al¬ representation that is compared to the memory,
though how or why still might be a puzzle. whether it is features of some sort, or spatial fre¬
Every day, practically every moment, you rec¬ quency components, or another system. We inves¬
ognize and identify perceptual objects like the one in tigate some of these proposals in the next section.
Figure 11-12, although they are seldom as novel. But
the ability to see a stimulus as an object is not suffi¬
Data-driven versus Conceptually-driven
cient. Our very survival may depend on our recogni¬
Processing
tion of an object as something we have seen before,
and on our labeling (really categorizing) that object Feature integration theory and most other modem
so that we can retrieve information about its likely be¬ theories of visual object recognition and identifica¬
havior or the behavior we should perform in its pres¬ tion assume at least two major types of psycholog¬
ence. In this section we discuss what we know about ical processing. One type is referred to as data-
how these feats are accomplished. driven processing, which begins with the arrival
 Object Recognition and Identification 325

of sensory information at the receptors. This type the physiology of sensory systems. In Chapter 4 we
of processing is characterized by the processing of discussed how neurons in the visual system interact
information, or data, directly in terms of some through lateral inhibition, and we showed how these
fixed set of rules or procedures. In a sense, the data interactions could be responsible for enhancement of
themselves drive the processing, since the rules contours, even creating some illusions. In Chapter 3
usually concern the registration of particular pat¬ we discussed cortical cells that respond to specific
terns in the data. In terms of visual object recog¬ features, such as contour orientation, in the retinal
nition, data-driven processing would be involved in image. These seem to be parts of the visual system
the registration of distinctive features, such as lu¬ that do data-driven processing.
minance differences in the image, line orientations, An interesting source of evidence for data-
intersections, or other attributes that distinguish the driven processing of patterns on the basis of ex¬
pattern from others. The first stage of preattentive tracted features from the retinal image comes from
processing in feature integration theory is supposed work on stabilized images (a technique we dis¬
to be data-driven. cussed earlier in the chapter). As you recall, when
The other type of processing is called concep¬ we stop the movement of the image on the retina it
tually-driven processing. In this type of processing, fades out of consciousness. This fading is not by a
higher level conceptual processes, such as memories gradual, uniform disappearance of the whole fig¬
of past experiences, general organizational strate¬ ure, nor is it by the disappearance of little random
gies, and expectations based on knowledge of the bits of the pattern; instead the stabilized image
world and previous events or the surrounding con¬ fades out by features, such as whole lines or angles
text, guide an active search for certain patterns in the (Evans & Wells, 1967; Heckenmueller, 1965). Fig¬
stimulus input. An example of conceptually-driven ure 11-13 contrasts the fadeout by features (A),
processing is the feature integration stage of feature which does happen, with a fadeout by random bits
integration theory, where focal attention selects a lo¬ (B), which doesn’t happen.
cus in space and integrates the features there into a
perceptual object, perhaps in conjunction with prior
Global versus Local Processing
hypotheses as to what to expect. Thus, a dark thing
flashing through the air on a playground might be Because there is considerable evidence for physio¬
seen as a ball someone has thrown, whereas in a quiet logical feature detectors, most approaches to data-
park it might be seen as a bird flying by, because of driven processing emphasize the role of local features
prior expectations. Both conceptually- and data- in object recognition and identification. Local fea¬
driven types of processing may occur together, or in tures may be viewed as the small-scale or detailed as¬
sequence, but both must occur. If only conceptually- pects of a figure, but there is a bit of a conceptual
driven processing occurred, we would see only what problem with our definition of what a local feature is.
we expected to see, and would make too many mis¬ Just how small-scale must a feature be to be called
takes to survive. If only data-driven processing oc¬ “local” ? What if a form has no features of this partic¬
curred, we would not be able to take advantage of our ular scale? Does it thereby lack local features? At
tremendous amount of experience with the visual least we can easily contrast local features with the
world to enhance our perceptual functioning, espe¬ overall or global aspect that gives the whole form its
cially in poor environments, and to distinguish the apparent shape. In Figure 11 - 14B, we see the global
relevant from the irrelevant in the flux of information shape of a letter H made up locally of small 5s. Each
on our retinas. small S is made up in turn of local features (e.g., the
Since in data-driven processing the emphasis is curved line segments) and each of these features
on detection of predefined patterns in stimulus input, could be subdivided into even smaller local features
one place to look for evidence of such processing is in (such as microdots of ink) if we had a large enough
326 Chapter 11/Form

Stabilized target-►Time

Disappearance

A Typical fadeout by features

-1
Disappearance

B Random fadeout

FIGURE 11-13 Typical feature-by-feature fading of a stabilized retinal image is shown in (A) as opposed
to what would be expected if disappearance were random in its time course, shown in (B).

magnifying glass. Thus, the terms local and global An interesting issue arises with respect to
are relative; we must specify what level of detail we global and local levels of detail in visual forms. In
are referring to when we use them. You can see for a seminal paper, Navon (1977) argued that the de¬
yourself the importance of local features in object tection of more global (larger-scale) aspects of a
identification by trying Demonstration Box 11-4. form with several levels of detail would always be
faster than the detection of the more local details.
This position resembles the Gestalt position, in that
H H S S
H H S S whole forms that are grouped by Gestalt processes
H H S S seem more immediately available to our conscious¬
H H S S
S S S S S S ness than do the more local constituents that have
H H H H H H
H H S S been so grouped. Navon (1977) did several exper¬
H H S S iments to test this proposition and one in particular
H H S S
H H S S was provocative. In this experiment, observers
were asked to name forms like those in Figures 11 -
A B 14A and 11-14B at either the global level (e.g., H)
or the local level (e.g., H for 11-14A, S for 11-
H H S S 14B) as fast as they could. In addition, sometimes
the name of the form was the same as that of its
H H S S constituents (11-14A) and sometimes the names at
H H H S S S the two levels were different (11-14B). The first
result was that regardless of what the stimuli were
H H S S
like observers were always faster in naming the
H H S S global level form than the local level forms. More¬
over, when naming the global level form, it didn’t
C D
matter whether the local constituents had the same
FIGURE 11-14 Examples of stimuli that have two name as the global form or not; observers were
distinct levels of detail. equally fast. However, when naming the local con-
 Object Recogni tion and Identification 327

DEMONSTRATION BOX 11-4. The Role of Local Features in Pattern Recognition

Harmon (1973) and Harmon and Julesz (1973) have inal photograph. To try this, look at the figure at normal
presented an interesting set of demonstrations that illus¬ reading distance. Do you recognize the person? Try
trate how local features can interfere with a more global again, viewing from 2 m this time. (It will also help if
percept. One of their demonstrations is presented in the you squint your eyes.) If you follow these instructions,
figure, a computer-processed block representation of a you should be able to identify this block portrait as a
photograph. The brightness information from this scan very famous historical person. If not, the name of the in¬
is then locally averaged, so that the brightness value dividual is printed upside down in the bottom right-hand
in each of the squares is an average of a number of corner of this page. (From Harmon & Julesz, 1973.
brightness samples taken in that area of the picture. Copyright 1973 by the American Association for the
This technique can be used to see if such local bright¬ Advancement of Science. Used by permission.)
ness information can elicit the percept of the orig¬

ujODun uniqTjqv
328 Chapter 11/Form

stituents, observers were greatly slowed down if faster reaction to groupings of local features of a
the global form had a different name. Based on particular size (e.g., Hughes et al., 1984).
these data, Navon argued for global precedence, All the factors above are affected by how an
the idea that detection of the more global aspects observer distributes attention to the figure. Observ¬
of a visual form will always be faster than detection ers are able to voluntarily direct their attention ei¬
of the more local aspects. ther to global or local aspects of the figure, and
Although Navon (1977) did demonstrate one thus give that level processing dominance (Hoff¬
set of conditions under which global forms have man, 1980; Kinchla, Solis-Macias & Hoffman,
processing dominance (see Ward, 1983) over lo¬ 1983). If they are forced to switch attention from
cal forms, other studies suggest that local and one level to the other, however, this tends to slow
global features are detected simultaneously (in par¬ and interfere with their ability to process either
allel) and at approximately equal speed (Boer & level of features (Ward, 1982b, 1985).
Keuss, 1982; Hughes, Layton, Baird & Lester,
1984; Paquet & Merikle, 1984). More specifically,
Integral versus Separable Stimuli
it seems that global processing dominance for vi¬
sual forms seems to hold only for a specific set of Processing dominance has also been studied from
conditions. When the global form is made a bit perspectives other than global versus local process¬
harder to see, for example, by spacing out the local ing. For example, Pomerantz, Sager, and Stoever
features as in Figures 11-14C and 11-14D, the re¬ (1977) showed that in groupings of simple linear
sults are reversed (Martin, 1979). Now the smaller features those that form closed, “good” figures are
letters are named more quickly and the naming of much easier to recognize than those composed of
the larger letter is slowed down when the smaller the same features in a different arrangement. These
constituents have different names, as in Figure 11- data seem to indicate some kind of direct (faster or
14D. The absolute size of the figure is also impor¬ earlier) perception of such good figures. Perhaps it
tant in determining whether global or local features is the case that figural goodness is itself treated as
are processed more easily. When the stimulus is if it were a feature or collection of features. If it
much larger (for instance, when you hold Figures were a perceptually dominant feature then it might
A or B close to your eyes), you will notice that be extracted earlier than nondominant features, and
now the smaller letters are much more salient and hence be responded to more quickly.
easier to see and the larger letter is more difficult There are some aspects of good figures that do
to see. In the laboratory, observers can tell which seem to make them special. For instance, they have a
of two smaller letters is present in such a display certain unitary wholeness, that is, they appear to be
more quickly than they can tell which of two larger coherent and not easily broken into components. One
letters is present when the display is larger than analysis of processing dominance that incorporates
about 7 deg of visual angle. When the display is this idea introduces the concept of an integral stimu¬
smaller than this (try holding the book at arm’s lus, which is a stimulus that is seen in all of its aspects
length to view A or B) the larger letters are more simultaneously. A light bulb is such a stimulus since
quickly discriminated (Kinchla & Wolfe, 1979). it is seen to have shape and color all at once (Lock-
Why is absolute size so important in determining head, 1966). Integral stimuli have many characteris¬
processing priorities? Perhaps because of the fact tics reminiscent of good figures. In contrast, we have
that various feature-sensitive cells in the cortex are the concept of a separable stimulus, which is a stim¬
“tuned” for specific-sized stimuli (see Chapter 3). ulus that has features that cannot be easily integrated.
Thus, the visual system may be biased toward a An example of a separable stimulus is one of the mat-
 Object Recognition and Identification 329

rices in Figure 11-9, in which the dots and the lines ples of perceptual organization is thus like that of a
forming the grid don’t seem to be parts of a coherent train engine to its track. Although the force and
figure. It has been argued that integral stimuli are pro¬ movement is provided by the engine (attention), the
cessed first as blobs at a global level before they are path that it must follow, if it is to go anywhere at
analyzed into their component parts (Lockhead, all, is determined by the pattern of the track (the
1972, 1979; Lockhead & King, 1977; Monahan & principles of organization). Of course, the fuel that
Lockhead, 1977). Some separable stimuli, however, powers the engine is data-driven feature extraction.
hang together better than others. These are said to
Content and Identification
contain a configural feature (usually something like
the Gestalt qualities of closure or symmetry). Thus, a As you have seen in the discussion above, we could
pair of parentheses like () is seen as more of a figure not talk about data-driven processing without men¬
than one like (( and is processed more quickly, even tioning attention, an aspect of conceptually-driven
though it still remains a separable configuration of processing. Any time your perceptual processes are
two components (Gamer, 1978). influenced by your expectations, previous experi¬
In our discussion of global versus local pro¬ ence, hypotheses, or any global organizing principle,
cessing we noted that processing dominance may you are demonstrating the operation of conceptually-
be affected by attentional factors. Attention also driven processing. There are many other examples of
seems to interact with figural goodness in determin¬ the importance of conceptually-driven processing in
ing how well we process certain patterns and pat¬ the formation, recognition, and identification of per¬
tern components. Many patterns are not really very ceptual objects. Look at Figure 11-15. Most people
good (and are certainly not very integral). These would see there two lines of characters, the top line
include the compound letter stimuli in Figure 11 - being A, B, C, D, E, F and the bottom one being 10,
14 (see Pomerantz, 1983). Such stimuli tend to be 11, 12, 13, 14. Now look closely at the forms you
synthesized into perceptual objects by attentional saw as B and 13. They are identical; the same form
factors, as well as by the Gestalt principles. Ac¬ was interpreted as a letter B in the context of other let¬
cording to feature integration theory we need to pay ters, and as a number 13 in the context of other num¬
attention to a particular place in visual space in or¬ bers. Notice that the data (the actual stimulus input)
der to join the separately extracted features of a are identical for both the perceptual organizations
pattern into a proper perceptual object. This would representing the B and the 13. Your identification of
explain why dividing attention between aspects of those perceptual objects, however, has been affected
a pattern slows identification at all levels. The less by conceptually-driven processing based on knowl¬
attention paid to features at a given level, the more edge and assumptions, mostly obtained here from the
difficult (and slower) it would be to combine those other stimuli that form the context.
aspects into a perceptual object. Since there is
much evidence suggesting that our attention is usu¬
ally “caught” by good figures, or the best aspects
of figures, this would also explain why better fea¬
tures are more easily and quickly joined into per¬
A, 13,0,13,1=1=
ceptual objects. Gestalt processes operating on the
entire pattern would produce groups of elements
that would be easy to join together into a perceptual
10,11,12,13,14
unit (Prinzmetal, 1981; Treisman, 1982). The re¬ FIGURE 11-15 The effect of context on pattern
lationship between attention and the Gestalt princi¬ recognition. The B and the 13 are identical figures.
330 Chapter 11/Form

Target objects

FIGURE 11-16

Another example of the effect of context on ulus evokes a series of expectations about objects
identification comes from a study by Palmer likely to be present. When the next object seen
(1975a). Palmer asked observers to identify objects matches these expectations identification is easier,
presented after they had seen either an appropriate whereas incongruous or unexpected items become
context or an inappropriate context for that object. harder to identify.
For example, in Figure 11-16, the loaf of bread (A) It should be clear by now that the final inter¬
would be appropriate in the context of the kitchen pretation of a visual scene depends on both data-
counter displayed there, but the mailbox (B) or the driven and conceptually-driven processing (cf.
drum (C) would be inappropriate. Objects pre¬ Norman, 1976; Palmer, 1975b). Palmer (1975b)
sented after an appropriate scene were more readily demonstrated this notion in relation to face percep¬
identified than were the same objects presented af¬ tion. Look at Figure 11-17. Notice that when seen
ter an inappropriate scene. It seems that what you as part of a face, any bump or line will suffice to
see immediately before the presentation of a stim¬ depict a feature. When we take these features out

A In context C

FIGURE 11-17 Facial components are easily recognized in context (A), but out of context they are much
less identifiable (B) unless they are made more detailed (C) (Palmer, 1975b). (From Norman, Rumelhart & the lnr
Research Group 1975. Copyright 1975 by W. H. Freeman and Co. Used by permission.)
 Theories of Object Identification 331

of context, they do not really portray the objects the input pattern. These shouts are listened to by
very well. We actually require more of a detailed the cognitive demons, each of which is listening for
presentation (such as in Figure 11-17C) to identify a particular combination of shouts from feature de¬
facial features unambiguously when presented in mons. As the information is analyzed by the feature
isolation. Thus, in this situation the conceptually- demons, the cognitive demons start “yelling”
driven contextual expectations compensate for lack when they find a feature appropriate to their own
of detail in the data-driven feature extraction. pattern, and the more features they find the louder
they yell. A decision demon listens to the “pande¬
monium” caused by the yelling of the various cog¬
THEORIES OF OBJECT nitive demons. It chooses the cognitive demon (or
IDENTIFICATION pattern) that is making the most noise as the one
that is most likely to be the pattern presented to the
Many different types of theories have been offered sensory system.
to explain how perceptual objects are identified. Pandemonium is one of many similar models
We have already encountered parts of a few of that depend on data-driven analysis of simple fea¬
them in our discussion. One recurring pattern, tures much like those to which cortical cells are
common to several of these theories, is the pre¬ tuned (see Chapter 3). In general, such models can
sumption that during the first stage of perception account for many aspects of object identification,
information in the retinal image is processed to de¬ such as the mistakes people make when trying to
tect simple features, which are then combined into identify alphabetic characters (see Ashby & Perrin,
perceptual objects at a second stage and compared 1988; Keren & Baggen, 1981; Townsend & Ashby,
to memories of other perceptual objects at a third 1982). Variations of the basic theory can be con¬
stage to determine the identity of the stimulus. Ex¬ structed to account for between-letter confusions
pectations and context play their role at the stage that either do (Sanocki, 1987) or do not (Friedman,
of comparing the input to memories; the match re¬ 1980) depend on minute details of the letters such
quired for identification doesn’t need to be as good as size, type style, and the like.
if a particular object is expected either because of
the context or because it is being searched for. The
Identification-by-Components
major disagreements arise in specifying just what
aspects of the perceptual object (the features) are Although pandemonium is a successful theory, it
used in the comparison process. has several shortcomings. One of these is that the
set of features used to characterize perceptual ob¬
jects is really rather arbitrary. This problem has
Pandemonium
been addressed by a more recent approach based on
One very successful theory of this type emphasizes the idea that objects can be represented by what
data-driven feature extraction processes. It is called these researchers feel is a nonarbitrary, primitive
pandemonium, because each stage of the analysis set of parts or modules (cf. Brooks, 1981; Guzman,
of an input pattern was originally conceived of as a 1971; Marr, 1982). According to this theory, there
group of demons shouting out the results of their are some simple properties of visual geometry that
analyses (Selfridge, 1959). Figure 11-18 shows generally remain constant even though the image
how the theory works. In the first stage, an image and the observer are moving and changing in var¬
demon passes on the contents of the retinal image ious ways. From these properties, the researchers
to each of a set of feature demons. These feature derived a set of simple components of which
demons shout when they detect “their” feature in all perceptual objects are said to be composed
332 Chapter 11/Form

Cognitive demons
Feature demons (“shout” when receive
(decode specific certain combinations
features) of features)

Image demon
(receives sensory input)

Processing
of signal

FIGURE 11-18 The pandemonium model in action. The number of each type of feature registered by
the feature demons is indicated by which circle is blackened in each box and by the number of times the
feature is printed in the box.
 Theories of Object Identification 333

(Biederman, 1987). All of the components, called jects (see also Figure 11-12). Biederman (1987)
geons, are (at a mathematical level) variations of a calculated that a very small set of such geons (no
generalized cylinder. Figure 11-19A displays some more than 36) can generate over 150 million 3-
of these geons and something of the variety of geon objects, ample to describe even the richness
shapes that can be derived from just one of them. of human object perception.
Figure 11-19B shows how various combinations of Not only does this theory provide a set of
the simple geons give rise to various perceptual ob¬ primitive features (the geons) with which to

FIGURE 11-19 (A) On the left is a partial set of geons, from any one of which many variants can be
created (as on the right for one of them) by varying the basic parameters. (B) Some of the objects that
can be created from geons. (Based on Biederman, 1987)
334 Chapter 11/Form

describe an object but it also accounts for some of culations made. Finally, the third level, hardware
the major phenomena of object identification. The implementation, describes the actual implementa¬
importance of geons can be seen from the fact that tion of a given algorithm in some device. The de¬
if you degrade a pattern but still permit an arrange¬ vice was originally meant to be a computer, but
ment of as few as two or three geons to be seen, it speculations are often made about whether the
doesn’t hinder object identification (Biederman, same principles might work in a brain.
1987). The idea is that the description of an object This computational approach is quite different
in terms of its geon components is compared to from the usual psychological approach. For exam¬
other remembered (geon) descriptions. Moreover, ple, perceptual psychologists typically deal only
even if you’ve never seen the form before, as in with the second level (representation and algo¬
Figure 11-12, it can still be analyzed into its geon rithm), and even then usually do not use mathe¬
components, and even tentatively placed into a cat¬ matical language to specify their theories. A good
egory. Finally, this approach also emphasizes that example of the psychological approach is that of
the law of Pragnanz applies to the geons that make Treisman (e.g., 1986b). She defines the task of ob¬
up an object and not to the whole object. We tend ject perception in terms of three domains: the phys¬
to see the best geons, which will then give rise to ical domain (physics’s description of the real
the best object. world), the phenomenological domain (what we
experience, the perceptual object), and the func¬
tional domain (the various processes that connect
Computational Approaches
the other two domains). Really, all of Treisman’s
There is a class of theories of object perception and domains are within Level 2 of the computational
identification that arises from a very different view¬ approach, as shown in Figure 11-20.
point and tradition. It began with the desire to de¬ Not all computational approaches to object
velop machines that can “see” and identify patterns. identification follow Marr, but all do have in common
Perceptual researchers soon recognized that this line the emphasis on being able to compute the solution to
of endeavor might lead to some useful insights about an identification problem (that is, having a computer
alternative ways in which biological organisms see do it). In Marr’s own approach, the computational
forms, since, at the very least, in order to program a theory stage discussed above is broken down into
machine to perform this function we must be very computable problems. First, a set of routines com¬
specific about the way information is selected and putes from the retinal image what Marr called a pri¬
processed. mal sketch. This is an abstract representation of
Probably the most influential version of this where various contours (described mathematically)
approach has been that of Marr (1982), who first are located, along with a rough grouping of contours
analyzed the problem of creating such a “seeing into shapes (blobs) that may belong together. From
machine” into three levels. The first required a the primal sketch is computed the 2V2-D sketch,
computational theory, which attempts to specify in which contains a description of the orientation and
formal terms (usually mathematical) exactly what approximate depth relationships of potential surfaces
the nature of the visual problem is, the information in relation to the viewer. Finally, a 3-D model is com¬
that may be available, and the information required puted from the 2V2-D sketch. This model, corre¬
to ultimately achieve the correct identification. The sponding to what we have called the perceptual
second level, called representation and algorithm, object, represents shapes and their spatial organiza¬
has to do with the various ways the required infor¬ tion in terms of their relationship to each other and in
mation could be represented and the necessary cal¬ terms of modules similar to geons. This set of de-
 Theories of Object Identification 335

Comparison of Object Identification Approaches

-r
i 1
Level 1 i Level 2 1 Level 3
Approach i 1
i 1
1 l
Computational Computational 1 Representation i Hardware
(Marr) theory J and algorithm l\ implementation
/
/
/

Psychological Physical Functional Phenomenological

(Treisman) domain domain domain

FIGURE 11-20 An indication of the different concerns, and levels of analysis, of computational and
psychological approaches to object perception.

scriptions constitutes the computational theory of ob¬ any insights as to how the brain actually processes the
ject representation (when it is expressed formally, visual information in a biological system, such as in
that is), although we don’t provide you here with the humans.
mathematical procedures for accomplishing these It is truly humbling to realize that understand¬
computations. Many investigators are still working ing the simple act of recognizing that you are look¬
both to refine the computational theory and to con¬ ing at a pencil, and that the pencil is not part of the
struct representations and algorithms that will do the desk on which it rests but is a separate object in its
computations (see, e.g., Ullman, 1986). Ultimately, own right, remains a problem about which there are
however, the usefulness to psychology of such ap¬ many theories but still no final answer.
proaches will depend on whether they provide us with
336 Chapter 11/Form

GLOSSARY Identification Naming a perceptual object and

knowing something about it.
The following definitions are specific to this book. Illusory conjunction Percept consisting of an incor¬
rect integration of features from separate objects into a
Backward masking The phenomenon whereby ex¬
single perceptual object (e.g., seeing a red X when only
posure to a second stimulus interferes with the perception
a red O and a green X are present).
of an initial stimulus presented at a critical interval pre¬
Integral stimulus A stimulus that is experienced in
viously.
all of its aspects at once and inseparably.
Conceptually-driven processing Perceptual infor¬
mation processing that is guided by conceptual pro¬ Law of closure The Gestalt law stating that contours
cesses, such as memories and expectations concerning that form a closed region tend to be attracted to each
the nature of the incoming stimulation. other and form a figure.
Law of good continuation The Gestalt law stating
Configural feature A stimulus whose aspects can be
seen separately but still produce an emergent feature, that figural elements that form smooth curves tend to be
such as closure, that dominates processing. grouped together.

Contour Any place in the retinal image where the Law of Pragnanz The Gestalt law stating that the
light intensity changes abruptly. psychological organization of the percept will always be
as “good” as prevailing conditions allow.
Data-driven processing Perceptual information pro¬
cessing that responds directly to properties of the incoming Law of proximity The Gestalt law stating that ele¬
stimuli according to fixed procedures and without influence ments close to one another tend to be grouped together.
from memories, expectations, or the like. Law of similarity The Gestalt law stating that the
Emergent feature A feature that characterizes a par¬ more similar figural elements are, the more likely they
ticular configuration of parts rather than others, and is at are to be grouped together.
least as perceptually salient as any of the parts. Local The detailed aspects of a figure as opposed to
Features Attributes of a shape that distinguish it the global aspects.
from other shapes. Metacontrast Interference (e.g., masking) between
Feature integration theory A theory of how fea¬ two contours that are adjacent but not necessarily over¬
tures are integrated to form perceptual objects; it assumes lapping.
that features are extracted in parallel and automatically, Pandemonium A computer model of pattern identi¬
but that attention must be paid to a particular spatial lo¬ fication based on a series of successive stages of feature
cus in order for perceptual objects to be formed from the analysis and recombination.
features. Perceptual object The perceptual experience of a
Figure Integrated visual experience that “stands part of the retinal image forming a whole entity, an “ob¬
out” in the center of attention. ject,” that usually corresponds to a real world object.
Focal attention Active attention focused on a partic¬ Preattentive processes Perceptual processes that op¬
ular spatial location. erate to produce shapes and register their features without
Ganzfeld A visual field that contains no abrupt lu¬ the need for focal attention.
minance changes and thus no visible contours. Processing dominance When some aspects of a
Geon One of the primitive components from which form are processed faster than, or are more immune to
perceptual objects are constructed in identification-by¬ interference from and interfere more with, other aspects
components theory. of the same form.
Gestalt A concept and school of psychology Recognition The experience of perceiving something
emphasizing the notion of a meaningful and coherent as previously known.
form, or “whole.” Retinal image The two-dimensional distribution of
Global The overall arrangement of parts of a figure, light of various intensities and wavelengths on the retina.
as opposed to the local details. Separable stimulus A stimulus that has single fea¬
Ground The background against which figures ap¬ tures that cannot be easily integrated.
pear.
 Glossary 337

Shape A region of the retinal image surrounded by ences between regions of the visual array in the textons
contours. they contain define textural contours.
Stabilized retinal image An image whose retinal Textural contour A contour created by a boundary
position remains constant regardless of eye movements. between two areas differing in visual texture.
Subjective (or illusory) contours Contours that are Visual field All the parts of the environment that are
consciously experienced, but not associated with physical sending light to the eyes at any moment.
stimulus change. Visual texture Aggregates of many small luminance
Texton Elongated blobs of a particular color, length, contours or dots of different colors in the retinal image.
width, or orientation, line ends, or line crossings; differ¬
CHAPTER
12
Speech and
Music
MUSIC
Musical Tones
Musical Forms
SPEECH
The Speech Stimulus
Consonants and Vowels
Phonemes
Acoustic Properties of Speech
Issues in Speech Perception
Ambiguity and Invariance
Is Speech Special?
Development
Context
Theories of Speech Perception
340 Chapter 12/Speech and Music

MUSIC lower pitch. In musical scales, however, there are

other relations between the pitches of sounds that
I f a classical pianist sits down to play a concerto
by Bach, the result is undoubtedly music. If a
help to describe what we hear.
One relationship that is important for musical
small child sits down at the piano, having never perception involves the concept of the octave. In
studied the instrument, his best efforts at playing the common scale—do, re, mi, fa, sol, la, ti, do—
produce auditory stimuli politely describable as the second do is one octave higher than the first do.
random sounds but perhaps more accurately called To get a tone one octave higher than another, you
noise. This is in spite of the fact that there are only simply double the frequency of the first sound.
88 notes on a standard piano, which is all that, say, Thus, middle C on a piano has a frequency of
Arthur Rubinstein had to work with, and that any¬ 261.6 hertz (Hz) and the C one octave higher is
one can sound each note, although not necessarily 523.2 Hz. Sounds separated by an octave seem
in the sequence suggested by Bach. What, then, more similar than do sounds separated by less than
distinguishes between a mere collection of sounds, an octave (such as C and G, or do and sol). In
varying in intensity, pitch, timbre, and duration, other words, as you ascend the musical scale all of
and the stimuli we experience as music? The an¬ the dos sound similar to each other, as do the res
swer seems to be that music is created by the and so forth, regardless of how many octaves apart
context, or relationship of each sound to those pre¬ they are.
ceding and following it. A child’s random striking This tendency for tones separated by octaves
of the keys on the piano, or your experience when to sound musically similar means that a simple one¬
presented single stimuli in a pitch judgment exper¬ dimensional scale (with low notes at the bottom
iment, lacks the relationships that would transform and high at the top) will not suffice to describe our
the stimuli into music. Once you perceive a se¬ perception of musical pitch. As we ascend the
quence of sounds as music, however, an entirely scale, each note seems to reappear periodically, cy¬
new set of phenomena emerges, and even the per¬ cling through again with a different quality. This is
ception of individual sounds will be different reminiscent of the situation we discussed in Chap¬
(Krumhansl & Shepard, 1979). ter 5 for color. There, when dealing with the di¬
mension of hue, we found that we seemed to be
moving around a circle, recycling from red through
Musical Tones
yellow, green, blue, and purple, then back to red.
One of the most striking examples of the perceptual In color we could independently vary the intensity
phenomena that differentiate musical perception dimension linearly, without changing the identity
from other forms of auditory perception is the dif¬ of the hue. In music we can vary the tone height
ference between musical pitch and acoustic pitch. without changing the identity of the tone. Thus we
In Chapter 7 we introduced the mel scale for acous¬ can have a high do and a low do that sound similar,
tic pitch. This scale shows that when subjects are although one is clearly of a higher pitch. To rep¬
asked to adjust a set of pure tones differing in fre¬ resent this relationship graphically we have to use
quency so that the intervals between “notes” are a three-dimensional scheme, much as we did for
equal steps, the results are not the same frequency color. In 1846, Drobisch proposed that musical
intervals that correspond to our common (equally pitch might be represented as a helix, an idea that
tempered) musical scale. For both the mel scale has persisted until the present (see Shepard, 1982).
and the musical scale, sounds vary along the di¬ Figure 12-1 shows this representation of tonal
mension musicians call tone height, which is sim¬ qualities.
ply whether a sound appears to be of higher or The additional aspect of musical pitch repre-
 Music 341

circle. It can be seen in Figure 12-1 as the circular

component of the helix, whereas tone height is rep¬
resented by the vertical component. One complete
turn of the helix (a 360-degree rotation in the hor¬
izontal plane) represents a single octave, and all
notes with the same name fall on a line drawn
down the helix onto the same point on the chroma
circle at the bottom and all sound similar. For con¬
venience, we have labeled the spiral with some
names of musical scale notes corresponding to the
white keys on the piano rather than using tone fre¬
quencies. Thus, Al is the lowest A note on the
piano, A2 is one octave above it, and so on.
Although the usual laboratory experiment in
pitch perception is able to isolate tones that seem
to vary only in tone height, it is also possible to
produce a series of sounds that vary only in tone
chroma (move around the chroma circle in Figure
12-1) yet seem to ascend in height (Shepard, 1964;
but see also Bums, 1981; Pollack, 1978). You can
try this for yourself in Demonstration Box 12-1.
An interesting variant of this illusion was reported
by Deutsch (1986). She found a circular pattern of
tones that is heard as ascending when played in one
musical key but descending when played in an¬
other, which is contrary to the experience in music
that a melody sounds the same when transposed to
another key.
Although the musical helix represents the ma¬
jor aspects of musical pitch, it doesn’t tell the
whole story. Although confirming the importance
of the chroma circle, much work has indicated that
other relations between musical tones are also quite
important (see Krumhansl & Kessler, 1982; She¬
pard, 1982). For example, when one tone is exactly
FIGURE 12-1 A regular helix represents the two
aspects of musical pitch, height and chroma.
1.5 times the frequency of the other (a 3-to-2 ratio,
or a perfect fifth musically) the two tones seem to
“go together’’ or sound better together than when
sented in the figure is called tone chroma, a name the tones are separated by other frequency steps
that shows its similarity to hue in the realm of color (save for the octave step where tones sound simi¬
perception. Thus, all dos have the same chroma, as lar). This and other relations between tones and
do all res, and so forth. More precisely, all tones between keys (a particular group of notes that
with the same name (e.g., C or G) share the same comprise the musical scale used for a particular
chroma. Like hue, tone chroma is represented by a melody) require a much more complicated
342 Chapter 12/Speech and Music

DEMONSTRATION BOX 12-1. The Tonal Staircase

Shepard (1964) invented a series of complex tones

generated by a computer that when listened to in se¬
quence seemed to continually increase in pitch. That
is, each step between tones was perceived as being a
step upward in pitch. Shepard, however, used a trick
in generating this series of sounds and in fact the se¬
ries ended where it had begun, completing a journey
around the chroma circle (see Figure 12-1). The con¬
tinuing rise in pitch was an illusion. It is rather diffi¬
cult to produce Shepard’s series of sounds without
complex equipment, but it may be possible for you to
hear the illusion anyway. First, fill a glass partially
full of water; a crystal glass would be best, perhaps a
wine glass, but any glass with a “ring” should do.
Now tap the glass gently with a knife or other imple¬ even though the sounds are highly similar and the
ment to make it ring. Continue tapping gently to pro¬ fundamental frequency probably does not change. In
duce a series of complex sounds. Each sound will be Shepard’s demonstration, this illusion is quite similar
slightly different from the others in its frequency to the visual staircase illusion shown here. The stairs
components because of variation in the way the knife seem to climb endlessly but never get anywhere. This
strikes the glass. The series of sounds produced this is probably the most striking demonstration of the
way can often be heard to ascend or descend in pitch reality of the quality of tone chroma in musical
continuously, much in the way Shepard’s sounds did, sounds.

representation, and agreement on what that might sionally misidentify the octave where it is located
be has not yet been reached (see Krumhansl & because of the perceptual similarity we have al¬
Kessler, 1982; Shepard, 1982). What is certain, ready discussed). However, when these same mu¬
however, is that such relations play an important sicians are presented with pure, sine-wave tones
role in music theory and do have a perceptual real¬ they can identify the notes correctly only about half
ity, at least for the musically inclined. Since these the time (Lockhead & Byrd, 1981). This is still a
relations must be taken into consideration in any lot better than those without perfect pitch, who tend
description of music perception, it is much more to be correct on only about 8 percent of the trials,
difficult to generate a musical space than it was to but it is nowhere near “perfect.” Remember from
generate a color space to represent our perceptions Chapter 7 that the perception of the pitches of a set
(see Chapter 5). of tones is predominantly determined by the fre¬
How well can people identify musical notes? quency of the fundamental, or lowest tone, in the
In music contexts, we frequently hear of individu¬ set. When a note is played on a musical instrument,
als who have perfect pitch. Such musicians are able though, different harmonic frequencies (multiples
to identify a musical note that is played on an in¬ of the fundamental) are also sounded, giving the
strument even when it is presented in complete timbre, or complexity, to the sound. Clearly, the
isolation from other notes (although they do occa¬ musicians who have perfect pitch are using more
 Music 343

than the fundamental frequency of the musical what octave they are from. For example, an E ma¬
notes to identify them; they are probably using jor chord is composed of the notes E, G#, and B
these higher harmonics (e.g., in a piano note) to no matter whether the notes are three octaves up
aid in identification. From their self-reports, these from the lowest on the piano (E4, G#4, and B4) or
musicians judge chroma by comparing the test note six octaves up (E7, G#7, and B7). This aspect of
with an internal (remembered) standard for each musical pitch also is consistent with the helix
note, whereas the people without perfect pitch sim¬ shown in Figure 12-1, if spacings between the
ply seem to guess at chroma. notes are logarithmic, since the intervals between
The spacing between tones or notes is called a the notes remain the same regardless of tonal height
musical interval. In music of the Western world, or absolute frequency.
scales have been arranged with logarithmic musical
intervals. This is because frequency intervals that
Musical Forms
are equal on a logarithmic scale are perceived as
being approximately equal intervals of musical So far we have described a few of the most impor¬
pitch. For example, if you heard an interval gen¬ tant local, or individual component, aspects of mu¬
erated by a pair of notes with frequencies of 200 sical sequences or combinations of notes. That is,
and 400 Hz and another generated by a pair with we have described music at the level of the partic¬
frequencies of 2000 and 4000 Hz, they would seem ular frequencies of the notes that make up the
to be about equally large (since log 400 —log 200 music. Any sequence of notes also has global prop¬
= log 4000 —log 2000 = 0.3). One implication of erties, however, that give an overall pattern to the
this is that any pair of steps between notes of the sound sequence and are very important to its mu¬
musical scale that are separated by the same num¬ sical character since they comprise the aspect of
ber of intervening notes will appear to be the same music called melody. These properties include the
size. In the preceding example we were dealing sequence of pitch changes, the proportion and sizes
with octaves, a separation of eight notes (from do of the various ascending and descending intervals,
to do, excluding sharps and flats), which are al¬ and so on. They are global properties because they
ways separated by 0.3 log units. When you subtract are perceived in relationship to one another, rather
logarithms of frequencies you are performing the than as individual features (Cuddy, Cohen & Mew-
equivalent of division on the frequencies. This hort, 1981; Deutsch, 1978). In fact, melodies can
means that when we are talking about the logarith¬ even be recognized on the basis of such global
mic intervals between musical notes we are also properties alone, although performance is better
talking about the ratios between frequencies when the local cues are also available. These global
of tones. cues are often collectively described as the contour
When three or more musical tones are played of a piece of music. Contour means much the same
at the same time, we have a chord. Chords give thing here as it does in visual pattern perception. It
much of the characteristic sound to what we call is the general shape of the musical sequence of
music (see e.g., Krumhansl, Bharucha & Kessler, sounds, defined in terms of rises and drops in fre¬
1982). Formal music theory provides a somewhat quency instead of in terms of changes in direction
complicated system for naming chords, which we of a line in space. Figure 12-2 shows some exam¬
won’t go into here. Suffice it to say that chords too ples of musical passages that share a contour even
are defined in terms of the ratios of the frequencies though they are in different positions on the musi¬
of the tones that constitute them. Chords whose re¬ cal scale.
spective components stand in the same frequency A typical piece of music consists of a rather
relationship are given the same name, no matter long sequence of different notes and chords, similar
344 Chapter 12/Speech and Music

Time

FIGURE 12-2 When a melodic sequence is transposed to different positions on the musical scale (A) it
still retains the same contour (B).

to a long string of sounds uttered by a person mak¬ There are three major principles that tend to
ing a speech. Just as we perceive a complicated group notes or chords together in a listener’s con¬
hierarchy of words, phrases, and sentences as we sciousness (see Deutsch, 1978). The first of these
listen to someone speaking, we also organize music is based on pitch range. This principle says that
in a hierarchical fashion. Combinations of notes notes that are close together in pitch are perceived
form motives (pronounced “mo-teevs"), combina¬ as part of the same perceptual unit, whereas notes
tions of motives form phrases, and so on (Deutsch, that are far apart in pitch are perceived in separate
1978). groups. This principle is related to the Gestalt law
How are these combinations formed percep¬ of proximity, and its visual equivalent can be seen
tually? It turns out that notes, motives, and phrases in Figure 11-6G. Whenever the same instrument
are grouped together according to principles that plays both a melody and an accompaniment, they
very strongly resemble those that govern visual pat¬ are played in different pitch or frequency ranges so
tern perception, suggesting that the perception of that the melody will be the figure (the part that
music and melody is a form of auditory pattern per¬ stands out perceptually) and the accompaniment
ception. Of particular interest in this regard are the will be the ground (the background against which
Gestalt laws of grouping that we discussed in the melody is imaged). An example is in folk-gui¬
Chapter 11. tar playing, where the performer often keeps a
 Music 345

steady accompaniment going on the bass strings of phrases were being played. A similar phenomenon
the guitar while playing a melody on the treble occurs for the music in C played by two pianos.
strings. Dowling and Harwood’s (1986) book includes a
Another principle of grouping is based on taped demonstration of this and many other musical
timbre (see Chapter 7). Different types of musical phenomena that you might wish to obtain from the
instruments play the same notes with different library and try for yourself.
timbres, which gives them their characteristic So far, we have concentrated on variations in
sounds and allows you to identify which instrument the frequency of musical notes or combinations of
is playing any given note. When several instru¬ notes, neglecting the other major dimension of mu¬
ments are playing simultaneously, the observer sical sounds, their duration. You probably learned
tends to group those of similar timbre into units, in in grade school that written sequences of musical
a way analogous to the Gestalt law of similarity notes indicate not only tone height (frequency) but
(see Figure 11-6B or C). In symphonic music, this also tone length—a quarter note is held for half the
principle is used to separate phrases that have a duration of a half note, and so forth. Both the du¬
similar fundamental frequency range but a different ration and height of notes are vital in determining
musical message. Also, timbre provides an addi¬ our perception of melody. You can clearly see this
tional principle of grouping to that of pitch range in Figure 12-4, which presents three musical ex¬
when different instruments play different parts of a cerpts. All have the same tonal contour. They dif¬
piece (as in the lead and rhythm guitar parts of a fer only in the time that the notes are held, but this
piece of modem rock music). causes a tremendous difference in the melody you
The third principle is based on the concept of perceive. The first is the beginning of the familiar
good continuation in direct analogy to the Gestalt American folk song “Red River Valley,” the sec¬
principle of that name (again, Figure 11-6D pro¬ ond is the opening of Mozart’s Serenade in D, and
vides a visual analogue of this principle). Se¬ the third is the beginning of the second movement
quences of frequency changes in the same direction of Beethoven’s Symphony No. 5. If you play a
(e.g., rising up the scale) tend to be perceived as musical instrument you might want to try these
part of the same sequence, whereas changes in di¬ phrases for yourself, just to hear how different they
rection between sequences of notes tend to act as sound.
boundaries between segments (Deutsch & Feroe, A sequence of sounds of various durations
1981). A good illustration of this is the illusion possesses rhythm and tempo. Tempo is the per¬
shown in Figure 12-3, first described by Deutsch ceived speed associated with the presentation of the
(1975). Two different sequences of tones are pre¬ sounds, and rhythm is their perceived organization
sented, one to each ear, as in the left panel of A. in time. When listeners are presented with a se¬
Seventy percent of Deutsch’s listeners heard the quence of sounds they spontaneously organize it
sound sequences represented in the right panel of into sub-sequences consisting of an accented sound
A; apparently good continuation created an audi¬ followed by at least one and sometimes several un¬
tory “stream” of notes (see Bregman, 1978) in accented sounds (Bolton, 1894). This is why the
each ear that was illusory. Later data indicated that ticking of a clock seems to go “tick, tock, tick,
such sequences are ambiguous figures, as in Figure tock,” despite the fact that every ticking sound
11-3, that have competing organizations (Smith, emitted by the clock is identical. This spontaneous
Hausfeld, Power & Gorta, 1982). However, some organization happens when the sounds are pre¬
cases are compelling enough to be used by com¬ sented at rates between 10 per second and 1 every
posers of music, as in Figure 12-3, B and C. In B, 2 seconds and seems to be optimal at rates of 2 to
two violin sections play the left phrases of music 3 sounds per second. Under some circumstances
but the audience hears the notes as if the right the percept may vary in the degree of accenting of
346 Chapter 12/Speech and Music

Stimuli presented What the listener hears

Right ear * J J J J A—
Left ear
J J J ..W J J J-H

0 fttf R • '• 0 m. - -n-

Violin section 1
r Eftr+rH
j ,n p,i f=h
Violin section 2
— ^

Piano 1 rf=n

Piano 2
c.-—
FIGURE 12-3 Three examples of the Deutsch illusion. In all cases, the left part of the figure shows what
is played, the right shows what is usually heard. (A) The original, from Deutsch, 1975. (B) From
Tchaikovsky, Sixth Symphony, last movement. (C) From Rachmaninov, Suite for Two Pianos, Opus 17,
second movement.

sounds so that a fairly complex rhythmic structure music (Deutsch, 1978; Handel & Oshinsky, 1981).
is perceived despite a physical stimulus that is ab¬ Actually, the most recent evidence indicates that
solutely regular. In the perception of music, these the principles associated with determining an ap¬
perceptual rhythms are superimposed on a deliber¬ parent melody and those determining your percep¬
ately manipulated rhythmic structure according to tion of rhythmic structure add together to produce
principles of grouping similar to the Gestalt-like a coherent musical pattern (Palmer & Krumhansl,
principles mentioned above. This rhythmic organi¬ 1987). Demonstration Box 12-2 shows how sounds
zation interacts with the organization induced by may be grouped together by varying the rhythm or
the variations in pitch of the musical notes, making timing between them.
it easier to perceive the melodic structure of the Music is an important part of every culture
 Music 347

Beethoven: Symphony
No. 5 (second movement)

FIGURE 12-4 Three musical phrases in which the melodic sequence of tone height is the same, and the
tones differ only in duration.

DEMONSTRATION BOX 12-2. Rhythmic Grouping

In this demonstration you will produce a series of tap¬ sible to hear this any other way. This is analogous to
ping sounds as stimuli. In order to indicate how your the Gestalt principle of similarity (the visual analogue
taps should be distributed in time, let us establish a is shown in Figure 11-6).
sort of rhythmic notation. Whenever we present a V While you are tapping, you can see that perceptual
it indicates a tap, whereas a hyphen indicates a brief groups or clusters can be formed by frequency or
pause. First, tap this simple sequence W-V. Listen timbre differences despite the absence of rhythmic
carefully, and notice that the first two taps seem to differences. Begin by steadily tapping a surface with
“go together” or form a unit, but(the last seems to your pencil. Make sure that the tapping rhythm is
stand alone. Now, repeat this sequence of taps several steady and unchanging. Now take a piece of paper
times and try to mentally change this organization so and slip it between the surface and your pencil and
that you have two groups with the first tap (VO form¬ notice that the sound quality changes. Without chang¬
ing one and the last two taps (V-V) forming the other. ing your rhythm, slip the paper in and out so that you
Notice that no effort of will allows you to do this. are tapping table, table, paper, paper, table, table,
The two taps that are close together in time seem to and so on. Notice that the sounds seem to take on a
go together and the other does not. This is analogous grouping, with the table taps together and the paper
to the Gestalt principle of grouping by proximity that taps together, and it seems, despite the fact that you
we discuss in Chapter 11. are tapping quite steadily and monotonously, that the
Next, try the sequence VW-V-V-VW-V-V-VW, sounds have a rhythm that goes table, table, pause,
and so on. Notice that this is a repetition of three paper, paper, pause, table, table, and so forth. Here,
quick taps, followed by two slow taps. Notice that grouping by perceived similarity has imposed an ap¬
now the three taps form one group, and the two slow parent rhythm on the sound sequence.
taps form another, perceptually. It is virtually impos¬
348 Chapter 12/Speech and Music

and it has many more aspects than those few we speech sounds when they occur at rates of less than
have space to cover here. Several recent books 1 unit every 1.5 seconds (2/3 unit per second)
(Dowling & Harwood, 1986; Howell, Cross & (Warren, Obusek, Farmer & Warren, 1969). Re¬
West, 1985; Pierce, 1983; Sloboda, 1985) and a member, we are discriminating not only various
special issue of the journal Perception & Psycho¬ units of sound but also their order of occurrence (as
physics (Dowling & Carterette, 1987) cover these when we discriminate between the words tab and
other aspects in some detail. In addition, we are bat). Advertisers often take advantage of our ability
now beginning to find that certain neural mecha¬ to process rapidly occurring speech sounds by hav¬
nisms may be critical to music perception. Perhaps ing announcers in commercials speak much faster
comparisons between the musics of different cul¬ than usual speech. Our ability to interpret speech
tures may yield some insight into which aspects of sounds much more quickly than the identity or or¬
music perception depend on learning the musical der of nonspeech sounds suggests to some re¬
vocabulary of a particular culture and which de¬ searchers that speech may involve some special
pend on mechanisms that characterize all human form of perceptual processing, although this sug¬
beings (Deutsch, 1982). Until such data are avail¬ gestion leads to controversy.
able we will continue to listen to a symphony of Simple knowledge of the nature of sound
theories and speculations. signals may well not be enough to allow us to un¬
derstand how people interpret speech sounds so ac¬
curately, since we can understand speech even
SPEECH when the signal is grossly distorted or transformed
(see Remez, Rubin, Pisoni & Carrell, 1981). Ex¬
“I just can’t understand it,” Consuela muttered to amples of distorted but understandable speech are
herself as she strolled down the Champs Elysees at found when you speak with an accent, with a
dusk. “I’ve given myself practically a Berlitz mouthful of food, or while holding your nose. De¬
course in French. I can read French fluently. Signs vices like telephones and radios also produce dis¬
and papers are no problem at all. Yet every time I tortions of the sound signal that don’t greatly affect
try to talk to someone in a store, or on the street, the intelligibility of speech. Telephones allow only
and especially when I go to a student party to try a limited range of frequencies to pass along the
to find out where the excitement is, I can’t under¬ wire, yet normal conversation is possible over the
stand a word anyone is saying. It all sounds like telephone (but try it in a foreign language that you
noise.” She sighed as she turned into the street don’t know very well and you will see how this
where her hotel was located. “I guess I just don’t distortion may render the conversation difficult to
have an ear for French.” understand). Other transmission systems can se¬
Consuela’s problem is not unique among those verely clip the speech signal (turn it into a series of
learning a second language. We seldom think about on or off pulses) yet still only marginally affect its
how remarkable an accomplishment speech percep¬ intelligibility (although it will sound different). You
tion is until we are in a situation like hers, where can also easily understand conversations despite a
we listen to a stream of nearly continuous speech background of noise, even when the noise level is
without understanding it. In our native language we only 6 decibels less than the speech intensity. In
can understand speech at rates of up to about 50 fact, even if the utterances and noise are the same
discrete sound units per second, although speech intensity we can identify about 50 percent of single
usually proceeds at about 12 units per second words, and we can understand speech even when it
(Foulke & Sticht, 1969). This is quite amazing, is less intense than the noise if it is about a familiar
since listeners can only determine the order of non¬ subject.
 Speech 349

As a first guess, you might think that speech bining them may be quite different (see Clark &
perception is merely another form of auditory pat¬ Clark, 1977; Ladefoged, 1975). Although our dis¬
tern perception and as such should follow princi¬ cussion will be limited to the English languge, you
ples similar to those of music perception. To a should be aware that a similar analysis can be con¬
certain extent this is true. However, because its ducted in any language.
function is to convey information, and because
speech plays such an important role in so many as¬
Consonants and Vowels
pects of human behavior, you will find that the way
we conceptualize speech identification problems In the English language, the vocal apparatus pro¬
and mechanisms will be quite different from the duces two basic types of speech sounds: vowels
ways we treated simple sound stimuli, or even mu¬ and consonants. They are produced by alternating
sic. Also, we should warn you in advance that sequences of opening and closing the vocal tract
there will be no final answers presented here. (the air passages in our throats, mouths, and nasal
Speech is one of the most controversial areas of areas) while air from the lungs flows through it.
perception research and we are moving only slowly Typically, closing movements produce consonants
toward the best way to think about the difficult and opening movements produce vowels.
problems that exist. Consonants can be classified along three major
dimensions, corresponding to the ways in which
they are produced. First, consonants can be voiced
The Speech Stimulus
or unvoiced. Voiced consonants consist of a con¬
When we dealt with visual pattern perception in striction of the flow of air out of the mouth fol¬
Chapter 11, we suggested that the major task for lowed very closely in time (less than 30 msec) by
the observer centered around the construction of vibration of the vocal folds (vocal cords). For an
perceptual objects that in turn carry information unvoiced consonant, the vocal folds don’t begin vi¬
about the environment. Speech perception is simi¬ brating until a longer time after the constriction,
lar, only in this case the objects are words, or usually more than 40 msec. Demonstration Box 12-
phrases, that carry linguistic information. As Lib¬ 3 allows you to experience an exaggerated version
erman and Mattingly (1985) put it, “the objects of of voiced and unvoiced consonants for yourself.
speech perception are the intended phonetic ges¬ The other two classifications are by how
tures of the speaker” (p. 2). This simply means (manner) and where (place) in the vocal tract the
that the ultimate goal of the speech perception pro¬ constriction of the airflow occurs. There are three
cess is to put into your consciousness an accurate manners in which the constriction can be produced.
representation of what the speaker intended to say. Stop consonants are formed by completely stopping
Given the intimate association between speech the flow of air from the lungs and then suddenly
and language it was probably inevitable that lin¬ releasing the flow. The p in “pea,” the t in “tea,”
guists were the first to set up procedures to describe and the k in “keep” are examples of stop conso¬
the speech stimulus. Their description depends on nants. Fricatives are formed by stopping the flow
the analysis of speech sounds in terms of how they through the nasal passages but leaving a small
are produced (phonetics) and how specific sounds opening in the mouth and forcing air through it,
distinguish words (phonemics). Such descriptions producing some variety of “hissing” sound. Ex¬
are universal, in the sense that speech production amples are the s in “best,” the z in “buzz,” and
and the methods of distinguishing linguistic units the/in “fricative.” Nasal consonants are produced
follow the same rules in every human language, through the nose, as you might have guessed. For
although the specific sounds and the rules for com¬ these sounds the mouth is closed and the air from
350 Chapter 12/Speech and Music

DEMONSTRATION BOX 12-3. Voiced and Voiceless Fricatives

Consonants in the English language are produced by of a z, such as in the word “zip.” Now try producing
a combination of vocal fold vibration and variations a tune (for example, “Oh, Susannah”) while you
in the passage of air through the oral cavity. Frica¬ sound the z. You should be able to do this easily; in
tives are a class of consonants that are formed when fact, it will sound something like a kazoo. However,
the mouth moves to a position that nearly blocks the now try the same thing while making the sound of an
flow of air, so air is forced through the tiny hole. /, such as in the word “fat.” F is a voiceless frica¬
However, fricatives also differ in whether they are tive; therefore, the lack of vocal fold vibration should
voiced (accompanied by vocal fold vibration) or make it impossible for you to produce a melody (re¬
voiceless (have no vocal fold vibration). To illustrate member, just produce the sound of /, do not hum si¬
this interaction between the flow of air and the quality multaneously). The quality of voicing, or vocal fold
of voicing, try the following demonstration suggested vibration, allows one type of fricative to be “me¬
by Brown and Deffenbacher (1979). Make the sound lodic,” whereas the other is lacking in that quality.

the lungs flows through the nasal passages. Exam¬ depends on the relative positions of various parts of
ples are the m in “mean” and the n in “nose.” the vocal tract. First, the position of the tongue in
There are two areas in the vocal tract where the mouth is important, whether front, center, or
most of the constrictions occur. In one, the lips or back, and so is its relative height. Changes in
the lips against the teeth control the flow of air tongue position and height change the shape of the
from the lungs; consonants produced by such con¬ resonating chamber in your mouth, resulting in the
strictions are called labial (labium is Latin for various vowels. For example, the ee in “beet” is
“lip”). Examples of labial consonants are the b in produced with the tongue at the front and quite
“bat” (voiced, stop), the v in “vat” (voiced, frica¬ high up in the mouth, the a in “sofa” is produced
tive), and the m in “mat” (voiced, nasal). The with the tongue central and of middle height in the
other area of constriction is inside the mouth. Here mouth, and the o in “pot” is produced with the
the tongue is positioned at various places, most of¬ tongue at the back and low down in the mouth. Try
ten at the ridge behind the teeth (alveolar), against saying these sounds and pay attention to where
the hard palate (palatal), or against the velum (the your tongue is. Second, the degree of rounding of
soft palate at the top of the throat, velar). Examples the lips is important in vowel production. The o in
are the t in “tin” (unvoiced, stop, alveolar), the n “who” is called a rounded vowel since the lips
in “gnat” (voiced, nasal, palatal), and the c in must be rounded in order to produce it, while the e
“cot” (unvoiced, stop, velar). Try producing these in “he” is unrounded since the lips are flat when
sounds and paying attention to where your articu¬ it is produced.
lators (the parts of the vocal tract used to produce
speech sounds, such as teeth, tongue, lips, and pal¬
Phonemes
ates) are while you are doing it.
Vowels are produced in a very different way Linguists have also worked out a descriptive sys¬
from consonants. In general, as we mentioned tem of speech units that is sufficient to describe any
above, vowels are produced by vibrating the vocal utterance in any language. In this system, the basic
folds as air moves out of the lungs through the unit of speech sound is the phone. A phone that is
open mouth. Here also, which vowel is produced used in a language to distinguish one word from
 Speech 351

another is called a phoneme. Every language has phonemic descriptions of words. Unfortunately, if
its own group of necessary phonemes; some have we look at the acoustic properties of the speech sig¬
only a few (Hawaiian has 11), whereas others re¬ nal itself, it is sometimes impossible to isolate the
quire as many as 60 (some African dialects) in or¬ sound features that correspond to a particular phone
der to distinguish all the words. North American in a particular uttered phrase. This is due to the fact
English has 40 basic phonemes (excluding regional that when we speak we often move our articulators
dialects such as drawls and nasal twangs). Table to produce sounds that provide information about
12-1 lists the major phonemes of American English several different phones simultaneously (this is
and the symbols used by the International Phonetic called coarticulation). It is therefore difficult (some
Association (IPA) to refer to them. Phonemes are say impossible) to find a clear correspondence be¬
usually set off from text by a pair of slashes (e.g., tween acoustic features and perceived phonemes
/p/ or /0/), but to make things more natural we will except in fairly specific, often simplified, instances.
simply italicize phonemes and give an example
word in quotes (e.g., the p in “pod”). You should
Acoustic Properties of Speech
try saying the various example words in Table
12-1, paying attention to how the sounds of the Since phones refer to speech sounds, it would seem
phonemes correspond to the way the sounds are useful to have a means of displaying the speech
produced (their articulatory features). sound signal so that we can better analyze it, per¬
Every phoneme has a unique description in haps with an idea of trying to determine some as¬
terms of its articulatory features (e.g., the b in pects of the relationship between articulatory and
“bat” is a voiced, bilabial, stop), so there is a one- acoustic features and the final perceived speech
to-one correspondence between the articulatory and units. One popular way is based on the fact that
any complex sound wave can be represented as the
Table 12-1. The Major Phonemes of North combination of a set of simple sine waves of dif¬
American English ferent amplitudes and frequencies (see Chapter 6).
For speech we must add the dimension of time,
Consonants Vowels since the speech waveform is not periodic (repeat¬
ing) but varies over time. The result of analyzing a
p pea e thigh i beet 0 go sequence of speech sounds into its simple sine
b beet 5 thy I bit D ought wave components on a moment-to-moment basis is
m man s see e ate a dot
displayed as a speech spectrogram. An example is
t toy 3 measure e bet 3 sofa
shown in Figure 12-5. In all such spectrograms,
d dog tf chip asbat 3 urn
the horizontal axis shows time in milliseconds from
n neat d3 jet u boot ai bite
k kill 1 lap u put au out the onset of the speech signal, and the vertical axis
good r rope A but or toy shows the frequency (in Hz) of the simple sine-
g
f foot y year D odd ou own wave components at a given moment. The intensity
S huge w wet of the sine-wave components is represented by the
h hot 0 sing darkness of the smudges on the spectrogram: the
V vote z zip darker the smudge the more intense the component.
M when / show Notice in Figure 12-5 that there are very intense
components from about 300 to 700 Hz in all of the
Note: The phonetic symbol is to the left of each column syllables. These components last about 200 msec
and its sound corresponds to the part of the word
represented in bold type. Some vowel and consonant
for the “bab” syllable and about 300 msec for the
combinations are also shown. “gag” syllable.
 352
Chapter 12/Speech and Music

(zh) Aouanbajj
Time (msec)

FIGURE 12-5 Speech spectrograms of the words “bah,’’ "dad,” and “gag,” spoken with a British accent.
(Based on Ladefoged, 1975)
 Speech 353

The dark smudges in a speech spectrogram are tinguished in speech spectrograms. The relative po¬
called formants. Formants are bands of especially sitions of the various formants roughly correspond
intense components that arise because the complex to the different vowel sounds. A special apparatus,
sound waves created by the passage of air from the called a vocoder, provides a sort of reverse spectro¬
lungs across the vocal folds and out through the graph, re-creating the sounds for any pattern of
mouth and/or nose are affected by the positions of sine-wave components it has been fed and even
the various parts of the vocal tract. For each mouth creating artificial speechlike sounds. Using artificial
posture and airflow pattern certain ranges of nearby speech stimuli it has been shown that only the first
frequencies are enhanced and others diminished, two formants are needed to create sounds that lis¬
and these show up as the pattern of smudges on the teners readily identify as vowels, hence in our
spectrogram. At least four formants can be distin¬ presentation we will indicate only the first two for¬
guished for each of the syllables in Figure 12-5. mants. Figure 12-6 shows the first two formants for
The one at the lowest frequency (around 500 Hz in a set of vowel sounds spoken by an adult male.
the figure) is called the first formant and is pro¬ Since people have different-sized mouths, noses,
duced by the shape of the pharynx (wall of the and throat passages, they have different ranges of
throat). Any change in the shape of the pharynx possible shape changes. This means that the first
produces a change in the frequency at which the and second formants will appear at a range of dif¬
first formant happens. The second formant, at ferent frequencies across different speakers. Figure
about 1400 to 1500 Hz in the different parts of 12-6 also shows, for comparison, one vowel spo¬
Figure 12-5, is produced by the shape of the oral ken by a child. Notice that both formants are cen¬
cavity. Higher formants are produced by complex tered on higher frequencies for the child than for
resonances of the vocal tract, including the nasal the adult.
passages. Consonants are generally indicated by changes
In general, vowels and consonants can be dis¬ of formants over short intervals of time (usually

the word

FIGURE 12-6 The first two formants for a series of vowel sounds made by an adult male voice. For
comparison, the last vowel sound is shown also as it would be made by a child’s voice.
354 Chapter 12/Speech and Music

FIGURE 12-7 Changes in the formant transition cause a systematic change in the consonant sound
heard for the same vowel. The vowel sound is the o in “sod.”

less than 100 msec) called formant transitions. A The rate of change in the formant transition is
typical formant transition is highlighted in Figure also quite important in the perception of the pho¬
12-5 and shown schematically in Figure 12-7. No¬ neme. Figure 12-8 shows that if we begin with the
tice in Figure 12-7 that as the formant transition e vowel as in “let,” a short formant transition is
changes the consonant changes from b to d to g, heard as the consonant b and a slightly longer one
although the vowel (defined as the two formants) as the consonant w; however, when it is made
remains the o as in “sod.” longer yet, it doesn’t sound like a consonant at all.

As in bet wet duet

the word
Time (each unit 100 msec)

FIGURE 12-8 The duration of the formant transition affects what is heard. When it is short a consonant
is heard; when it is long a shift between two vowels is heard.
 Speech 355

but rather like the change between two vowels, as computer scientists as well as psychologists and
the ue in “duet.” speech and hearing scientists—there is a tremen¬
Simple sounds of the sort described here do dous diversity of approaches to the field. One way
produce systematic perceptual responses, yet in to integrate these diverse views is to focus on the
natural speech the signal is not so regular. The the¬ major problems, or issues, common to most ap¬
oretically expected components are often missing proaches. We have chosen a few of these for dis¬
or distorted. Some individuals (such as Victor Zue) cussion here.
seem to be able to “read” speech spectrograms
with an accuracy of about 90 percent (Cole, Rud-
Ambiguity and Invariance
nicky, Zue & Reddy, 1980), but this is a rare qual¬
ity. In our usual conversational exchanges there is Although we suggested earlier that we can find
seldom a precise correspondence between the consistent features of the speech signal that corre¬
acoustic properties of the stimulus and the speech spond to our conscious perception of speech ele¬
signal as it is heard. As we will see below, whereas ments, this is actually an oversimplification that
it is possible to obtain a precise acoustic description only holds under certain controlled circumstances.
of any utterance in terms of a speech spectrogram, In natural speech we do not simply add together the
it is not always possible to say exactly what aspects various simple vowel and consonant components to
of that spectrogram are meaningfully related to produce the final complex utterance. Conversely,
speech perception. In some respects this is quite when we are listening to normal speech we may
reasonable, suggesting that we should be looking perceive a consonant in the absence of the expected
for relationships or patterns rather than individual formant transition, or hear a vowel that is not the
speech units. In music we can recognize a melody one predicted by the actual formants present. Put
played on a piano in one key as being the same simply, the problem is that specific features in the
melody when played on the clarinet in another key acoustic signal do not always predict specific per¬
even though no two acoustic signals corresponding ceptual interpretations of the speech stimulus.
to parts of the melody are ever the same. The There are a number of ways our perception of
melody is carried in the relationships among the the speech signal differs from its acoustic proper¬
individual sounds rather than in any sound pattern ties. For instance, we hear speech in segments that
itself. Such a global analysis based on relationships we interpret as phonemes, words, or phrases sepa¬
instead of on individual acoustic components may rated by pauses. In actual fact, the acoustic signal
be needed in speech perception as well. is often a continuous stream, without any obvious
breaks or other “markers” corresponding to these
perceived subdivisions (e.g., Chomsky & Miller,
Issues in Speech Perception
1963). Another feature the speech signal lacks is
Our orientation thus far has been to view speech linearity. In terms of phonemes, linearity means
perception as a form of auditory pattern perception that for each phoneme in an utterance we should be
in which the listener’s immediate task is to isolate able to find a corresponding segment of the physi¬
phonemic units and integrate them into a meaning¬ cal speech signal. Furthermore, linearity requires
ful whole, much like the perceptual object task that the order of the segments in the physical signal
we discussed in the chapter on visual form (Chapter must correspond to the order of phonemes. Neither
11). To this end, we have been looking at some of these criteria are met in the natural speech
features of the speech signal and their character¬ signal.
istics. However, because speech perception is Another source of ambiguity comes from the
studied by investigators from so many different fact that the speech signal lacks acoustic-phonetic
disciplines—linguists, electrical engineers, and invariance. Invariance refers to the notion that for
356 Chapter 12/Speech and Music

each phoneme there must be some constant set of in the second formant is a rise; when we hear d in
acoustic features associated with it whenever it is combination with o, to form the sound do as in
heard. In concrete terms, this might mean that there “dope,” the second formant transition is a sharp
is a specific feature, say the formant transition rep¬ drop; and when we hear the d in combination with
resenting a particular consonant, that must be pres¬ e, forming the de in “deck,” there is no second
ent if we are to hear this consonant in the utterance. formant transition at all! This situation seems to vi¬
If this acoustic feature is present it means that the olate linearity, since the information about d does
associated phoneme was intended, whereas its ab¬ not correspond to a specific segment of the speech
sence means that the phoneme was not intended. signal. It also violates invariance, since the acous¬
Unfortunately, such invariance is not found in the tic cue for d depends on the context (here the vowel
speech signal. The actual state of affairs is much it is paired with) rather than a specific set of invar¬
more complex. iant acoustic features. This lack of clear correspon¬
Figure 12-9 illustrates how very different dence between phonemes and acoustic features
acoustic signals can be perceived as the same makes it quite difficult to determine exactly what
phoneme. In this case the phoneme is d. Formant you will perceive given only the information pres¬
transitions usually convey information about con¬ ent in the acoustic input. You can see this for your¬
sonants; the formants themselves contain informa¬ self in Demonstration Box 12-4.
tion about the vowels. Here we have a set of
consonant-vowel combinations in which the con¬
Is Speech Special?
sonant remains d. As we saw in Figure 12-7, it was
the change in the formant transition of the second In some respects, we must recognize that speech is
formant that determined which consonant we quite special and different from other stimuli. The
heard. If that is the case, then at the acoustic level speech stimulus is produced by humans to com¬
something strange is going on here: When we hear municate linguistic information, something only
d in combination with the vowel ee, to form the humans do naturally. Speech sounds special to the
complex dee as in “deep,” the formant transition listener; there is a distinct difference between our

As in deep date deck dot dog dope

the word
Time

FIGURE 12-9 Despite the fact that the formant transition in the second formant changes from a rise to
a drop, depending on the vowel, the same consonant phoneme d is heard. The fact that the first formant
is largely irrelevant to the perception of a consonant suggests that there is no single feature
corresponding to this phoneme, hence it demonstrates the lack of acoustic-phonetic invariance in the
speech signal.
 Speech 357

DEMONSTRATION BOX 12-4. Segmenting the Speech Signal

To see how difficult it is to segment the nearly con¬ gram of the phrase “Peter buttered the burnt toast.”
tinuous speech signal into words purely on the basis There are five occurrences of the phoneme t as in
of acoustic criteria, look at the two speech spectro¬ “toy” in this phrase. If there is invariance and lin¬
grams in this box, both of which are adapted from earity in the speech signal you should be able to find
Pisoni and Luce (1986). Before reading the next par¬ five repetitions of the same signal representing this
agraph, look at each segment and try to figure out phoneme t. Can you find them? It may be of some
how many words or syllables are in phrase A. comfort to know that speech researchers have the
Phrase A has five words and six syllables. It was same difficulties, and they haven’t solved them yet.
made from a recording of the sentence “I owe you a A recent book devoted to this problem illustrates the
yoyo.” How successful were you? Even knowing the variety of approaches that is being tried and provides
phrase in advance doesn’t seem to help very much an excellent summary of the current state of knowl¬
does it? edge on this subject (Perkell & Klatt, 1986).
Now look at phrase B. It represents the spectro¬

perception of speech and of other sounds. How¬ “speech mode” of processing, followed by the
ever, speech researchers usually mean something “speech is just a form of auditory pattern percep¬
else by the question of this section. They refer to tion” forces showing that the same finding can be
the proposal by a group of researchers that the per¬ obtained using nonspeech stimuli and proposing a
ception of speech is accomplished by a specialized purely acoustic explanation for the phenomenon
set of neural mechanisms in the human (e.g., Lib¬ with both speech and nonspeech stimuli. The re¬
erman, 1982). This is a major point of controversy sults have thus been much like a tennis match, with
within the speech research community and re¬ the players lobbing research findings and interpre¬
searchers often classify each other by where they tations.
stand on this issue. Several lines of research have At the physiological level, there is a reason to
been used to argue for or against the special quality suggest that speech is different from other aspects
of speech processing. We discuss a few of them in of auditory processing. It has long been known that
what follows. The arguments usually take the form in most humans the two hemispheres of the brain
of the “speech is special” forces obtaining a dra¬ show some degree of specialization in their func¬
matic finding that appears to demonstrate a special tioning. Damage to the left side of the brain is
358 Chapter 12/Speech and Music

more likely to produce disruption in speech com¬ tity of the phoneme was ambiguous or a combina¬
prehension or production (e.g., Kolb & Wishaw, tion of the two. Perceptually, such a graded change
1985), and a series of behavioral studies has also does not occur.
confirmed that speech processing is predominantly As the voice onset time is changed you hear
left hemisphere in nature. Such studies have either one consonant until, when the voice onset time ex¬
used dichotic listening procedures (where different ceeds some critical value, there is a shift in the
messages are simultaneously presented to the two percept and you now hear the other consonant—
ears, and hence differentially activate the two rather like what you might expect if there had been
hemispheres) or direct recording of brain activity an abrupt shift from one category of stimulus to
while individuals are presented with various speech another. In studies using artificial speech sounds
and nonspeech stimuli. The general pattern of data we find that for every voice onset time up to some
obtained shows that the left hemisphere is most particular value listeners report hearing only the
strongly involved in speech processing, whereas voiced consonant (e.g., b in “ba”), as shown in
the right hemisphere is most strongly activated Figure 12-10A. When the voice onset time is just
when musical or other patterned auditory stimuli a little longer than this they suddenly start hearing
are presented (Bryden, 1982: Springer & Deutsch, the unvoiced consonant (e.g., p in “pa”), as is
1985). Although this specialization is not exclusive also seen in Figure 12-10A, and hear only that for
and some language processing is also undertaken in all longer voice onset times. The value at which
the right hemisphere (Millar & Whitaker, 1983), this change in the phoneme seems to occur is called
and simple auditory stimulus processing seems to the phonemic boundary and is around 35 msec in
be bilateral, such specialization of function in dif¬ the curve in Figure 12-10A. In other words, there
ferent areas of the brain suggests that speech might is a sudden shift in the percept from the phoneme
receive a different form of perceptual processing b to the phoneme p at the phonemic boundary.
than other forms of auditory pattern perception. Moreover, if listeners are presented with pairs of
sounds with voice onset times on the same side of
Categorical Perception From a percep¬ the phonemic boundary (say 10 vs. 20 msec before
tual level, one of the first phenomena to suggest the voice onset) they have a very hard time discrimi¬
possibility of a special speech mode of processing nating them at all, whereas if the stimuli come
was that of categorical perception (Liberman, from opposite sides of the boundary (say 30 and
Harris, Hoffman & Griffith, 1957). An example of 40 msec) discrimination is very good. Conso¬
categorical perception uses the continuum of voice nants show this categorical perception, but vowels
onset time for stop consonants discussed above. do not.
Basically, this refers to the fact that if voicing oc¬ When these findings were first presented, it
curs shortly (say 20 msec) after a constriction of was thought that nonspeech stimuli didn’t segregate
the airflow from the lungs is released, a voiced into perceptual categories but rather produced much
consonant (such as the b in “bad” ) is heard. If more gradual perceptual transitions as the stimulus
the voicing is delayed somewhat after the constric¬ changed. This led many researchers to assume that
tion is released, an unvoiced consonant (such as the some special kind of mechanism must exist to pro¬
p in “pad”) is heard. The time at which the onset cess and categorize acoustic cues defining the var¬
of voicing occurs can be viewed as a stimulus con¬ ious consonants. The data are not so clear now,
tinuum, ranging from about 0 to 70 msec or so af¬ however (see Pisoni & Luce, 1986, for a review).
ter the constriction is released. This might lead you Several studies have shown that similar dramatic
to expect that as we varied the voice onset time we categorization effects can occur for nonspeech
should get a gradual shading of the consonant from stimuli. For example. Cutting (1976) found a fairly
b to p, with perhaps some region in which the iden¬ sharp categorical boundary for “plucked” versus
 Speech 359

Phonemic boundary

Normal
category
boundary

After
adaptation
to “ba”

FIGURE 12-10 Typical results of an experiment on categorical perception of phonemes. Before

adaptation the phonemic boundary is at about 35 msec voice onset time. After listening to “ba” for 2
minutes, the phonemic boundary is shifted to about 25 msec voice onset time.

“bowed” sounds (like those of violins), which dif¬ point in any perceptual continuum, which serves as
fered only in their onset times. Other studies have a subjective reference point for judgments. Stimuli
shown similar categorization effects in nonhuman above the adaptation level seem to be qualitatively
species such as macaque monkeys and Japanese different from those below the adaptation level
quail, neither of which should be expected to have (e.g., hot vs. cold, loud vs. soft). For a variety of
a special mechanism for speech processing (Kluen- stimulus continua, discrimination of stimuli is con¬
der, Diehl & Killeen, 1987; Kuhl & Padden, siderably worse when stimuli are selected either all
1983). Such data, although still quite controversial, from below or all from above the adaptation level.
seem more consistent with the notion that speech This pattern of data is much like that observed
perception is a special case of auditory pattern per¬ in categorical speech perception (Streitfeld & Wil¬
ception, rather than a special perceptual process. son, 1986).
Categorical perception effects can actually be Just as the adaptation level can be made to
obtained for any perceptual continuum. The trick change because of previous experience with partic¬
seems to be to use the adaptation level as the cat¬ ular sets of stimuli, the phonemic boundary can be
egory boundary. You should recall from Chapter 2 made to shift around. The phonemic boundary can
that the adaptation level represents a sort of neutral be moved by presenting listeners repeatedly with a
360 Chapter 12/Speech and Music

good example of a phoneme at one end of a contin¬ present a group of synthetically generated for¬
uum such as voice onset time, say the syllable ba mants, called the base stimulus, to one ear and
with a voice onset time of 10 msec. After 2 min¬ present an isolated formant transition to the other
utes or so of listening to ba, listeners are now pre¬ ear. Typically, the base stimulus is heard as a par¬
sented with the other voice onset time stimuli as ticular syllable, say da, which either is heard this
before, such as those used to generate the curve in way or modified by the presence of the formant
Figure 12-10A. This results in a category shift, as transition in the other ear (e.g., it might now be
though the effectiveness of the original stimulus heard as ga). Duplex perception manifests itself
has been weakened. Thus, some stimuli that pre¬ when, in addition to hearing the speech sound,
viously were heard as ba are now classified as pa. most observers also hear a nonspeech “chirp.”
The boundary has been shifted in the direction of This chirp is actually the same sound you would
the adapting stimulus (ba) so that now a stronger hear if the isolated formant transition were pre¬
(shorter voice onset time) stimulus is necessary to sented alone. If observers are attending to the
perceive ba, as shown in Figure 12-1 OB. Similar speech sound they show many of the phenomena
results have been obtained with other characteris¬ usually associated with speech processing, such as
tics of speech stimuli, such as place of articulation categorical perception; if they are attending to the
and vowel pronunciation (see Diehl, 1981). nonspeech chirp they do not show these phenom¬
Originally, these findings were interpreted (by ena. This result has been interpreted as suggesting
proponents of the “speech is special” view) as that when you attend to speech stimuli you invoke
meaning that there were special speech feature de¬ a different mode of perceptual processing than you
tectors that were being fatigued or adapted (Abbs do for nonspeech stimuli (see Liberman, 1982).
& Sussman, 1971; Eimas & Corbit, 1973; Lisker It is also possible to produce duplex percep¬
& Abramson, 1970). As a counter to this argu¬ tion by presenting the same specifically constructed
ment, the “speech is just auditory pattern percep¬ stimulus to both ears (Whalen & Liberman, 1987).
tion” group obtained similar shifts in categorical In this case, the speech percept alone is heard at
boundaries with nonspeech stimuli, depending on lower intensities, while at higher intensities both
the acoustic similarity of the nonspeech stimulus speech sounds and nonspeech chirps are heard. In¬
used for adaptation to the speech stimulus used for terestingly, the “duplexity threshold,” at which
testing (Samuel, 1986; Sawusch, 1986). These re¬ both sounds are heard, is about 20 dB higher than
sults have been taken to suggest that no special the threshold for discriminating speech sounds in
speech feature detectors, or special speech process¬ the same stimuli. At the least this indicates that
ing, need to be postulated to explain the data; speech processing dominates nonspeech process¬
however, a fine-grain analysis of the patterns of ing, and is consistent with the idea of a separate
adaptation actually obtained suggest that the situa¬ speech processing mode. However, duplex percep¬
tion may be more complex than originally thought tion does occur for nonspeech stimuli such as mu¬
(Samuel, 1986). sical chords (Pastore, Schmeckler, Rosenblum &
Szczesiul, 1983), providing evidence for an argu¬
Duplex Perception Yet another finding ment against interpreting these phenomena as aris¬
that has been interpreted as evidence for a special ing from a special speech processing mode.
speech processing mode is a phenomenon called
duplex perception, in which the same sound can Cross-modal Integration A final phe¬
be perceived as having both speech and nonspeech nomenon that has implications for the “is speech
qualities (Rand, 1974). The most common tech¬ special?” controversy is the McGurk effect
nique used to demonstrate duplex perception is to (McGurk & MacDonald, 1976). This involves a
 Speech 361

form of cross-modal integration in which nona¬ there is about an equal amount of information from
coustic stimuli affect what the listener hears when each cue does the conflict produce the “in be¬
listening to speech. In this effect, you listen to a tween” percept.
string of speech sounds, like da, and at the same Of course, in normal speech the visual and au¬
time look at a movie or video of a face articulating ditory cues are usually consistent, hence the sight
speech sounds in synchrony with the actual speech may be a useful means of augmenting the intellige-
stimuli. Everything is fine as long as the sounds bility of the sound. It is reasonable to suppose that
and the facial movements refer to the same sylla¬ speech perception would involve such a mecha¬
ble. Suppose, however, the sound reaching your nism, since most linguistic communication takes
ear is ba but the face is making the articulatory and place, and is learned by the child, in a face-to-face
mouth movements associated with saying ga. The mode where both types of cues are available. How¬
visual input seems to alter your interpretation of the ever, we don’t need visual cues to understand
sound, and you hear a compromise sound, da. This speech. For example, we understand speech on the
suggests that in some fashion nonauditory infor¬ radio, even at higher than normal presentation rates
mation is used to aid in our interpretation of as in commercials. Also, other theories can account
speech, and perhaps to resolve some of the ambi¬ for such cue integration (e.g., Massaro, 1987).
guities inherent in the acoustic signal. There is You can experience a similar effect caused by the
some evidence that the McGurk effect is strongest cross-model integration of speech cues by trying
for syllables, not for complete words (Easton & Demonstration Box 12-5.
Basala, 1982). This has been taken to mean that if The McGurk effect has been interpreted as
there is a great deal of auditory and semantic infor¬ evidence that visual and auditory cues about what
mation about what is being said, the visual cue is is being said converge at some abstract (special?)
not powerful enough to overcome it. Only when level of speech processing. Some data seem to be

DEMONSTRATION BOX 12-5. Cross-Modal Integration of Cues

For this demonstration you need access to a TV set the visual cues as to what is being said, you will find
and a radio. Bring the radio into the same room as that you now can understand the TV speaker when
the TV set and turn them both on. Tune the radio to you see his or her face. The effect will be similar to
a point in between channels so that a hissing or roar¬ turning down the noise slightly, except that all you
ing sound comes from the speaker. Tune the TV set did was to add the visual cues. When you close your
to a newscast or other show where there is a person eyes again, you should find it again impossible to un¬
talking steadily, looking directly into the camera. Set derstand the TV speaker’s speech. The additional in¬
the volume of the TV set to a medium setting, so that formation you are obtaining visually by watching the
you can comfortably understand what is being said, speaker talk is clearly having an effect on the intelli¬
but low enough so that when the radio noise is turned gibility of the speech. These visual cues are particu¬
to a high level you can’t hear the TV. Now, close larly important anywhere the intelligibility of speech
your eyes and turn up the radio noise until you can’t is reduced by the presence of noise, such as at a noisy
understand what the speaker on the TV is saying, then party or on a noisy downtown street or if your hearing
lower the radio noise until you can just barely under¬ is not very acute. Thus, many hearing-impaired peo¬
stand the speaker, and finally raise it a bit so you ple find it easier to understand speech when they are
can’t again. Now, open your eyes. In the presence of looking at a speaker’s face.
362 Chapter 12/Speech and Music

inconsistent with this interpretation, however. For criminations similar to those of adults has thus been
instance, Roberts and Summerfield (1981) set up taken as evidence of some innate mechanism for
conditions like those we described above (the speech processing.
sound was ba, the visual image mouthed ga, and Another very impressive demonstration of in¬
the subject heard da) and repeated the presentation fants’ speech perception ability was made by Kuhl
until subjects had adapted. Next the phonemic and Meltzoff (1982). They presented 4-month-old
boundary was measured, as we illustrated in Figure infants with two video displays of the same per¬
12-10. The question was, Does the phonemic son’s face speaking two different vowel sounds, as
boundary shift in a manner consistent with the pho¬ illustrated in Figure 12-11. At the same time, the
neme the subject perceives (as a special process baby was presented (from a loudspeaker midway
might predict) or in a manner consistent with the between the faces) with the sound of the person’s
actual acoustic signal (as auditory pattern theory voice producing one or the other of the vowel
might predict). The answer was that the phonemic sounds. The voice and the two faces were all in
boundary seems to shift according to the acoustic synchrony with each other, but the voice corre¬
signal ba, rather than the perceived signal da. sponded to only one of the faces’ articulatory
Let us now reiterate our original question, “Is movements. From video recordings of where the
speech perception a special process?” As you can baby looked during the tests, it was determined that
see from the give-and-take nature of the data we the baby spent more time (about 73 percent of the
have presented, a special process is a possibility, total looking time) looking at the face that was ar¬
but a simple extension of auditory pattern percep¬ ticulating the vowel sound it was hearing. Thus,
tion processes cannot be rejected. Hence, this ques¬ very young infants also demonstrate cross-modal
tion is far from decided, and the controversy will integration of speech cues, just like adults do (e.g.,
probably go on for a while yet. the McGurk effect), even though they have had
only limited experience with such cues and do not
yet speak. Moreover, this ability is apparently a
Development
function of the left hemisphere of the brain in in¬
Infants are bom with a remarkable ability to re¬ fants, just as speech production is a function of that
spond to human speech in a special way (see Jus- hemisphere in adults (MacKain, Studdert-Kennedy,
czyk, 1986; Kuhl, 1987). At birth, infants move Spieker & Stem, 1983).
their limbs in synchrony with connected adult Although there is a good deal of evidence sug¬
speech but not with other sounds, such as tapping gesting that from birth infants can discriminate the
sounds or disconnected vowel sounds (Condon & entire set of possible phonemes that the human vo¬
Sander, 1974). In addition, infants show much the cal apparatus is capable of making, adults can not.
same patterns of responses that adults do. For ex¬ Thus, a native Japanese speaker may say “ararm
ample, Eimas, Siqueland, Jusczyk, and Vigorito crock” when he actually means “alarm clock,”
(1971) found that infants as young as 1 month of not because of sloppy speech but because the Jap¬
age discriminated speech stimuli better across pho¬ anese language does not have the two separate pho¬
nemic boundaries than within phonemic categories, nemes r as in “run” versus / as in “look” and the
showing categorical perception much the way that adult speaker no longer has the ability to make this
adults do. Since, at 1 month of age, infants have perceptual discrimination. As we discuss later (in
had only a little bit of exposure to speech sounds Chapter 16), infants maintain their ability to dis¬
and their utterances consist only of cries, screams, criminate phonemes present in the language that
and random babbles, they clearly have not yet they hear in their environment. However, they soon
learned language. Their ability to make speech dis¬ begin to lose the capacity to discriminate phonemes
 Speech 363

Video camera
to record

FIGURE 12-11 Illustration of the setup of an experiment showing that infants can integrate phonemic
information across auditory and visual modalities. (Based on Kuhl & Meltzoff, 1982)

they are not regularly hearing. For instance, 6- make sounds that sound to their ears as if they have
month-old infants can still make discriminations in crossed the boundary from z to th, but that to native
languages other than the one their parents speak, English speakers still appear to remain on the z
but by the time they are 1 year old they have lost side. That is why a French speaker is perceived by
this ability (Werker & Tees, 1984). It is as if our an English speaker as saying that she is going out
linguistic surroundings tune the mechanisms that to “walk ze dog.”
will later be used to discriminate speech, with the
end result that we no longer have the ability to per¬
Content
ceive phonemes not found in our own tongue.
This inability may simply be due to the fact We mentioned above that a major problem in
that different languages have different phonemic speech perception is the ambiguity of the speech
boundaries. Phonemes that sound different in one signal due to the lack of an invariant set of acoustic
language might then be heard as the same phoneme features that correspond to the perceived speech
in another and our experience may alter our place¬ units. We encountered this fairly graphically in
ment of this boundary and hence our categorization Demonstration Box 12-4. One factor that helps in¬
of the phoneme. For instance, if you make a con¬ terpret patterns, whether visual, auditory, or
tinuous z sound as in “zzzzz-oo,” and then bring speech, is the context in which these patterns ap¬
the tip of your tongue close to your teeth, the sound pear. The other stimuli in the environment assist
will change into an extended th as in “that.” The our identification of the present stimulus. We saw
distinction between z and th is determined by the this for visual patterns in Figure 11-15, where the
tongue position, which sets the phonemic boundary same pattern elements were seen either as a letter
conditions. This boundary is different for different or a number, depending on the context formed by
languages. Thus, adult native French speakers will the surrounding stimuli. In terms of the ultimate
364 Chapter 12/Speech and Music

function of speech, which is to convey meaning, ance was concerned with. There are a great many
context is vital, since the same sound unit may sig¬ other context effects in speech perception, many of
nify different things. This occurs in the case of them depending on the meaning attached to the
homophones, which are words that sound alike speech sounds (see also Liberman & Mattingly,
when spoken—such as be and bee, rain and reign, 1985; Sawusch, 1986). One interesting demonstra¬
no and know—although they convey different tion of this type of effect was provided by Day
meanings. Although there may be acoustic cues (1968, 1970). She presented sound sequences si¬
that differentiate these words, most often the sur¬ multaneously to both ears of listeners. For exam¬
rounding context helps us tell which word was ple, if the left ear received b-a-n-k-e-t the right ear
meant by the speaker. received l-a-n-k-e-t. Many of the listeners fused the
Interestingly, there are other words that sound two sequences into the word blanket, even when
quite different to us, such as married and buried, lanket preceded banket by several msec; other lis¬
that are produced by speakers using nearly identical teners heard only the separate sound sequences.
movements of the mouth and lips. These are called But no one heard Ibanket, which is a sequence of
homophenes, and are difficult for lip-readers to phonemes that does not occur in English. The ex¬
discriminate out of context. Clearly, if we are pectations as to which sounds can occur in speech
watching a speaker’s face for some cross-modal in¬ clearly provide a context that influences what is
formation in a difficult listening context and if the perceived.
word uttered is a homophene, we will not get much The importance of context in speech percep¬
help from the visual image, but we may get help tion is demonstrated by the fact that if trained lis¬
from the context provided by other words in the teners are asked to provide a phonetic transcription
sentence. Demonstration Box 12-6 gives an exam¬ of spoken passages in an exotic language they do
ple that demonstrates how context affects our abil¬ quite poorly, despite their training, simply because
ity to extract the meaning of speech stimuli. of the absence of an adequate semantic and syntac¬
The context effect in Demonstration Box 12-6 tic context (Shockey & Reddy, 1974). Even if you
was based on identifying the general topic an utter¬ know the language, your ability to identify isolated

DEMONSTRATION BOX 12-6. Context and Speech Perception

In the absence of an appropriate context, even com¬ time, you can compare the sentences (or nonsent¬
mon words are often difficult to identify. To see how ences) that were heard with and without the context.
context interacts with speech perception, read the fol¬ Without the context you might find responses such as
lowing phrase in a smooth, rapid, conversational style “In middies, sar, in clay nanar” or “In may deals
to a friend: “In mud eels are, in clay none are.” Ask are, en clainanar” (Reddy, 1976). Here the words in
your friend to write down the phrase exactly as he or the sentence are difficult to identify when presented
she heard it. Now you should provide a context by rapidly, and a strange and largely meaningless set of
telling your listener that you are going to read a sen¬ segments is generated. When the proper context is
tence from a book that describes where various types supplied, however, the same speech sounds are cor¬
of amphibians can be found. Then read the above rectly segmented into words, and interpreted as mean¬
sentence again at the same speed that you did before. ingful elements.
After your listener writes down what was heard this
 Speech 365

words taken from a stream of recorded speech is weakness in speech perception research (e.g., Pi-
quite poor. Generally in such tests, listeners are ca¬ soni & Luce, 1986). The many candidate theories
pable of identifying less than half of the items pre¬ of speech perception seem to be subdivided on the
sented in isolation (Pollack & Pickett, 1964). If the basis of their level of analysis, whether they are
same words are presented surrounded by longer oriented toward the identification of phonemes or
strings of the words in the original recorded utter¬ of words, and on the basis of whether they utilize
ance, identification is much better. Thus, the more active or passive processing (Nusbaum & Schwab,
acoustic, syntactic, or semantic context provided, 1986). Passive processing involves a filtering or
the better the observers are at identifying the feature detection sequence of events that is rela¬
words. tively fixed in nature and works at a sensory level.
The context of a sentence can actually induce After the message is sensed and filtered, it is then
a listener to supply missing parts of the stimulus to mapped fairly directly onto the acoustic or articu¬
fill in a gap in continuous speech so that it is heard latory features of the language. Active processing
as if it were uninterrupted. Warren (1970) pre¬ involves a much more interactive sequence, in
sented listeners with a taped sentence: “The state which the acoustic features are sensed but then a
governors met with their respective legislatures series of higher level processes involving analysis
convening in the capital city.’’ The acoustic infor¬ of the context (either meaning or phonetic) is con¬
mation corresponding to the first s in “legisla¬ sidered, and in which expectations or even knowl¬
tures’’ was deleted and replaced by the sound of a edge of speech production and articulation may
cough. Nineteen of 20 listeners reported nothing play a role. In active processing the sequence of
unusual about the sentence. They restored the miss¬ processing steps may not be fixed, but may vary
ing phoneme, hence the effect was named the pho¬ depending on the results of earlier computations.
nemic restoration effect. Again, the context of the Both active and passive models may use general
sentence determined the speech sequence actually acoustic processing rules or invoke “special”
perceived. Apparently both meaning and acoustic speech analysis units.
cues play a role in such restoration of missing or Many passive theories incorporate the notions
obliterated components (Bashford & Warren, 1987; of feature detectors or template matching. Feature
Samuel, 1981). This means that if we delete por¬ detectors for speech are usually conceptualized as
tions of common words or phrases, or present only neurons specialized for the detection of specific as¬
the distinctive parts of words, the listener is more pects of the speech signal, much the way specific
likely to “hear” the speech as being continuous neurons in the visual cortex selectively respond to
and to accurately restore the missing portions in aspects of the visual stimulus such as line orienta¬
consciousness. In effect, we hear the speech units tion (see Chapter 3). The concept of an auditory
that the immediate context suggests should be in template may be viewed as a stored abstract repre¬
the phrase, even if they are not physically present. sentation of certain aspects of speech that develops
as a function of experience and serves the same
function as a feature detector. The use of feature
Theories of Speech Perception
detectors or templates in speech identification is of¬
A great number of different theories have been pro¬ ten viewed as a process similar to the visual feature
posed to explain various subsets of the phenomena extraction model called “pandemonium” we dis¬
we have discussed above. It is probably safe to say cussed in Chapter 11, in that the signal is perceived
that no one theory has yet achieved a consensus of on the basis of the template that it most closely
researchers that it is the most useful. In fact, many matches or the phoneme or word that has the most
researchers feel that theory is an area of significant features in common with it.
366 Chapter 12/Speech and Music

Some theories that are predominantly passive of emphasis on these various mechanisms, active
in nature stress that ordinary auditory processes are theories often differ dramatically from one another.
sufficient to explain speech perception at the level For example, Marslen-Wilson (1980) offered
of phonemes (e.g., Fant, 1967; Massaro, 1987). an active model of word identification called cohort
These auditory theories usually postulate several theory. In this model the early passive stages of
stages of processing of speech sounds. The first analysis initially extract the first phoneme(s) of a
stage consists of “ordinary” auditory processing, word. On the basis of this information all the words
including analysis of a complex sound into its in memory that have the same beginning (e.g., all
simple sine wave components, auditory feature words beginning with st, such as “stop,” “stall,”
analysis, and auditory pattern processing. For some “stride”) are activated. These words constitute the
theorists feature analysis occurs first, then pattern “cohort” or group of possibilities to be considered.
analysis, whereas for others they occur at the same After the cohort is activated, other acoustic or pho¬
time (in parallel). The next stage then applies more netic information as well as higher-level expecta¬
specialized (but not “special”) rules to the out¬ tions operate to eliminate all the candidates except
puts) of the first stage(s), integrating them to pro¬ one, which is identified.
duce perception of phonemes. A good example of Phonetic refinement theory (Pisoni, Nusbaum,
this approach that uses words, rather than pho¬ Luce & Slowiaczek, 1985) is also concerned with
nemes, as the unit is Klatt’s (1980) Lexical Access word identification. It resembles cohort theory in
From Spectra (LAFS) model. In this model the lis¬ that the initial feature processing activates candi¬
tener does a spectral analysis of the input signal, date words that all sound alike along some dimen¬
matching the results of this analysis to a set of tem¬ sions (not just the first phoneme) that form what
plates of features stored in memory. Words are they call a phonetic space. The set of candidates is
then identified from the set of features detected in narrowed down through the application of addi¬
the input. In this model, there is no need to de¬ tional information, both phonetic and contextual,
scribe segments or phonemes or other linguistic en¬ until a single word is identified because it satisfies
tities; the speech input is directly matched to words all the available constraints best and is hence most
in memory by a fixed process. strongly activated. In this model, words can be
It is possible also to have a passive model that identified even if there is only partial information
uses “special” speech units (for example, Eimas available, since even that partial information may
& Corbit, 1973). In such a model, the first stages activate the correct word more than it does any
would consist of detection of speech features by other words.
“special” feature detectors followed by integration Another active theory, which has been imple¬
of these features into percepts by (possibly) “spe¬ mented as a computer model, is McClelland and
cial” rules. Figure 12-12A gives a schematic rep¬ Elman’s (1986) TRACE model. This model begins
resentation of a general passive speech perception with passive feature detection in three levels: (1)
theory that contains elements similar to many cur¬ acoustic feature detectors whose output is the input
rent models. to (2) phoneme detectors whose output is the input
Active models of speech perception are some¬ to (3) word detectors. The unique aspect of this
what more variable since they often involve analy¬ model is that the various detectors and other pro¬
sis of the context in which the speech is occurring, cessors, referred to in the computer model as
the expectations of the listener, the distribution of nodes, are highly interconnected. Activating one
attentional resources, and memory components. node tends to activate all the nodes to which it is
Since different researchers place different degrees connected, both at the same level and at other
 Speech 367

Primary
auditory Feature extraction Unit

)))) Identification

Speech
signal

A. Passive processing

Primary
auditory Feature extraction Active

))))
Speech
signal

FIGURE 12-12 Diagrammatic representation of the difference between a typical passive model of
speech processing (A) and a generalized active model of speech processing (B).

higher or lower levels. This is actually quite a com¬ of speech perception is difficult to classify as active
plex model in which various levels may interact or passive for it seems to have elements of both
with one another in a looping fashion, with higher approaches. This is the motor theory proposed by
levels “tuning” or altering the weighting given to Liberman, Cooper, Shankweiler and Studdert-
specific features. Kennedy (1967) and recently revised by Liberman
All active theories have one thing in common. and Mattingly (1985). It is oriented toward the
They all have higher-level decisional processes su¬ identification of phonemes considered as the in¬
perimposed on the initial feature extraction results. tended phonetic gestures of a speaker, that is, what
Thus, the speech you “hear” may be determined the speaker intends to say. In this theory, speech is
by factors other than the actual acoustic signal. A clearly “special,” in the sense that perception of
diagram of a general active speech processing the¬ speech sounds is accomplished by a specialized
ory is shown as Figure 12-12B. processing mode that is both innate and part of
One of the first, and most influential, theories the more general specialization for language that
368 Chapter 12/Speech and Music

humans possess. In particular, the theory assumes gradual blurring of some of the distinctions. For
that the same adaptations of the mammalian motor instance, we have seen that active processing mod¬
system that made speech possible for humans also els begin with a passive-processing, feature-extrac¬
made possible a system for perceiving the sounds tion component; some of the passive processing
produced by the speech motor system based on the models may well allow active processing when the
actual movement commands issued to produce the signal is degraded or conditions are difficult (e.g.,
speech sounds. This system uses complex calcula¬ Fant, 1967; Massaro, 1987); and motor theory
tions to deduce the intended speech gestures from doesn’t fit especially well into either the active or
the acoustic signal based on an abstract represen¬ passive category. In addition, some investigators
tation of the articulatory movements the listener have proposed that both general auditory processes
herself would have used to produce just such a and special speech processes are necessary to ac¬
speech signal. count for all the data (e.g., Pisoni, 1973; Werker
An example of how this system might be in¬ & Logan, 1985). Certainly there can be no argu¬
volved in speech perception can be seen by looking ment as to whether an auditory mode of processing
back at Figure 12-9 and noting that the d in the exists. The problem speech theorists still must con¬
composite dee as in “deep” is perceived as the tend with is just how “special” speech perception
same d as in do (in “dope”) although they are is, and how much of what is heard is in the signal
quite different acoustically. According to motor and how much is constructed in the mind of the
theory, both are heard as d because the listener listener.
would use equivalent articulatory movements to
produce the d phoneme for both utterances. The
face that the actual sounds of the two d's are not GLOSSARY
the same is irrelevant, just as the pitch of voice,
speed of speech, and other sound-distorting factors The following definitions are specific to this book.
are irrelevent.
Acoustic-phonetic invariance The idea that there
A recent further development of this approach must be some set of acoustic features associated with
emphasizes the modular nature of the special and each phoneme in all contexts.
distinct system proposed to process speech sounds Active processing Models of speech perception that
(Liberman & Mattingly, 1985). A module consists incorporate the effects of expectations, context, memory,
of neural circuits that perform special processing and attention. Processing may vary depending on pre¬
vious computations.
that provides higher cognitive processes with rep¬
Articulators Parts of the vocal tract that are used to
resentations of events that have particular biologi¬
produce speech, such as teeth, tongue, lips, and palates.
cal, ecological, or behavioral significance (see
Categorical perception A phenomenon in which
Fodor, 1983). We can have little awareness of what discrimination of stimuli within a perceptual category is
a module is doing, and its processing is automatic worse than that across a category boundary.
and dominates other processing of the same stim¬ Chord Simultaneous presentation of three or more
uli. In this sense the theory is a passive one: in it musical notes.
speech perception cannot be greatly influenced by Consonant A basic speech sound produced by clos¬
ing the vocal tract.
conscious analysis. However, the processing in¬
volved is complex and flexible, taking account of Contour The general “shape” of a musical se¬
quence of sounds defined by the rises and drops in fre¬
the effects of expectations, context, and other im¬
quency of the notes.
portant factors, just as the more active theories do.
Duplex perception Perception of both speech and
No one theory or level of analysis has come to nonspeech sounds simultaneously from a single auditory
dominate speech perception. In fact, there is a stimulus.
 Glossary 369

Formants Bands of especially intense components Phonemic boundary The point on a speech feature
of a speech signal, seen as dark smudges on a spectro¬ continuum where the perception of the phoneme changes
gram. from one category to another (e.g., from “ba” to “pa”
Formant transitions Changes in formants over rel¬ as voice onset time increases past 25-35 msec).
atively short intervals of time (less than 100 msec) that Phonemic restoration effect When a listener hear¬
are related to consonant sounds. ing a spoken sentence fills in a missing phoneme based
Homophenes Different words that are produced by on the context.
almost identical patterns of lip movements. Rhythm The perceived organization in time of a se¬
Homophones Words that are pronounced similarly, quence of sounds.
but spelled differently. Speech spectrogram A representation of the speech
LAFS Acronym for Klatt’s (1980) theory of word signal in terms of the frequencies and amplitudes of its
recognition: Lexical Access From Spectra. simple sine wave components as these change over time.
Linearity The idea that for each phoneme in an ut¬ Tempo The perceived speed with which a sequence
terance there must correspond a segment of the physical of sounds is proceeding.
speech signal. Tone chroma A “circular” dimension of musical
McGurk effect The perception of an “intermediate” pitch that connects similar notes of different octaves.
phoneme when auditory and visual speech cues conflict. Tone height The simple “vertical” dimension of
Musical interval The perception of the separation in pitch in musical scales.
musical pitch between two musical sounds. TRACE McClelland and Elman’s (1986) theory of
Passive processing A model of speech perception word identification.
based on filtering the signal for features only, with no Voice onset time The latency in producing a vowel
higher level interactions. sound following a stop-consonant sound.
Phone The basic sound unit used by linguists to de¬ Vowel A basic speech sound produced by opening
scribe speech. the vocal tract and vibrating the vocal folds.
Phoneme A phone used in a language to distinguish
one word from another.
CHAPTER
13
Time and
Motion
TIME
Biological Clocks
Circadian Rhythms
Short-Term Timers
Biological Pacemaker
Cognitive Clocks
Change
Processing Effort
Temporal versus Nontemporal Attention
MOTION
Physiological Motion Detectors
Stimulus Factors in Motion Perception
Apparent Motion
Biological Motion
Eye Movements and Motion Perception
The Vestibular Sense
The Vestibular Stimuli and Receptors
Neural Responses in the Vestibular Sense
Self-Motion
372 Chapter 13/Time and Motion

T he Nobel-prize-winning physicist Albert Ein¬

stein often thought about the meaning of time,
qualities are an awareness of a present moment and
the impression that time passes. Fet us call these
ultimately to conclude that “the distinction be¬ the concepts of now and flow, respectively (Mi-
tween past, present, and future is only an illusion, chon, 1985).
however persistent” (Einstein & Besso, 1972). The concept of now is fairly unitary, and has
From the psychological point of view, a better con¬ been described by William James (1890) as the
clusion might be that time and space are modes by “saddle-back of time with a certain length of its
which we perceive and think about our world, own, on which we sit perched, and from which we
rather than physical conditions under which we look in two directions into time.” Sometimes
live. called the subjective present, it is the few seconds
Your sense of the passage of time is a percep¬ of our current experience of ongoing conscious¬
tual experience. Up to now we have seen that most ness; all else is either past or future. Although flow
perceptual experiences are more or less related to is an equally fundamental perceptual attribute of
some physical stimulus continuum. However, in time, it can be further subdivided into several mea¬
the case of time the situation is less clear. There surable aspects of experience (cf. Poppel, 1978).
certainly is no readily visible “time organ” for Each of these additional aspects of time perception
such stimuli to impinge on. Our notion of time may may be different from the others and may be main¬
well be associated with some form of internal clock tained by different physiological or information
that we consult like a wristwatch to determine the processing mechanisms. First we have duration es¬
span of an event, but it also seems to be tied to our timation, which is a report of the experience of
experience of successive change (Fraisse, 1963). It how much time has elapsed between two events
is through the concept of change that time and mo¬ (for example, between when you turn on the heat¬
tion become intertwined. Changes that occur in a ing element under a pot of water and when the wa¬
sequence are often associated with a perception of ter boils). We usually use units such as seconds or
time passing, whereas changes in location may, un¬ minutes to describe duration. Next, we have the
der the proper conditions, be perceived as move¬ perception of order or sequence, which involves
ment. Time and motion thus represent dynamic the determination of which event came first, sec¬
qualities of perceptual experience. ond, and so forth (such as the sequence of digits in
a phone number someone has just read to you).
There is a special case of the perception of se¬
TIME quence that involves determining the minimum
time interval that must separate two events before
The concept of time is fundamental to human they are perceived as occurring one after the other,
beings. For example, every language thus far ana¬ rather than at the same moment. This judgment in¬
lyzed has separate tenses for past, present, and fu¬ volves the discrimination between the experience of
ture, plus innumerable modifiers to specify when simultaneity versus successiveness. The last aspect
more precisely—yesterday, today, recently, in an of flow is somewhat less perceptual but still re¬
hour, while, during, after, and hundreds more quires time estimation. It is the anticipation or
(Bentham, 1985). Despite this position as a funda¬ planning of an ordered sequence of events before
mental experience the study of time is complex, they occur, which is especially important in playing
since “time is not a thing that, like an apple, may musical instruments, or in actions such as speech
be perceived” (Woodrow, 1951). In fact, there are production where we automatically plan and exe¬
two qualities of our perception that seem to be cute an ordered sequence of sounds to produce
added to our consciousness, and do not seem to meaningful utterances.
correspond to simple physical dimensions. These When we look for the mechanisms by which
 Time 373

we perceive time, we find that two general pro¬ ular area 24 hours after a successful hunt or food
cesses have been suggested. We may call these find there (Groos & Daan, 1985; Rijnsdorp, Daan
“clock” theories, since each gives a mechanism & Dijkstra, 1981). There are also more subtle
that determines how we monitor the passage of physiological processes that have their own peri¬
time. The first involves a biological clock, and as¬ odic changes. For example, the pulse, blood pres¬
sumes that there is a biological or physiological ba¬ sure, and temperature of the body show day-night
sis to our perception of time. Just as we have a variations in humans as well as in many other ani¬
sense organ that is sensitive to light (the eye), we mals; there is a more than 1 deg C difference in
also have a sense organ that accounts for our ability body temperature between the coolest point, which
to keep track of time. The second involves a cog¬ occurs during the night, and the warmest point,
nitive clock, where time is viewed as a purely cog¬ which occurs during the afternoon. These are all
nitive process that is not tied to any objective or examples of a circadian rhythm, which comes
“clock” time, but rather is based on how much from the Latin circa, meaning “approximately,”
sensory information is processed, how many events and dies, meaning “day.” Thus, a circadian
occur within a given interval, or how much atten¬ rhythm is one that varies with a cycle of roughly
tion is paid to ongoing cognitive events. In this lat¬ 24 hours.
ter viewpoint time is constructed rather than simply So much rhythmic activity in behavior sug¬
monitored. Both of these types of clocks may exist, gests control by some internal biological clock. Al¬
and each may be used for different types of time ternatively, it may be that these repetitive 24-hour
perception. changes are simply a function of the regular
changes in light and temperature that occur in the
day-night cycle. Thus, an animal might become ac¬
Biological Clocks tive in the presence of daylight when it can see
Many physical phenomena have their own rhythms more clearly and the temperature is a bit higher,
or timing—there are day-night cycles, cycles of the and it is this activity that then alters the physi¬
moon, cycles of the seasons, and many others. Liv¬ ological function. The “built-in” approximately
ing organisms often display similar rhythmic activ¬ 24-hour cycle, however, can be demonstrated ex¬
ities—many flowers open and close at particular perimentally in the absence of light or temperature
times of the day. In animals there are physiological changes. For example, suppose that we find our¬
and behavioral processes that cycle regularly. One selves in a constant-light environment, where there
proposal about the way time is perceived is based are no changing cues that indicate the passage of
on the idea that the flow of subjective time is re¬ time. Under these conditions our biological clock
lated to some body mechanism that acts in a peri¬ will “run free,” gaining or losing time like a not-
odic manner, with each period serving as one too-accurate clock. Although different people will
“tick” of the biological timer. Anything that alters have different cycle lengths, most of us will begin
the speed of our physiological processes would to live a “day” that is approximately 25 hours long
then be expected to alter our perception of the (e.g., Aschoff, 1981; Wever, 1979).
speed at which time passes. If the internal biological clock is set for about
25 hours, why do our internal and behavioral
rhythms continue on a 24-hour cycle? Why doesn’t
Circadian Rhythms
our daily activity cycle drift out of phase with local
One of the most obvious examples of an apparently time? This is because there is a mechanism that
timed behavior is the sleep-wakefulness cycle that synchronizes the internal timer with local time.
runs through a regular daily rhythm. Another is the From the behavioral point of view, the most salient
return of hunting and foraging animals to a partic¬ aspect of local time is the alternation of light and
374 Chapter 13/Time and Motion

dark cycles. To be an accurate reference against 13-1. The timing function of this brain structure is
local time, a biological clock must be synchronized easily demonstrated. For example, rats are noctur¬
with the local day-night cycle, and it must have a nal animals, sleeping during the day and foraging
stable period that is relatively free of unpredictable at night. Destroying the SCN abolishes this pattern.
environmental fluctuation. This process of synchro¬ The animal still sleeps the same amount of time,
nization is called entrainment. If there were no but the circadian pattern is gone and it sleeps in
such mechanism, traveling across the continent, random periods throughout the day and night (Ste¬
where the sun might rise 3 hours earlier relative to phan & Nunez, 1977). Tumors in this region have
the current setting of your biological clock, would the same effect in humans (Fulton & Bailey, 1929).
leave you 3 hours “out of step’’ with your new Furthermore, electrical stimulation of the SCN in
environment. There is, of course, some disruption animals will reset the biological clock, in much the
of your time sense from such trips in the form of same way that brief flashes of light do for dark-
jet lag, which accounts for the sight of newly ar¬ reared animals (Rusak & Groos, 1982). Because
rived Europeans wandering through the lobbies of light is the primary zeitgeber for the circadian
New York hotels at 4 or 5 a.m., looking for an clock, we would expect that the SCN would re¬
open restaurant to have breakfast in. Because of ceive inputs from the visual system, and it does
their great speed of travel, their circadian rhythms (e.g., Groos & Meijer, 1985).
are still set to Paris, Moscow, or some other Eu¬
ropean time. Body time does eventually adapt to
Short-Term Timers
the new time zone at a rate of 1/2 to 1 hour per
day. This adaptation comes about through entrain¬ It seems clear that our circadian rhythms are main¬
ment of the biological clock to the local environ¬ tained by an internal biological clock, but we often
mental sunlight-to-darkness cycle. make estimates of times that are considerably
To use the scientific term, we would say that shorter than 24 hours. We can even accurately de¬
light is the primary zeitgeber (German for “time termine which of two time intervals was longer,
giver’’). There is much evidence, based on several when each was less than a second in duration. A
species of animals including humans, that shows “slow” circadian clock would probably be quite
the internal clock is synchronized to light (e.g., useless for this task, which suggests that there are
Johnson & Hastings, 1986). A brief flash of light probably several biological clocks in animals. For
will reset the biological clocks of animals reared in example, destroying the SCN does not affect the
constant darkness, either advancing it or retarding cyclic change in body temperature (Fuller, Lydic,
it, depending on when the flash occurs (Aschoff, Sulzman, Albers, Tepper & Moore-Ede, 1981), nor
1979). If there is no regular light cycle, however, does it seem to affect some shorter-cycling biolog¬
other environmental stimuli, such as daily fluctua¬ ical rhythms. In much the same way that we might
tions in temperature, may serve as zeitgebers to set use a stopwatch to measure short intervals, our
the internal timer. wristwatch to measure longer ones, and a calender
Is there a single physiological structure that to measure even longer periods of time, there seem
might serve as the biological clock? Researchers to be different biological clocks for different as¬
have isolated a region in the hypothalamus that pects of behavior. Heartbeats, electrical activity in
seems to be the basis of the circadian rhythm in the brain, breathing, hormonal and metabolic activ¬
animals and humans (e.g., Rusak & Zucker, 1979). ities, and even walking steps have at one time or
It is called the suprachiasmatic nucleus (which we another been suggested as candidates for an internal
will abbreviate as the SCN), and it is located very biological timing mechanism (Aschoff, 1981; Om-
near the optic chiasm, as can be seen in Figure stein, 1969; Poppel, 1978; Treisman, 1963). Some
 Time 375

FIGURE 13-1 The location of the suprachiasmatic nucleus of the hypothalamus, which is thought to be
the basis of the biological clock that maintains circadian rhythms.

of these might be useful as biological clocks to msec in duration, which would thus be the shortest
measure intervals shorter than the 24-hour circa¬ perceived duration a stimulus can have. In addi¬
dian period. Hence, we might view the perception tion, stimuli presented within the same moment
of time as occurring in a “clock shop” rather than either would be perceived as occurring simulta¬
in a single biological timer. neously or, depending on the nature of the stimu¬
Rather than looking at long time intervals, lus, would not be distinguishable from each other.
some researchers have gone to the other extreme Stimuli presented in different moments would be
and asked what is the shortest time interval we can perceived as being successive. Efron (1967, 1973)
sense. Experimentally, they asked the question, demonstrated this aspect of the perceptual moment
What is the minimum time separation needed for by looking at micropatterns, which are variations
two events to be perceived as occurring at different in a stimulus that occur so quickly that there is no
times (successively) rather than at the same time corresponding change in the perception. For in¬
(simultaneously)? In effect, they were searching for stance, a 20-msec stimulus composed of 10 msec
the basic time unit in perception. This idea was dis¬ of red light followed by 10 msec of green light is
cussed in some detail by Stroud (1955), who sug¬ not perceptibly different from one in which the
gested that psychological time is not a continuous green comes before the red—both appear yellow (if
dimension but rather consists of discrete bits. These visual persistence is eliminated; Yund, Morgan &
perceptual moments are the psychological unit of Efron, 1983).
time. Based on a number of research findings, White (1963) attempted to measure the per¬
Stroud estimated that each moment is about 100 ceptual moment by having observers estimate the
376 Chapter 13/Time and Motion

number of clicks they heard. He presented the perceptual moment comes from Intraub (1985),
clicks at different rates up to 25 per second. Ob¬ who presented a series of pictures to subjects at a
servers were fairly accurate at rates of up to 5 per rate of 1 every 111 msec. One of these pictures
second; at the highest click rates, however, observ¬ always had a frame around it, and observers were
ers still estimated a presentation rate of about 6-7 simply asked to indicate which picture had the
clicks per second. This corresponds to a perceived frame. On 54 percent of the trials, subjects reported
rate of 1 stimulus every 150 msec. Thus, informa¬ that the frame was around the picture that appeared
tion could not be processed in “chunks” smaller before or after the correct one, probably because
than 150 msec, which would be the resolution limit the two pictures fell within the same perceptual
of the internal timer. In another study, Efron moment. All these data suggest that, depending on
(1967) presented two brief pulses of light and asked the specific task, the minimum perceptual duration
observers to say which one was longer. One of the (or the time between ticks of the fastest biological
flashes was always 1 msec in duration; the other clock) is probably between 25 and 150 msec.
was of a variable duration. Both flashes were al¬
ways seen as being of the same length until the
Biological Pacemaker
exposure time of the variable flash exceeded a
value of 60 or 70 msec. At this duration, the vari¬ To the extent that there is a biological timer that
able flash was seen as being longer than the 1 -msec serves as a sort of pacemaker, ticking away inter¬
flash. Efron concluded that the minimum duration nal time, it would be reasonable to expect that it
of a stimulus in consciousness (which should be 1 would speed up or slow down along with other
perceptual moment) was around 60 or 70 msec. physiological processes in the body. Hoagland
It seems likely that the perceptual moment is (1933) verified this when his wife became ill with
different for different tasks and, perhaps, for dif¬ a high fever. He asked her to estimate the duration
ferent sensory modalities (e.g., Kolers & Brewster, of 1 minute by counting to 60 at a rate of one num¬
1985). For example, reaction time studies (where ber per second. When her body temperature was
observers are asked to react as quickly as possible approximately 39 deg C (103 deg F), her perceived
to a stimulus input) have indicated that short-term minute was only 37.5 sec by objective clock time.
memory can be scanned at about the rate of 25-30 This suggests that at higher body temperatures the
msec per item (e.g., Sternberg, 1975). The timing speed of physiological activities increases, and this
of well-trained motor tasks, such as typing or piano causes the pacemaker to tick more rapidly than
playing, also seems to support a 30-msec internal usual. Thus, when asked to reproduce a given
timing organization (Augenstine, 1962; Shaffer, physical time interval, a person with a high body
1985). Eriksen and Collins (1968) used a set of temperature produces an interval that is too short.
patterns that, if seen by themselves, seemed ran¬ An alternate way of looking at this is to note how
dom. If, however, two patterns were superim¬ our perception of physical (clock) time will seem
posed, either physically or psychologically, they to change when psychological time is running
contained a word. They found that when observers quickly. A given physical duration will appear to
were shown patterns sequentially, recognition for be too long if the psychological clock is ticking
the word was highest when the interval between the faster than the physical clock, hence giving more
presentations was about 25 msec. This implies that ticks per unit time than normally occurs (see Figure
the perception of simultaneity is maintained over 13-2).
only a 25 msec interval rather than one that ap¬ If an increase in body temperature increases
proaches 100 msec. apparent duration, then lowering body temperature
An interesting “reverse” demonstration of the may have the opposite effect. This was found by
 Time 377

Start of time End of clock

interval time interval
I I
i M 1 II I 1 II I 1 I [ A Passage of clock time

Fast biological clock; fast

g psychological time or decrease
in the size of the psychological
time unit

Slow biological clock; slow

| | | | | | | | | !| | | | n psychological time or increase
i i in the size of the psychological
1 i time unit

FIGURE 13-2 Each tick mark on these lines represents a unit of time. Those in A are clock time; those
in B and C are ticks of the biological clock. Notice that in B the internal clock is faster, so that the same
amount of clock time seems psychologically longer and time “drags by.” For C, the interned clock is
slower and so the same amount of clock time seems much shorter psychologically.

Baddeley (1966), who tested scuba divers diving in ing time using the slower ticking rate of their
cold water off the coast of Wales. Like Hoagland, internal timer led them to underestimate the pas¬
he asked his subjects to count to 60 at a rate of one sage of time. In other words, when your internal
number per second. After the dive, when their clock is too slow, physical time seems to whiz by
body temperature was approximately 1 deg C lower (see Figure 13-2). You may demonstrate the effects
than it had been prior to entering the water, his of temperature on your own time sense by trying
subjects required approximately 70 sec to count to Demonstration Box 13-1.
60. This indicates that their pacemakers were tick¬ If we have an internal biological clock, then
ing at a slower rate than the external clock. Count¬ anything that affects the rate of physiological

DEMONSTRATION BOX 13-1. Body Temperature and Time Perception

This demonstration is based on an experiment per¬ minute. Then take your temperature. (Do not take
formed by Pfaff (1968). We know that our body tem¬ your temperature before you count, otherwise it may
perature can fluctuate as much as 1 deg C during the bias your counting rate.) Do this several times
course of a day. It is at its lowest point early in the throughout the day, and keep a record of your results.
day and tends to rise throughout the afternoon. Given If the theory is supported, you should find that your
this, try the following observations. On rising in the counting time will shorten (relative to a clock minute)
morning, try counting to 60 at the rate of what you as your body temperature increases. Thus, as the
perceive to be one number per second. You will prob¬ body clock speeds up, the passage of time tends to be
ably need a friend to keep track of clock time for you overestimated and “clock” time seems to pass more
so that you can relate your perceived minute to a clock slowly.
378 Chapter 13/Time and Motion

function might also affect our estimates of time. shown to affect the subjective perception of dura¬
For instance, fatigue usually slows physiological tion are: (1) the number of events occurring during
functioning. Thus, the longer you are awake (or the the interval (e.g., Adams, 1977; Block, 1974;
greater the pressure for sleep) the slower your bio¬ Poynter & Holma, 1985), (2) the complexity of
logical clock, and the more likely that when asked stimulus events (e.g., Block, 1978; Omstein,
to estimate the passage of an hour your estimate 1969), (3) the type of cognitive or information pro¬
will be longer than a physical clock hour because cessing required (e.g., Hicks, Miller & Kins-
of shortened time experience (Aschoff, 1984; boume, 1976; Thomas & Weaver, 1975), and (4)
Daan, Beersma & Borbely, 1984). Similarly, gen¬ the amount of attention given to the passage of time
eral anesthetics lead to a shortening of time expe¬ (e.g, Brown, 1985; McClain, 1983). All of these
rience (Adam, Rosner, Hosick & Clark, 1971; affect the perception of the flow or passage of time,
Steinberg, 1955). Conversely, a number of inves¬ and all are consistent with the idea that the rate at
tigations have found that drugs like amphetamines which the cognitive clock ticks is affected by how
and caffeine (both of which are stimulants) lead to internal events are processed.
a lengthening of time experience (Frankenhauser,
Change
1959; Goldstone, Boardman, & Lhamon, 1958).
Drugs such as marijuana, mescaline, psilocybin, One notion is that the ticking rate of the cognitive
and LSD also seem to produce a lengthening of clock is dependent on event processing or change
perceived time relative to a nondrug state (Fisher, monitoring. The greater the number of events, or
1967; Weil, Zinberg & Nelson, 1968). It has been the more changes that occur, during an interval, the
argued that all these changes in time perception are faster your cognitive clock ticks, and thus the
caused by acceleration or deceleration of the phys¬ longer is your estimate of the amount of time that
iological pacemaker that serves as our internal has passed. Several studies seem to support this
timer. idea. A duration filled with stimulus events is per¬
ceived as being longer than an identical time period
empty of any external events, a phenomenon
Cognitive Clocks
known as the filled duration illusion. For exam¬
When you say that two minutes sitting on a hot ple, if we fill a time interval with brief tones, this
stove feels like two hours, but two hours sitting interval will be perceived as being longer than an
with your loved one seems like two minutes, you identical time period in which no tones (or fewer
actually are expressing the central aspect of most tones) were presented. This is also true for such
cognitive clock theories of time perception. These events as light flashes, words, or drawings (e.g.,
theories are based on the presumption that the per¬ Avant, Lyman & Antes, 1975; Hicks, Miller, Gaes
ception of the passage of time is based not on phys¬ & Bierman, 1977; Omstein, 1969; Poynter &
ical time but rather on the mental processes that Holma, 1985). Conversely, observers engaging in
occur during an interval. In effect, time is not di¬ restricted environmental stimulation technique
rectly perceived, but rather “constructed” or “in¬ studies (where they remain anywhere up to 24
ferred” (Fraisse, 1963; Woodrow, 1951). This hours or more, reclining in a soundproof, darkened
suggests that the tasks a person engages in will in¬ chamber with essentially all typical environmental
fluence that person’s perception of the passage of stimulation removed) tend to drastically underesti¬
time. Given the variety of potential cognitive activ¬ mate the amount of time they have spent in the
ities, it is perhaps not surprising that there are a chamber (Suedfeld, 1980). This underestimation
number of different variables that affect the cogni¬ occurs, presumably, because so few stimulus
tive clock. Among the variables that have been events have transpired during the interval.
 Time 379

Processing Effort are told in advance that we will have to judge the
time that a task takes, we tend to judge the duration
How difficult stimuli are to process, and the
as longer than if we are unexpectedly asked to
amount of memory storage they require, have also
judge the time after the task is completed (e.g.,
been shown to affect our perception of the duration
Brown, 1985; McClain, 1983). Simply telling ob¬
of a time interval. For example, we tend to judge
servers that they will later have to estimate the time
the brief presentation of a word to be longer in du¬
that has passed causes them to pay attention to, and
ration than a blank interval of the same length
perhaps to order, internal events and external phys¬
(Thomas & Weaver, 1975). Furthermore, the pre¬
ical events in a manner that increases the perceived
sentation interval of familiar words is judged to be
duration of the task.
shorter than the presentation interval of meaning¬
Conversely, anything that draws our attention
less verbal stimuli (Avant & Lyman, 1975; Avant,
away from actually monitoring the passage of time
Lyman & Antes, 1975) and presentations of non¬
should shorten our sense of ‘‘time passing.” For
familiar words appear to take longer than familiar
instance, making the task we are working on more
words (Warm & McCray, 1969). In both instances,
difficult makes it harder to attend to time directly.
an increase in the amount of information processing
For this reason, we find that estimates of the dura¬
required during the interval (a word versus a blank
tion of difficult tasks are usually shorter than esti¬
and a meaningless group of letters versus a word)
mates of the duration of easy tasks (e.g., Arlin,
leads to an increase in the estimated duration of the
1986; Brown, 1985; McClain, 1983). Sometimes,
interval. This is consistent with a processing effort
directing attention toward or away from the passage
model of time perception. Similarly, the more
of time may even reverse the filled duration illu¬
items you store in memory during an interval of
sion, which we discussed earlier, since it is more
time, the longer you judge the time to be (Block,
difficult to process many events in an interval while
1974; Mulligan & Schiffman, 1979), a notion
at the same time attending to the flow of time itself
sometimes referred to as the storage size model of
(e.g., Miller, Hicks & Willette, 1978; Zakay, Nit-
time perception. Both are based on the presumption
zan & Glicksohn, 1983). Demonstration Box 13-2
that the ticking rate of the cognitive clock is depen¬
shows how attention to time and task difficulty in¬
dent on the amount of cognitive activity actually
teract to affect our perception of the passage of
engaged in.
time.
It should be clear from the discussion above
Temporal versus Nontemporal Attention
that in the same way that there are a number of
Both the event processing and processing effort biological clocks that can interact in complex ways
mechanisms seem to affect our cognitive clock to give us a sense of the flow of time, there are
time, but the results are complicated by the way the also a number of cognitive clocks, or at least a
observer is attending to the task. A simple example number of ways to set the speed of a single cogni¬
of this is given by the old homily “A watched pot tive clock. Even level of motivation can affect the
never boils,” which suggests that the more atten¬ apparent passage of time. For example, in one
tion you pay to the passage of time, the longer the study subjects who were told that a desirable goal
time interval appears to be (e.g., Block, George & would be obtained on completion of their tasks
Reed, 1980; Cahoon & Edmonds, 1980). This may tended to overestimate the passage of time relative
be called the temporal processing model of time to subjects who did not have this expectancy (Filer
perception. & Meals, 1949). In another study, observers who
One of the best examples of the temporal pro¬ were placed in a stress situation (they were re¬
cessing model is the fact that, in general, when we ceiving electrical shocks) also overestimated the
380 Chapter 13/Time and Motion

DEMONSTRATION BOX 13-2. Time Perception and Attentional Factors

For this demonstration you will need a stopwatch or out loud. When you feel that 30 seconds has passed,
a watch with a sweep second hand. Do each step be¬ stop counting and note the amount of time that has
fore you read the instructions for the next one. elapsed.
1. Sitting quietly, note the time and then, with your 3. Compare the two time estimates. The first one
eyes closed and with no counting, estimate the pas¬ should be shorter than the second one, because your
sage of 30 seconds. Then open your eyes and note the cognitive clock was moving slower when you were
actual amount of time that has passed. attending only to the passage of time and faster when
2. Next, note the time, look away from the watch, you were dividing your attention between the count¬
and start to count backwards from 571 by threes ing task and the monitoring of time (see Figure
(e.g., 571, 568, 565, etc.). Be sure that you count 13-2).

passage of physical time (Falk & Bindra, 1954). we follow a moving car with our eyes and the im¬
Given all these factors, the perception of time re¬ age of the car remains fixed on the same retinal
mains a complex and only partially understood location. There are also times when we should see
phenomenon. movement but do not. When our eyes move from
one location to another the images of objects that
are stationary in the environment are sliding across
MOTION our retinal receptors, yet we perceive the world as
remaining stationary. Thus, movement of the ret¬
Perception is not static, but changes continually inal image does not fully account for the perception
over time. Some of these changes are like succes¬ of motion. Although it will be important for us to
sive snapshots, such as glancing from one page to understand the visual stimulus conditions that eli¬
another or shifting your gaze from one building to cit the perception of motion, you will soon see that
another as you stand in the street, but many other there are also important nonvisual factors to
changes are more continuous in nature, such as the consider.
sight of a car moving in the street beside you or a
bird flying through the air. These latter perceptual
Physiological Motion Detectors
experiences have the added quality of perceived
motion. It seems reasonable to begin our discussion of the
Your initial feeling might be that the percep¬ perception of motion by first seeing if there are
tion of motion is really quite trivial. You might ex¬ specific neural units for the detection of motion,
pect that all you need for motion to be perceived is much as there are for the detection of colors. The
the image of a visual stimulus moving across your existence of such physiological mechanisms is sug¬
retina. Actually, motion perception involves some gested by clinical cases where patients can no
fairly complex interactions among a number of dif¬ longer perceive motion. Consider one case of bilat¬
ferent systems (cf. Sekuler, Ball, Tynan & Mach- eral brain damage where a woman reported that,
mer, 1982). For instance, it is possible for us to although she could still recognize cars when she
perceive movement when the image of the stimulus saw one, she could no longer judge their speed.
is not moving across our retina at all, such as when The simple act of pouring a cup of coffee became
 Motion 381

virtually impossible since she could not see the dy¬ response to moving stimuli may be quite strong.
namic flow of the fluid, nor the rise of the liquid Thus, there seem to be cells that respond not only
level (Zihl, von Cramon & Mai, 1983). to particular directions of movement but also to
In Chapter 3, we described how certain cells particular velocities of the moving targets (Maun-
in the visual system respond differently to different sell & van Essen, 1983; Orban, Kennedy & Maes,
stimulus properties. For example, we noted that 1981a,b). There are also some suggestions that the
there were two types of retinal ganglion cells, an X entire tectopulvinar pathway in vision may be spe¬
type, which seems particularly suited for detail per¬ cialized for the perception of movement and to
ception in stationary targets, and a Y type, which control the direction of responses, such as eye
responds to any change in stimulation. The Y type movements, toward moving stimuli (Flandrin &
of cell seems particularly suited for the detection of Jeannerod, 1981; Guitton, Crommelink & Rou-
motion. The fact that X and Y cells are differen¬ coux, 1980; Von Essen, 1979). Furthermore, there
tially distributed across the retina (Y cells are more is recent evidence suggesting that there is a specific
abundant in the peripheral retina) helps to explain region of the cortex that contains many cells with
why the apparent speed of a moving target might very specific velocity tuning and directional char¬
depend on where in the visual field it is (Campbell acteristics. This region is not in the occipital cor¬
& Maffei, 1981). For example, you will probably tex, which is the primary visual center, but is
recall from Chapter 4 that visual acuity diminishes located in the temporal lobe of the brain (Allman,
with distance from the fovea (the number of X cells Miezin & McGuinness, 1985; Maunsell & van Es¬
diminishes with increasing distance from the sen, 1983).
fovea). In a similar fashion, our ability to detect Some of the information about the nature of
slow target movements (up to about 1.5 deg/sec) motion-specific cells in humans comes from the use
decreases with distance from the fovea (Choudhurt of a psychophysical technique called selective ad¬
& Crossey, 1981; Lichtenstein, 1963; McColgin, aptation (Sekuler, 1975). The rationale of this
1960). For higher target velocities, however, this technique is the same as the adaptation procedures
relationship reverses. At moderate to fast veloci¬ we discussed in the spatial frequency section of
ties, the peripheral retina seems better able to de¬ Chapter 4. It involves exposing the eye to a moving
tect movement (because of the increased proportion pattern such as a field of stripes. Prolonged viewing
of Y cells) even though the decrease in acuity may of such a stimulus causes the motion-specific cells
be so great that the observer may not be able to that have been responding to the direction of the
recognize what is moving (Bhatia, 1975; B. pattern’s movement to become less sensitive (e.g.,
Brown, 1972). Hunzelmann & Spillmann, 1984). When this hap¬
At the level of the visual cortex, we find even pens, an observer’s ability to detect the movement
clearer evidence for specialized cells tuned to stim¬ of other patterns, moving in the same direction and
ulus motion. Specifically, there are numerous com¬ at the same speed as the previously exposed adapt¬
plex cells in the cortex of mammals that respond ing pattern, is reduced. However, this drop in sen¬
only to moving targets. Not only do these cells re¬ sitivity does not carry over to faster or slower
spond to motion, but they are tuned to the direction movements, nor does it generalize to movements in
of stimulus movement, discharging strongly when the opposite direction (Sekuler, 1975; Sekuler &
a properly oriented stimulus drifts in one direction Ganz, 1963; Wright & Johnston, 1985). Thus, the
across the visual field and responding less strongly selective adaptation procedure gives results consis¬
(or not at all) when the same stimulus moves tent with the idea that our brains contain move¬
through the field in the opposite direction (see Hu¬ ment-sensitive cells that are tuned to a particular
bei & Wiesel, 1979). The degree of specificity of direction and velocity. To experience an interesting
382 Chapter 13/Time and Motion

illusion that researchers believe is caused by the parts of the visual image; the second involves using
fatigue of motion-sensitive cells, try the procedure our eyes to follow a moving target. Presumably
in Demonstration Box 13-3. these involve different perceptual systems, and for
convenience we call the system that responds to
image changes the image-retina system and the
Stimulus Factors in Motion Perception
one that interprets motion from our eye and head
Generally speaking, there are two principal mech¬ movements the eye-head system (cf. Gregory,
anisms by which we perceive motion. The first in¬ 1978). Since the image-retina system involves
volves detecting shifts in the relative positions of stimulus relationships, let us consider it first.

DEMONSTRATION BOX 13-3. Motion Aftereffect

To experience a motion aftereffect, cut out (or trace) movement, since the stimulus is no longer in motion.
the accompanying stimulus and place it on the turn¬ It is probably caused by fatiguing, or selective adap¬
table of a record player as if it were a record. Let the tation, of physiological motion detectors, produced by
stimulus rotate for about a minute, while you stare at prolonged stimulation in one direction of movement.
the center. Stop the turntable and hold it so that it is You can demonstrate that the cells are tuned for dif¬
completely stationary. While the turntable was mov¬ ferent stimulus velocities by changing the speed of
ing, the spiral appeared to expand. Now it should ap¬ your turntable and repeating the demonstration. You
pear to be (paradoxically) shrinking. This shrinking will notice that this will change the rate of shrinking
(without any apparent change in size) is an illusory in the aftereffect.
 Motion 383

Perhaps the first question to ask is how much change” in the visible pattern and hence may in¬
movement in the image is needed before we can volve processes similar to form perception (e.g.,
perceive motion. To answer this question, we usu¬ Mack, Heuer, Fendrich, Vilardi & Chambers,
ally measure a movement threshold just as we mea¬ 1985; Wallach, Becklen & Nitzberg, 1985). In
sured thresholds for the minimum amount of light terms of the detection and discrimination of mo¬
or sound needed for sensation (see Chapter 2). Our tion, we appear to be much more sensitive to ob¬
sensitivity to the movement of an external target ject-relative change.
depends on several variables. In experimental set¬ We have just seen how the addition of a visual
tings, movement thresholds have usually been stud¬ context or background, in the form of stationary
ied using a small point of light that moves against stimuli in the field, can increase our sensitivity to
some sort of stationary background, as in one of motion. Under certain conditions the relationship
the earliest studies by Hermann Aubert (1886). He between the visual context and a target stimulus
found that observers could detect the movement of can also distort our perception of movement. For
a luminous dot in the dark, 50 cm from the eye, instance, Duncker (1929) displayed a bright dot in
when it was moving at about 2.5 mm per second a dark room. When the dot was moved very
(which is about Vs of a degree of visual angle per slowly, observers were not certain whether or not
second). it was moving. However, when a stationary dot
Target movement alone, however, is not was placed near the moving dot (in effect becoming
enough to allow us to describe the motion thresh¬ the visual context), it became quite clear that one
olds. Our ability to detect that a target has moved of the dots was in motion (due to the object-relative
is much more sensitive when there is some motion¬ changes). Curiously, observers could not identify
less reference, such as stationary features associ¬ which of the two dots was moving. Duncker next
ated with the background, which we call the visual changed the context stimulus by making it a rectan¬
context. An example might be a stationary square gular luminous frame that was stationary and sur¬
frame surrounding the target. Under these circum¬ rounded the dot. Under these circumstances there
stances we find that observers are much more was no ambiguity and observers were able to tell
sensitive (e.g., Palmer, 1986). The minimum move¬ that the dot rather than the frame was in motion.
ment that can be detected in the presence of a Duncker next varied the conditions so that the dot
stationary visual context is about .25 mm per sec¬ was stationary and the surrounding rectangular
ond (or Vioo of a degree of visual angle per frame was moving. Under these circumstances an
second). This is an incredible degree of movement illusion appeared, in that observers reported that
sensitivity. If a snail were to crawl across a desk the stationary dot was moving rather than the
1.5 m wide at this rate, it would take it 1 hour and frame. Duncker called this induced motion, since
40 minutes to go from end to end. the perceived movement of the dot was induced or
Our ability to judge the difference between brought about by the real movement of the sur¬
two velocities is similarly facilitated by the pres¬ rounding context. This is similar to the perception
ence of other stimuli that are stationary (Bonnet, that the moon is moving behind the clouds, when
1984). Some researchers contend that the image- actually the clouds are moving quickly and the
retina system really involves two different sources moon moves much more slowly (relative to the
of motion information. The first is subject-relative earth) than we can detect. The clouds provide a
change, where the only information is the move¬ surrounding context that is in motion and, consis¬
ment of the target relative to the observer’s position tent with the principle that Duncker discovered in
in space. The second is called object-relative the laboratory, they induce an apparent motion of
change, which is the movement of one target rela¬ the not-detectably-moving moon.
tive to others and creates a sort of “configurational Induced movement effects are most dramatic
384 Chapter 13/Time and Motion

when the context is moving slowly rather than Max Wertheimer (1912). Beginning with two lines
quickly (Wallach & Becklen, 1983). Square frame separated in space that could be flashed on and off
shapes are more effective than circular frames and sequentially, he varied the time interval between
large surrounds are more effective than small ones the offset of the first line and the onset of the sec¬
(Michael & Sherrick, 1986). Furthermore, in order ond (we call this variable period the interstimulus
to induce motion, the target and the background interval). When the interstimulus interval was very
must be at the same distance from the observer brief, observers saw two lines appear simulta¬
(that is, apparently near each other). If the frame neously. If the interval was long, the observers saw
that supplies the context is too far in front of or a line appear followed by a second line in a differ¬
behind the target, no motion will be induced (Go- ent location. However, for some intermediate inter¬
gel & Koslow, 1972). Actually, the appearance of stimulus interval Wertheimer’s observers reported
induced motion seems to be controlled by the part that they saw a line appear and then move from the
of the visual context or frame that is closest to the first position in space to the second. Although ini¬
target (Schulman, 1979). You can produce induced tially called phi movement, we now refer to this
motion yourself by following the instructions in experience of movement between successively pre¬
Demonstration Box 13-4. sented stationary stimuli as apparent movement,
to distinguish it from real movement, where the
stimulus actually moves in space. You can dem¬
Apparent Motion
onstrate this type of apparent movement for your¬
Illusions of movement, such as induced motion, self by following the instructions in Demonstration
might seem to be merely interesting but not very Box 13-5.
useful curiosities. Yet every time you go to the cin¬ The magnitude of the apparent movement ex¬
ema, you are paying to see two hours of a motion perience is dependent on the interstimulus interval
illusion. Each frame in the film you watch is ac¬ and also on the distance between the positions of
tually stationary, being exchanged for a new frame the stimuli. Generally speaking, when the stimuli
about 24 times a second; television works in much are separated by larger distances, longer time inter¬
the same way, with static frames changing about vals between the stimuli are needed for apparent
30 times a second. motion to be perceived (Farrell, 1983).
One of the early psychological researchers Several researchers have suggested that there
who systematically studied this phenomenon was are actually two separate perceptual systems that

DEMONSTRATION BOX 13-4. Induced Movement

To induce movement in a stationary target, all you the dot and slowly move the cellophane across the
need is a sheet of clear cellophane or glass, and a paper. You will notice that the dot appears to move
sheet of white paper. In the middle of the white paper in the direction opposite the motion of the rectangle.
draw a small dot. On the clear cellophane draw a The effect is strongest when the dot is near the sides
large rectangle, about 10 to 16 cm (4 by 6 in.), using of the rectangle, where object-relative change plays a
a felt-tip marker or a grease pencil. Now lay the clear role. Increasing the speed of movement should reduce
sheet over the paper so that the dot is enclosed by the the amount of induced motion you perceive. Why?
rectangle and near one of its sides. Look steadily at
 Motion 385

DEMONSTRATION BOX 13-5. Apparent Movement

To see apparent movement similar to that described close each eye in turn (remembering to keep your
by Wertheimer, simply hold your index finger verti¬ eyes relaxed). At slow rates you should see your fin¬
cally a short distance in front of your nose. Look at ger “jump” from side to side; however, at some
any distant target (such as a mark on the far wall of moderate rate of winking you should see the finger
the room). Relax your eyes and alternately wink each appear to actually “move” from one position to the
eye. You should see your finger in a different place other.
with each eye. Now, begin to rhythmically open and

bring about apparent movement. They are called, The second system, the long-range process,
respectively, the short-range process and the responds to stimuli that are spatially separated by
long-range process (Anstis, 1978; Braddick, distances greater than 15 minutes of visual angle;
1980). The short-range process encodes only small this may include separations of many degrees
spatial target jumps as motion, perhaps 15 minutes across the visual field. The interstimulus interval
of visual angle or less. It also responds only to needed to produce the perception of motion is also
fairly short interstimulus intervals, usually less than much longer, over 100 msec. This process seems
100 msec (e.g.. Baker & Braddick, 1985). This to be based on more complex inferential proce¬
perceptual process probably detects only simple dures—for instance, the brain deciding that it is
shifts of a contour, and is probably mediated by improbable that the disappearance of one stimulus
activity of some of the physiological motion detec¬ should be followed so quickly by the sudden and
tors we discussed in the section of that name. This independent appearance of an identical stimulus
is supported by the observation that if we expose nearby, and therefore concluding that the original
an observer to such short-range apparent motion in stimulus must have moved to this new location. In
one direction for a while, a motion aftereffect much some ways this process seems to involve a form of
like the one produced in Demonstration Box 13-3 “logic” or simplifying principle, which derives ap¬
will appear (Vautin & Berkley, 1977). parent motion as a reasonable interpretation of the
386 Chapter 13/Time and Motion

stimulus changes observed (Hatfield & Epstein, transformations can predict the motion of differing
1985; Rock, 1983). objects, and how individuals can later identify
In some respects, the apparent motion ob¬ these objects based on schematic movement pat¬
tained from the long-range process demonstrates a terns.
good deal of “tolerance” in its interpretation of Perhaps the most interesting work is with
movement. For example, suppose we present an biological motion (e.g., Johansson, 1976a). This
apparent movement display, alternately flashing refers to the intricate and coordinated set of
spatially separated stimuli at a rate we know pro¬ movement patterns accomplished by the skeletal
duces the sensation of motion. Now suppose that structure of the human body, for example, when
the target on the right is red and the one on the left walking across the room. Johansson began by ask¬
is green. Will we still see motion? The answer is ing, “Will an observer be able to identify these
that we will see a target both moving and changing motions as the act of walking even in the absence
color as it moves. We can get apparent motion not of any other information, such as sight of the per¬
only between targets of different colors but be¬ son?” To answer this question Johansson and his
tween targets with different shapes, sizes, bright¬ co-workers used the following technique. They at¬
nesses, and orientations, and in most of these tached small flashlight bulbs to the shoulders, el¬
situations the target seems to be transformed while bows, wrists, hips, knees, and ankles of an
it is moving (Anstis & Mather, 1985; Bundesen, individual (see Figure 13-3A). They then made a
Larsen & Farrell, 1983; Kolers & Green, 1984; motion picture film of the person as he moved
Kolers & von Grunau, 1976). Similarly, apparent around in a darkened room. When observers later
motion is “flexible” in that it will adapt to the con¬ watched the film they only saw a pattern of lights
ditions in the visual field. If you place an object in moving about in total darkness. Nonetheless, ob¬
the pathway of the apparent movement, the per¬ servers were able to identify the pattern as a person
ceived path of motion will seem to deflect around walking or running, even when they only got to see
that object (Berbaum & Lenel, 1983). If a particu¬ the motion for as short an exposure as 200 msec
lar pathway is suggested, by, for instance, briefly (Johansson, von Hofsten & Jansson, 1980). Ob¬
flashing a curved path between the two flickering servers were also easily able to detect abnormali¬
stimuli, the apparent motion will seem to follow ties, such as the simulation of a small limp. In
that pathway (Shepard & Zare, 1983). All these another experiment, two people with similar arrays
factors suggest that higher-level cognitive process¬ of lights were filmed while performing a spirited
ing mechanisms play a role in the perception of folk dance. Figure 13-3B shows a series of posi¬
apparent motion, at least when we are dealing with tions from the folk dance in which the black dots
the long-range process (Rock, 1983). mark the positions of the lights. Once again, even
with only a moving pattern of lights, observers had
no difficulty identifying the motion as a dancing
Biological Motion
couple (Johansson, 1976b). Infants as young as 4
All the experiments we have thus far described months of age seem to notice that biological motion
tended to use rather simple stimuli, and we have is different from other forms of motion, and prefer
mostly been concerned with the process by which to watch patterns of the sort we have been discuss¬
we see motion in these stimuli. Movement patterns ing, rather than random patterns of lights moving
themselves, however, serve an important function (Fox & McDaniel, 1982).
in helping us identify objects in our environment. Our precision in identifying individuals based
In a series of studies, Gunnar Johansson and his only on their biological motion patterns is really
co-workers have shown how various perspective quite striking. For example, in one study research-
 Motion 387

FIGURE 13-3 An example of the type of displays used by investigators to study patterns of humans in
motion. A indicates the positions of lights affixed to individuals and B shows a sequence of movement
positions made by a dancing couple.

ers photographed a group of people who were ac¬ speed, bounciness, and rhythm of the walker, the
quainted with one another. These people were amount of arm swing, or the length of steps as fea¬
photographed with only lighted portions of several tures that allowed them to make their identification.
joints visible. Several months later these same in¬ In other studies, these same investigators found that
dividuals were invited to watch the films and at¬ observers could tell, even under these conditions,
tempt to identify themselves and their friends in the whether a person was a male or a female, despite
motion picture. People were able to identify them¬ seeing only a moving pattern of dots. In fact, it was
selves and others correctly on a fair percentage of not necessary for all the body joints to be repre¬
trials, although their performance was not perfect. sented in the light display for people to make cor¬
The investigators also asked their observers how rect identifications. Even when only the ankles
they went about making their identifications of var¬ were represented, observers could detect the sex of
ious individuals in the film. People tended to men¬ the walker. They could also make these gender
tion a variety of motion components such as the identifications within about 5 sec of viewing
388 Chapter 13/Time and Motion

(Barclay, Cutting & Kozlowski, 1978; Cutting & the body parts in an individual fashion making
Kozlowski, 1977; Cutting & Proffitt, 1981; Koz¬ identification possible. Perhaps it is patterns of bi¬
lowski & Cutting, 1977). Thus, different motion ological motion such as this that enable us to iden¬
patterns characterize each sex and each individual. tify people in light too dim to allow us to see their
Much work is being done to determine the na¬ faces. It also probably explains how you can iden¬
ture of the information used to identify individuals, tify people walking down the street, even though
and from this some fairly sophisticated computer they may be too far away for you to make out their
programs have been developed to create simulated features or may have their backs to you.
biological motion patterns (Cutting, 1978; Runeson
& Frykholm, 1983; Todd, 1983). For instance,
Eye Movements and Motion
Cutting, Proffitt, and Kozlowski (1978) proposed
Perception
that the torso of the body acts like a flat spring with
the limbs in symmetrical motion around it. This, Up to now we have focused our discussion mainly
along with certain individual differences in bodily on the visual stimulus factors that contribute to our
dimensions (such as the relative widths of the perception of motion, such as movement of the im¬
shoulders and hips), provides a center of movement age across the retina. To that extent we have been
that is not necessarily associated with any body concerned with the image-retina movement system
part; however, it organizes the coherent motion of (see Figure 13-4A). There are, however, aspects of

Initial target Initial target

position position

Perception: Target moves Perception: Target moves

to right to right
A B

FIGURE 13-4 (A) The image-retina movement system. The image of the moving object stimulates the
retina when the eyes are held stationary. This gives information about object motion, possibly as a result
of the involvement of movement-detecting cells. (B) One of the functions of the eye-head movement
system. When the eye pursues a moving target, the image remains stationary on the fovea of the eye but
we still perceive the movement of the object.
 Motion 389

motion perception based on information from Actually, there are two forms of smooth-
sources other than the visual image. The most well pursuit eye movements. The first is reflex pursuit
researched of these alternative modes of motion movement, which keeps images of objects rela¬
perception involves the eye-head movement system, tively fixed in one place on your retina despite the
which we mentioned earlier. This system enables fact that your head may be moving. An example of
us to detect the movement of external objects even this is shown in Figure 13-5, where the individual
when the image remains in a fixed position on the is steadily looking at the lens of the camera. Notice
retina. This most commonly occurs when we move that the eyes seem to remain stationary while the
our eyes to follow the path of a physically moving head seems to rotate around them, whereas actually
object, as when we track an automobile moving the eyes are tracking in the direction opposite to the
down the highway. Such eye movement, called head movement. This automatic reflex movement
smooth-pursuit movement and illustrated in Fig¬ is controlled by the vestibular system (see Parker,
ure 13-4B, functions to keep the image of the tar¬ 1980), which we will discuss later in this chapter.
get on the fovea (the most acute part of the retina). The second type of smooth eye movement is vol¬
Suppose that the image of the target remains on the untary pursuit movement. This type of eye
fovea and that there is no patterned background to movement (the one we showed in Figure 13-4B)
provide object-relative information as to move¬ tries to keep the image of the object fixed on the
ment. This means that the only way the observer fovea, despite its physical movement across the vi¬
can know the path or the speed of an object is to sual field. This system is actually found only in
monitor the speed and direction of the tracking eye animals that have foveas. It now seems clear that
movements (e.g., Epstein & Hanson, 1977; Rock voluntary pursuit eye movements provide most of
& Halper, 1969). It should not surprise us to learn the information about target movements (Post &
that spatial information is conveyed by eye move¬ Leibowitz, 1985; Raymond, Shapiro & Rose,
ments, since we have previously (see Chapter 10) 1984).
discussed how eye movements can provide us with To the extent that our perception of the motion
information as to the size and location of objects of an object depends on information about the
(e.g., Coren, 1986). movements our eye has used to track the object, it

FIGURE 13-5 Reflex-pursuit eye movements are used to keep the image of an object fixed on your
retina even though you move your head. Here the individual is looking at the camera while rotating her
head. Notice how these vestibularly controlled movements keep the eyes fixed while the head seems to
rotate around them.
390 Chapter 13/Timf, and Motion

seems reasonable that anything that alters the direc¬ is the Aubert-Fleischl effect (named for the two
tion or speed of motion of the eye might also alter researchers who explored it). Aubert and Fleischl
our perception of the movement of the object. This noted that when we track a target with our eyes
routinely occurs, since the eye does not pursue while it moves relative to a stationary background,
moving targets with perfect accuracy, but rather it appears to move more slowly than it would if we
tends to follow some distance behind the target. were to fixate steadily on the stationary back¬
The degree to which the eye lags behind is depen¬ ground. This phenomenon is associated with sev¬
dent on the speed of the target (Fender, 1971; eral predictable perceptual distortions. Because of
Puckett & Steinman, 1969), and under some cir¬ the lag in tracking, an observer will not only un¬
cumstances the eye never really catches up to the derestimate the velocity of a target that is tracked
stimulus (Young, 1971). This may cause distor¬ with the eye but will also tend to underestimate the
tions in the size or the shape of the path the eye distance the target has moved (Mack & Herman,
follows (e.g., Festinger & Easton, 1974). For ex¬ 1972). Demonstration Box 13-6 allows you to see
ample, Coren, Bradley, Hoenig, and Girgus (1975) this effect for yourself.
have shown that the size of the circular path traced There are some circumstances where our own
out by a rotating spot of light seems to shrink as eye or head movements produce movements of the
the speed of the target increases. At slow speeds, visual image across our retina that are very similar
where the eye can track accurately, or at speeds to those that might occur if the scene were actually
much too fast for even an attempt at tracking, the in motion. Another important function of the eye-
judgments are reasonably accurate. head movement system is to compensate for such
Another example of the effect of tracking lag movements, so that we continue to see the world

DEMONSTRATION BOX 13-6. The Aubert-Fleischl Effect

To experience the underestimation of both speed and

distance moved when you track a target, begin by
practicing a movement that will serve as your tracking
target. With your eyes closed, swing your hand back
and forth in front of you at a moderate speed and in
a rhythmic fashion as shown in the figure. When you
are moving with a nice regular tempo, open your
eyes, look straight ahead, and judge the speed and the
distance your hand is moving. Now look directly at
your finger and track it. Notice that your hand seems
to be moving more slowly, and the size of the back-
and-forth movement (the path length) appears to be
shorter. The apparent path length seems shorter prob¬
ably because when you are tracking the target, infor¬
mation about the extent of the motion comes from the
eye-head system, and your tracking movements lag
behind the physical target, resulting in a slower over¬
all velocity and a shorter eye-movement path length.
 Motion 391

as being stationary despite the fact that we are the brain initiates an eye movement, efferent (mo¬
moving. This process of compensating for eye tor) signals are sent out commanding the eyes to
movements is called position constancy, and the move. Copies of these signals, sent to central re¬
fact that objects seem to maintain a fixed position gions of the visual system, cancel the movement
relative to us, despite rotations of both our head information coming from the retina as the eyes
and eyes, is called direction constancy. We are move. Since the interpretation of the origin of
quite accurate in our ability to distinguish image movement is based on information from the mes¬
movements caused by target movements from those sage sent out from the brain that initiates an eye
caused by our own movements (see Wallach, movement, this is called an outflow theory. As in
1987). Although we do not know exactly how this the case of inflow information, there seem to be
movement compensation system works, it must in¬ some cells in the cerebellum and the cortex of
clude a system that monitors the changing position monkeys that contain information about eye posi¬
of the eye relative to the position of the head, either tion. Since these cells respond before the actual
when we are tracking a moving target or when we movement takes place, they could represent the
are making eye movements to scan a stationary source of outflow information registering the inten¬
scene (Howard, 1982). Two classes of theory have tion to move, rather than the movements them¬
been proposed to explain how the eye-head move¬ selves (Miles & Fuller, 1975; Wurtz & Goldberg,
ment system accomplishes this. 1971).
Sir Charles Sherrington (1906) suggested that To see how outflow information might com¬
motion is detected via feedback information from pensate for eye movements, try this little experi¬
the six extraocular muscles that control eye move¬ ment suggested by Helmholtz. Place your hand
ments. This feedback information, called propri¬ over one eye and try tapping or pushing (through
oceptive or position information (see Chapter 9), the eyelid) the side of your other (uncovered) eye
enables the observer to monitor eye position. The very gently with your fingertip. This rotates the eye
proprioceptive information tells the brain that the in a movement similar to one that could be initiated
eyes have moved, and in turn this information by the brain. However, in this case the brain has
allows the brain to interpret movement across the not sent a signal to move the eye. When the eye is
retina as being observer-generated rather than ob¬ rotated in this passive fashion, the visual field will
ject-generated. In support of this theory, there is seem to swing around in the direction opposite to
evidence that there are cells in the superior colli¬ the movement of the eye, and to the same extent
culus and visual cortex of the cat that monitor eye that the eye actually moved (e.g.. Miller, Moore &
position (Berkley, 1982; Kurtz & Butter, 1980). Wooten, 1984). Thus, the stability of the visual
These cells fire at different rates depending on the field holds only for eye movements initiated by sig¬
extent and direction of eye movement (Donaldson nals from the brain. Passive eye movements result
& Long, 1980; Kasamatsu, 1976; Noda, Freeman in an apparent movement of the visual field. It
& Creutzfeldt, 1972). Sherrington’s theory is often seems as though the action of the eye-head system
called an inflow theory because it is the informa¬ requires that the signals to and/or from the eye
tion “flowing in” from the eye muscles to the muscles be compared to signals arriving from the
brain that constitutes the crucial message for the retina indicating changes in retinal image position
interpretation of movement. (Matin, 1982). Figure 13-6 illustrates the differ¬
A different theory about how the eye-head ence between the inflow and outflow theories; it
system compensates for self-generated movements seems likely that both inflow and outflow may be
of the visual image was offered by Hermann von needed to provide a full explanation of the eye-
Helmholtz (1909/1962). He suggested that when head movement system. Table 13-1 gives a
392 Chapter 13/Time and Motion

FIGURE 13-6 A major function of the eye-head movement system is to maintain the stability of the
visual world during eye movements, thus achieving position or direction constancy. The inflow theory (A)
states that this is accomplished by comparing movement signals from the retina with proprioceptive
feedback from the eye muscles, which indicates that the eye has moved. The outflow theory (B) maintains
that commands from the brain to initiate voluntary eye movements cancel the movement information
coming from the retina as the eyes move.

summary of the relationship between the image-ret¬ The Vestibular Sense

ina and eye-head movement systems. Demonstra¬
tion Boxes 13-7 and 13-8 give other interesting There is another sensory system that is primarily
examples that illustrate the relationships between concerned with motion and motion-related matters,
eye movements, visual stability, and the perception and tends to interact with visual stimulation to sep¬
of motion. arate bodily motion from stimulus motion. This is
 Motion 393

Table 13-1. The Effect of Retinal Image Change and Voluntary Eye Movements on the Perception of
Motion

Commands to
Physical Target Action of Eye Retinal Image Eye (Head) Perception

t'-fmage-retina
Moving Stationary Moves None Movement
Eye-head
Moving Tracks target Stationary on fovea Yes Movement
Stationary Moves Moves Yes No movement
/stationary Passively pushed Moves None Movement in direction
opposite to eye
motion
i^fmage Moves Stationary on retina Yes Movement in same
stabilized direction as eye
on retina motion

DEMONSTRATION BOX 13-7. Afterimages and Apparent Movement

You can readily experience one of the ways in which Perhaps the easiest way to generate an afterimage
the eye-head movement system differentiates external is to look at a rather bright but small source of light
from observer movement. The first thing that is for a brief period of time. Make a 1-cm hole in an
needed is to generate a stabilized retinal image. Or¬ index card and hold it up in front of a light bulb.
dinarily, the retinal image is in constant motion and Look at the hole for a few moments and this should
stimulates varying groups of receptors at a rapid rate. provide a clearly visible afterimage when you look
However, by quickly satiating or fatiguing a single away from the light. Now notice that each time you
group of retinal receptors, we can generate an image move your eyes the afterimage seems to jump in the
that maintains its position regardless of eye move¬ same direction. This apparent movement is due to the
ments. Many of you are probably familiar with the action of the eye-head system.
technique used to give rise to such an image if you Commands have been issued to the eye to move,
have ever had your picture taken with a flashbulb at¬ yet the image remains on the same place on the ret¬
tached to the camera. If you looked at the light while ina. This could only occur if the image had moved as
it flashed, you may have noticed a purple dot that much as the eye (see Table 13-1). You may also no¬
tended to linger in your field of view for some time tice that the image sometimes seems to drift smoothly
after the picture was taken. This purple dot is called from place to place. Again, the image never moves;
an afterimage. It is one example of a stabilized retinal the movement is signaled from the movements of
image. The afterimage does not shift position on the your eyes. This is one example of how the action of
retina. It stays in a constant position regardless of the eye-head movement system can lead to illusions
how we move our eyes. We can use the afterimage to of motion.
demonstrate the operation of the eye-head movement
system.
394 Chapter 13/Time and Motion

DEMONSTRATION BOX 13-8. The Autokinetic Effect

There is an interesting phenomenon in which move¬ may have guessed, these are not monitored by the
ment is seen in the absence of any physical motion of visual system. One type of involuntary eye movement
the target. The word used to describe the occurrence is eye drift, and this is the mechanism that has been
is autokinesis, which means “self-moving.” For this implicated in the autokinetic effect (Matin & Mac¬
experiment, you will need a very dark room. No stray Kinnon, 1964). When we steadily fixate or stare at a
light of any sort should be visible. In addition, you target, it is difficult for the eyes to maintain steady
will need a small dim point of light (a lighted ciga¬ and accurate fixation on that one point in space
rette works fine). Place the point of light about 2 m (Ditchbum, 1973). The eyes will tend to drift off of
away from you and look at it steadily. After a few the fixation point; however, the visual system does
minutes it should appear to move, perhaps slowly not monitor this movement until it exceeds a critical
drifting in one direction or another. Of course the point. In the autokinetic situation, retinal image
light is still stationary; hence the movement is an il¬ movement has been signaled in the absence of com¬
lusion, which is called the autokinetic effect. mands to initiate voluntary eye movements. This is
The autokinetic effect demonstrates the outflow the situation under which the movement of the retinal
principle that operates in the eye-head movement sys¬ image is attributed to an externally moving object (see
tem. The visual system only monitors commands to Table 13-1). There is no information that the eyes
initiate voluntary eye movements. However, these are have moved, so illusory movement of the dim spot of
not the only types of eye movements possible. Our light is seen.
eyes also exhibit involuntary movements. As you

the vestibular system, which functions to inform the stone rolls along over a number of different
us about the motion of our body through space, to hairs, bending them and generating a different re¬
assist in the maintenance of an upright posture, and sponse indicating tilt. The function of such organs
to control eye position as we move our heads while is to signal the animal’s orientation with respect to
viewing various stimuli. For the most part, these gravity. More advanced invertebrates, such as the
operations take place outside of consciousness. Be¬ squid or octopus, have statocysts that approach ver¬
fore we consider the higher-level interactions that tebrate vestibular organs in complexity, with mul-
involve this system, let us first consider its basic tichambered statocysts and the ability to detect
structure and physiology. acceleration in several planes (Stephens & Young,
Some of the most primitive organisms have 1982).
organs that are sensitive to changes in motion of Primitive vertebrates have organs that have a
the body. In primitive invertebrates such as the similar function; they are called otocysts and the
crayfish these are called statocysts. Each consists bones they contain are called otoliths. Notice that
simply of a fluid-filled cavity that is lined with hair these terms each contain the root oto, meaning
cells. In the cavity is a tiny stone, called a statolith “ear.” These organs are usually closely associated
or “still stone,” that rests on the hairs. When the with the ears, since both the auditory receptors and
animal accelerates, the stone tends to lag behind the vestibular organs probably evolved from pits on
because of its inertia, thus bending the hairs on the surface of hairy skin. In mammals, these organs
which it rests. This action generates an electrical are protected by the skull from possibly damaging
response to the movement. If the animal is tilted. outside forces. In humans, the bony labyrinth in
 Motion 395

the head contains the cochlea (which is the auditory each semicircular canal (see Figure 13-7). The
organ) and the semicircular canals, the utricle, crista consists of an array of sensory cells from
and the saccule, which comprise the vestibular or¬ which tiny hairs protrude, as shown in Figure 13-8A.
gans (see Figure 13-7 and look back at Figure 6-4). These hairs are embedded in a jellylike material
called the cupola. When your head accelerates, the
inertia of the fluid in the canals causes the cupola
Vestibular Stimuli and Receptors
to move in the opposite direction. This in turn
The effective physical stimulus for any vestibular causes the hairs to bend, generating neural re¬
organ is change of rate of motion, or acceleration, sponses. As your head continues to move at a par¬
which occurs whenever we move through space, ticular rate of speed, the cupola gradually comes
whether we jump up and down, take off in a jet back to its resting position, no longer bending the
plane, or simply stand up and walk. The semicir¬ hairs, and no longer causing a response in the sen¬
cular canals and their associated receptor organs sory cells. This is why the effective stimulus is ac¬
seem particularly well suited for monitoring rotary celeration rather than steady movement.
acceleration (as when turning around or falling The receptor organ found in the utricle and the
down). The other two organs, the utricle and the saccule is called the macula, shown in Figure
saccule, seem mainly to respond to linear acceler¬ 13-8B. It functions much like the statocyst we dis¬
ation (as when taking off in a plane). cussed before. As in the crista, tiny hairs protrude
The movement-receptive portion of the semi¬ from the sensory cells in the macula. These hairs
circular canals is called the crista, which is found are embedded in a jellylike substance covered by a
in a swelling (called an ampulla) at the base of membrane containing otoliths, which lags behind

Anterior
canal
Bony
labyrinth
Ampulla
Branches of
Membranous vestibular nerve
labyrinth

Lateral
canal
Cochlea

Posterior
canal

Round window Membranous duct

of cochlea of cochlea
Connecting duct
of Hensen

FIGURE 13-7 A diagram of the right inner ear showing the cochlea (which houses the auditory
receptor), the semicircular canals, the utricle, and the saccule. (From Geldard, 1972)
396 Chapter 13/Time and Motion

Crista Macula

Membrane with
Cupula imbedded otoliths
Gelatinous
substance

Hair fibers
Hairs of sensory
of sensory cells
cells

Sensory
cells Nerve
fibers
Sensory cell

Nerve fibers

A B

FIGURE 13-8 (A) Diagram of the crista, the receptor found in the ampulla of each semicircular canal.
(B) Diagram of the macula, the receptor found in the utricle and the saccule. (From Geldard, 1972)

when the head is accelerated, bending the hair cells circular canal. The arrangement indicates that mus¬
and generating an electrical response. When the cles that move the eye in a particular plane are
jelly-and-hairs catch up to the rest of the head, controlled by nerve fibers that originate in the one
which would happen if the acceleration ceased and of the semicircular canals that responds to acceler¬
motion became steady, the hairs are no longer bent. ation in that plane. Acceleration in a particular di¬
This means no response would be generated, even rection causes compensatory eye movements in the
though the head could be traveling at thousands of opposite direction. This allows the eyes to remain
kilometers per hour relative to the earth. fixed on an object even though the head is turning
in various directions. The relationship between eye
Neural Responses in the Vestibular movements and vestibular stimulation is shown in
Sense Demonstration Box 13-9.
The hair cells from both the crista and the macula Lowenstein and Sand (1940) performed a clas¬
send their information to the brainstem via the sic study that illustrates the electrophysiology of
eighth cranial nerve. From there most of the nerve the vestibular system. They recorded the electrical
fibers go to the vestibular nuclei (still in the brain¬ activity of single nerve fibers from the crista of a
stem). After this the sensory pathways become ray (a kind of fish) while the entire labyrinth was
complicated and somewhat obscure. There are pro¬ rotated on a turntable. They found that as long as
jections to the cerebellum and to the cortex, but the head was accelerating, the fibers responded.
they are different in different animals (see Correia The fibers increased their firing rate above the rest¬
& Guedry, 1978). It is important to note that most ing rate for acceleration in one direction, and de¬
of the fibers leaving the vestibular nuclei are motor creased it below the resting rate when the
or efferent fibers. One major group of these fibers acceleration was in the opposite direction. Thus, as
forms a pathway to the muscles that move the eyes. in other sensory systems, both excitatory and inhib¬
Szentagothai (1950) discovered that each pair of itory responses to physical stimuli occur. Lowen¬
eye muscles receives fibers from a different semi¬ stein and Sand also showed that the magnitude of
 Motion 397

DEMONSTRATION BOX 13-9. Vestibular Stimulation and Eye Movements

For this demonstration you will need a friend and a that the eyes drift steadily in one direction, and then
little space. Have your friend hold her arms out and snap back and start to drift again. This type of repet¬
spin around (like a whirling ice skater) until she be¬ itive eye movement is called nystagmus. It is a reflex
comes dizzy. This continuous rotation sets up currents movement that is evoked automatically by the vesti¬
in the semicircular canals that trigger the compensa¬ bular stimulation caused by fluid currents in the sem¬
tory eye-movement system. Now stop your friend icircular canals.
from turning and look into her eyes. You will notice

the response (impulses per second in single fibers) Self-Motion

varies directly with the magnitude of the stimulat¬
ing acceleration. So stimulus intensity seems to be Back in the nineteenth century there was a fair¬
encoded in a manner similar to that in other sensory ground ride called the Haunted Swing. In this ride,
systems. There are at least two other types of nerve people entered a boat-shaped enclosure and artifi¬
fibers: one always responds to acceleration (regard¬ cial scenery was slowly swung backward and for¬
less of direction) with an increase in firing rate, and ward outside the windows. This resulted in an
the other only responds with a decrease in firing incredibly strong illusion that the chamber was
rate. rocking, and people felt all of the bodily sensations
The fibers connected to the hair cells of the of real motion, including a feeling of loss of their
macula respond to their stimuli somewhat more postural stability that made them sway, and even
simply. Two types of responses have been de¬ vertigo (Howard, 1982; Wood, 1985). There is an
scribed. The first is an increase in the rate of neural everyday example of this effect. Probably most of
firing when the head is tilted; the second is a rate you have had the experience of sitting in a bus or
increase when the head is returned to its original a train parked next to another vehicle. All at once
position (Wybum et al., 1964). Although there the adjacent vehicle starts to move. However, in¬
have been some studies of cortical responses to ac¬ stead of correctly attributing the movement to the
celeration of the head, little is known in detail vehicle beside you, you have a powerful sensation
about these responses. One fact that has emerged is of yourself in motion. This is a case of induced
that inputs from vestibular, kinesthetic, and visual movement, such as we discussed earlier, however
systems converge in the cortex, so that our sensa¬ it is an induced movement of the self. It is impor¬
tions of “turning” and the like depend in a com¬ tant since it reflects the interchangeability of visual
plex way on all of these inputs (Mergner, and vestibular factors in the perception of body
Anastasopoulos, Becker & Deecke, 1981; Parker, motion.
1980). One striking example of this complex inter¬ If we were not aware of dynamic changes in
action is the phenomenon of motion sickness, the visual image as we move, we would probably
which is often caused by a mismatch between vi¬ bump into things much more often as we walk
sual and vestibular or kinesthetic inputs. A great around, or not notice that our body has swayed or
deal of effort is being put into studying this aspect leaned until we actually incline too far and topple
of human reaction, especially because of its impor¬ over. Our perception of self-motion depends on an
tance in space travel, which involves zero-gravity analysis of the continually changing aspects of the
conditions. retinal image as we move. Consider our most
398 Chapter 13/Time and Motion

typical motion, which is forward in depth. For this which is called the focus of expansion, indicates
kind of motion the visual array in front of us is a the direction of your movement. Although the spe¬
radially expanding pattern in the center of our vi¬ cific patterns shown by the streaming perspective
sual field and a laterally translating pattern in our of targets in the field will vary as you move your
periphery. For example, consider the pattern shown eyes (Andersen, 1986; Cutting, 1986; Regan &
in Figure 13-9A. Here the arrows represent the Beverly, 1982), it is still easy to direct your move¬
flow of the visual array as if you were moving to¬ ments by keeping the door you wish to reach in the
ward the door marked A. Images of objects around center of the outward flow of stimuli.
Door A, that is, those stimuli to its sides or above If we present you with a steadily moving pat¬
or below it, expand radially outward and into the tern that is the equivalent of a natural streaming-
periphery as you move forward. This flow of stim¬ perspective pattern, you will feel as though you are
uli has been called streaming perspective (Gibson, moving. If the pattern is radially expanding, as in
1979). If your path were angled so you were going Figure 13-9, you will feel as though your body is
toward the door marked B, the optical transforma¬ moving forward. If the pattern is moved steadily to
tion pattern would be similar to that in Figure 13-9B. the side you will feel that you are moving (or start¬
In both cases, the center of this outward flow, ing to lean or tilt) sideways, or even rotating if the

FIGURE 13-9 Streaming perspective. A shows the pattern of flow or expansion of stimuli as it would
appear if you were moving toward Door A. The length of the arrows indicates the rate of change or speed
of expansion, with longer arrows meaning faster change. B shows how the pattern of stimulus flow
changes if you were now moving toward Door B, rather than straight down the hallway.
 Motion 399

pattern surrounds you. Such induced motion of the the projected pictures associated with riding down
self is usually called vection (Dichgans & Brandt, a roller coaster, flying in a helicopter, or hurtling
1978). Generally speaking, there is a consensus down a raceway at high speed. Most viewers feel
that the central visual field is more specialized for all the bodily effects normally associated with the
object motion, whereas stimulation of the periph¬ equivalent self-motion, and they are sufficiently in¬
eral visual field is necessary to induce the feeling distinguishable from actual movement that it sug¬
of self-motion. Thus, patterns that extend into the gests that the visual and vestibular inputs must have
periphery tend to produce strong feelings of vection some common neural pathways and centers. This
(e.g., Delorme & Martin, 1986; Held, Dichgans & would be a sensible arrangement, since the vesti¬
Bauer, 1975). If the speed of flow is not too fast, bular system only responds to accelerations or de¬
and the pattern is correct, visually induced self- celerations of body motions. As we saw in the
motion can be experienced even for smaller central previous section, after any prolonged period of
patterns (Andersen & Braunstein, 1985; Stoffregen, constant velocity the vestibular system would cease
1985). Demonstration Box 13-10 shows how you to respond and the only remaining indication of
may experience a form of induced self-motion. movement would be the motion in the visual array.
The relatively greater contribution of the pe¬ To the extent that vestibular and visual infor¬
ripheral retina to vection may explain why the feel¬ mation can each produce similar feelings of self-
ings of self-motion can be so strong when you view motion, it should not be surprising to find that there
motion pictures with an oversized or wraparound are cells in the vestibular nuclei whose rates of fir¬
screen. In fact, a modem version of the “haunted ing are influenced by signals suggesting bodily mo¬
swing” illusion can be found in some fairs and tion, whether such signals come from the vestibular
amusement parks, where viewers are surrounded by organs themselves or from visual motion (e.g.,

DEMONSTRATION BOX 13-10. Induced Self-Motion

For this demonstration you should have two small

light sources (lighted candles will work fine) and a
darkened room. Hold the candles out at arm’s length
and about at eye level, as shown in the figure. Look
straight forward (remove your glasses or squint your
eyes a bit so that you don’t see the surrounding room
too clearly). Now slowly move the candles back to¬
ward the sides of your head (not too close!). As you
do so, you should experience an induced motion of
your body so that you now feel that you are leaning
forward slightly. If you move the candles slowly for¬
ward, you should get the impression that you have
straightened up, or are now leaning backward some¬
what. Next try the same arm movements with your
eyes closed to see that this effect does not occur in
the absence of the visual stimulation. This means that simulated by the movements of the lights, which sug¬
the feeling of body tilt produced in this situation is a gested that at least the upper portion of your body was
form of vection caused by the streaming perspective moving forward.
 400 Chapter 13/Time and Motion

Henn, Young & Finley, 1974; Waespe & Henn, Biological clock A physiological mechanism that
1977). There seems to be a complex interaction be¬ underlies our experience of time.
Biological motion The movement patterns of the
tween the visual and the nonvisual inputs to give
skeletal structure of a human engaged in activities such
us this feeling of self-motion (DiZio & Lackner,
as walking or mnning.
1986; Henn, Cohen & Young, 1980).
Bony labyrinth The structure inside the head that
The perception of motion, and to some extent contains the cochlea and the vestibular organs. It is made
the perception of time, are good examples of a pro¬ of very hard bone.
cess we will call sensory convergence. This is a Change monitoring The theory that the tick rate of
process by which a number of inputs all combine the cognitive clock reflects the number of events, or
to produce a single coherent perception, in which changes, that occur in an interval.
you may not be conscious of all the components Circadian rhythm A rhythmic biological cycle oc¬
curring over an approximately 24-hour period.
that have gone into the computation. Thus, we
Cognitive clock A cognitive mechanism that deter¬
have seen that the apparently visual experience of
mines our experience of time.
motion may contain proprioceptive inputs from eye
Crista Movement-perceptive organ found in the am¬
movements and the vestibular sense, or copies of pulla of the semicircular canal of the ear.
efferent commands issued to the eye muscles, even Cupola Jellylike material in which the hairs of the
though your conscious impression remains strictly crista are embedded.
visual. The mechanism of sensory convergence Direction constancy The stability of an object’s per¬
may be quite simple, with various peripheral ceived direction despite changes in eye position.
sources of information converging on a single Efferent fibers Neural fibers that carry outgoing
group of neurons, or it may involve higher-level, commands from the brain to muscles or other action sys¬
tems.
central or cognitive integration, as we saw in our
Entrainment The process by which the biological
consideration of the factors influencing our percep¬
clock is synchronized to physical time cycles.
tion of the passage of time. The final percept, as
Event processing See Change monitoring.
experienced in our consciousness, may be an amal¬
-'''Eye-head system A movement-perception system
gam of many sources, though to us it seems quite
that monitors and differentiates eye- or head-generated
simple and pure. from object-generated movement of the visual image on
the retina.
Filled duration illusion An interval filled wth stim¬
ulus events is perceived as being longer than an identical
interval without stimulus events.
GLOSSARY
Flow The perception of time passing.
Focus of expansion A point in space around which
The following definitions are specific to this book.
all other stimuli seem to expand. If you are moving for¬
Acceleration Change in rate of motion. ward, it represents the place toward which you are mov¬
Ampulla Swelling at the base of a semicircular canal ing.
cpntaining the crista. ^Tmage-retina system A movement-perception sys¬
4pparent movement Perceived movement of spa¬ tem that detects movement within the retinal image.
tially separated static stimuli flashed successively ^t ap¬ Induced motion An illusion of movement of a sta¬
propriate interstimulus intervals. tionary object created by movement of the background or
^/Aubert-Fleischl effect A moving target being tracked surround context.
with the eyes seems to move more slowly than when a Inflow theory The suggestion that motion is detected
stationary background behind the moving target is fix¬ via feedback information from the six extraocular mus¬
ated. cles controlling eye movement.
Autokinetic effect The illusion of movement of a Interstimulus interval The time span between the
stationary point of light viewed in an otherwise totally end of one stimulus presentation and beginning of the
dark field. next.
 Glossary 401

Long-range process The perceptual system respon¬ Semicircular canals Vestibular organs contained in
sible for apparent motion when stimuli are separated by the bony labyrinth, next to the cochlea of the ear.
more than 100 msec in time and more than 15 min of Sensory convergence The notion that several differ¬
visual angle in space. ent sensory inputs, from different modalities and sources,
Macula The receptor organ of the utricle and sac¬ combine to form an apparently simple perceptual expe¬
cule. It is responsive to linear acceleration. rience.
Motion aftereffect An illusion of movement that oc¬ Short-range process The perceptual system respon¬
curs in the direction opposite to a moving stimulus that sible for apparent movement when stimuli are separated
an observer has viewed for an extended time. by less than 100 msec in time and less than 15 min of
Now The interval of time that we interpret as “the visual angle in space.
present.” Smooth-pursuit movement The continuous eye
Nystagmus A reflexive, jerky eye movement caused movement involved in following a smoothly and steadily
by stimulation of the vestibular organs (cristas) in the moving object.
semicircular canals. Statocysts In primitive invertebrates, motion-
Object-relative change The movement of objects in sensitive cavities lined with hair cells.
relation to one another. Statolith A tiny stonelike body resting on the hairs
Otocysts In primitive vertebrates, fluid-filled cavities of statocysts and causing them to bend in response to the
functioning to maintain balance and attitude to gravity. motion or change of position of an animal.
See Statocysts. Storage size model A model that contends that time
Otoliths Bony bodies in otocysts. is measured internally by the number of items processed
and stored in memory during an interval.
Outflow theory The suggestion that brain commands
initiating eye movements enable the eye-head movement Streaming perspective The optical flow of stimuli
system to differentiate object-generated from observer¬ as we move through space, centered around the direction
generated movement of the retinal image. of movement.
Pacemaker The mechanism that sets the speed of Subject-relative change The movement of objects
the biological clock. relative to the body.
Perceptual moment The hypothetical basic psycho¬ Suprachiasmatic nucleus (SCN) A center in the hy¬
logical unit of time, about 100 msec. pothalamus that is believed to be responsible for circa¬
dian rhythms.
Position constancy Stable perceived position of ob¬
jects despite body, eye, or head movements. Temporal processing model The theory that your
experience of the passage of time depends on the amount
Processing effort model A theory that time is mea¬
of attention that you direct toward monitoring time.
sured internally by the amount of cognitive work or pro¬
cessing that an individual does during an interval. Utricle A vestibular organ contained in the bony lab¬
yrinth.
Real movement Physical movement of a stimulus.
Vection An illusion of induced motion, in which you
Reflex pursuit movements Smooth eye movements,
experience your body moving through space or tilting be¬
under vestibular control, that are made to keep the image
cause of changes in the optical flow.
of a target on the fovea despite head movements.
Vestibular nuclei In the brainstem, way stations
Restricted environmental stimulation technique A
along the route of nerve fibers from the crista and macula
procedure in which a subject remains in a dark, sound-
to the cerebellum and cortex.
deadened room for specified intervals of time.
Vestibular system The system that monitors the
Saccule A vestibular organ contained in the bony
body’s movement and orientation in space. Its receptors
labyrinth.
are located in the bony labyrinth.
Selective adaptation A psychophysical technique in
Visual context Visual stimuli that surround or ac¬
which motion-specific cells in the visual system are fa¬
company other stimuli.
tigued, causing reduced sensitivity to other visual stimuli
moving in the same direction and at the same speed as Voluntary pursuit movements The eye movements
by which you voluntarily track a moving target.
the previously exposed adapting pattern.
Zeitgeber The stimulus used to calibrate, or entrain,
-'Self-motion The perception that the body is moving
the biological clock.
through space.
' CHAPTER
14
The
Constancies
•

THE TASK OF PERCEPTION

PERCEPTUAL CONSTANCIES
SIZE CONSTANCY
Size Constancy and Distance Cues
Direct and Constructive Aspects of Size
Constancy
Size Constancy and Illusion
SHAPE CONSTANCY
LIGHTNESS OR WHITENESS CONSTANCY
COLOR OR HUE CONSTANCY
OTHER CONSTANCIES

403
404 Chapter 14/The Constancies

W hen you look around at the world, your per¬

ception is of objects and surfaces. Each of
these has a relatively enduring set of properties that
are apparent to you, such as size, shape, and color.
Now consider a very simple problem. Suppose you
are presented with two rectangles made of card¬
board and are asked to say which one is larger. If
the difference in their physical size is not too small,
you would probably have little trouble giving the
correct answer. Now consider a second problem.
How did you reach your conclusion? Many people
would probably give an answer like this: “The
larger rectangle produces a larger image in my
eye.” However, this answer would be quite wrong,
as can be seen from Figure 14-1. There we have
three different rectangles; the image of each is the
same size on your retina, yet each is a representa¬
tion of a diffent-sized cardboard rectangle “out
there” in “the real world.” The smallest of these
pictured objects would be only a few centimeters
on each side “out there,” whereas the largest
would be over a meter in width and length.
To paraphrase Albert Einstein, “We like to
keep things simple, but not too simple.” Up to
now we have been treating perception in a fairly
simple fashion, in that we have adopted the general
position that for every distinct kind of perceptual
experience—color, brightness, distance, size, or
the like—there is some unique stimulus or type of
stimulus information that affects our sense organs.
Although we may not fully understand what that
stimulus is, we believe that we could potentially
discover it. We also have been assuming that our
FIGURE 14-1 Three rectangles whose retinal
perception is related directly to the stimulus infor¬
images are all the same size, although each
mation that reaches us, with, perhaps, room for appears to be different in actual size from the
some minor adjustments caused by interactions others.
with or limitations imposed by the nature of our
sensory receptors, or the neural processes used to
encode the information they receive. However, this THE TASK OF PERCEPTION
assumption that a given identifiable stimulus di¬
mension (such as the size of a retinal image) is re¬ Before we go any further in our discussion we must
lated uniquely to a particular perceptual dimension make some distinctions. To begin with, stimuli can
(such as the apparent size of the object we are be divided into two general classes: a distal stim¬
looking at) is “too simple.” ulus is an actual object or event “out there” in the
 Perceptual Constancies 405

world; a proximal stimulus is the information our unconscious inference, which is to say that it is
sensory receptors receive about that object. For ex¬ derived or pieced together from a number of differ¬
ample, a tree falling in the forest would be a distal ent sources of information. This information may
stimulus, whereas the sound of its fall at our ears come from other sensory inputs, such as feedback
and the changing light reflected from it to our eyes from eye movements, from prior experience that
would be proximal stimuli. The task of perception has given us a concept as to the usual size or shape
is to characterize accurately the distal stimulus, of an object, or even from expectations and guesses
since that corresponds to an actual object or event as to the nature of the distal stimulus.
in the real world (Brunswick, 1956; Coren, 1984a). It is important to notice that both direct per¬
The problems associated with doing this come from ception and constructive theories involve some
the fact that the proximal stimulus is, by itself, not form of multidimensional interactions. The major
always an ideal source of information about the difference is that in direct perception theories the
distal stimulus, as we saw when we judged the size interactions do not involve any computation or in¬
of the real-world objects depicted in Figure 14-1. ference and all occur within the same modality as
Other factors such as the context in which the distal the percept, whereas in constructive theories the in¬
stimulus occurs, must be taken into account. In this teractions may involve other sensory modalities and
sense, our perception of objects may be viewed as some fairly complex deductive or inferential pro¬
being a form of multidimensional interaction (Ut- cedures. It seems likely that both a direct and a
tal, 1981). constructive process are ultimately involved in
There is a bit of controversy about the nature achieving the task of perception, which is to ascer¬
of multidimensional interactions. Consider how we tain the actual properties of objects in the environ¬
go about judging visual size. One approach is ment (e.g., Gogel & DaSilva, 1987a).
called direct perception, a notion that began as
early as the writings of Hering (1878/1964) and
that has been well elaborated since then (Gibson, PERCEPTUAL CONSTANCIES
1966, 1979; Michaels & Carello, 1981). In direct
perception theories, the perception of visual size is If the only information we had about nature were the
completely based on information from the proximal proximal stimulus, such as the retinal image of an ob¬
stimulus, here the distribution of light on the retina. ject, our world would be as chaotic as the Wonder¬
Certain features in the proximal stimulus determine land Alice found at the bottom of the rabbit hole.
perceived size. These features may be higher-level Since the retinal image is larger the closer an object
and fairly complex, such as the comparison of the is, an approaching friend would appear to grow larger
size of the image of the object to the sizes of im¬ as she grew nearer. A piece of white paper would ap¬
ages of other objects that form its visual context. pear black when viewed in the moonlight, since the
Presumably, if we look carefully enough, we amount of light in the retinal image is no greater un¬
should be able to isolate a set of purely visual stim¬ der these conditions than in the image of a piece of
ulus factors that determine our perception of the coal viewed in normal room light. This same piece of
size of an object. paper would appear to change shape continually—
The alternative approach is of equal vintage. the retinal image changing from rectangular to trape¬
It can be traced back to the works of Helmholtz zoidal as the paper’s angle of tilt was varied—and
(1909/1962), has many active supporters (e.g., Ep¬ its color would appear to be blue under fluorescent
stein, 1973; Rock, 1983; Uttal, 1981), and refers lighting and yellow under incandescent light¬
to a constructive theory of perception. In con¬ ing. Fortunately our perception of objects is much
structive theories, perception arises as a form of more constant than would be expected if the only
406 Chapter 14/The Constancies

information available were the proximal stimulus. process by which the changes in the proximal stim¬
Your friend remains the same size but changes her ulus are encoded for processing. There is no need
distance from you. The piece of paper remains a for the individual to be consciously aware of this
white rectangle, although you might sense the fact registration process. The second phase involves ap¬
that the color or intensity of the light falling on it, or prehension, the actual subjective experience. This
its angle of tilt relative to you, has changed. This il¬ is the conscious component that is available for you
lustrates a very basic aspect of perception, which is to describe. Normally, registration is oriented to¬
that the properties of objects tend to remain constant ward a focal stimulus, which is simply the object
in consciousness although our perception of the you are paying attention to. In addition, you also
viewing conditions may change. The fact that our register a set of stimuli that are nearby, or occur¬
perception of the world does not vary as much as fluc¬ ring at the same time, which form the context
tuations in the proximal stimulus would lead us to ex¬ stimuli. During apprehension you become aware of
pect is what we mean by the perceptual constancies. two classes of properties, the object properties of
Although there are many varieties of percep¬ the focal stimulus, which tend to remain constant,
tual constancies, they fall into three general and the situation properties, which indicate more
classes. The first pertains to object properties, such changeable aspects of the environment, such as your
as an object’s size and shape; the second to certain position relative to the focal object or the amount
qualities, such as the whiteness or color of sur¬ or color of the available light, and which are de¬
faces; and the third to the locations of objects in rived from cues found in the context. The way
space relative to the observer. these categories interact is shown in Table 14-1,
Each constancy has two major perceptual which describes these variables for a number of
phases. The first phase involves registration, the constancies. If all this appears a bit complicated in

Table 14-1. The Relationships between the Registered and Apprehended Variables in Some of the More
Common Perceptual Constancies.

Registered Stimulus Apprehended Stimulus

(may be unconscious) (conscious)

Focal
Constancy Stimulus Context Constant Changes

Size constancy retinal image size distance cues object size object distance
Shape constancy retinal image shape orientation cues object shape object orientation
Lightness intensity of light on illumination cues surface whiteness apparent
constancy the retina intensity illumination
Color constancy color of retinal image illumination cues surface color apparent
illumination color
Position constancy retinal location of sensed head or object position in head or eye position
image eye position space
Loudness intensity of sound at distance cues loudness of distance from sound
constancy the ear sound
Odor constancy amount of odorant in proprioception intensity of smell strength of sniff
the nose from sniff
 Size Constancy 407

theory, in practice it is really quite straightforward. walk down the street. Suppose we perceived size
Let us look at some of the more common constan¬ only in terms of retinal image size. If such were
cies to see how they work. the case, a man who is 180 cm (about 6 ft) tall
would look like a small child when he was at a
reasonable distance but would appear to “grow” as
SIZE CONSTANCY he approached us. Of course this does not happen.
Instead, we perceive him to be “man size” regard¬
Before you can understand size constancy, you must less of his distance from us. This stability of per¬
understand what happens to the retinal images of ob¬ ceived size despite changes in objective distance
jects as our distance from them varies. As the dis¬ and retinal image size is called size constancy. In
tance between the eye and the object grows larger, the essence, it involves assigning a constant size to an
size of the retinal image grows smaller. This relation¬ object in consciousness, no matter what its distance
ship is shown in Figure 14-2. As you probably recall or retinal size may be.
from Chapter 4, retinal image size is usually ex¬
pressed and measured in terms of visual angle. The
Size Constancy and Distance Cues
visual angle for (Stimulus 1) is aj and for S2 is a2.
Like other angles, these are expressed in degrees, Size constancy, we have seen, is a process by
minutes, and seconds of arc. As an example, the im¬ which we “take into account” the apparent dis¬
age size of a quarter (25-cent piece) held at arm’s tance of an object, in order to “adjust” the per¬
length is about 2 deg, whereas at a distance of about ceived size to more accurately represent the actual
80 m the quarter would have a visual angle of 1 min of physical size of the object. For example, in Figure
arc. At a distance of 5 km it would have the tiny ret¬ 14-3A we see three men standing in a courtyard.
inal image size of only 1 sec of arc. Thus, as its dis¬ All appear to be about the same size, even though
tance from an observer increases its retinal image size the retinal image size of the apparently most dis¬
decreases. tant individual is only about one-third that of the
Consider what happens as you watch someone apparently nearest one. In other words, we estimate

FIGURE 14-2 The visual angle. Although the physical size (S) of the object does not change, changes in
distance (D) will result in changes in the size of the visual angle (a). Here a, (the image of object S, close
to the eye) is larger than a2 (the image of object S2 farther from the eye).
408 Chapter 14/The Constancies

FIGURE 14-3 (A) Three men, whose retinal images grow smaller as they appear to be more distant;
however, because of size constancy they appear to be the same size. (B) Here, when the images remain
the same size, the more distant man appears to be larger.

distance and size together and adjust our perception tributes to the maintenance of size constancy, be¬
of size in accordance with our distance judgment, ginning with the early work by Hoi way and Boring
perceptually “enlarging” more distant objects. (1941). Simply stated, the more cues, the better the
You can see how this constancy scaling works by size constancy (e.g., Chevrier & Delorme, 1983).
looking at Figure 14-3B, where all the images of For example, Harvey and Leibowitz (1967) asked
men are exactly the same size. Here the constancy observers to choose a size match for a standard tar¬
scaling correction makes the apparently farthest get that was placed at various distances. There
appear much larger. This suggests that changes in were two viewing conditions. One corresponded to
apparent distance should alter our perception of natural situations with many distance cues, and the
apparent size, which is a reasonable suggestion, as second involved the removal of the surrounding
can be seen from Demonstration Box 14-1. context by having the observers view the standard
How does the perceptual system take depth target through a small opening that blocked the
into account? The simplest answer is that we utilize view of everything but the target to be observed.
cues to the distance of the target that are available Under natural viewing conditions, size matches
as part of the visual context. This would suggest conformed very well to the predictions based on the
that when we increase the number of distance cues efficient operation of the size constancy mecha¬
constancy should be better, whereas reducing the nism. This was the case for all the observer-target
number of cues should reduce our tendency toward distances used in this experiment. However, when
constancy. Many experiments have demonstrated the context was removed, size matches conformed
how the availability of cues in the visual array con¬ to constancy predictions only at viewing distances
 Size Constancy 409

DEMONSTRATION BOX 14-1. Size Constancy and Apparent Distance

An easy way to demonstrate how apparent distance Again you will see the dark square projected against
affects apparent size requires that you carefully fixate the wall, but now it will appear to be much larger in
the point marked X in the accompanying white square size. Because of the nature of an afterimage, its visual
while holding the book under a strong light. After a angle does not change. But as you project it against
minute or so, you will form an afterimage (see Chap¬ surfaces at varying distances from you, its apparent
ter 5) of the square. If you now transfer your gaze to size changes. It appears to be larger when it is pro¬
a blank piece of paper on your desk, you will see a jected on a distant surface. This is an example of how
ghostly dark square floating there. This is the after¬ size and distance perception interact by means of the
image, which will appear to be several centimeters size constancy mechanism. The quantitative expres¬
long on each side. Now shift your gaze so that you sion of the relationship is often called Emmert’s law.
are looking at a more distant, light-colored wall.

up to about 120 cm. After that, size matches began to produce stable binocular foveal fixation. When
to deviate from the predictions based on size con¬ targets are relatively close to you, information
stancy. This experiment shows the role of distance about their distance can be gotten from feedback
context in the maintenance of size perception. from the accommodation and convergence actions
Why was there still constancy at the closer of the eyes. This suggests that these physiological
distances even though the depth cues were re¬ cues to distance have been incorporated into the
moved? Probably because the manipulation used multidimensional mix of information resulting in
removed only visible distance cues. As we dis¬ the size constancy.
cussed in Chapters 3 and 10, when we fixate ob¬ One important thing to notice in this example
jects at different distances, the lens of the eye is that the source of the distance cues is not really
changes shape to accommodate for changes in fix¬ important. The cues do not have to come from the
ation distance. Simultaneously, the eyes either con¬ visual array per se, but can come from other
verge for near objects or diverge for distant objects sources. We can show this by demonstrating that
410 Chapter 14/The Constancies

in the absence of any other information about the some way, this is one way in which structures
distance of the target (such as the pictorial cues to within the visual system “take distance into ac¬
distance catalogued in Chapter 10), changes in ac¬ count” in computing the size constancy correction.
commodation and convergence result in changes in Actually, any source of distance information
perceived size. This was confirmed by Leibowitz can be used to obtain accurate size constancy.
and Moore (1966). Their observers viewed a white We have seen that visual depth cues trigger size
triangle in an otherwise completely dark field. They constancy, as do accommodation and conver¬
matched the size of this stimulus by making size gence. Adding binocular disparity, the depth cue
adjustments in a similar triangle. Accommodation involved in stereopsis (see Chapter 10), can also
and convergence were varied by inserting prisms strengthen size constancy, as shown in Demon¬
and lenses before the eyes. This forced the subjects stration Box 14-2.
to adjust their convergence and accommodation to
closer or farther distances when viewing the target,
Direct and Constructive Aspects of
although the retinal size remained constant. If these
Size Constancy
depth cues help to stabilize the perception of size,
this experimental manipulation should have re¬ Earlier in this chapter we contrasted the direct and the
sulted in changes in perceived size. constructive theories of perception. Direct theories of
The prediction of these investigators was con¬ perception are based on the presumption that all of the
firmed. The size of the target judged to be equal to information we need for such things as constancy can
that of the standard triangle increased as accom¬ be found in the proximal stimulus. Geometric regu¬
modation and convergence changes were manipu¬ larities, such as the convergence of parallel lines with
lated to indicate increasing target distance (much as increasing distance (linear perspective) or the in¬
the size of the man increased with increasing ap¬ creasing textural density of more distant fields of ele¬
parent distance in Figure 14-3B). Hence, it seems ments, serve as reliable cues for distance. Such cues
that the state of the oculomotor (eye muscle) sys¬ may be sufficient to maintain size constancy (Gibson,
tem, which varies with the distance of the distal 1950,1979; Michaels &Carello, 1981).
stimulus, conveys information that helps to stabi¬ For example, we can imagine two objects of the
lize size perception. Since the perception of size same size sitting on a surface at different distances
seems to be linked to the perception of distance in from the observer. As you know from Chapter 10, the

DEMONSTRATION BOX 14-2. Additional Depth Cues Strengthen Size Constancy

Hold out both of your hands with their backs toward side until its image appears to be just next to the near
you. One hand should be relatively near you (about one. Now when you compare the size of the two
20 cm or 8 in. should do), and the other should be hands it should be clear to you that the more distant
out at arm’s length. At first glance, both hands should one appears smaller than the near one, showing a
appear to be about the same size. Now, remove the clear weakening of size constancy. You can restore
binocular disparity depth cue by closing one eye. the size constancy by adding additional depth cues—
Keeping your hands at these different distances, and open both eyes and swing your head from side to
your head very steady (to prevent motion parallax as side, and your hands will again appear to be the same
a further depth cue), move your distant hand to the size.
 Size Constancy 411

fact that textured surfaces show denser gradations of different, as for object D. Different sizes are associ¬
coarseness as they recede into the distance is a very ated with different ratios between the textures and the
powerful distance cue. If two objects at different dis¬ objects themselves. Thus A, B, and D appear to be the
tances appear to cover the same number of textured same height (about 1 texture element), whereas C
elements (in other words, their relationship to the (whose height in the picture is actually smaller than
textured surface remains constant), they will remain A) appears to be a much taller object since it is about 4
perceptually the same size. This principle was first texture elements high. Thus, according to direct per¬
described in detail in 1604 by the artist Jan de Vries. ception, the observer could extract the physical size
In Figure 14-4 we have modified one of the drawings of the objects (which is simply our size constancy
he used. You will notice that here we have two rec¬ correction) by comparing the relative size of the ob¬
tangular structures (A and B) that appear to be about jects to the size of the surrounding texture elements.
the same size, but that appear to vary in distance. De¬ Constructive theories of perception allow for
spite the fact that their retinal image size differs, no¬ sources of information other than the proximal
tice that each rectangular object is 3 texture elements stimulus to shape the final percept. We have al¬
long and 3 wide (here the texture elements are the ready seen how feedback from the accommodative
square “tiles”), regardless of its location. Notice that and convergence movements of the eye might serve
this relationship holds even when the viewing angle is such a function. More important than this type of

RC ,

....-

FIGURE 14-4 Rectangles A, B, and D all appear to be the same size, since each covers the same number
of texture elements. (Based on de Vries, 1604/1968)
412 Chapter 14/The Constancies

information for constructive theories, however, is 1957), presumably because adults have had more
information generated by cognitive judgments and experience with a greater variety of environmental
operations or from learned factors and expectations viewing conditions.
(e.g., Epstein, 1973; Rock, 1983; Uttal, 1981). In A number of other considerations seem to sup¬
this type of theorizing it is probably inappropriate port some constructive theory factors in size con¬
to speak of “cues.” Rather, any factor that results stancy. Some evidence suggests that you do not
in a change in one aspect of perception (here the actually have to perceive the distance. Simply know¬
distance of the object) can bring about a change in ing the distance, such as being told how far an object
another aspect of perception (here the size of the is from you, seems to be enough to elicit the size con¬
object), regardless of the source of that information stancy adjustment (Pasnak, Tyer & Allen, 1985).
(see Hochberg, 1974). Let us see how some non¬ Furthermore, if your attention is directed elsewhere,
visual sources of information can affect our percep¬ so that the full measure of cognitive processing is not
tion of size. available, your size constancy processing begins to
Our experience with the world has already break down (Epstein & Broota, 1986). Demonstra¬
provided us with much information that assists us tion Box 14-3 allows you to see how attention inter¬
in maintaining size constancy. For example, we acts with size constancy.
learn that particular objects have typical visual Neither direct nor constructive theories of per¬
sizes. This familiar size information can be used in ception seem adequate to explain all aspects of size
the absence of any other information to judge the constancy (nor any constancy, for that matter). It is
size of an object once it has been identified, or to more likely that both processes combine to produce
judge the distance of the object based on its retinal the final perception of size and distance (e.g.. Go-
and familiar size, similarly to the situation in Dem¬ gel & DaSilva, 1987a).
onstration Box 10-1. Simply put, we have an ex¬
pectation that familiar objects, such as playing
Size Constancy and Illusion
cards or coins, have typical sizes. If we were pre¬
sented with a very tiny image of a playing card, we Size constancy provides stability in our perception
would maintain our size constancy by seeing this of the world by giving our conception of particular
as a normal-sized playing card viewed from a long objects a consistent set of properties despite varia¬
distance rather than a playing card that is much tions in the retinal image size. This is usually use¬
smaller than usual (cf. Gogel & DaSilva, 1987b; ful, but under special circumstances it can lead to
Higashiyama, 1985; Ono, 1969). errors or illusions. To see how this comes about,
Our experience and expectations also explain consider a variation of Figure 14-3: in Figure
an interesting breakdown of size constancy that oc¬ 14-5A we see two logs lying in the middle of a
curs at very large target-observer distances. You road. Although they have been drawn to be two
need only to climb to the top of a tall building and different sizes on the paper, the distance cues in the
note that people below appear to be tiny dolls and context (perspective, texture, and others) indicate
cars appear to be little toys. It seems likely that the that the upper log is more distant. Because of size
unusual viewing conditions and exceptional dis¬ constancy we see it as being the same-sized object
tances are so unfamiliar that they simply do not as the closer log. In Figure 14-5B we have two
trigger the size constancy mechanism in this in¬ logs that appear to be different in size, with the
stance (Day, Stuart & Dickinson, 1980). This is more distant one seemingly longer than the closer
supported by the fact that the range of distances one, although they have been drawn to be exactly
over which size constancy works is greater for the same physical size. Once again, this merely
adults than for children (e.g., Zeigler & Leibowitz, represents the operation of size constancy. To the
 Size Constancy 413

DEMONSTRATION BOX 14-3. Attention and Size Constancy

Begin with your right hand in front of you at arm’s move your eyes from your finger, but try also to pay
length. Now look directly at your hand and move it some attention to the image of your hand as you move
toward your face and away again several times. Al¬ it closer and farther from your face. Under these con¬
though the retinal image size is changing, your hand ditions, where your attention is divided and pulled
appears to be the same size because of the operation away from simply viewing the target, your size con¬
of size constancy. stancy should break down. Now the hand seems
Next hold the index finger of your other hand up in smaller when farther and larger when nearer to you,
front of your face (at about 20 cm) as shown in the demonstrating that when attention is somewhat di¬
figure. Look steadily at the finger. Now, while main¬ verted we are more apt to experience the proximal
taining fixation on your fingertip, bring your right (retinal) stimulus changes rather than apply the con¬
hand toward and away from your face. Try not to stancy correction.

extent that the picture mimics conditions in the real ure 14-5B, the two lines are physically equal in
world, the upper log appears to be more distant. In length, this perceptual difference is a visual-
the real world, it could cause the same-sized retin¬ geometric illusion (usually called the Ponzo illu¬
al image as the lower log only under conditions sion. Actually, this is an illusion only in the sense
where it was physically longer. Since in the pic¬ that no context for depth or distance has been
ture the logs have been drawn the same size, the drawn into the figure. It is caused by the fact that
constancy scaling mechanism has correctly adjusted there are registered cues for distance here (the con¬
our perception. verging perspective lines) that are sufficiently
The perceptual problem arises with Figure strong to evoke the size constancy mechanism, yet
14-5C, where we see two converging lines and two are not sufficiently strong to evoke the conscious
horizontal lines. Notice that the upper line appears apprehension of distance (Coren & Girgus, 1977;
to be slightly longer. Because, like the logs in Fig¬ Gillam, 1980; Gregory, 1966). In some instances
414 Chapter 14/The Constancies

FIGURE 14-5 (A) The two logs lying on the road appear to be at different distances. Therefore their
apparent size is the same despite the fact that the apparently more distant log is physically smaller (on
the page) than the other one. (B) The logs are identical in size; however, the one that appears to be more
distant looks larger. The application of size constancy can lead to illusions of size, as seen in (C), which is
the Ponzo illusion and is similar to (B) except that the context indicating distance and depth has been
greatly reduced. (Based on Coren &. Girgus, 1978)
 Size Constancy 415

there is a very fine distinction between a picture application of size constancy is inappropriate in
and a simple array of lines that produces a visual this situation.
illusion. It is possible to imagine a child with poor Several other illusion distortions also seem to
drawing ability producing a figure like 14-5C when result from observers responding to implied depth
asked to draw 14-5B, where the converging lines cues in the configuration (Coren & Girgus, 1978;
were really meant to be depth cues! Ward, Porac, Coren & Girgus, 1977). Perhaps the
Cues that are registered in sufficient strength to most spectacular of these is the moon illusion, where
inappropriately elicit size constancy are often quite the moon on the horizon appears to be larger than the
subtle. Consider Figure 14-6, where the vertical line moon when it is high in the sky, despite the fact that it
marked A appears to be shorter than the vertical line is optically always the same-sized disk. One explana¬
marked B although they are equal in length. This dis¬ tion of this phenomenon is based on size constancy
tortion, which is called the Mueller-Lyer illusion, (Kaufman & Rock, 1962). The notion is that the
has also been explained in terms of size constancy moon is seen as if it were on the ‘ ‘ surface ’ ’ of the sky.
(Coren & Girgus, 1978; Eijkman, Jongsma & Vin¬ If the sky were a uniform hemisphere there would be
cent, 1981; Gregory, 1966; Madden & Burt, 1981). no illusion. However, the sky actually appears to be a
For example, the wings turned toward the vertical flattened bowl, with the horizon farther away than the
line might mimic the perspective cues of the outside zenith, as shown in Figure 14-7. This is because
of a building (shown in dark lines in C), whereas the when you look toward the horizon you have many
wings turned away from the vertical line might mimic depth cues for distance (texture gradients, familiar
an interior comer of a room (shown in D). Since the objects, etc.), whereas when you look up there is only
closest point in the array is the plane of the paper, it is sky. This means that the moon appears farther away
easy to see that if the wings imply increases or de¬ when on the horizon than when at zenith and, hence,
creases in distance away from that plane, the vertical by the action of size constancy, its apparent size is ap¬
shaft in B is more distant than that pictured in A. prehended as being larger. Although size constancy
Hence, the operation of size constancy would enlarge does not account for all the effects in the moon illu¬
the apparent length of B relative to A. Notice again sion (see Hershenson, 1982; McCready, 1986; Reed,
that this is an illusion of size only in the sense that 1984), nor totally for all other illusions of size (see
no depth or distance was intended; therefore, the Coren & Girgus, 1978), it certainly seems to play a

FIGURE 14-6 (A) The underestimated segment of the Mueller-Lyer illusion. (B) The overestimated
segment. (C and D) The corresponding perspective configurations.
416 Chapter 14/The Constancies

FIGURE 14-7 The fact that the moon is registered to be more distant when it is at the horizon than
when it is overhead helps to explain the moon illusion.

role in producing some interesting illusions of size, as stant while the environment changes or our own
well as correcting for normal variations in size of the relative condition changes. Thus, Epstein and Park
retinal image that occur with the changing distance of (1964) have defined shape constancy as the rela¬
viewed objects. tive constancy of the perceived shape of an object
despite variations in its orientation. To see why
such a form of constancy correction is necessary,
SHAPE CONSTANCY consider what happens when you view a rectangu¬
lar card from different angles, as in Figure 14-8.
We have spent a good deal of time describing size As we increase the tilt of the card, the retinal image
constancy because the other constancies have much becomes more like a trapezoid with the formerly
in common with it. Each involves the registration vertical sides tapering outward. Yet the object still
of either environmental cues or cues about our re¬ “looks” rectangular. The same happens when we
lationship to an object in the environment, and the swing a door outward. The large changes in the
apprehension of the object properties as being con¬ shape of retinal image are ignored, and the door
 Shape Constancy 417

What the situation Image seen from Shape of the

is from the side the front perceived surface

FIGURE 14-8 Shape constancy. Changes in the tilt or slant of objects will cause changes in the shape of
the retinal image; however, perceived shape remains constant. (Based on Lindsay & Norman, 1977)

still appears to be rectangular. In achieving shape available, observers tend to perceive the shape and
constancy, the perceptual system appears to com¬ slant of objects with remarkable accuracy (Lappin
pensate for changes in slant in a way analogous to & Preble, 1975). As in the size constancy situation,
the compensation for distance changes in size con¬ when observers are prevented from using contex¬
stancy. Follow the instructions in Demonstration tual or depth information that would indicate the
Box 14-4 to see the operation of shape constancy degree of slant, the operation of shape constancy
for yourself. becomes less effective and the percept comes to
There is an intimate relationship between size reflect the retinal situation rather than the actual ob¬
and shape constancy—both are related to distance ject (Leibowitz & Bourne, 1956; Leibowitz, Wil¬
perception. However, for shape constancy the dis¬ cox & Post, 1978).
tance information pertains to the relative distance Observers use several strategies to assist in the
of different parts of the object from the observer, judgment of orientation and to supplement contex¬
in other words, its orientation in space or its slant. tual information. For example, in Figure 14-9A we
In unrestricted viewing, with many contextual cues have a shape (the letter E) that has a common or a
418 Chapter 14/The Constancies

DEMONSTRATION BOX 14-4. Shape Constancy

Look at the box in the accompanying figure. Most that its real physical shape is a parallelogram, not a
people believe that a dime will fit inside the top of square.
this box. Try placing a dime (flat on one face) into
the box. Does it fit?
The reason that the top surface of this box appeared
to be large enough to accommodate the dime is
because you made a shape (and size) constancy cor¬
rection. The shape constancy correction changed the
appearance of the top of the box into a square; the
size constancy correction made the sides of the box
top appear equal. Look back at the box and notice

normative orientation based on our experiences

with this form. We can use such normative infor¬
mation to infer whether the shape is upright, ro¬
tated, or tilted (e.g., Braine, Plastow & Greene,
1987; Rock, 1973). In the absence of such prior
experience, we make certain presumptions about
shapes. For example, we presume that the longest
dimension (sometimes called the principle axis)
represents the upright dimension. Thus, we are apt
A to consider the rectangle shown in Figure 14-9C as
more tilted than that shown in Figure 14-9B (cf.
Humphreys, 1983, 1984).
The way we approach the task of looking at
objects may also affect the degree of constancy we
obtain. For example, Carlson (1977) and Kaess
(1980) asked observers either to report the sizes
and shapes of the objects they were viewing (the
objective instruction) or to report the sizes and
shapes of their retinal images (the projective in¬
B C struction). The objective instruction is closest to
normal viewing, where the perceptual task is to de¬
FIGURE 14-9 (A) We can tell how some shapes
rive what is “out there.” The projective instruction
are tilted in space because of our familiarity with
is similar to what an artist must do in trying to
them. For less common shapes we assume that
the longest axis indicates “upright”; hence, we will translate the scene being viewed onto a canvas con¬
see (C) as more tilted than (B), although they are sisting of sizes and shapes of colored regions that
both tilted the same amount from the horizontal. will represent objects in space when viewed by an
 Lightness or Whiteness Constancy 419

observer. It has been shown many times (e.g., Gil- Thus, if a surface that reflects 90 percent (or 0.9)
insky, 1955; Leibowitz & Harvey, 1969) that when of the incident light receives that light from a
observers are asked to adopt the projective viewing source with a physical intensity of 100 units, we
set they show less constancy, although it appears would calculate a “light at eye” value of 90 units.
that they cannot completely turn off the constancy Although not all of the light reaching the eye from
correction. Observers continue to make constancy a surface actually reaches the retina itself, since
corrections although the judgments become more some is absorbed or reflected by the cornea, lens,
like the retinal image (Lappin & Preble, 1975; and the like, the proportion of the “light at eye”
Lichte & Borresen, 1967). There are other viewing that does reach the retina is a constant. Thus, ret¬
factors that affect the degree of constancy obtained. inal illuminance does depend directly on the
For instance, Epstein, Hatfield, and Muise (1977) amount of “light at eye.”
showed that, much like any cognitive task, the Your impression of how light an object is,
more processing time available, the more shape however, is relatively independent of the amount of
constancy is found. Epstein and Lovitts (1985) light reaching your eye and, thus, of the retinal il¬
showed that, much like size constancy, the more luminance. A piece of white paper appears to be
attention you pay to the task, the better the shape approximately the same shade of white whether it
constancy correction. is viewed in dim light or bright light. A piece of
coal viewed in bright sunlight will still appear
black even though it may be reflecting a greater
LIGHTNESS OR WHITENESS amount of light to the eye than would a piece of
CONSTANCY white paper viewed in ordinary room light (i.e.,
0.05 X 1,000 units of sunlight falling on the coal,
How light or white an object appears to be is also which equals 50 units of light reaching the eye
affected by a constancy mechanism. The amount of from the coal, is greater than 0.9 x 50 units of
light at different points in our retinal image coming room light falling on the paper, which equals only
from an object (the retinal illuminance) is deter¬ 45 units of light reaching the eye from the paper).
mined by two things. The first is the amount of This is an example of lightness or whiteness
light from any source, such as the sun or a light constancy.
bulb, that falls on the object. This is called exter¬ Two types of explanation have been given for
nal illuminance. The second, called reflectance, lightness constancy. The first is based on direct
is the proportion of light falling on the object that perception theory, in which constancy is main¬
is reflected to the eye of the observer. The reflec¬ tained by stimulus relationships. In size constancy,
tance is the object property that most closely cor¬ the ratio or relationship between texture element
responds to how light or white a surface appears. size and the object size is significant. In lightness
For example, a white surface will reflect most (per¬ constancy, a similar ratio is suggested between re¬
haps 80%-90%) of the light that falls on it. In con¬ gions of illuminance on the retina.
trast, a black surface will absorb a great deal of This constant-ratio notion works as follows.
light, and the proportion reflected will be quite Suppose you are looking at two pieces of paper;
small (often less than 4%-5%). Roughly speaking, one is white with a reflectance of 80 percent, and
the amount of light reaching the eye can be ob¬ the other is gray with a reflectance of 40 percent.
tained from the simple formula They are illuminated by a light source of 100 units
of intensity. The amount of light reaching your eye
light at eye = reflectance x external illuminance would then be 80 and 40 units, respectively.
420 Chapter 14/The Constancies

Suppose that we now reduce the intensity of the the gray surface, regardless of the intensity of the
light by half to 50 units. Now the amount of light external illumination.
reaching your eye is 40 and 20 units, respectively. The constant-ratio explanation of lightness
We might now predict that the white patch would constancy has received experimental support from
appear to be gray, and similar in lightness to the a number of researchers, such as Wallach (1948,
gray patch viewed under the intense illumination. 1972), who verified it using the experimental
This, however, is not the case—the white still ap¬ setup shown in Figure 14-10. There have been a
pears white and the gray appears gray. It is not the number of elaborations of this theory (cf. Corn-
total retinal illumination that matters, but rather the sweet, 1970, 1985; Richards, 1977; Marr, 1974),
ratio of the intensities of the two patches of light and some suggestion that the mechanism might
on the retina, which remains constant with that work via the operation of lateral inhibition, which
from the white surface twice as intense as that from we discussed in Chapter 4. Such a theoretical

Screen Screen

Projector Projector Projector Projector

A B

FIGURE 14-10 Relational determinants of lightness constancy. Wallach (1948, 1972) found that
regardless of absolute physical intensity, equal lightness would be perceived when targets stood in an
equal intensity ratio with their surroundings. In this example, which demonstrates the experimental
technique used by Wallach, the central circles will appear to be equally bright because they both stand in
a 1/2 to 1 intensity ratio with their respective surroundings. The numbers in the figure represent relative
light intensity after a subject has made a match by adjusting the central target in B to match the central
target in A.
 Lightness or Whiteness Constancy 421

mechanism would work if we assume that the

amount of lateral inhibition increases as we in¬
crease the intensity of retinal illumination at any
one point on the retina. This would result in light¬
ness constancy, since the amount of stimulation
and the amount of inhibition would both increase
equally, leaving the overall response of the eye rel¬
atively unchanged.
One prediction made by the constant-ratio the¬
ory is that lightness constancy should depend on
having several different levels of reflectance under
the same illumination in close proximity in the vi¬
sual field. This prediction is confirmed in the re¬
sults of a classic experiment by Gelb (1929). He
used a concealed light source to illuminate an ob¬
ject that was placed in a dimly lit field, as shown
in Figure 14-11. The illuminated object was a
black disk, but observers reported seeing a white
disk in dim light rather than a very brightly lit
black disk. Of course, this represents a complete
failure of constancy. If constancy were operating,
the observers would see the black disk as black
even though it was very brightly lit. Gelb then tried
a second manipulation. He placed a piece of white
paper in front of the brightly lit black disk. As soon
as this was done, observers reported that the disk Observer
looked black. In other words, with the addition of
FIGURE 14-11 Experimental situation used by
the reference white paper constancy returned.
Gelb (1929) to test for lightness constancy when a
However, as soon as the piece of white paper was light source was hidden from view.
removed, the black disk returned to its former
white appearance. Although these results appear to
strongly support the direct perception notion of Thus, in the Gelb experiment it could be argued
constant intensity ratios as the basis for lightness that the introduction of the white piece of paper
constancy, Gilchrist and Jacobsen (1984) have provides a cue indicating that there is an intense,
shown that other factors still must play a role, since hidden light source. This information evokes the
we do maintain lightness constancy even in a world formerly inoperative constancy correction.
of one reflectance level, such as an all-white room! Several such cues seem to be important in
A second explanation based on constructive triggering lightness constancy. For instance, the
theory would add a number of factors to the con¬ presence of visible shadows produces a lightness
stant ratio explanation of lightness constancy. Such correction (Gilchrist & Jacobsen, 1984; MacLeod,
a theory would argue that the observer responds to 1947). Also, cues as to the location of the object
cues indicating the nature of the illumination falling relative to the light source provide information to
on the object, and adjusts the apprehended light¬ allow us to correct our perception of the whiteness
ness of the object in consciousness accordingly. or lightness of the object, despite the retinal illu-
422 Chapter 14/The Constancies

DEMONSTRATION BOX 14-5. Lightness Constancy

To a certain extent, lightness constancy depends on Also notice one other interesting aspect of the tube
assumptions that the observer makes about the nature shown here, namely that either the right- or the left-
of the world. Consider the gray tube shown here. No¬ hand portion can be viewed as the interior or the
tice that the gray of the interior of the tube appears to exterior surface. A figure that can assume several dif¬
be lighter than the gray of the exterior. In fact, they ferent orientations depending on one’s point of view
are the same gray. Coren and Komoda (1973) sug¬ is called a reversible figure. Notice how the apparent
gested that this apparent lightness difference involves lightness difference between the two sides changes
a cognitive adjustment based on presumptions that we depending on whether you see the right or the left
make about the environment. If the tube were real, its side as the interior surface. The apparent inner sur¬
interior would be likely to receive less light than its face, regardless of whether it is the right or the left,
exterior. In the tube pictured here, however, the same appears to be the lighter one.
amount of light reaches the eye from both the appar¬
ent interior and the apparent exterior surfaces. This
could only happen if the interior surface reflects a
greater proportion of the light that reaches it; in other
words, the internal surface must have a greater reflec¬
tance. This demonstration shows one way in which
lightness constancy operates. The visual system
makes presumptions about the amount of light reach¬
ing surfaces and adjusts the perceptual experience so
that the apparent lightness corresponds to the assumed
relative reflectances rather than to the actual distribu¬
tion of light reaching the eye.

mination intensity (e.g., Beck, 1965; Flock & be registered as color or hue changes. Nonetheless,
Freedberg, 1970; Hochberg & Beck, 1954). Even within limits, we can identify red as red whether it
information about the relative spatial relationships is viewed under fluorescent lights that have a dom¬
among objects seems to contribute to this effect inant blue hue or incandescent lights that are basi¬
(Gilchrist, 1980; Mershon & Gogel, 1970). Dem¬ cally yellow. Color constancy refers to our ability
onstration Box 14-5 shows how your presumptions to maintain the percept of a particular hue through¬
about the illumination falling on a surface can af¬ out variations in the quality of the illuminance and
fect its apparent lightness. the reflectance properties of an object’s surface pig¬
ment. Based on our discussion of lightness con¬
stancy, we know that such variations will change
COLOR OR HUE CONSTANCY the properties of the retinal image in terms of both
luminance and wavelength. However, just as in
Similar to the size and intensity changes of the ret¬ lightness constancy, our perception of color is not
inal image, there are systematic changes in the based completely on the properties of the image on
spectral composition of the retinal image that could the retina but on the relative composition of sur-
 Other Constancies 423

DEMONSTRATION BOX 14-6. Color Constancy

To carry out this demonstration you will need a re¬ same color under both light sources. Now look at the
duction screen, which is simply a Vi cm hole cut into paper through the reduction screen, making sure that
the center of a dark piece of cardboard. In addition, nothing else is visible through the hole except the
you will need a fluorescent and a tungsten light source patch of color. If you have used green paper, it
(such as your desk lamp and your room light, respec¬ should appear to be yellow-green under the tungsten
tively) and a piece of colored paper (green works light but blue-green when under the fluorescent lamp.
best). Hold the paper under the tungsten light. Notice In a full-cue situation with context information avail¬
its color. Next hold the colored sheet under a fluores¬ able, the operation of color constancy prevented you
cent lamp and notice that it still appears to be the from seeing this hue change.

rounding colors in the field (e.g., Brou, Sciancia, in which despite our head and eye movements the
Linden & Lettvin, 1986; Land, 1977) and also on egocentric direction of objects (where they lie rel¬
our perception of the nature of the illuminant fall¬ ative to our bodies) remains constant.
ing on a surface. An example of color constancy is Position and direction constancy can be distin¬
shown in Demonstration Box 14-6. guished from each other in the following way. Eye
movements do not change egocentric direction, but
head and body movements can. For example, look
OTHER CONSTANCIES at an object that is straight ahead of your body.
Now shift your head to one side. Since we tend to
There are many other constancies, some well- use the head as the reference point for egocentric
known, some less obvious. For example, there is direction (see Chapter 10), the object no longer
an auditory version of size constancy called loud¬ seems to be directly straight ahead. Although the
ness constancy. In this situation, the perceived object is now perceived to lie in a different direc¬
loudness of a sound source remains constant, even tion, its position in space is the same as it was be¬
though the sound level at your ear is diminish¬ fore the head movement. Position and direction
ing because you are moving farther away from constancy are related but still separable phenomena
the source. (Shebilske, 1977).
There are also several other visual constan¬ There is even a kind of odor constancy.
cies. We already encountered one class of these in When you are sniffing an object, a deep sniff will
Chapter 13. There we saw that despite the fact that tend to pull more of the odorous molecules into
the retinal image moves we do not experience the your nose. We know that if we artificially give a
world as moving but recognize that this change is large puff of some vapors to you, it will smell more
due to our eye movements. This phenomenon intense than a smaller puff (e.g., Rehn, 1978). Yet
is known as position constancy, which we saw is when you actually sniff something, its “smelli¬
controlled by feedback of some sort from our eye ness’ ’ remains constant despite the strength of
and head movements combined with the actual your sniff, hence demonstrating odor constancy
movements of the visual image across the retina. (Teghtsoonian, Teghtsoonian, Berglund & Berg-
Similar, but not the same, is direction constancy, lund, 1978).
 424 Chapter 14/The Constancies

As we go through this list of constancies, a "■^tHstal stimulus An object or event in the environ¬
pattern ought to be emerging. The purpose of per¬ ment.

ception is to derive information about the nature of Emmert’s law The quantitative expression of the re¬
lationship between apparent size and apparent distance;
the external environment and the objects that in¬
objects appear larger when projected on a more distant
habit it. The viewing conditions, our relationship surface.
to objects, and our own exploratory behaviors will External illuminance The amount of light falling on
very frequently change the pattern of the proximal an object.
stimuli at our receptor surface. The constancies, Focal stimulus The stimulus you are perceiving.
then, are a complex set of “corrections” that take Lightness constancy The process by which the ap¬
into account the ongoing conditions and allow us parent lightness of a stimulus remains unchanged, despite
to extract a stable set of object properties from the changes in physical illumination.

continuous flow of sensory inputs at our receptors. Loudness constancy The process by which the ap¬
parent loudness of a sound source remains unchanged de¬
Were it not for such constancy corrections, objects
spite changes in your distance from it.
would have no permanent properties in conscious¬
Moon illusion An illusion in which the moon ap¬
ness at all. They would continually change size, pears larger when near the horizon than when high in the
shape, lightness, color, and direction, with every sky.
move we make. Consciousness and sanity would be “''Mueller-Lyer illusion An illusion of length caused
difficult to sustain in a world of such sudden and by placing inward or outward facing wings on the ends
frequent changes. of lines.
Multidimensional interaction Interactions between
several stimulus channels or sources of information.
Object properties The physical properties character¬
izing an object, such as size, shape, etc.
GLOSSARY
Objective instruction Instruction directing an ob¬
server to report the sizes and shapes of real-world ob¬
The following definitions are specific to this book. jects.
Apprehension The conscious representation of a Odor constancy The process by which odor inten¬
perceptual experience. sity appears to be unchanged, despite variations in the
Color constancy The phenomenon whereby the strength of sniffing.
color of an object does not appear to change despite "■^Ponzo illusion A length illusion produced by regis¬
changes in the spectral composition of the light falling tration of converging lines as distance cues.
on it. Position constancy Stable perceived position of ob¬
Constancy scaling The process by which the size of jects despite body, eye, or head movements.
a target is changed in consciousness in order to correct Projective instruction Instruction directing an ob¬
for registered viewing distance. server to report the sizes and shapes of parts of his or her
Constructive theory An approach to constancy con¬ retinal image.
tending that perception may be altered by experience, or •-'Proximal stimulus Information about a distal stim¬
other factors not in the proximal stimulus. ulus that reaches the receptors, such as visual image on
Context stimuli Stimuli that are near to, or occur at the retina, or sound at the ears.
the same time as, the stimulus you are perceiving. Reflectance The proportion of light reflected from a
'^Direct perception An approach to constancy main¬ surface.
taining that all aspects of the percept must be directly Registration Perceptual representation of informa¬
derived from components of the proximal stimulus. tion extracted from the proximal stimulus, but not nec¬
✓ Direction constancy The stability of an object’s per¬ essarily consciously apprehended.
ceived egocentric direction despite changes in eye or Retinal illuminance The amount of light falling on
head position. the retina.
 Glossary 425

^Shape constancy The process by which the apparent spite changes in objective distance and retinal image
shape of an object remains constant despite changes in size.
the shape of the retinal image. Unconscious inference A process by which infor¬
Situation properties The conditions affecting the mation from a number of sources is unconsciously put
way objects are viewed, such as your distance from them together to create a perception.
or the amount of ambient light. Whiteness constancy See Lightness constancy.
^Size constancy The stability of perceived size de¬
CHAPTER
15
Attention
VARIETIES OF ATTENTION
ORIENTING
Orienting Reflex
Covert Orienting
The Attentional Gaze
FILTERING
The Cocktail Party Phenomenon
The Video Overlap Phenomenon
Divided Attention
SEARCHING
Eye Movements and Visual Searching
Feature versus Conjunction Searching
Automatic versus Controlled Searching
Vigilance
EXPECTING
Costs and Renefits of Information Cues
THEORIES OF ATTENTION
428 Chaffer 15/Attention

My experience is what I agree to attend to. caused you to look toward the door to the room
Only those items which I notice, shape my where your roommate was bringing you a midnight
mind. snack. In all of these cases important environmen¬
tal events demanded an orienting response. That
—WILLIAM JAMES (1890, p. 402)
is, your attention was drawn, or pulled, to the
O ur ability to perceive, process, and interpret
the numerous and diverse stimuli that contin¬
source of a sudden change in your sensory world.
Some events you gave only brief attention, as when
you initiated the foot-stretch. Other events you lis¬
ually bombard our receptors is limited. For this
tened to (the fire siren) or looked at (your room¬
reason, we are forced to choose among all that is
there to be looked at, listened to, felt, smelled, or mate bringing the snack) for longer periods of time.
tasted. The various ways by which we select While you attended to some stimuli you were also
among stimuli as we sense our world are often excluding many others, hence you were probably
grouped together under the general label of atten¬ unaware of the goldfish swimming in its bowl or
tion. the humming of the refrigerator in the next room.
Technically we would say that you were filtering
out the extraneous events, attending to only one, or
VARIETIES OF ATTENTION perhaps a few, information channels (distinct and
separable sources of stimulus information).
Attention has several different aspects. For exam¬ As you continued to study this book, perhaps
ple, think back to the last time you were reading your attention wandered for a moment and you
this book and the various extraneous events that thought of the exam scheduled for tomorrow in cal¬
distracted you. You got a cramp in your foot and culus. You wondered where your notes were and
initiated a stretching movement; a fire engine looked up, scanning your room for the green binder
screamed by outside and you listened until the siren you would be poring over in a few minutes. You
stopped somewhere down the block; a flicker of were searching for a relevant stimulus in the en¬
movement in the periphery of your visual field vironment, scanning your sensory world for partic-

Task
Searching Expecting
Filtering
Orienting

Visual
/
ra //
>*>
T3
O
S
Auditory
y
' //
/ ys

/ /y
Divided
vyjL~~-
--
Conditions

FIGURE 15-1 A representation of various perceptual attention situations dealt with in this chapter.
 Orienting 429

ular features or combinations of features. Finally, sponses such as flicking the eye in the direction of
just as you were about to start studying your math a sound or peripheral movement occur automati¬
notes, you paused, realizing that it was at about cally and are collectively referred to as the orient¬
this time every night that the wolves in the zoo next ing reflex. This is such a reliable reflex (e.g.,
door began to howl at the moon. You listened for Butterworth, 1981; Muir & Field, 1979) that eye-
a few moments. Yes, there they were, right on and head-turns toward sounds have been used to
time. You were expecting something to happen test the hearing of newborn infants. The best stim¬
and momentarily attended to “empty space” until uli to elicit the orienting reflex are often loud
it did. In all of these situations, different forms of sounds and suddenly appearing bright lights,
attention were called on and each played a major changes in contours, or movements in the periph¬
role in determining your conscious perceptual ex¬ eral visual field that are not regular, predictable oc¬
perience. currences. When these stimuli happen, the animal,
In what follows, we describe in more detail or human, turns its eyes or ears so as to fixate the
some of what is known about how attention oper¬ visual object or sound source, and often orients the
ates in these four tasks: orienting, filtering, search¬ head and body to face toward the event as well. A
ing, and expecting. Our attention is restricted to variety of other behaviors also occur, such as pos¬
just the visual and auditory modalities, since the tural adjustments, skin conductance changes, pupil
most work has been done on these, but our conclu¬ dilation, decrease in heart rate, a pause in breath¬
sions apply to the other modalities as well. By way ing, and constriction of the peripheral blood vessels
of definition, it must be clear that each of the tasks (see Rohrbaugh, 1984, for more details). In short,
can involve a single target or event (focused atten¬ every resource is focused on the stimulating event,
tion), or several (divided attention). Figure 15-1 to allow its significance to be understood as quickly
is a summary of all the situations we discuss in this as possible.
chapter. The orienting reflex seems to be triggered by
unexpected changes in our environment. It is as
though we had an internal “model” of the imme¬
ORIENTING diate world of stimuli around us. When we notice
a departure of stimulus input from that model, we
The simplest form of selecting among the stimulus reflexively attend to that stimulus in order to update
inputs is to orient your sensory receptors toward that model (Donchin, 1981; Sokolov, 1975). This
one set of stimuli and away from another. In this is consistent with the fact that the orienting reflex
sense we might say that you do not passively see rapidly habituates. This means that it becomes
or hear, but rather you actively look or listen. weaker and less likely to happen if the same stim¬
ulus, no matter how strong and sudden, is repeated
again and again. The response, however, recovers
Orienting Reflex
to full strength with any change in the nature or
Whenever a sudden movement or loud sound hap¬ form of the stimulus.
pens in your environment, your attention tends to
be drawn toward that event. You have all seen a
Covert Orienting
dog or a cat prick up its ears and turn its head to¬
ward a sudden sound. The animal is performing the Up to now we have been dealing only with overt
most primitive form of orienting response, which orienting responses, which actually involve looking
involves adjusting the sense organs so they can op¬ at or turning toward a stimulus. Although this is
timally pick up information about the event. Re¬ one of the most direct signs that we are attending
430 Chapter 15/Attention

to something, a number of researchers have pointed ance of a stimulus in the visual field captures visual
out that it is possible to attend to an event or stim¬ attention and gives that stimulus an advantage in
ulus without making any overt sign that we are terms of how quickly or accurately it can be re¬
doing so. For example, Helmholtz (1909/1962) ob¬ sponded to. They asked observers to say whether a
served that he could direct his attention voluntarily letter target, whose identity was indicated at the be¬
to any portion of an absolutely featureless dark ginning of each trial, was present in a field of other
field at will without the necessity of an eye move¬ distractor letters. On each trial, one letter appeared
ment or change in accommodation or convergence. abruptly in the visual field while either one or three
This shift of attentional focus, which is dissociated others appeared gradually by the fading of selected
from any visible change in overt eye/head/body ori¬ lines in figures that had been displayed previously,
entation, is called covert attention. A common ex¬ as shown in Figure 15-2. Sometimes the target let¬
ample of covert attention is when you become ter was the abruptly appearing one, sometimes it
aware of a familiar voice and the conversation it is was one of the gradually appearing letters. When
having somewhere else at a party even while you the target appeared abruptly, observers detected it
look at, nod at, even say “uh-huh” every once in significantly more quickly than when it faded on.
a while to the person in front of you. Most modem Moreover, it didn’t matter whether there were two
research on attention takes for granted that overt or four distractor letters to check when the target
orienting is not necessary for paying attention. appeared abruptly. When the target was one of the
Typically, in an experiment eye/head/body move¬ gradually appearing letters, however, it took signif¬
ments are strictly controlled, or they are made ir¬ icantly longer to detect when there were four letters
relevant by using headphones or such short in the display than when there were two. The
stimulus presentations that there is no time to shift abruptly appearing stimulus seems to have been
the eyes to fixate the stimulus. Such controlled pre¬ drawing attention to itself. If it was the target, a
sentations allow researchers to separate the effects positive response could immediately be made; if it
of overt orienting from more covert shifts of atten¬ wasn’t the target, then the other figures had to be
tion. When your attention is involuntarily drawn to checked for the target before a response could be
a stimulus, without any overt orienting response, made, slowing down that response. Even if the
we refer to this as covert orienting. It is similar abruptly appearing figure wasn’t the target, observ¬
to the orienting reflex in that in such instances ers still covertly oriented toward it first, before they
we seem to have no choice about what we pay checked any of the other figures.
attention to. A similar phenomenon happens in the auditory
There are many ways your attention can modality. In the early experiments on divided at¬
be seized by certain stimuli. A dramatic example tention, subjects were given a different message in
capitalizes on the fact that visual stimuli seem each of their two ears by means of headphones.
to be more capable of drawing our attention than They were asked to pay attention to and repeat the
auditory stimuli. This is the phenomenon ventrilo¬ message they were hearing in one ear (this proce¬
quists depend on, called visual capture. Demon¬ dure is called shadowing) while another message
stration Box 15-1 allows you to experience this for was delivered to the other ear. When an abrupt or
yourself. distinctive sound was presented or an unexpected
An interesting experimental demonstration of change, such as a switch from a male to a female
how attention can be drawn to a stimulus was made voice, occurred in the nonshadowed ear, subjects
by Yantis and Jonides (1984, Jonides & Yantis, tended to “stumble” or lose the continuity of their
1988). They demonstrated that the abrupt appear¬ shadowing (see e.g., Kahneman, 1973). Appar-
 Orienting 431

DEMONSTRATION BOX 15-1. Visual Capture

Visual capture is a phenomenon in which attention the sound is highly correlated with the picture. (You
is caught by a visual stimulus in a way that results in could also use a radio and a television set, tuning in
an illusion of auditory localization. Whenever you are to a simulcast show like some concerts.) Now turn
listening to a sound, such as a voice talking, there is off the sound on one of the sets and turn off the pic¬
a tendency to try to identify visual events, or objects, ture on the other. Move back a short distance and
that could be causing the sound. When the ventrilo¬ look at a place between the two sets while paying
quist’s dummy is moving its mouth and limbs, and attention to the picture-displaying set. The sound
the ventriloquist is talking without moving his mouth, seems to come from that set, even though its sound is
then your visual attention is “captured” by the dum¬ turned off. It actually doesn’t matter where you look,
my’s movements and you hear the ventriloquist’s the sound will seem to come from the set with the
voice coming from its mouth, even though it is really picture. You could also try moving the sets apart to
the ventriloquist speaking. see how powerful the phenomenon is. You will be
You can demonstrate this effect for yourself by ob¬ surprised at how far apart the sets can be before the
taining two television sets (or going to a store that actual sound source dominates. By the way, this also
sells them and asking to use two of theirs for a “sci¬ explains why when you are watching a film the sound
entific demonstration”). Place them side by side, seems to come from the actors’ mouths, even though
about 500 centimeters apart, and tune both sets in to the speakers may be located at the side of the film
the same newscast, talk show, or other show in which screen, or even in the back of the room.

ently their attention was drawn involuntarily to the and searching all depend on the presence of one or
nonshadowed message, causing them to fail to hear more stimulus cues toward which attention is either
what they were supposed to be attending to and drawn (orienting) or directed (filtering or search¬
thus to interrupt the smooth flow of the shadowing. ing). When we receive information in advance
In both of the examples above, attention was about where or when something is likely to happen,
drawn to some conspicuous stimulus somewhere in but there is no stimulus cue available toward which
the visual or auditory field. We refer to such a to orient our attention (we call this an information
stimulus as a stimulus cue. Orienting, filtering, cue), attentional phenomena appear to be quite
432 Chapter 15/Attention

Fade-on
sequence

Parts fade
Target
figure

Sudden-
onset
sequence

t
Start
t
Test
Time

FIGURE 15-2 The upper sequence shows how a stimulus is gradually presented by fading lines as
opposed to sudden onset of the stimulus in the Yantis & Jonides (1984) study. Sudden-onset stimuli were
more easily detected.

different. This suggests that our expectations inter¬ though you are swinging your gaze through space
act with how we direct our attention, a matter dis¬ with a fixed velocity of about 125 degrees per sec¬
cussed in a later section of this chapter, Expecting. ond—that’s only 8 msec to move 1 deg (see Tsai,
1983). Notice that this implies that your attention
does not “jump” from one stimulus to another, but
The Attentional Gaze
rather that in shifting your attention from point A
A useful way of conceptualizing some of the find¬ to point B your attentional gaze actually moves
ings in covert visual orienting, and in visual search through all the intermediate positions (Shulman,
as well, is a metaphor we will call the attentional Remington & McLean, 1979; but see also Murphy
gaze. Other terms have been suggested for this & Eriksen, 1987).
concept (e.g., a zoom lens by Eriksen & Hoffman, Auditory attention also acts as if it has a direc¬
1972, Eriksen & St. James, 1986, and Eriksen & tion of gaze. It can be drawn to particular spatial
Murphy, 1987; a spotlight by Hernandez-Peon, locations in a way similar to that of visual atten¬
1964, and Treisman, 1982; the mind’s eye by Jon¬ tion. However, auditory attention moves at about
ides, 1980), but attentional gaze is the most gen¬ 233 deg/sec, almost twice as fast as the shift of
eral. In this metaphor, we imagine that your visual attention (Rhodes, 1987). Another difference
attention can “gaze” about independently of where is that for shifts of gaze greater than 90 deg the
your eyes are looking. In the case of orienting, at¬ time it takes to move attention no longer increases
tention can be drawn to a stimulus cue anywhere in with increasing distance; instead, auditory attention
the visual field where there is adequate acuity and seems to “pop up” at the new location rather than
sensitivity to register it, either by an abrupt onset, systematically traveling through space.
as above, or by movement or other conspicuous One major limitation on both the visual and
changes (but see Jonides & Yantis, 1988). Such auditory attentional gaze mechanisms is that, like
covert attentional shifts seem to take place as the movement of your eyes, your attention cannot
 Filtering 433

be drawn simultaneously to more than one location of detail at any one time. In the visual modality,
in the visual field at any one instant (Eriksen & for example, this detail set tends to direct the focus
Yeh, 1985; Van Der Heijden, Wolters, Groep & of attention to elements of a particular relative size.
Hagenaar, 1987; Yantis & Jonides, 1984). Thus, Several studies have shown that observers can fo¬
divided covert orienting would seem to be as diffi¬ cus selectively on either the more global (relatively
cult to accomplish as divided overt orienting, given larger) aspects or the more local (relatively smaller)
that you can’t point your head, body, or eyes in features in a visual form. Thus, if we have a large
more than one direction at the same time. figure made up of smaller distinct components
There are three aspects of the attentional gaze (such as the large letter made up of smaller letters
that are important in the processing of sensory in¬ that we saw as Figure 11-14), you may be set to
formation. At any one moment attention may be attend to either the larger figure or its smaller ele¬
described as having a locus, an extent, and a detail ments. When you are set to attend to one level of
set. As noted above, the attentional gaze shifts detail, your processing of features at the other level
around much as your eyes move to take in visual is poorer (Hoffman, 1980; Kinchla, Solis-Macias &
information. Once attention is located at a particu¬ Hoffman, 1983; L. M. Ward, 1985). An example
lar place, or locus, in the visual field, processing of the effects of detail set is given in Demonstration
of stimuli occurring at or near that locus is im¬ Box 15-2.
proved. It is improved more if the locus is near the
fovea than if it is in the more peripheral regions of
the retina (Shulman, Sheehy & Wilson, 1986). The FILTERING
extent of the area over which attention is spread
can be controlled by making the stimulus cue larger Once we have oriented, either covertly or overtly,
or smaller (LaBerge, 1983; Podgomy & Shepard, to a source of stimulation, we may continue to at¬
1983). The greater the extent, the less the process¬ tend to (look at, listen to) that source to the exclu¬
ing efficiency. Processing also becomes less effi¬ sion of other things happening in our environment.
cient for stimuli that are farther away from the Specifically, we are filtering out all stimuli except
center of the attended region (Eriksen & St. James, those that come from the region or channel we are
1986). attending to. How well can we do this? What fac¬
Finally, there is some evidence that the atten¬ tors affect how well we can select one information
tional gaze is set or calibrated for a particular level source and filter out others? And what happens to

DEMONSTRATION BOX 15-2. Level of Detail and Attention

How many times in your life have you looked at a material you include and your placement of it. The
penny? Probably thousands of times. Take a piece of reason for this is that it is possible to recognize a
paper and, from memory, draw both sides of a penny. penny based on a fairly global set of characteristics,
There is no need to be artistic, just try to represent all namely its size, shape, and color. So your detail set
the figures, all the words, numbers, and dates on a when attending to pennies has probably seldom been
penny, each in its proper place. Next, compare your small enough to pick out the local characteristics, re¬
drawings to an actual penny. It is likely that you will gardless of the thousands of times that you have
find at least one, and probably several, errors in the looked at one.
434 Chapter 15/Attention

information we don’t attend to? At the level of of the messages as it is presented. If the observer
awareness, it is apparent that things we attend to is allowed to lag slightly behind the message, and
seem sharp and clear while things outside the atten- repeat entire phrases at once, the technique is
tional field are less distinct and more difficult to called phrase shadowing. If the requirement is to
remember. It turns out that research confirms these repeat each syllable as it is presented, it is called
informal impressions. phonemic shadowing. Cherry was the first to dem¬
onstrate that observers could orient to one message
and filter out the other.
It is not equally easy to shadow all messages.
The Cocktail Party Phenomenon
For example, using the dichotic listening technique
Let us consider a noisy cocktail party. This is a it has been shown that if the selected message is
good example of many possible sources of stimu¬ prose, such as a selection from a story, shadowing
lation occurring at one time, which require atten- is relatively easy. Shadowing random lists of words
tional filtering for you to make any sense out of the is more difficult, and shadowing nonsense syllables
proceedings. For example, there are often many (e.g., orp, vak, bij) is the most difficult of all.
conversations going on simultaneously, including Phrase shadowing is considerably easier than pho¬
the one in which you are involved. If you hear a nemic shadowing. Clearly meaning and grammati¬
significant or familiar voice you may orient to it. cal structure help us attend to one message and
You may follow that conversation for a while, nod¬ filter out others. Shadowing is also easier if the
ding occasionally to your own conversational part¬ messages come from two different places in space,
ner. You may also suddenly be startled to notice are different in pitch (e.g., one male voice and one
the person to whom you had been “talking” give female voice), or are presented at different speeds.
a sniff and walk away rapidly, obviously angry For an example of how this works, try Demonstra¬
with you. You are puzzled, because you can’t re¬ tion Box 15-3.
member a thing that person has said in the last five What happens to the inputs you don’t attend
minutes. However, you remember perfectly what to, which we earlier suggested were “filtered out”?
your former sweetheart said in the conversation to Cherry (1953) found that listeners could remember
which you were listening. Apparently you very ef¬ very little of the rejected message in the shadowing
fectively filtered out everything else, including task. Moray (1959) found that in difficult shadow¬
whatever it was that caused your conversational ing tasks, even though they knew that they would
partner to walk away. later be asked about it, listeners were unable to re¬
Colin Cherry (1953), in a now-classic article, member words that had been repeatedly presented
investigated some of the problems exemplified in in the unshadowed message. Did the listeners sim¬
the behavior we described above. He introduced ply not hear the unshadowed message, or did the
the experimental technique called shadowing in or¬ shadowed message somehow interfere with their
der to control how his observers oriented their at¬ memory of the unshadowed message? Both Cherry
tention. In this technique, an observer is presented and Moray had waited a little while after the shad¬
with two messages through two different informa¬ owing task was completed to ask about the unsha¬
tion channels. For example, the two channels could dowed message. Perhaps the unshadowed message
be the two ears (one message to each ear, a tech¬ was heard, maybe the words were actually recog¬
nique called dichotic listening), or one message nized, but they were forgotten quickly because they
could be presented visually and the other audito¬ weren’t committed to a long-lasting memory. Per¬
rily, or the two messages could be presented at dif¬ haps we must pay attention to an input in order to
ferent locations in space. The observer must repeat remember it for longer than a few seconds. This
aloud (that is, follow along with, or shadow) one idea was tested by interrupting listeners’ shadowing
 Filtehing 435

DEMONSTRATION BOX 15-3. Selective Attention and the Precedence Effect

You may remember our discussion of the precedence

effect from Chapter 7, where we listed some variables
Friends—2 m apart
that affect our ability to localize the position of sound
sources in space. When sounds are emitted in en¬
closed spaces, they tend to cause echoes as they
bounce from walls, ceilings, and floors. However, we
can still make a correct localization of the sound
source because the sound emanating directly from this
source will reach our ears before its echoes. The au¬
ditory system is sensitive to these time differences
and can use this information in the localization of
sound-producing sources. The direction of the sound
emanating directly from the sound source takes pre¬
cedence over other sounds in localization, hence the
name precedence effect.
The precedence effect can also be helpful in selec¬
tive attention, when we are attempting to process one
of many simultaneously occurring stimulus events. A
good example of this is found in cocktail party situ¬
ations, where you may try to follow one of many
competing conversations. This aspect of selective at¬
tention is helped by the spatial and temporal separa¬
tion of the auditory inputs. You can demonstrate this
for yourself with the aid of two friends (preferably of
the same sex) and a doorway. First have your friends
stand as shown in Figure A, while each reads pas¬
sages from a book or newspaper simultaneously. No¬
tice that even with your eyes closed you can easily
separate and locate the two messages. Now stand out
of the direct line of sight (and sound) of each friend,
as shown in Figure B. In this situation the messages
must travel indirectly out through the open door. This
means that they will tend to reach you at the same
time and come from the same direction. Now, again
with your eyes closed, notice how difficult it is to
locate the voices and to separate their messages.

to ask them to report what had just been presented seems that material in the unshadowed ear is ac¬
to the unshadowed ear (Glucksberg & Cowen, tually perceived at some level and is available for
1970; Norman, 1969). When this happens, listeners processing, and attention, for a short while after it
can usually recall the last five to seven words, occurs, but unless it is attended to it is not entered
numbers, or whatever units are being shadowed. It into a long-lasting memory.
436 Chapter 15/Attention

The Video Overlap Phenomenon of one program when the other one was superim¬
posed on it, although they did make a few more
Although it happens rarely in this electronically so¬ errors in this condition. Moreover, the odd events
phisticated age, sometimes two powerful TV sta¬ in the shadowed programs were almost always no¬
tions may be near enough that their signals ticed, whereas the odd events in the unshadowed
encroach on each other, making it impossible to programs were rarely noticed. For example, only 1
tune a television set to a single channel in some of 24 subjects noticed the handshake in the hand
geographical areas. When this happens, there will game while shadowing the ball game; no subjects
be an overlap of video broadcast of two different noticed the ball disappear in the ball game while
programs, with one usually looking somewhat they were watching the hand game. The reports
ghostly, or like a negative image. If you have ever that did occur were vague and uncertain, and usu¬
experienced this, but wanted to watch one of the ally not correct. Generally, those few subjects felt
channels enough (your favorite soap opera), you that there might be something unusual about the
may have experienced a visual phenomenon similar unshadowed program but they didn’t know what it
to the cocktail party phenomenon for sound. What was. Most subjects noticed nothing unusual at all.
you probably found was that it was possible to This indicates that, like auditory filtering, visual
watch “your” program and filter out the other one, filtering allows little of the filtered-out information
although of course it wasn’t pleasant because you to make a lasting impression, a result that has been
had to make an unaccustomed effort to do so. verified in many different situations (e.g.. Rock &
An analog of the shadowing task described Guttman, 1981). You can experience a similar type
above has been used to study this kind of visual of visual shadowing, and its effect on memory of
filtering (Neisser & Becklin, 1975). Overlapping the unshadowed message, by trying Demonstration
video programs, one of a hand game and the other Box 15-4.
of a ball game, were presented to observers. In the
hand game, the players tried to slap each other’s
Divided Attention
hands, and observers who “shadowed” this game
had to report each attacking stroke (but not feints). An important question relevant to perception is
In the ball game, players threw a basketball to one whether it is possible to attend to more than one
another while moving about irregularly. Observers source of information at the same time, and if so,
who shadowed the ball game had to report each whether performance suffers. In the visual filtering
throw of the ball from one player to another (but experiment just discussed, observers were asked
not fakes and dribbles). “Odd” events were also also to divide their attention and to shadow both
sometimes inserted in the programs (e.g., the hand- programs simultaneously. That is, they had to re¬
game players shook hands then resumed play, or port both attacks in the hand game and throws in
the ball-game players threw the ball out of the pic¬ the ball game. When they tried to do this, perfor¬
ture, played with an imaginary ball for a few sec¬ mance deteriorated dramatically. Observers missed
onds, and then resumed playing with the real ball). many more events and typically said the task was
Figure 15-3 shows examples of single frames from “demanding” or even “impossible” (Neisser &
each game and the two frames superimposed. Becklin, 1975). Moreover, presenting the two pro¬
The results of this study were remarkably sim¬ grams to different eyes (dichoptic presentation)
ilar to those from auditory shadowing experiments. made the divided attention task no easier. The con¬
Observers could easily follow the events in one clusion is that dividing visual attention between
program presented alone, as would be expected. two (or more) sources is very difficult; we can look
They also had little difficulty following the events at only one thing at a time. Another example of
 Filtering 437

FIGURE 15-3 Outline tracings of isolated frames from the video overlap experiment. (A) Hand game
only. (B) Ball game only. (C) Hand game and ball game superimposed. (From Neisser & Becklin, 1975)

what happens when we must divide our attention Box 15-5 allows you to experience this failure of
visually is demonstrated by the phenomenon of divided visual attention for yourself.
binocular rivalry. When the views presented to Attempts to divide attention between two au¬
the two eyes are different enough, it is virtually ditory information channels are similarly difficult.
impossible to hold both views in consciousness si¬ Of course, it is impossible to shadow two messages
multaneously. The two views rival each other for at once because we cannot say two things at once.
our attention. Thus we first see one eye’s image Experiments have been done, however, in which
while suppressing the other, and then the view al¬ people were asked to listen to messages (in this
ternates to that of the other eye. Demonstration case word lists) in both ears and later distinguish
438 Chapter 15/Attention

DEMONSTRATION BOX 15-4. Visual Shadowing and Memory

In the accompanying passage, the relevant message is

shaded and the irrelevant message is printed in the man Uttikktititi car 'it house U boy tttltltttMfy hat M-
normal fashion. You are to read the shaded passage pMMiiit shoe fSM candy Wit old tmtdtfM horse
aloud as rapidly as possible, ignoring the irrelevant Wit tree U pen M.'W£ phone MM cow tit/ book ttit
(unshaded) message. Now without cheating and look¬ hot iiitljjitit tape Ml pin IWi stand HtMUMt view
ing back, write down all the words you remember tbiMI sky M read ddtidiUJt man Mft car piMtttiMI-l
from the irrelevant message. Go back and read the tmy house dddtfMtd boy Wit hat MiWdiit shoe
shaded passage again, but this time stop after each ihtMdt candy Witip horse id tree dtiity pen Wit
line to write down the words you recall from the ir¬ phone KNM cow MMtiiidii book id hot titit tape M&J-
relevant message (without looking back at it). You cjiiiiM pin Ah stand WWdtl view id sky iMM red it
should find that the list of remembered words is mmiMMMi
longer when your reading is interrupted and you are
not asked to recall all the irrelevant message at once.
(From Lindsay & Norman, 1977)

words they had heard from distractors (Levy, 1971, there may be attentional resources that are unique
cited in Kahneman, 1973). Recognition perfor¬ to certain sensory modalities or perhaps even par¬
mance was far poorer when listening to both ears ticular tasks (allowing better performance in divid¬
than when listening to only one ear and filtering out ing across modalities or tasks). However, in the
the other. You may have had similar experiences at vast majority of cases, divided attention perfor¬
a cocktail party when trying to listen to two inter¬ mance is considerably worse than focused attention
esting conversations at once. It is possible to switch performance.
back and forth between them, but if they are at all Finally, it is worth mentioning that dividing
demanding a great deal of the information of each attention between two demanding tasks does be¬
one will be lost. come easier with extensive practice. You all have
Divided attention is easier if the information experienced doing two, or more, things at once,
channels are in different modalities, such as vision such as driving a car and carrying on a conversa¬
and audition, although performance still suffers in tion, or reading a book, chewing on a sandwich,
comparison to attention focused on only one chan¬ and scratching your head. It has been shown that a
nel whenever the filtering task is at all difficult skilled typist can type at a high rate and shadow a
(e.g., Treisman & Davies, 1972). Only when the message at the same time with almost no loss of
task is very easy, such as responding to a simple efficiency at either task (Shaffer, 1975). What
signal as soon as it occurs in either of two modali¬ seems to be happening here is that extensive prac¬
ties, is there no decline in performance under di¬ tice in typing has made that skill somewhat auto¬
vided-attention conditions (e.g., Miller, 1982). An matic for the typist. Automatic processing does
apparent exception to this rule is a study by All¬ not require conscious effort and little conscious at¬
port, Antonis, and Reynolds (1972) in which sub¬ tention is allocated for it, thus allowing more atten¬
jects sight-read music and shadowed a message at tion to be allocated to the less-automatic skill
the same time without loss of efficiency compared (shadowing, in this case). We discuss automaticity
to doing either task alone. This seems to imply that more fully in the next section.
 Filtering 439

DEMONSTRATION BOX 1.5-5. Binocular Rivalry

You may demonstrate binocular rivalry by using the should overlap, with one eye viewing the vertical and
accompanying figure and a pocket mirror. Place the the other the horizontal stripes. Now look at the su¬
mirror on the center line of Figure B and hold your perimposed lines for a few moments. At first you will
head relatively close as shown in Figure A. Adjust see one set of lines. Then they will be replaced by
the mirror, and your head, so that the half of the fig¬ the other set as they rival each other, alternating in
ure seen in it appears to be at the same distance as and out of your consciousness.
when seen directly. The two halves of the figure

^ Eyes

Pocket

A How to view the stimulus given below

in order to experience binocular rivalry

Center line

B Binocular rivalry stimulus

440 Chapter 15/Attention

SEARCHING hundred milliseconds may allow people to retrieve

previously stored search strategies that can then be
Imagine you are waiting in an airport for a loved used to find the important areas within the scene
one to arrive home from a trip abroad. The plane (Rabbitt, 1984).
arrives and a flood of people comes through the There certainly is evidence that meaning and
exit door. Your eyes flick back and forth across the expectation can direct what we look at in a visual
mass of faces, searching for that familiar face. You scene (Antes & Penland, 1981; Findlay, 1981;
don’t bother looking at people’s clothes because Stark & Ellis, 1981). For example, Yarbus (1967)
you know she is wearing a new outfit, and you recorded eye-movement patterns while observers
don’t know what it looks like. You are looking in¬ looked at pictures with different intentions in mind.
stead for the peculiar combination of longish, coal Figure 15-4 shows the eye-movement patterns ob¬
black hair, large nose, and close-set eyes that you servers made for a typical picture (A) when asked
remember so well. Someone near you suddenly to estimate either the ages of the individuals in the
shouts to a large man in a bright orange suit, who picture (B) or their wealth (C). Clearly, people
waves in reply. The shouter confides to you that looked at different places in order to find informa¬
she always has an easy time finding her husband at tion relevant to the different questions. Also, peo¬
the airport because he always wears that silly suit ple have a harder time locating a target object, such
and “stands out like a sore thumb.” We often as a bicycle, in a photograph of a natural scene if
know what we are looking or listening for and must the scene has been jumbled by randomly inter¬
search a field of “distractors” to find it. This situ¬ changing different areas as has been done in Figure
ation has become one of those most used for study¬ 15-5 (Biederman, Glass & Stacey, 1973). More¬
ing attention, partly because it is easy to implement over, we look at unusual objects in a visual scene
in the laboratory and partly because it has impor¬ longer when we find them (Antes & Penland, 1981;
tant implications for everyday life. Friedman, 1979), even when we are motivated to
distribute our attention equally over the scene
(Friedman & Liebelt, 1981). Because of this, un¬
Eye Movements and Visual Searching
expected objects tend to be remembered and rec¬
It is much easier to study looking-for than listen- ognized more easily, and exchanges of one unusual
ing-for, since there is an obvious external indicator object for another (a cow for a car in a living room)
of visual searching—eye movements. Our eyes are are noticed far more often than are exchanges of
constantly exploring the visual field with high¬ one usual object for another (a chair for a table in
speed ballistic movements called saccades. “Peo¬ a living room) (Friedman, 1979).
ple scan the visual world actively and purposively.
They know where to look and they know what to
Feature versus Conjunction Searching
look for” (Rabbitt, 1984, p. 287). People can rap¬
idly learn to inspect spatial locations in optimal or¬ A common laboratory task to study visual search¬
der to detect targets that may be present. Although ing involves asking an observer to scan an array of
this is a fairly automatic process for adults, it does letters, or other visual forms, in order to find a tar¬
not appear to be fully developed (for some tasks) get letter or form. One relatively early study of this
until children are about 6 or 7 years of age (e.g., type was by Neisser (1967). Observers scanned,
Cohen, 1981; Green, Hammond & Supramaniam, from top to bottom, an array of letters arranged in
1983) and becomes much more difficult for the el¬ 50 lines of 6 letters each, and came to perform this
derly (Rabbitt, 1984). In typical situations, people search at great speed (as fast as 60 letters/sec). A
use their knowledge of the world to guide their number of factors, however, affected their search
searching. The ability to recognize a scene in a few speed. For instance, when the target was an angular
 Searching 441

FIGURE 15-4 Eye-movement patterns made when viewing the picture (A) vary depending on whether
the viewer was asked the ages of the individuals in the picture (scan pattern B) or their wealth (scan
pattern C). (From Yarbus, 1967. Copyright Plenum Publishing Company, reprinted by permission.)

letter (W, Z, X) and the other letters (distractors) tentive” level of processing that segregates regions
were roundish (O, Q, C), observers searched much of a visual scene into figure and ground, a distinc¬
more quickly than when the target was more simi¬ tion we discussed in Chapter 11. This suggests that
lar to the distractors (e.g., a W when distractors when there are clear feature differences between
were K, Z, X, 7). When the target differs from all the target and the distractor items, the target be¬
of the distractors by possessing a feature they don’t comes readily visible because the distractors are
have (e.g., an angled line), we call this a feature lumped together as ground and the target stands out
search. When the only way to detect the target is as a figure by the action of this preattentive process
to detect a conjunction (or particular combination) alone, without any further processing. The notion
of features (such as the particular angles and their is that certain (usually similar) elements are
orientation that distinguish between a W and an A/), grouped together automatically and the ones that
we call this a conjunction search. In general, fea¬ don’t fit seem to leap into consciousness. This isn’t
ture searches are much easier than conjunction possible when the target and background items
searches. Thus, Neisser’s subjects typically re¬ closely resemble each other as in a conjunction
ported that when they were searching the list, par¬ search. Here, closer attention and scrutiny are
ticularly when the target was very different from needed to detect specific elements (e.g., Julesz,
the distractors, the nontarget letters were just a blur 1980).
and they did not “see” individual letters. In fact The differences between feature and conjunc¬
the target often just “popped out” of the array. tion searching have been extensively explored
Neisser (1967) argued that there is a “preat- (e.g., Treisman, 1982, 1986a). The really striking
442 Chapter 15/Attentio,\

FIGURE 15-5 It is easier to find a target object in a coherent, natural scene (above) than in the same
scene randomly jumbled (opposite page). iFrom Biederman et at., 19731

result is that when feature search is possible the cation at a time. Recalling our discussion of atten-
number of distractor items doesn’t seem to affect tional gaze, we might say that features occurring in
searching speed. The target simply pops out of the a single attentional “glance” are combined to form
display. However, when conjunction search is re¬ an object. This combination and comparison takes
quired the number of distractors does affect search¬ time and effort. If your attention is diverted or ov¬
ing speed. This can be seen clearly in some erloaded, errors may occur, and you may attribute
prototypical data illustrated in Figure 15-6. In con¬ the wrong features to a particular item and either
junction search we seem to be comparing each of miss your target or select a wrong target (Prinz¬
the distractors, one at a time, with the image of the metal, 1981; Treisman & Schmidt, 1982). Dem¬
target and responding only when they match. Such onstration Box 15-6 gives you an opportunity to try
an orderly and sequential set of comparisons is of¬ feature and conjunction searches for yourself.
ten referred to as a serial search. To explain this
kind of data, Treisman offered a feature integration
Automatic versus Controlled Searching
theory. It suggests that each feature of a stimulus
(such as its color, size, or shape) is registered sep¬ Searching can be improved by particular strategies
arately. When an object must be identified from a that an observer adopts, and by practice. For ex¬
combination of features, a correct analysis can only ample, a conjunction search may be treated as two
be achieved if you focus your attention on one lo¬ simple feature searches under some circumstances
 Searching 443

(Egeth, Virgi & Garbart, 1984). This would mean a very long time on a task that always demands the
that in Demonstration Box 15-6 you might be able same response under the same conditions, the na¬
to look only at the white letters in array C, ignoring ture of the search process seems to change—
the black letters, while searching for the white O. searching time for a stimulus becomes much faster
In a sense, the figure-ground preattentive process is and seems to be relatively independent of the num¬
used to reject as ground all distractors that can be ber of distractors present. Perhaps the most care¬
ruled out on the basis of a simple feature difference fully designed research demonstrating this result
from the target, leaving you with only a second was done by Shiffrin and Schneider (1977; Schnei¬
simple feature search to complete. Another strategy der & Shiffrin, 1977). In a typical study, some ob¬
that sometimes helps is to group items together into servers searched for a fixed set of targets (say the
smaller sets of stimuli. If stimulus sets are small letters H, S, and T) among a fixed set of distractors
enough (say two to eight items), attention operates (say all of the digits). At first, the more distractors
as if all items are checked at the same time (par¬ in the display, the longer it took to find the target.
allel search), rather than sequentially as in serial However, after practicing for 14 days (over 4,000
search. Using such a grouping strategy, the smaller searches) on the same task, the number of distrac¬
arrays may be searched in parallel for both features tors in the display ceased to matter. It took the
and conjunctions (Pashler, 1987). same amount of time to find the target regardless
When observers have been able to practice for of the number of distractors. Figure 15-7 shows
444 Chapter 15/Attention

FIGURE 15-6 The relation between response latency to report the presence of a target and the number
of distractor items that must be checked for feature and conjunction searches. (Based on Treisman, 1982)

this result graphically. Shiffrin and Schneider So we see that with practice the direction of
(1977) argued that at first the searching was a typ¬ attention shifts from controlled to automatic pro¬
ical serial conjunction search (they called it con¬ cessing (Schneider, Dumais & Shiffrin, 1984), and
trolled processing), but that after much practice in that this is accompanied by several changes in the
a consistent environment the searching became au¬ nature of the searching process. For instance, it be¬
tomatic, rather like a simple feature search. A sim¬ comes more difficult to stop yourself from respond¬
ilar result has been obtained for an auditory ing to targets that you are automatically set to
detection task (Poltrock, Lansman & Hunt, 1982), search for, even if you wish to ignore them. You
which indicates that automatic and controlled pro¬ will be able to do other tasks at the same time
cessing are not limited to vision but occur in other that you are engaging in this automatic search,
modalities as well. however, and these added tasks won’t materially

DEMONSTRATION BOX 15-6. Feature and Conjunction Search

In each of these arrays of visual forms there are tar- A 0 N 0 N N O N 0 N N 0 0 N

o 0 N N O N N 0 N O N N O 0
gets to find. The target is a white 0. Scan each array
N 0 0 N 0 N N N 0 N N O O
quickly, only once, and write down how many targets N 0 N 0 N N 0 N 0 0 N N 0
you see. Notice in each array how difficult or easy it
is to find the targets. Do this before reading further.
N N N N N N N N
Now you can know that arrays A and B required a B N N N N N N N
feature search (in A the feature was brightness, in B N N N N N N N
it was shape), whereas array C required a conjunction N N N N N N N
search (for both brightness and shape). There were
three targets in each array. Did you get them all?
N NO OON 00
Most people find the conjunction search to be the NOON O N O
most difficult of these tasks, and if they are apt to NO O N N 0 0
miss any targets it will be in array C. 00 NO O ONO
 Searching 445

Number of distractors

FIGURE 15-7 The relation between response latency to report the presence of a target and the number
of distractors to be checked for unpracticed and practiced observers. (Based on Glass & Holyoak, 1986)

interfere with your automatic processing. Unfortu¬ in the same place (see Kahneman & Treisman,
nately, you won’t be able to remember as well the 1984 for these and other similar data). Strong
things you found and responded to under automatic automaticity would require that as long as the
control. Many of these results have been docu¬ word could be read automatically it wouldn’t
mented in other studies (see Schneider et al., matter where it was, yet it seems that it is possible
1984). You can experience for yourself the power¬ to filter out the incompatible color word if it isn’t
fully automatic nature of reading words, and how part of the same perceptual object but not if it is.
this can interfere with other tasks, by trying the This kind of result favors a weaker notion of auto¬
demonstration of the Stroop effect in Demonstra¬ matic processing.
tion Box 15-7. Actually, many lapses of attention There are other similar data that confirm
in everyday life can be traced to such seemingly weaker automaticity. For example, it is easier to
automatic processes and their inevitable effects (see detect another stimulus while searching for an
Reason, 1984). “automatic” target in a task like that of Shiffrin
We might take an extreme position, for the and Schneider (1977) if that stimulus is close to the
sake of argument, and suggest that when a process spatial location of the automatic search target. Con¬
is truly automatic it requires no attentional re¬ versely, it is easier to detect the automatic target if
sources at all (cf. Schneider et al., 1984). This is it occurs near another stimulus that is also to be
called strong automaticity. A number of results detected (Hoffman, Nelson & Houck, 1983). Also,
have discouraged this extreme position. For exam¬ it has been shown that extensive practice can dra¬
ple, reading in the Stroop effect demonstrated in matically improve visual searching even when the
Demonstration Box 15-7 is supposed to be auto¬ conditions and responses are not always kept con¬
matic. However, the Stroop effect is much weaker stant, such as when the task demands that different
when the colored ink and the incompatible color responses have to be made to the same stimuli un¬
name are spatially separated than when they occur der different conditions and these conditions keep
446 Chapter 15/Attention

DEMONSTRATION BOX 15-7. The Stroop Effect

The Stroop effect is an interesting example of how tion is difficult to ignore, and the automatic expecta¬
well-learned material can interfere with our ability to tions that have come to be associated with the pres¬
attend to the demands of a task. In 1935, Stroop ence of words often take over, resulting in difficulties
found that observers had difficulty screening out in focusing attention.
meaningful information even when it was irrelevant Color Plate 8 is an example of the Stroop Color
to the task. He devised three situations. In the first he Word Test, so you can try this for yourself. Have a
recorded how long it took individuals to read a list of friend time you with either the second hand of a
color names, such as red and green, printed in black watch or with a stopwatch, if you have one, as you
ink. He then took an equal number of color patches read each group. Start timing with the command
and recorded how long it took observers to name each “Go” and read across the lines in exactly the same
one of the series. Then he took a color name and fashion for each group. When the last response is
printed it in a color of ink that did not coincide with made in each group, stop timing and note your re¬
the linguistic information (for example, the word blue sponse time. You should find that reading the color
printed in red ink). When he had observers name the names will take the least amount of time, whereas
ink color in this last series, he found that they often naming the colors of the ink when the printed word
erroneously read the printed color name rather than names a different color will take you the most time.
the ink color name; therefore, it took them much Naming the color patches will fall in between these
longer to read through this last series. The Stroop ef¬ two.
fect demonstrates that meaningful linguistic informa¬

changing (Cooke, Breen & Schvaneveldt, 1987). Vigilance

However, even extensive practice in a consistent
environment sometimes fails to produce automatic Sometimes we are asked to maintain a sustained
searching (e.g., Fisher, Duffy, Young, & Pollat- level of attention, for prolonged periods of time,
sek, 1988). while we search for targets that may only rarely
A somewhat different approach to these same appear. An example is an observer watching a ra¬
data emphasizes the development of skill in accom¬ dar screen for the presence of a particular type of
plishing various perceptual tasks (Neisser, 1976). aircraft that flies by only occasionally, or a quality
Here the suggestion is that the effects of practice control inspector examining an assembly line where
do not simply involve a switch from controlled to damaged or substandard items seldom appear. In
automatic processing, but rather that the task is these cases the observer is said to be performing a
being restructured (Cheng, 1985). That is, a differ¬ vigilance task. Research into vigilance began after
ent strategy is being used to accomplish the same it was noticed that radar operators during World
task. A nonperceptual example would be adding a War II tended to become fatigued after a time on
group of identical numbers, such as 2 + 2 + 2 + duty, resulting in a decrease in their ability to de¬
2 + 2. This could be accomplished by adding each tect enemy planes. After the war, experiments be¬
of the numbers to a running sum, or more quickly gan in an attempt to understand how attention
and easily by simply multiplying 5 X 2. In this sustained itself, particularly in boring search tasks
view, extensive practice allows a new strategy to with infrequent stimuli.
be developed and learned, rather than causing a The original experiments on vigilance required
transition from controlled to automatic processing. observers to watch a display similar to a clock face
 Expecting 447

around which a clock hand moved in steps. They seems to be a certain degree of physiological
had to press a key each time the hand took a double arousal. We adopt certain body positions, tense
step. After only Vi hour of watching, observers be¬ specific muscle groups, and have the feeling of
gan to report fewer and fewer double steps, missing “concentrating” whenever we are vigilant. Appar¬
almost 25 percent of them (Mackworth, 1948). ently most of us already believe that if we are
Physical fatigue didn’t seem to be a reasonable ex¬ highly aroused physiologically we will be better
planation of the drop in performance since the work able to sustain attention, since we often attempt to
load was very light. Perhaps the visual system itself raise our arousal level in vigilance situations with
was becoming fatigued and thus less sensitive, or stimulants such as coffee. In order to understand
perhaps the observer was just as sensitive to the how arousal affects vigilance, we should look at
double steps but simply failed to respond on some how arousal affects performance in general.
occasions. It was important to decide whether one The relation between arousal and performance
or the other or both of these explanations were is perhaps most elegantly expressed in the well-
correct. known Yerkes-Dodson law (Yerkes & Dodson,
The scene was set for the application of signal 1908). Figure 15-8 shows this relation graphically.
detection theory (see Chapter 2). If the visual sys¬ Contrary to what common sense might tell us, per¬
tem was becoming less sensitive it would be re¬ formance doesn’t always get better the more highly
flected in a decrease in d’, the measure of the aroused we are. In fact, overall performance of any
observer’s sensitivity. If there was some change in task peaks at an intermediate level of arousal. This
how willing the observer was to report the double intermediate level is lower for difficult tasks than
step it would be reflected in a change in (3, the for easy tasks, perhaps because as arousal increases
observer’s criterion that indicates response bias. attention narrows, thus decreasing the number of
When this type of analysis was applied, it was inputs that can be monitored. This tends to de¬
found that sensitivity, or d’, did not change over crease the number of distracting inputs processed,
time but (3 did. Observers were becoming less will¬ which would help in easy tasks. However, difficult
ing to say yes, meaning that the rare event they tasks usually require that more inputs be attended
searched for had actually occurred, the longer they to, so a narrowing of attention would actually hurt
had to maintain vigilance (Broadbent & Gregory, performance in such tasks (Easterbrook, 1959;
1963, 1965). More recent research has indicated Hockey, 1970). How does this relate to vigilance?
that extensive training can decrease or eliminate The modem consensus is that physiological arousal
such vigilance decrements (Fisk & Schneider, is mainly responsible for the overall level of vigi¬
1981; Parasuraman, 1984). Apparently the setting lance but pretty much unrelated to the decline in
of a criterion for responding in such tasks is a func¬ performance over time (Parasuraman, 1984).
tion of alertness, or the way available attention is
allocated to the task at hand (Parasuraman, 1984).
Does sensitivity ever change in a vigilance EXPECTING
task? Yes it does, for the worse, but only when the
numbers of events per unit of time that must be Knowing exactly when or where an important sig¬
monitored for the targets is very high and the tar¬ nal will occur is often difficult. For this reason we
gets are difficult to discriminate from the nontargets have orienting mechanisms that draw our attention
(Parasuraman, 1984; Parasuraman & Mouloua, to conspicuous stimuli. We also have search strat¬
1987). Declines in sensitivity seem to be associated egies that allow us to investigate likely locations
with prolonged high demand on attentional re¬ where important stimuli might be. However, some¬
sources and do indeed result from a kind of fatigue. times we get an advance cue about where or when
A major part of maintaining overt attention something will happen. We call such a cue an
448 Chapter 15/Attention

trials. In 80 percent of the cued trials the flash oc¬

curred on the side to which the arrow pointed (valid
trials), and in the other 20 percent it occurred on
the opposite side (invalid trials). The observers
were not allowed to move their eyes away from
either the plus or the arrow; they could only orient
their attention. Figure 15-9A shows a summary of
these conditions.

Start Target
trial presentation

■
FIGURE 15-8 Yerkes-Dodson law. Performance
Neutral 50% to
is best at intermediate levels of arousal, and +
trial either side
performance peaks at lower levels of arousal for
difficult than for easy tasks.

Valid 80% of
trial — trials
information cue. For example, imagine you are
back in the airport, this time trying to monitor two
doors at once, through only one of which your be¬ Invalid 20% of
— trials
loved will arrive. Suddenly the loudspeaker an¬ trial

nounces that passengers disembarking from that

flight will arrive through Gate 21, the left one of A
the two doors. Although you know that still doesn’t
guarantee it will be the door, you find yourself 350
more often shifting your attention to the left door.
You are actively expecting something to happen
there, and it has affected your attentional state. I 300

CD
E
Costs and Benefits of Information
c 250
o
Cues T5
co
CD
Probably the best demonstration of the effects of rr

information cues on performance was done for an¬ 200

other reason. Posner (1980) was trying to demon¬

strate covert orienting by asking observers to press
a key when they detected a flash of light either to Invalid Neutral Valid
the right or the left of a fixation point. In half of Position uncertainty
the trials (the neutral trials), observers fixated a
B
plus sign in the middle of the visual field and the
flash occurred randomly on one side or the other. FIGURE 15-9 (A) Stimulus presentations used to
In the other half of the trials, observers received an study the effects of expectation on detection. (B)
Results of reaction-time study of expectation
information cue in the form of an arrow pointing
showing the costs (invalid cue) and benefits (valid
either right or left and located where the plus was cue) of advance knowledge of stimulus location.
located on the neutral trials. These were the cued (Based on Posner, 1980)
 Expecting 449

Figure 15-9B shows the results of Posner’s the same. A similar result was obtained for costs
experiment. Using the neutral trials as a baseline to on the invalid trials. Moreover, if both cued and
indicate what performance level you would expect uncued locations were in the same hemifield, there
without any information cue, you can see that it were no costs or benefits at all! Finally, it was
took over 50 msec more to respond to the flash on found that with practice the benefits of the valid
invalid trials (a cost of the information cue being cue disappeared, whereas the costs of the invalid
wrong), but about 30 msec less to respond to the cue remained. Hughes and Zimba (1985) suggested
flash on the valid trials (a benefit of the valid in¬ that “attending to a spatial location acts primarily
formation cue). The costs and benefits of informa¬ to inhibit the processing of signals that originate in
tion cues have been interpreted by Posner (1980), the unattended hemifield” (p. 428). The costs re¬
and others, as indicating that attention can be co¬ sult from this inhibition. This suggests an interac¬
vertly oriented by an information cue, even in the tion between the physiological organization of the
absence of a stimulus on which to focus the atten- visual system and the way in which expectations
tional gaze. Of course, it is possible that a more affect the distribution of our attention.
complex interpretation is called for; for instance, Our expectations can help or hinder our detec¬
observers may be using different strategies on the tion of stimuli in sensory modalities other than vi¬
neutral trials than they are on the cued trials (cf. sion. For instance, pretend you are expecting your
Jonides & Mack, 1984). But for whatever reason, mother to come home any minute now. You are
the data do demonstrate that a valid information expecting to hear her cheery “Hello” in her usual,
cue allows you to detect a stimulus more rapidly rather high-pitched voice. At this moment your fa¬
than when there is an invalid information cue. ther shouts to you to come help him in the base¬
The effects of expectation and information ment. You don’t hear him calling and a minute
cues on attention are somewhat complicated. For later he storms into the room, demanding to know
instance, we noted earlier that when stimulus cues why you weren’t responding to him. You might ex¬
are present it takes time to shift the attentional gaze plain that you were listening for your mother’s
from one location to another, and that the time it voice and simply didn’t hear his much-lower-fre¬
takes is longer when the distances between the quency voice (if he doesn’t believe you, you can
starting and the new locations are greater (e.g., show him this book). There is lots of evidence that
Tsai, 1983). When attention is shifted because of detection of sounds is more difficult, that is, there
the presence of an information cue, however, it al¬ is a cost, when they are of uncertain frequency
ways takes the same time to shift attention to a new (Swets, 1963; Scharf, Quigley, Aoki, Peachly &
location (Remington & Pierce, 1984), at least un¬ Reeves, 1987). Moreover, if observers are told
der certain circumstances. which frequency to listen for by an information
Hughes and Zimba (1985) repeated Posner’s cue, such as lights, tones, or even the early parts
experiment but altered the display so that the flash of tonal patterns, detection of the target sounds im¬
didn’t always occur in the same place in the visual proves if the cue was valid but remains poor if it
field. Imagine a hypothetical line drawn vertically was invalid (Howard, O’Toole, Parasuraman &
through the fovea that divides the field of view into Bennett, 1984). Again, the information cue seems
a left and right half, or hemifield. In Chapter 3 we to lead to inhibition of the detection of the unex¬
saw that each such hemifield projects its informa¬ pected stimulus rather than better detection of the
tion to a different side of the brain. On both valid expected one.
and invalid trials the flash could occur either in the Apparently, information cues act differently
usual position or sometimes in other positions in on attention than do stimulus cues (cf. also Briand
the same hemifield. It didn’t matter where in the & Klein, 1987). When a stimulus cue is present,
cued hemifield the flash occurred, the benefit was the attentional gaze can be drawn or moved
450 Chapter 15/Attention

voluntarily to a particular spatial location, where it would be accomplished based on the physical char¬
facilitates processing of perceptual objects and their acteristics of the stimulus, rather than its meaning.
properties. However, an information cue in the ab¬ While the sensory qualities, such as frequency and
sence of a stimulus cue may cause us to suppress loudness, are registered for all of the voices, only
inputs from unexpected sources. Such suppression the words associated with some particular physical
would enhance processing of expected inputs by characteristics, such as voice quality and location,
lessening interference from unwanted inputs rather are passed on to be processed for content and com¬
than by directly facilitating processing of expected prehension. This is called early selection and is
inputs. depicted schematically in Figure 15-10.
The early selection models had difficulty with
the fact that there is evidence that at least some
THEORIES OF ATTENTION analysis is done on information coming through un¬
attended perceptual channels. This processing may
Ever since attention was first studied, investigators affect our responses even if we are unaware of it
have attempted to construct a coherent theoretical (see Cheeseman & Merikle, 1984; Holender,
account of the major phenomena. Unfortunately, 1986). A striking example is when someone in a
because the concept of attention has meant so many conversation that you are not paying direct atten¬
different things to different people and has been tion to mentions your name—you immediately be¬
studied in so many different ways, this has not yet come aware of the fact. This kind of evidence led
been accomplished. At present there are several to a set of structural theories that emphasized late
different approaches to understanding attention. We selection. They hypothesized that all information
will try to give you the flavor of a few of the major entering sensory systems gets preliminary analysis
approaches here, but you must remember that no and the bottleneck occurs at the stage of more or
one of these approaches seems adequate to explain less conscious processing, when material is being
all of the data described above, let alone the vast entered into a longer-lasting memory (e.g.,
array of data we did not describe. Deutsch & Deutsch, 1963; Norman, 1968). A
All the models attempt to explain why atten¬ schematic representation of this kind of model is
tion lets some information reach consciousness shown in Figure 15-10. The debate between early
while screening other information out. Probably the and late selection still rages (cf. Pashler, 1984) and
earliest theoretical approach to have survived until has spilled over onto other approaches as well.
now is the group of structural theories. As A second general approach to attention has
pointed out by Kahneman and Treisman (1984), the grown mostly from studies of search and expecta¬
studies of stimulus filtering that were popular in the tion (cf. Kahneman & Treisman, 1984), especially
1950s and 1960s seemed to imply that perceptual studies involving comparisons of focused and di¬
attention was structurally limited. The notion was vided attention. The general finding that dividing
that there is a bottleneck or a filter somewhere in attention between two tasks or searching for more
the information processing system beyond which than one target usually is more difficult than focus¬
only one, or at most a few, stimulus inputs can ing on one task or target has led to the notion that
pass at one time. The earliest studies suggested that there are attentional resources that can be “used
this bottleneck occurs very early in the perceptual up” by a task. If there is more demand than there
process, just after registration by the sensory sys¬ are resources available, then performance suffers.
tem and before the meaning of an input can be de¬ The first theories that assumed a limited capacity to
termined (e.g., Broadbent, 1958). For example, if attend thought of attention as a single “pool” of
you are trying to listen to only one person in a capacity (e.g., Kahneman, 1973). The operation of
crowded room full of talking people selection such a model is shown in Figure 15-11 A and B.
 Theories of Attention 451

Early selection Late selection

Selection of
appropriate appropriate
response response

FIGURE 15-10 Schematic diagrams of two types of bottleneck models of attention. Both models assume
that perceptual selectivity stems from a structural limitation on our capacity to process all the incoming
stimulation (stimuli are indicated by S, and S2). The early selection models see the limitation as occurring
at the earliest stages; only limited amounts of stimulation are channeled along for further processing.
The late selection models conceptualize stimulus selection (attention) as occurring at later stages of
information processing.

All of the available capacity is used for one task in ears will probably require that the same set of re¬
A, whereas in B involving divided attention the ca¬ sources and analyzers be utilized, hence these two
pacity must be shared, leaving less processing re¬ tasks would interfere with each other. In contrast,
sources for each task. drawing a picture or doodling while monitoring
Recently the attentional resource models have someone speaking probably involves different types
had to be revised. There have been some demon¬ of mental capacity and one task will not compete
strations of near-perfect divison of attention, for in¬ with the other for mental resources. Recent re¬
stance, when sight-reading music and shadowing at search suggests that the bottleneck and capacity
the same time. This has led some theorists to sug¬ models can be combined, and it may make sense to
gest that there may be multiple resources, as shown think of selectivity and capacity limitations at both
in Figure 15-11C (Navon & Gopher, 1979; Wick- early and late processing stages (Dark, Johnston,
ens, 1984). Some of these resources are probably Myles-Worsley & Farah, 1985).
specific to a particular modality, whereas others Where does this leave us with the notion of
may be attributable to an “executive” that moni¬ attention and its role in perception? Simply put,
tors inputs from the various modalities and controls that which we do not attend to, or which does not
access to response selection. Whether attention to force us to orient toward it, has no more effect on
one task interferes with attention to another would us than a subthreshold stimulus. Attention is the
then depend on the characteristics of the tasks and gateway through which only selected stimuli, a few
the processing required. For example, monitoring out of endless hordes impinging on our receptors,
and analyzing two prose passages read into the two enter into the limited realm we call consciousness.
452 Chapter 15/Attention

Stimulus Stimulus Stimulus

inputs inputs inputs

Mental ' AllocatedX Mental /

/Allocated Allocated\ Mental
( Modality \
resources I to ) resources 1 resources
. t0 t0 2 /1
available 'V Task 1 / available \\ Task 1 Task available

Possible activities Possible activities Possible activities Possible activities

selected selected selected selected

Actual Actual Actual

responses responses responses

FIGURE 15-11 Attentional resource models. These models suggest that attention is limited by the
mental resources available. Parts A and B show a single resource model. Notice that in B, where attention
must be divided between two tasks, fewer resources are available for Task 1 than in A where attention is
undivided. This would suggest decreased performance for Task 1 under divided attention. In C, we have a
model where there are separate resources available for different modalities or task types. Whether tasks
interfere with each other thus depends on whether they require the same or different modalities.

GLOSSARY Controlled processing A type of information pro¬

cessing characterized by a serial, capacity-limited, vol¬
untary comparison of stimulus items with target repre¬
The following definitions are specific to this book. sentations.
Attentional gaze A metaphor for how attention is Cost When an invalid information cue results in
drawn or directed to stimulus cues. poorer perceptual performance.
Attentional resources The Rapacity for processing Covert orienting When attention is focused on a
stimulus inputs that can be “used up” by a task, result¬ stimulus in the absence of an overt orienting response.
ing in poorer performance in divided-attention situations. Detail set The level of detail, for example the rela¬
Automatic processing A type of information pro¬ tive size of elements, for which the attentional gaze is
cessing characterized by parallel, capacity-free, and in¬ set.
voluntary comparison of stimulus items with target rep¬ Dichoptic presentation Presentation of different in¬
resentations. formation to each of the two eyes.
Benefit When a valid information cue results in im¬ Dichotic listening A technique in which two differ¬
proved perceptual performance. ent messages are simultaneously played through ear¬
Binocular rivalry If the views presented to the two phones, with a different message to each ear.
eyes are different enough, we see only one or the other Divided attention Attention directed toward more
of them but not both. than one source of stimulus information or more than one
Conjunction search A type of search for a target perceptual task.
defined by a conjunction, or particular combination, of Early selection Attentional selection occurs imme¬
features each of which is also possessed by some detrac¬ diately following sensory registration, before the mean¬
tors. ing of an input can be determined.
 Glossary 453

Expecting When an observer is in possession of ad¬ Phonemic shadowing When a listener must repeat
vance information about where or when a stimulus event each syllable of a shadowed message as it occurs.
will happen. Phrase shadowing When a listener is allowed to lag
Extent The area of the visual or auditory field over somewhat behind a shadowed message and repeat entire
which the attentional gaze is spread. phrases at once.
Feature search A type of search for a target when Saccades High-speed ballistic eye movements that
the target differs from all distractors by possessing a fea¬ facilitate exploration of the visual field.
ture they don’t have. Searching Scanning the environment for particular
Filtering Attending to a single information channel features or combinations of features.
and attempting to ignore others. Serial search A pattern of visual search in which
Focused attention Attention directed toward only a items in an array are compared one at a time with a target
single source of stimulus information or a single percep¬ representation.
tual task. Shadowing When listeners are asked to repeat the
Hemifileld One half of the visual field, usually as di¬ verbal input they are receiving, usually in a particular
vided vertically through the fovea. ear; used to study filtering and divided attention.
Information channel A separable source of stimulus Skill An approach to attention that emphasizes learn¬
information, such as each of the two ears or a particular ing how to process stimuli optimally, rather than shifting
spatial location in the visual field. between modes of processing, as an explanation for good
Information cue Advance information about where divided-attention performance.
or when a stimulus event will happen. Stimulus cue A conspicuous stimulus somewhere in
Late selection All information entering a sensory the visual or auditory field toward which attention can be
system gets preliminary analysis and selection occurs at drawn or directed.
the stage where material is entered into longer-lasting Stroop effect The difficulty of observers to eliminate
memory. meaningful but conflicting information from a task even
Locus A particular spatial location to which the at¬ when that information is irrelevant to the task.
tentional gaze has been drawn or directed. Structural theories Theories of attention that em¬
Orienting When attention is drawn toward a sudden phasize a structural limitation on the ability to attend to
change in the environment. Often accompanied by an multiple perceptual inputs.
orienting reflex. Vigilance Maintaining overt attention to a perceptual
Orienting reflex A constellation of responses to a task, often with infrequent stimulus events, for prolonged
novel or dramatic stimulus, including the orienting re¬ time periods.
sponse and various physiological changes such as pupil Visual capture When sound seems to be originating
dilation and heart-rate decrease. from a spatial location where visual movement is occur¬
Orienting response When an observer turns toward ring, as in ventriloquism.
and orients sensory receptors toward a novel or dramatic Yerkes-Dodson law The principle that arousal and
stimulus. performance are related, with the best performance oc¬
Parallel search A pattern of visual search in which curring for a medium amount of arousal.
all the items in an array can be compared to a target
representation at the same time.
CHAPTER
16
Development
PERCEPTION IN INFANTS
Development of the Visual System
Infant Visual Psychophysical Methods
Eye Movements and Spatial Vision
Visual Acuity
Brightness and Color
Pattern Discrimination
Infant Hearing
Touch and Pain in Infants
Taste and Smell in Infants
PERCEPTUAL CHANGE THROUGH
CHILDHOOD
Attention and Search
Filtering
Encoding and Memory
PERCEPTUAL CHANGE IN ADULTS
Visual Function and Aging
Age Effects on the Other Senses
Global Changes in Perceptual
Performance
456 Chapter 16/Development

T he camp counselor turned to the newest arrival

and asked, “And how old are you, son?”
to both viewpoints is the conclusion that, despite
the fact that you may not be aware of it, your per¬
“Well,” said the boy, “it all depends. Ac¬ ceptual behavior is continually changing. Your ex¬
cording to my latest set of anatomical tests I’m 7. perience of the world differs from individuals who
According to my physical dexterity test I’m 10. are 10 years older or 10 years younger than you.
I’ve got a mental age of 11, a moral age of 9, and Because the developmental and perceptual learning
a social age of 10. If you are referring to my chron¬ approaches use different techniques and often ad¬
ological age though, that’s 8, but nobody pays any dress somewhat different theoretical issues, we deal
attention to that these days.” with these areas in separate chapters, beginning
Although you might not relish the thought of here with the developmental approach and proceed¬
spending a summer with this child, his comments ing to the effects of learning and experience in
point out that there are significant changes in many Chapter 17.
of our physical and psychological characteristics as
we age. Each of these changes has its own time
course. Some changes simply represent physiolog¬ PERCEPTION IN INFANTS
ical transformations occurring as the body matures
(such as a person’s anatomical age). Others repre¬ Before speaking about how perception changes as
sent patterns of behavior that are learned as the in¬ we age and develop, we must first know what per¬
dividual grows older (such as social or moral age). ceptual capacities we had at the moment of birth.
Still others may represent a combination of both Unfortunately, newborn infants (neonates) are dif¬
learning and maturation (such as mental age). Al¬ ficult to test. They sleep most of the time, and they
though no one refers to a perceptual age, there are do not respond to instructions or answer our ques¬
also changes in perceptual characteristics that tions in any direct verbal fashion. Finally, they
occur as an individual develops and matures. produce only a limited range of observable behav¬
These changes are usually improvements producing iors. These problems require experimenters to be
perceptual experiences that more accurately repre¬ rather ingenious in devising measures of the per¬
sent the physical environment. However, there are ceptual abilities of the very young, and different
also some perceptual capacities that deteriorate techniques may produce somewhat different pic¬
with age. tures of the developmental process (Teghtsoonian,
In considering how an individual’s perceptual 1987; Trehub & Schneider, 1987). These same
functioning changes we can adopt two different problems also often force researchers to use animal
perspectives. The first is long-term, viewing people subjects rather than humans, especially if direct
over their entire life span. This is the developmen¬ physiological measures of functioning are desired.
tal approach, which assumes that knowledge of a
person’s chronological age will allow us to predict
Development of the Visual System
many aspects of perceptual behavior. The other
approach is short-term, viewing the changes that Let us begin by looking at the physiology of the
occur in perceptual responses as a result of a cir¬ infant’s visual system. In comparison to the rest of
cumscribed set of experiences. This is the the body, the size of the eye changes very little
perceptual learning approach. It is based on the after birth. The body may increase in size about 20
presumption that our interactions with the world times, but the eye merely doubles in volume, with
can shape our percepts. These two approaches are the length from the cornea to the retina growing
not mutually exclusive; understanding the nature of from about 16 to about 24 mm (Hickey & Peduzzi,
perception often requires us to use both. Common 1987). The infant’s retina contains rods and cones
 Perception in Infants 457

as does the adult’s. Electrical measures indicate data. If we measure the physiological functions of
that these receptors are functioning from birth, al¬ the various sites in the visual pathways of the cat
though the responses may not yet exactly match at the time when the animal first opens its eyes, we
those of older children or adults (Aantaa, 1970; get results like those in Table 16-1 (see Hickey &
Maurer, 1975). Anatomically, however, the retina Peduzzi, 1987; Imbert, 1985; Norton, 1981a). The
still seems immature (Banks & Salapatek, 1983). table shows that a number of adultlike and imma¬
For instance, the region of the central fovea is not ture response patterns coexist in the newborn cat.
well defined in a 1-week-old infant (Abramov, Thus, in the retinal ganglion cell, we find the ex¬
Gordon, Henderson, Hainline, Dobson & La Bos- pected center-surround arrangement of excitatory
siere, 1982). Visual development seems to take and inhibitory responses; however, the receptive
place first in the central retina, with the peripheral fields differ in size from those of the adult and
portion maturing several weeks later (Banks & Sa¬ there is a general sluggishness in the response
lapatek, 1983; Russoff, 1979). (e.g., Russoff & Dubin, 1977).
Knowledge of the status of the visual path¬ In Chapter 3 we discussed two different visual
ways in newborns and infants comes mostly from response types, sustained and transient, char¬
animal studies, with the cat providing most of the acterizing X- and Y-type retinal ganglion cells,

Table 16-1. The Functional Condition of Various Sites in the Visual Pathways of
the Newborn Cat

Adultlike Responses Immature Responses

Retinal Ganglion Cells

Center-surround organization of receptive Low activity level

fields Overly large receptive fields
Adult percentage of on/off center Slow responses to light and weak inhibition
X vs. Y responses not clear

Lateral Geniculate Nucleus

Normal visual-field mapping Low activity and silent areas

Binocular separation of inputs Large receptive field diameter
Slow, sluggish, fatigable responding

Superior Colliculus

Normal visual-field mapping Slow, sluggish fatigable responses

Center-surround receptive fields Large receptive fields
Adult percentage of on/off center No movement direction sensitivity

Striate Cortex

Normal visual-field mapping Sluggish, fatigable responses

Adult separation of responses by eye of Many silent cells
input Fewer or absent orientation and direction-
selective cells with broader tuning
No binocular disparity cells
458 Chapter 16/Development

respectively. These appear to involve different the full adult pattern of response clearly is not pres¬
types of information processing, with the sustained ent (Banks & Salapatek, 1983). Some of these
(X-cell) responses associated principally with detail statements are species specific, and humans appear
vision and the transient (Y-cell) responses special¬ to develop somewhat more slowly than cats do.
ized for movement and rapid response. These two Thus, whereas cats show separation of the two eyes
systems are characterized both by response pattern into ocular dominance columns from birth, humans
differences and by different pathways to the cortex. may take 4 to 6 months to develop similar complex
In the infant cat, however, the two response types neural structures (Hickey & Peduzzi, 1987).
are not as well defined, and the difference between It should be clear from this discussion that
the sustained response and transient response sys¬ many of the characteristics of the visual system
tems is much reduced (Hamasaki & Sutija, 1979; mature at different rates. The X pathways to the
Mooney, Dubin & Russoff, 1979). cortex may mature more quickly than the Y path¬
At the lateral geniculate nucleus, we do find ways (Maurer & Lewis, 1979). Also, the responses
the adult correspondence between retinal response of the neonate are slower and less vigorous than
location and geniculate response location, the sep¬ those of the adult, suggesting that the quality of
aration of the inputs from the two eyes into clearly visual information reaching the higher centers of
defined layers, and some evidence of the X-like the newborn’s brain may be somewhat poorer (see
sustained response types. However, many of the Movshon & van Sluyters, 1981). This leads to dif¬
cells in the geniculate simply don’t seem to respond ferent perceptual functions appearing at different
to any sort of visual input, responses are gener¬ times during development, as we will see shortly.
ally slow and fatigue easily, and the transient Y-like In humans, we can determine how well the
responses appear to be absent (Daniels, Pettigrew visual cortex of the infant is functioning by mea¬
& Norman, 1978). A somewhat similar pattern suring what is called the visually evoked potential
emerges for the superior colliculus, with the gen¬ (often abbreviated VEP). This is a change in the
eral topography resembling that of the adult and electrical activity of the brain in response to a vi¬
with a center-surround organization of responses. sual stimulus, which is correlated with some as¬
However, again, receptive field size is too large pects of visual detection and pattern identification
and responses are slow and weak and not particu¬ (Cannon, 1983). The VEP is usually recorded by
larly direction sensitive (Norton, 1981a). fixing electrodes (generally fiat pieces of silver) to
Finally, at the level of the primary visual cor¬ the scalp and connecting them to very sensitive am¬
tex, we find that the two eyes’ inputs do separate plifiers. Almost all newborn infants (and even most
into the expected columnar arrangement discussed premature infants) show some VEP, although it
in Chapter 3, and that directional and orientation- differs somewhat from the adult response in its pat¬
sensitive cells (both simple and complex) are some¬ tern, size, and speed (Ellingson, 1968; Umezaki &
times present. However, there appear to be fewer Morrell, 1970). Over a period of about 3 months,
of these feature-specific cells, and even when they the infant’s electrical responses to visual stimuli
are found, responses are slow and easily fatigable come to look more and more like those of adults
(Imbert, 1985). Binocular-disparity-sensitive cells (Atkinson & Braddick, 1981; Banks & Salapatek,
seem to be almost absent until several weeks of age 1983; Harter & Suitt, 1970; Jensen & Engel,
(Blakemore & van Sluyters, 1975; Fregnac & Im¬ 1971). It is generally agreed that during the first
bert, 1978; Held, 1985). Overall, many of the year of life the visual system may continue to ma¬
characteristics of the adult system seem to be pres¬ ture, and that, although it begins to show many
ent or anticipated in the newborn visual system, but adult capabilities by the end of the second year
 Perception in Infants 459

(Ellingson, Lathrop, Nelson & Donahy, 1972; chamber (either on its back or in an infant chair).
Movshon & van Sluyters, 1981), the brain contin¬ Visual stimuli are then placed on the walls or the
ues to develop until the child is 10 years of age or roof of the chamber, and there is a tiny hole
older (Imbert, 1985). through which the experimenter can watch the
baby. An apparatus similar to Fantz’s is shown in
Figure 16-1. When the baby views one of a pair of
Infant Visual Psychophysical Methods
stimulus patterns placed in the chamber, the exper¬
Methods of testing infants’ visual capacities must imenter determines which one by noting where the
be very carefully devised, since we can’t use verbal eyes turn and then triggers a timer that records how
instructions or obtain verbal responses from them. long the infant views each of the two stimuli. If the
The researcher’s only recourse is to use existing baby looks at one target longer than the other, this
behaviors, which, for perceptual research, usually is taken as indicating a preference for it. The exis¬
involves attending behaviors. These include eye tence of a preference implies that the infant can
movements, head turns, visual following behavior, discriminate between the patterns. Unfortunately,
and other subtle indicators (Banks & Dannemiller, this simple result does not tell us why the baby
1987). Given the limited response repertoire avail¬ preferred to look at one stimulus rather than the
able to a young baby, we can appreciate the meth¬ other, nor can we be sure that the absence of a
odological breakthrough accomplished by Fantz preference means that the baby cannot discriminate
(1961). His procedure, called preferential look¬ between the two stimuli.
ing, involves first placing a young baby in a special There have been many elaborations of this

FIGURE 16-1 An apparatus for monitoring how long infants view particular stimuli.
460 Chapter 16/Development

technique, such as the one by Teller (1981) called much better, as shown by the fact that infants can
the forced-choice preferential looking technique. move their eyes so as to bring targets onto or close
Her procedure allows the investigation of the detec¬ to their foveas. Thus, if we present a young infant
tion of stimuli as well as discrimination between (about 2 weeks of age) with a target that suddenly
stimuli. Here, the infant is presented with only a appears 15 deg or 20 deg from the fovea, he will
single stimulus, while its response is monitored by turn his eyes in the direction of the stimulus (Tron-
a hidden observer or TV camera. If, on the basis ick, 1972). As the infant grows older, he will direct
of the infant’s head and eye movements alone, the his eyes toward targets that appear even farther
observer can correctly tell whether the test target away in the periphery (Aslin, 1987; Harris &
was presented to the left or right side of the screen, MacFarlane, 1974). Furthermore, 3-month-old in¬
it is presumed that the information concerning the fants seem to be able to identify targets in the pe¬
position of the target has been transmitted from the riphery of their visual field well enough to guide
screen through the infant’s visual system and be¬ their eyes to selected or preferred stimuli (Maurer
havior to the observer. At the minimum, this sug¬ & Lewis, 1979).
gests that the infant can see the stimulus. Although infants will look at a target that
A further variation of monitoring an infant’s flashes on to the side, or suddenly moves to one
looking behavior allows us to see if an infant can side, infants’ eye movements are not exactly like
notice any difference between stimuli. Again, only those of adults. There are two types of voluntary
one stimulus is presented and the viewing behavior eye movements, each taking time to develop fully.
is monitored. At first the infant will spend a good The first is saccadic eye movements, which are
deal of time looking at the stimulus, but as time fast, sharp movements from one target to another
passes it will cease to pay any attention to it. This that occur when you direct your attention toward a
process is called habituation. If we now present a target. In adults, a saccade will start the eye mov¬
different stimulus, the baby will once again look. ing toward a target displaced to the side within
The presence of renewed looking at the stimulus 200 to 250 msec, and usually come within 5 or 10
suggests that the infant can see that something has percent of the distance needed to center the image
changed, and that the present stimulus is different on the fovea in a single movement (e.g., Komoda,
from the previous one (e.g., Kellman & Spelke, Festinger, Phillips, Duckman & Young, 1973). A
1983; McCall, 1979). typical adult eye movement to a target 30 deg to
These looking techniques in conjunction with the side is shown in Figure 16-2. Infants are much
the monitoring of eye movements and physiolog¬ slower to begin the saccade, and tend to make a
ical responses (such as the VEP), and a few more series of small saccades, often not reaching the tar¬
esoteric methods, provide the main methods for get for well over a second or more, as shown in
measuring infant visual capacities (Banks & Danne- Figure 16-2 (see Aslin, 1987).
miller, 1987). The other type of eye movement is the smooth-
pursuit eye movement used to track a steadily
moving object, such as a ball flying through the air
Eye Movements and Spatial Vision
or a person on a swing, where the eyes track the target
In Chapter 10 we saw that certain aspects of spatial with a uniform and even motion. Smooth-pursuit eye
vision, such as the binocular perception of depth or movements do not appear in newborns, they use
distance, are not present at birth but take time to short, jumpy, saccadic eye movements to track
develop (e.g., Held, 1985; Yonas & Granrud, smoothly moving objects. Thus, rather than keeping
1985a). The perception of direction, however, is pace with the moving target infants seem to attempt to
 Perception in Infants 461

O)
0>
T3,
C
o
'55
o
Q.
<D
>.
LU

to one side Time (msec)

of viewing
position

FIGURE 16-2 A typical adult eye movement to a target appearing 30 deg to one side of fixation will
involve a single, fast, large saccade and a small corrective flick, whereas an infant will have a longer delay
before moving, and the movement will involve a series of short eye movements.

grab a fixed glimpse of it, wait till it drifts from A generally similar (albeit not as smooth) pattern
view, and then attempt to look at it again, so pro¬ of eye movements is found in infants younger
ducing a set of little stepwise movements rather than 5 days (Kremenitzer, Vaughan. Kurtzberg &
than a smooth tracking. The more adultlike pattern Dowling, 1979). In fact its appearance is so reli¬
of smooth movement begins to emerge at 8 to 10 able that the absence of optokinetic nystagmus is
weeks of age (Aslin, 1981a). used as an indication that there may be neurological
The fact that infants move their eyes in re¬ problems (Brazelton, Scholl & Robey, 1966). This
sponse to moving or displaced stimuli can be used eye-movement pattern seems to be automatic or re¬
to measure other capacities in the newborn. For in¬ flex in nature, rather than voluntary, and it seems
stance, if we show an adult observer a continuously to be controlled by the tectopulvinar system we de¬
moving pattern (such as a screen full of stripes all scribed in Chapter 3 (Atkinson & Braddick, 1981;
moving in one direction) we get a characteristic Hoffmann, 1979). The perceptual scientist can use
eye-movement pattern. The eye will smoothly track these reflex movements to measure other aspects of
in the direction of the movement for a distance, and visual function. If an infant cannot see a pattern of
then snap back in the opposite direction. After this moving stripes (because they are not large enough
return movement the observer’s eyes lock onto an¬ or lack sufficient contrast), it will not be able to
other stripe and follow it, and this process repeats track the moving pattern. This technique has been
itself while the observer views the array. This re¬ used to study brightness discrimination, visual acu¬
petitive eye movement sequence in the presence of ity, and motion perception (see Banks & Salapatek,
a moving pattern is called optokinetic nystagmus. 1983).
462 Chapter 16/Development

Visual Acuity Brightness and Color

Many studies have shown that the visual acuity of A variety of techniques has been used to assess the
infants is rather poor, but improves steadily with basic sensitivity of infants to brightness and color.
age. A number of different methods have been used The findings show that 3-month-old infants are
in these studies. For instance, the optokinetic re¬ about 10 times less sensitive to light, both under
sponse can be used to test the visual acuity of in¬ dark-adapted (scotopic) and light-adapted (pho-
fants by finding the narrowest width of stripes that topic) conditions, than are adults, and 1-month-old
will still produce the tracking response. Alterna¬ infants are about 50 times less sensitive (Peeples &
tively, one of the preferential looking procedures Teller, 1978; Powers, Schneck & Teller, 1981).
can be used. The levels of acuity found for infants However, despite this difference, infants like adults
may vary with the technique (Teller & Movshon, are still exquisitely sensitive to small amounts of
1986) or the specific acuity stimuli used (Shimojo light. For example, in Chapter 4 we found that an
& Held, 1987), but there is a general agreement adult can detect an input of as little as 6 quanta of
that visual acuity is originally quite poor, often light hitting anywhere in a patch of 1,300 rod re¬
around 20/800 (6/240 in metric units) for neonates ceptors. In comparison, a 3-month-old infant
(which is less than that needed to see the single would need an input of about 40 quanta of light
big E on a standard Snellen acuity chart). New¬ over the same region, and a 1-month-old infant
borns act as if they had limited accommodative would detect an input of about 100 quanta of light
ability, with their lenses fixed somewhere around a (see Teller & Bomstein, 1987), still a very minimal
focus point of 20 cm (White, 1971). The child’s amount of light.
acuity increases steadily with age, finally reach¬ A number of studies show that despite differ¬
ing average adult levels at around 3 or 4 years of ences in absolute sensitivity, the relative sensitivity
age, as shown in Figure 16-3. If you have access of infants and adults to different wavelengths of
to a young infant you can see the effect of this light is about the same. Both are most sensitive to
limited accommodation by trying Demonstration middle wavelengths, and exhibit a gradual decrease
Box 16-1. in sensitivity to longer and shorter wavelengths

C
0
(0
>
‘d
cr
<D

<D
C

3
o
<

Age (months)

FIGURE 16-3 Steady improvement in visual acuity from birth to 3 years of age.
 Perception in Infants 463

DEMONSTRATION BOX 16-1. Infant Accommodation

To demonstrate that an infant’s accommodation is the child’s head and eyes and notice that the infant
limited to close objects, you will, of course, need an will track, or at least try to track, the pencil. Now
infant, preferably 2 months of age or younger. If you repeat this, only vary the distance to 1 or 2 meters
can find one, catch its attention and then slowly move away from the child. At this distance you should have
a pencil from side to side near the infant’s face. Use exceeded the ability of the infant to accommodate and
a distance of about 20 cm, or around 8 inches. Watch you should notice that little, if any, tracking occurs.

(Dobson, 1976; Moskowitz-Cook, 1979; Werner, nate between different patterns (Fantz & Miranda,
1979). This does not mean, however, that infants 1977). This means that the optical and neural bases
have color vision equivalent to that of adults. In of pattern vision do not abruptly become functional
general, young infants do show the ability to dis¬ after the full term of pregnancy, the age at which
criminate between colors (Bomstein, 1985; Werner babies can first be ordinarily observed, but rather
& Wooten, 1979). However, color vision matures, have already matured to a reasonable degree of
and discrimination performance improves. Varner, function prior to the normal birth time.
Cook, Schneck, McDonald, and Teller (1985) re¬ Preferential-looking studies have shown that
ported that most 1-month-old infants fail to make a young infants can discriminate among a variety of
discrimination between a pair of short-wavelength different types of patterns. For instance, in one ex¬
stimuli, thus acting much like tritanopic color¬ periment newborn infants were shown pairs of tar¬
blind individuals (see Chapter 5), whereas most 2- gets. These neonates showed a clear preference for
month-old infants do make this discrimination. viewing patterns of stripes over a simple square,
Although there is good color discrimination be¬ and also preferred patterns with high contrast
tween the long and middle wavelengths of light between the figures and the background. They
(red and green) for the 1-month-old infant, the showed a preference for larger patterns, indicating
short-wavelength (blue) discriminating mechanism that they can discriminate size, and also preferred
seems to be immature (see Teller & Bomstein, patterns containing many rather than few elements.
1987). Thus, at birth, infants do have brightness In addition, they showed some ability to discrimi¬
and color vision, but brightness sensitivity lags be¬ nate certain aspects defining contours, such as
hind that of the adult and color perception is still curvature, by preferring curved to straight-line ele¬
in the process of developing. ments. Figure 16-4 shows some sample forms, and
the star indicates those most preferred by newborns
for each pair (Fantz & Yeh, 1979).
Pattern Discrimination
Generally, infants prefer moderately complex
The preferential-looking technique has been used stimuli over those that are very simple or very
extensively to explore pattern perception in infants. complex, although preferences do change with age
Using this procedure it is possible to show that (Karmel & Maisel, 1975). Younger infants prefer
premature infants, bom 1 to 2 months prior to a simple patterns with highly contrasting elements,
full-term gestation, still often preferentially look at whereas 5-month-olds can make more subtle dis¬
patterned stimuli rather than plain ones of equal tinctions in contrast and configuration (Fantz &
average brightness, and also sometimes discrimi¬ Yeh, 1979). Banks and Salapatek (1983) suggested
464 Chapter 16/Development

■ ■

□
□
■ ■

□
□
★
Pattern presence Higher contrast

■ ■ ■

■ ■

■ ■ ■

★
Larger size Larger number Curved vs. straight

FIGURE 16-4 Patterns most looked at by newborns are indicated with a star for each pair of stimuli.
(Based on Fantz & Yeh, 1979)

that pattern perception in infants reflects the devel¬ habituated, and then some test patterns. The pat¬
oping ability to discriminate various spatial fre¬ terns accompanied by a plus sign are changes that
quencies (see Chapter 4). the infant would be expected to notice; that with
Preferences in viewing also show that some minus sign is a change that would not be noticed
higher-level aspects of pattern perception are pos¬ (see Dodwell, Humphrey & Muir, 1987).
sible for the young infant. Infants can discrimi¬ Certain meaningful patterns receive special at¬
nate the orientation of patterns within the first few tention, even from neonates. A number of re¬
weeks (Maurer & Martello, 1980), and perhaps searchers have studied the response of infants to
even on the first day of life (Kessen, Salapatek & targets that approximate the human face. One com¬
Haith, 1972). Furthermore, they seem to be aware mon procedure is to use some targets that are only
of certain forms of symmetry, or its absence (Born- head-shaped, others containing only some facial
stein, 1981). Although infants respond to both the features (such as a hairline or eyes), some contain¬
size and position of stimuli, at age 4 months they ing scrambled facial features, and others that ac¬
are relatively insensitive to changes in the configu¬ tually look like faces. Samples of such stimuli are
ration of the stimuli (Humphrey, Humphrey, Muir shown in Figure 16-6. In general, it is found that
& Dodwell, 1986). Furthermore, 3- to 4-month-old by 2 months of age infants prefer to look at stimuli
infants seem to pay attention to specific features that contain facial features arranged in the normal
(such as whether the dots making up a pattern are configuration rather than scrambled facial stimuli,
square rather than round) instead of to global whereas children younger than 1 month do not
changes in the configuration (such as the pattern the make this discrimination (Carey, 1981; Haaf, 1977;
dots make); by 6 or 7 months they are responding Maurer & Barrera, 1981). Between 1 month and 4
to these global changes as well (Dineen & Meyer, months infants begin to take note of certain features
1980). Figure 16-5 summarizes the sensitivity of in the facelike stimulus. By about 10 or 12 weeks,
the 4-month-old infant to various aspects of visual infants notice and recognize changes in the hairline
patterns. It shows a pattern to which the infant is and eyes, although changes in mouth and nose con-
 Perception in Infants 465

Habituation pattern

Size change
+ change + change change
+

FIGURE 16-5 In a habituation experiment where 4-month-old infants habituate to the top pattern, they
act as if they recognize changes in the pattern dimensions indicated by a plus (+) sign, but do not for the
change in pattern indicated by the minus (—) sign.

figurations go unnoticed (Caron, Caron, Caldwell Infant Hearing

& Weiss, 1973). However, the configurational and
specific features picked up by infants only 1 month The ears of infants are functional at birth, but the
of age do seem to be sufficient to permit the infant auditory cortex is still rather immature and contin¬
to discriminate its own mothers face from that of ues to develop over the first year (Kuhl, 1987). A
a stranger (Maurer & Salapatek, 1976), which sug¬ number of studies have suggested that infants less
gests that young infants can discriminate among than 6 months of age have higher absolute thresh¬
certain classes of fairly complex patterns. olds than those of adults (Berg & Smith, 1983;
A general summary of the child’s increasing Trehub, Schneider & Endman, 1980). An interest¬
visual competence, when certain skills or the abil¬ ing feature of these data is that the differences are
ity to resolve certain features or cues in the envi¬ most noticeable in the frequency range below
ronment appear, is presented in Figure 16-7. 10000 Hz. The ability of adults to detect tones in

FIGURE 16-6 Schematic and scrambled facelike stimuli.

466 Chapter 16/Development

Pictorial interposition Familiar size Shading cue

Gestalt organization JT Size constancy

<D Relative size jr Pictorial depth

o
c
®
<» Kinetic depth Jr Pattern symmetry Biological motion
CL
E jt Convergence
o
o Binocular disparity

Smooth pursuit Face recognition

Collision avoidance jr Accommodation

jr Directional saccades
Motion w Red-green color vision
Red-areen Scotopic
Sc vision Gross brightness perception

Birth
Age (months)

FIGURE 16-7 Various visual abilities, and sensitivity to various environmental or stimulus dimensions,
appear at different ages.

this range is nearly twice as good as that of infants You probably recall from Chapter 7 that there
(Olsho, 1984). At the higher frequencies, however, are several binaural cues that help to indicate the
infants show more adultlike sensitivity (Kuhl, direction of a sound relative to the listener. The
1987). two most important of these are the time differ¬
Newborns seem to be able to indicate their ences in the arrival of low frequency sounds to the
ability to localize the direction of a sound source two ears (first to the closer ear), and the intensity
by turning either their head or eyes (Butterworth, differences between the two ears caused by the lack
1981; Muir & Field, 1979). Probably the youngest of bending of higher frequencies of sound around
child tested for this ability was by Wertheimer the head (see Green, 1976; Moore, 1977). Which
(1961). A mere 3 minutes after birth, with the in¬ of these cues is most effective for the infant? By
fant lying on her back, a loud click was sounded directly controlling both the time differences be¬
next to her right ear or left ear. Two observers tween the ears and the intensity of sound reaching
noted whether the eyes moved to the infant’s right the two ears, Clifton, Morrongiello, and Dowd
or to her left, or not at all. On 18 out of the 22 (1984) demonstrated that newborn infants, and
times when the child’s eyes moved, they moved in those up to about 9 weeks of age, respond to inten¬
the direction of the click. When the experiment was sity differences between the two ears by turning in
completed the child was still only 10 minutes old; the direction of the sound. At this age, the more
hence these data allow us to conclude that some complex time discrimination cue is not adequate to
directional aspects of auditory stimuli are accu¬ induce the child to turn its head in the appropriate
rately processed and are capable of guiding behav¬ direction, but by age 5 months both cues are effec¬
ior from birth. tive and cause the child to look in the direction of
 Perception in Infants 467

DEMONSTRATION BOX 16-2. Auditory Localization in Infants

If you have access to an infant, auditory localization or a toy “clicker.” Watch the infant’s head and eyes.
is easily demonstrated. Simply look squarely at the You should see the eyes flick in the direction of the
child and then make a sharp sound near one ear. sound, or you may see the head turn in the direction
Good sound sources are a rattle, a snap of the fingers. of the stimulus.

the sound source. If you have access to an infant. impact believed to persist as long (e.g., Eland &
Demonstration Box 16-2 will show you how to Anderson, 1977). This has led to the practice of
demonstrate auditory localization. giving little treatment for pain to babies during or
after major medical procedures and operations (see
Liebeskind & Melzack, 1987; Owens, 1984). Un¬
Touch and Pain in Infants
fortunately, recent evidence (e.g., Grunau & Craig,
Generally speaking, touch and heat sensitivity ap¬ 1987) suggests that this is not the case, and infants
pear to be among the first sensory modalities to appear to be just as susceptible to the perception of
emerge during the course of fetal development pain as adults.
(Hall & Oppenheim, 1987). This can be demon¬
strated through certain reflexes, which show the
Taste and Smell in Infants
ability to feel and to localize touch stimuli imme¬
diately after birth. For instance, there is the rooting Taste receptors start to form early in fetal life and
response, in which a child will reflexively turn its are apparent as early as 13 weeks after conception
head in the direction of a touch to the cheek. This (Bradley & Stern, 1967). In general, then, neonates
response helps the child to locate the breast for appear to be as well equipped with taste receptors
feeding. Demonstration Box 16-3 shows you how as adults. However, they respond to the taste pri¬
to elicit this directional response. maries differently (Crook, 1987). Using sucking
There is a widespread belief among many cli¬ responses as an indicator, Lipsett (1977) found a
nicians and other investigators that because the cor¬ preference for sweet stimuli in newborns. Even
tex is not fully developed in the neonate infants do small differences in the concentration of sweetness
not experience pain as severely as adults, nor is its produced differences in neonatal reactions. How-

DEMONSTRATION BOX 16-3. The Rooting Response

The easiest method to show tactile sensitivity and lo¬ in older infants). To demonstrate tactile localization
calization in infants is to elicit the reflex called the ability you should stroke the infant’s cheek lightly
rooting response. To see how early this ability exists, with your finger. If you stroke the right cheek the
a very young baby of less than 2 months of age infant should turn to the right. If you stroke the left
should be used (although the response can be elicited cheek the infant should turn to the left.
468 Chapter 16/Development

ever, infants less than about 4 months of age seem around the age of 2 months (Atkinson & Braddick,
to be insensitive to the taste of salt (Beauchamp & 1981; Maurer & Lewis, 1979), when there is a sud¬
Cowart, 1985). If we use strong enough concentra¬ den improvement in the child’s visual abilities.
tions, we can get some indication of responses to Acuity increases markedly (Braddick & Atkinson,
sour and bitter stimuli by watching for differences 1979), tracking behavior becomes more adultlike
in facial responses (Ganchrow, Steiner & Daher, (Atkinson, 1979), the ability to recognize individ¬
1983). Thus, we can probably summarize neonatal ual elements surrounded by an enclosing contour
gustatory perception as clearly showing sweet sen¬ appears (Milewski, 1976), and infants begin to
sitivity, with less sensitivity for sour and bitter, and show more adultlike eye-movement patterns when
very weak or absent salt sensitivity (Crook, 1987). viewing figures (Hainline, 1978). By 3 months of
Work on infant smelling ability has involved age stereoscopic depth perception appears (Shea,
presenting newborns with a cotton swab saturated Fox, Aslin & Dumais, 1980), and this ability con¬
with some olfactory stimulus. The swab is placed tinues to improve over the first 2 years (Fox, Aslin,
under the infant’s nose and activities such as heart Shea & Dumais, 1980; Held, 1985). The ability to
rate, respiration, and general bodily activity are discriminate depth based on binocular disparity also
monitored using a polygraph (i.e., Engen, Lipsitt seems to improve throughout childhood and into
& Kaye, 1963). These studies have shown that in¬ early adolescence (Romano, Romano & Puklin,
fants can detect a number of strong odorants, such 1975).
as anise oil, asafetida, alcohol, and vinegar. How¬ Other basic visual processes also seem to de¬
ever, infants seem to have more sensitivity to velop rapidly over the first 2 years. Thus, visual
smells than originally thought. For example, new¬ acuity, which is originally quite poor, improves
born infants turn away from noxious odors and to¬ steadily into early childhood (Gwiazda, Brill, Moh-
ward pleasant ones (Rieser, Yonas & Wikner, indra & Held, 1980) and early astigmatic problems,
1976). This turning response has been used to show which serve to lower visual resolution in infants,
that infants respond to odorants of human body also disappear (Atkinson, Braddick & French,
origin. Babies less than 2 weeks old will orient 1979; Ingram & Barr, 1979). By 5 years of age
toward an object carrying their mother’s scent, children seem to have fully developed scotopic and
such as a breast pad (Cernoch & Porter, 1985; Rus¬ photopic visual systems, which show adaptation ef¬
sell, 1976). There is even the suggestion that, in fects and sensitivities equivalent to those of adults.
contrast to some of the other limitations on infant A similar pattern is found for the other senses.
sensory capacities, children actually may be con¬ Consider hearing as an example. Infants begin with
sidered to be more responsive than adults to human a substantial low-frequency hearing deficit and a
body odors (Filsinger & Fabes, 1985). lesser high-frequency deficit. Over the first 2 years
hearing improves quickly, especially for the low
frequencies, and the improvement then continues
PERCEPTUAL CHANGE THROUGH more gradually until about 10 years of age (Kuhl,
CHILDHOOD 1987; Trehub et al., 1980; Yoneshige & Elliott,
1981).
Throughout childhood there is a general improve¬
ment in perceptual discrimination, identification,
Attention and Search
and information processing. Many of these changes
occur fairly rapidly within the first year or two, and In addition to changes in basic sensory processes,
others continue over much longer time spans. there appear to be changes in the patterns of atten¬
The most dramatic changes seem to occur at tion and information encoding, which show up as
 Perceptual Change through Childhood 469

developmental changes in perception. Theorists Kessen, 1973). Their eyes seem to be “captured”
such as Hochberg (1981, 1982) suggest that what by the feature since they dwell on it for prolonged
is happening during the developmental process is periods. Because the gaze of a 1-month-old infant
that the way information is integrated over time is is caught by the first contour encountered, most of
changing. This notion of integration involves the the viewing time is spent focused on the external
construction of mental models of the perceptual sit¬ contours of a form. If the stimulus has internal fea¬
uation called schemata. In addition, it involves the tures, they are ignored or missed. This changes by
ability to select relevant information and to retain the age of 2 months. Now the infant scans the con¬
and compare this new perceptual information to the tours a little more and shorter periods are spent on
existing schemata. Simply put, this means that at¬ each feature (Banks & Salapatek, 1983; Hainline,
tention and memory come to play a role in the per¬ 1978; Salapatek, 1975). In addition, the infant
ceptual process. dwells almost exclusively on the internal features
As we learned in Chapter 15, one of the ways of the stimulus, seemingly ignoring the overall pat¬
we can observe the pattern of overt attention is to tern. These differences are shown in Figure 16-8.
monitor eye movements. Developmental theorists, The patterns of 3- and 4-year-olds are similar
such as Piaget (1969), have argued that patterns of to those of a 2-month-old infant. Children of this
eye movements provide some clues as to which
stimuli are being selected and compared by individ¬
uals of different ages. For instance, we know that
adults display a strong tendency to look at forms
that are informative, unusual, or of particular func¬
tional value (Antes, 1974; Friedman & Liebelt,
1981; Loftus & Mackworth, 1978). Thus, by mon¬
itoring eye-movement patterns in children, we are
observing the search component of visual atten¬
tion. Information about search should be helpful in
determining how children are viewing, and hence
constructing, their visual world.
We have already seen that infants from birth
to 2 months do make a variety of eye movements—
such as fixating stationary stimuli, tracking moving
stimuli, or guiding their eyes toward stimuli in the
periphery—although not as precisely as do adults.
More importantly, they typically do not move their
eyes to the most informative parts of the stimulus
(at least by adult standards) but rather seem to view
only limited parts of the stimulus, usually around a
border or comer (Day, 1975; Mackworth & Bru¬
ner, 1970). For instance, when there is a distinct
contour within the visual field of an infant, its eye
is drawn toward it. Infants of 1 month tend to di¬
rect their eyes toward one distinctive feature of a
visual stimulus, such as the comer of a triangle FIGURE 16-8 Eye movements typical of 1- and 2-
(Haith, Bergman & Moore, 1977; Salapatek & month-old infants.
470 Chapter 16/Development

age spend most of their time dwelling on the inter¬

nal details of a figure, with only an occasional eye
movement beyond the contour boundary. The 4- or
5-year-old child begins to make eye-movement ex¬
cursions toward the surrounding contour. At 6 and
7 years of age, there is a systematic scan of the
outer portions of the stimulus with occasional eye
movements into the interior. This development is
shown in Figure 16-9 (Zaporozhets, 1965).
Eye-movement patterns have important con¬
sequences for certain perceptual discrimination tasks.
Vurpillot (1968) monitored the eye movements of
children between the ages of 2 and 9 years. They
were presented with pictures of houses with differ¬
ent kinds of windows and were asked to indi¬
cate whether or not the houses appeared to be the
same, a task that required systematic compari¬
son of the windows. She found that the youngest
children did not conduct a systematic search.
Rather, they often continued searching through the
houses even after looking at a pair of windows that
were quite different. This lack of systematic view¬
ing was accompanied by a low degree of accuracy
in the discrimination judgments of the younger
children. Older children, with more regular and
systematic viewing patterns, were much more ac¬
curate. Similarly, Cohen (1981) found in a figure¬
matching task that 5- and 8-year-old children take
longer to decide where to move their eyes than
adults do in the same task. In addition, they make
more eye movements and are less likely than adults
to look directly at the matching target in their first
eye movements. It is likely that these differences
reflect differences in strategies of attention and in¬
formation pickup, rather than differences in visual FIGURE 16-9 Changes in eye movements from
capacity, since eye movements seem to be strongly ages 3 to 6. (From Zaporozhets, 1965. Copyright The Society
for Research in Child Development, Inc.)
affected by task demands, meaning, context, and
expectations (Antes & Penland, 1981; Findlay,
1981; Stark & Ellis, 1981). sists of a number of objects (fruits and vegetables)
Such differences in observing strategy may that are organized into a larger figure (a bird).
explain why as a child becomes older there is a Children 4 and 5 years old report seeing only the
gradual change in the way it comes to view patterns parts (“carrots and a pear and an orange”). By
and the elements that make them up (Elkind, the age of 7, children report seeing both the parts
1978). For instance, consider Figure 16-10. It con- and the global organization (“fruits and carrots
 Perceptual Change through Childhood 471

There is an interesting set of phenomena that

may show age changes in stimulus filtering more
graphically. These are responses to visual-geomet¬
ric illusions, which are simple line drawings in
which the actual size, shape, or direction of some
elements differs from the perceived size, shape, or
direction (see Coren & Girgus, 1978). We have
already encountered some of these illusions in
Chapters 1 and 14; two of them are shown in Figure
16-11. Figure 16-11A shows the Mueller-Lyer
illusion, in which the line marked jc appears to
be longer than the line marked y. Figure 16-1 IB
shows the Ponzo illusion, in which the line marked
w appears to be longer than the line marked z. This
is the case in spite of the fact that x and y are phys¬
ically equal in length, and w and z are also physi¬
FIGURE 16-10 A vegetable-fruit-bird figure used cally equal to each other.
to measure part-versus-whole perception in
One explanation for certain visual-geometric
children.
illusions is that the lines that induce the illusion
become confused with the test lines, thus causing
the distortion (Coren & Girgus, 1978). For in¬
and a bird”). By 8 or 9 years, the majority of
stance, in the Mueller-Lyer illusion (16-11 A), the
children respond in terms of both the parts and the
upper figure is actually longer if you measure from
global organization (“a bird made of fruits and
wing tip to wing tip. Thus, confusing the wings
vegetables”).
with the horizontal line might add to the distortion.
This idea is supported by the fact that focusing at¬
tention on the lines, and ignoring the wings, re¬
Filtering
duces the strength of the illusion (Coren & Girgus,
Attention involves more than simply searching for 1972b), whereas directing attention to parts of a
targets and scanning the environment. Several of figure can produce stronger illusions (Coren &
these other aspects have been shown to change sys¬ Porac, 1983a). If we accept this explanation, and
tematically with age (Enns & Cameron, 1987). For if children are poorer at filtering out extraneous
instance, selective attention involves the compo¬ stimuli, then we should expect that children will
nent we call filtering. This refers to the ability to show stronger visual illusions than adults. In gen¬
ignore irrelevant stimuli in the environment while eral, we do find that visual illusions are larger for
more task-relevant stimuli are being processed. A children and decrease with age, suggesting that
number of studies have shown that children are children are less able to ignore the inducing lines
more easily distracted by irrelevant stimuli (i.e., when making their judgments (Coren & Girgus,
Day & Stone, 1980). Thus, in a card-sorting task, 1978; Enns & Girgus, 1985; Pick & Pick, 1970).
both children and adults show poorer performance In one study, illusion magnitude (the degree to
if there are irrelevant as well as relevant features which individuals are susceptible to the perceived
present; however, children show a much greater re¬ differences in line length) was found to decrease
duction in efficiency than do adults (Well, Lorch & until about the age of 25 years (Porac & Coren,
Anderson, 1980). 1981).
472 Chapter 16/Development

FIGURE 16-11 Two illusions that show age-related differences in their perception: (A) the Mueller-Lyer;
(B) the Ponzo.

Encoding and Memory recognize faces continues for many years (Carey,
1981; Flin, 1980). Thus, we find that a rapid in¬
Another factor that can affect perceptual develop¬ crease in the ability of children to recognize unfa¬
ment is the change in encoding ability. For a child miliar faces occurs between the ages of 6 and 10.
to integrate and compare stimuli, he or she must To estimate the magnitude of this change we might
have the ability to encode them and to remember note that under conditions where a 6-year-old will
other instances of a stimulus or the schemata asso¬ only recognize a little more than half of the faces
ciated with a particular class of stimuli. Thus, it is previously shown her (60%), a 10-year-old will
not surprising that a number of studies have shown recognize nearly all of them (95%). Furthermore,
that the ability to discriminate between visual pat¬ improvement in ability can be observed even
terns improves with age (Cratty, 1979). through adolescence to the age of 16 (Carey, Dia¬
An interesting example of the improvement of mond & Woods, 1980).
pattern recognition with age, which is probably de¬ There is one form of pattern discrimination
pendent on improvements in encoding ability, in¬ error that seems to be characteristic of young
volves the recognition of human faces. The very children. This involves mirror reversals. Children
impressive achievements we noted above for 5- to confuse lateral mirror-image pairs (such as p and q,
7-month-old infants in discriminating face from or b and d) more frequently than up-down mirror-
nonface stimuli should not be interpreted to mean image pairs (such as p or b, or q and d) (Springer
that these infants have adultlike abilities in this & Deutsch, 1985). These confusions are quite
area. The development of the ability to encode and common in young children (around 3 years old)
 Perceptual Change in Adults 473

and gradually decrease until about the age of 10 or to the orientation of a stimulus. There is an inter¬
11 (Gaddes, 1980; Serpell, 1971). Some of the im¬ esting quirk associated with this issue. Consider a
provement seems to be associated with educational stimulus, such as a human face, that has a familiar
processes, since between the ages of 5 1/2 and 6 orientation. When a face is inverted it seems to lose
1/2 there is a sudden improvement in the ability to much of its facelike quality, and even very familiar
make these discriminations. It is likely that the im¬ individuals are difficult to identify when their pho¬
provement is caused by the formal instruction in tographs are turned upside down (Rock, 1974).
reading and writing that usually begins at about that Thus, it is not surprising to find that adult observ¬
age. With appropriate training, kindergarten-aged ers, who have had thousands of exposures to up¬
children can learn the left-right discrimination quite right faces, show greater accuracy of identification
well, although it still seems to be more difficult when faces are upright than when they are inverted
than the up-down discrimination (Clark & White¬ (Yin, 1970). However, the ability of 6-year-olds to
hurst, 1974). Enns (1987) has demonstrated that identify faces is the same regardless of whether the
the ability to make discriminations based on sym¬ face is presented in a normal or in an inverted po¬
metries in patterns depends on the development of sition. By the age of 10 years, children’s facial
memory encoding ability, rather than simply on vi¬ identification responses begin to look like those of
sual function. adults; in other words, identification ability is dis¬
When a child continues to have left-right con¬ rupted when the faces are inverted (Carey & Dia¬
fusions, he can experience problems later with mond, 1977; Carey, Diamond & Woods, 1980). It
reading. The specific term used for such reading seems that the adults’ sensitivity to orientation dif¬
disability (when it is not associated with other ferences makes it more difficult for them to identify
disturbances such as mental retardation, sensory inverted familiar faces than for 5- and 10-year-olds
impairment, or emotional problems) is dyslexia. tested on the same task. You can explore the adult
Estimates of the incidence of dyslexia vary widely, sensitivity to orientation in yourself by trying Dem¬
but it seems that the problem affects no less than 2 onstration Box 16-4.
percent of all children in Western countries, with
the incidence perhaps being as high as 10 percent
(Bannatyne, 1971; Gaddes, 1976; Spreen, 1976). PERCEPTUAL CHANGE IN ADULTS
This is a problem that seems to have a perceptual
rather than an intellectual basis. The dyslexic indi¬ Perceptual and sensory functions continue to
vidual may be highly talented in all other respects, change throughout the life span, although the rate
except for the reading problem. There are many of change is usually slower in adults than during
case histories of exceptional people who have been infancy and childhood. Also, the earlier changes
dyslexic, among them the inventor Thomas A. Edi¬ are toward increasing efficiency in perceptual pro¬
son, the surgeon Harvey Gushing, the sculptor Au¬ cessing, whereas the later changes, beginning
guste Rodin, United States president Woodrow around age 40, are toward decreased functioning,
Wilson, and the author Hans Christian Anderson. as sensory receptors age and neural efficiency drops
One characteristic of all children who have been (Corso, 1981).
diagnosed as dyslexic is that they show confusions
between the left-right mirror images of targets, al¬
Visual Function and Aging
though they have no problem with the up-down
mirror images (Gaddes, 1980; Newland, 1972; Sid- A number of structural and neural changes occur in
man & Kirk, 1974). the aging individual that might be expected to re¬
Children’s inability to discriminate letter re¬ duce visual sensitivity. For instance, the aging eye
versals suggests that they are relatively insensitive generally shows a smaller pupil size, hence less
474 Chapter 16/Development

DEMONSTRATION BOX 16-4. Orientation and Stimulus Recognition

Adults are more rigid in their reliance on normal ori- to identify, but if you invert the page you will see
entations than are children. Look at the two figures them suddenly become identifiable. This effect is es-
here and you will probably find them quite difficult pecially striking for faces and for handwritten script.

A B

light enters the eye (Corso, 1981; Weale, 1982).

The optics of the eye become less efficient since
the crystalline lens continues to become yellower
and darker (Coren & Girgus, 1972a), and even the
cornea yellows somewhat (Lerman, 1984). Both of
these factors also limit the amount of light reaching Dark adaptation
the retina. Of course, in the presence of reduced as a function of age
Age range: 16-89
effective light input due to such factors, we would
N, 240
expect a decrease in sensitivity. Between 20 and 70
years of age, we find a consistent decrease in
threshold sensitivities for the detection of spots of
Age
light (Fozard, Wolf, Bell, McFarland & Podolsky,
80-89
1977). This is particularly evident in dark adapta¬
tion. Although the time to reach minimum thresh¬ 70-79

old remains the same, the maximum sensitivity

eventually achieved decreases with age. This is
shown in Figure 16-12 (McFarland, Domey, War¬
ren, & Ward, 1960).
Some deterioration in vision may be due to 0 10 20 30 40 50 60

loss of receptors with age, which may account for Time (min)
the gradual deterioration in color vision. It seems
FIGURE 16-12 Age changes in dark adaptation.
that sensitivity to the shorter wavelengths of light (From McFarland, et al., 1960)
 Perceptual Change in Adults 475

(those that appear bluish) continually diminishes that the contrast threshold is higher for these older
from early childhood until death (Bomstein, 1977; observers (Leibowitz, Post & Ginsburg, 1980).
Lakowski, 1962). Also, some reduced function Furthermore, it appears that the neural system that
may be due to an age-related loss of neurons in the responds to transient stimulation (remember the Y
visual cortex. This neuron loss is quite extensive. cells discussed in Chapter 3) shows the greatest
Consider, for instance, the area of the cortex re¬ loss of sensitivity (Sekuler & Hutman, 1980).
ceiving foveal projections. In a 20-year-old this These findings suggest that in tasks such as night
area contains about 46 million neurons per gram of driving, where good acuity and response to rela¬
tissue, whereas in an 80-year-old the neuronal den¬ tively fast moving stimuli are required yet illumi¬
sity is reduced by nearly one-half, to only 24 mil¬ nation levels are low, older individuals might be
lion neurons per gram of tissue (Devaney & quite inefficient, and may even be at risk.
Johnson, 1980).
Although the number of remaining neurons is
adequate for most visual tasks, we might expect
Age Effects on the Other Senses
this reduction in the number of responding units to
show up as a reduction in visual acuity as well as There is much evidence, from the clinic as well as
in reduced sensitivity (Weale, 1986). Many studies the laboratory, indicating that hearing ability de¬
have shown that visual acuity decreases with age; clines with age. Generally speaking, hearing im¬
at age 40 nearly 94 percent of individuals have 20/ pairments will begin to appear during middle age
20 visual acuity or better, whereas by age 80 only and increase rapidly after age 60. About 15 percent
6 percent of the population will have this level of of all people over 65 could be classified legally as
acuity (Richards, 1977; Woo & Bader, 1978). The deaf, and as many as 75 percent of all 70-year-olds
relationship between age and visual acuity can be have some hearing problems (Schaie & Geiwitz,
seen in Figure 16-13. 1982). The loss of hearing ability is much more
The actual pattern of the acuity loss with age marked for high-frequency stimuli (Corso, 1981).
is quite interesting. Older observers are still able to As can be seen from Figure 16-14, at age 70 there
resolve visual details, however the light level nec¬ is still very little loss of threshold sensitivity for a
essary for them to do so is greatly increased. In 1000-Hz tone, but an 8000-Hz tone will show a
terms of our discussion in Chapter 4, we would say reduction in threshold sensitivity of nearly 50 db.

FIGURE 16-13 Age-related decrease in visual acuity.

476 Chapter 16/Development

FIGURE 16-14 The age-related decrease in auditory sensitivity is particularly noticeable for the higher
frequencies.

The sensitivity in a number of other modalities twice as well, although on some common items,
also decreases with advancing age. The pattern of such as coffee, performance is about the same for
change, however, is not always consistent. Thus, both age groups.
although older observers show a reduced sensitivity
to touch (Thombury & Mistretta, 1981), they do Global Changes in Perceptual
not show any decrease in their sensitivity to pain
Performance
(Harkins & Chapman, 1977).
Some of the most noticeable changes with age Several types of change seem to affect all the sen¬
occur in the realm of taste and smell. Odor sensi¬ sory modalities. The most important of these is a
tivity is greatly diminished, although the reduction general slowing of neural responses, accompanied
is not uniform across all stimuli. For instance, el¬ by an increasing persistence of the stimulus (ac-
derly subjects seem best able to discriminate among
fruity odors, as compared to other classes of scents Table 16-2. The Percentage of 20-
(Schiffman & Pasternak, 1979). There is also a di¬ Year-Olds versus the Percentage of
minished sensitivity to the primary tastes. For in¬ Elderly Individuals (Mean Age of 73
Years) Correctly Identifying Some
stance, thresholds rise measurably, although not
Common Foods in Pureed Form
dramatically, for both salt and sugar (Grzegorczyk,
(Based on Schiffman, 1977)
Jones & Mistretta, 1979; Moore, Nielson & Mis¬
tretta, 1982). The combination of diminishing sen¬ Food 2,0-Year-Olds Elderly
sitivity to odor and taste in the elderly greatly
reduces their ability to identify foods, especially Apple 81 55
when blended or pureed so that they are not rec¬ Lemon 52 24
ognizable by sight (Schiffman, 1977). Just how Strawberry 78 33
large a deficit results can be seen by comparing the Broccoli 30 0
performance of a group of 20-year-olds to a group Carrot 63 7
of elderly subjects (average age of 73), and seeing Corn 67 38
how poorly the older do at recognizing foods by Beef 41 28
Coffee 89 70
taste and smell alone. As can be seen in Table
Sugar 63 57
16-2, in most instances the younger individuals do
 Glossary 477

tually slower recovery or clearing time). This Filtering Part of selective attention that involves
means that older individuals have more trouble screening out irrelevant stimuli while attending to rele¬
with briefly presented stimuli (Hoyer & Plude, vant stimuli.
1980), show slower reaction times to stimulus on¬ Forced-choice preferential looking An experimen¬
sets (Stem, Oster & Newport, 1980), and cannot tal method in which observers watch infants and attempt
to determine on the basis of their responses on which
readily identify stimuli arriving in a rapid sequence
side of the visual field a stimulus was presented.
(Birren, Woods & Williams, 1980). This slowing
Habituation The process by which an observer
of perceptual processing of the elderly individual ceases to respond to a repeated stimulus.
becomes most apparent when the perceptual tasks Integration The process by which an observer at¬
are complex (Cerella, Poon & Williams, 1980; tempts to build an internal model of objects or relation¬
Cunningham, 1980). ships based on incoming stimulation and knowledge from
previous interactions.
Another general change that accompanies ag¬
Mueller-Lyer illusion An illusion of size in which
ing involves the distribution of attention to percep¬
the apparent length of a line is affected by the direction
tual tasks (Botwinick, 1984). Much of this can be of its contextual wings.
traced to the idea that a fixed amount of attentional Optokinetic nystagmus An eye-movement sequence
resources can be divided among various tasks (see in which there is a smooth movement and a quick return
Chapter 15 and Kahneman, 1973). Older individu¬ in the presence of a moving pattern.
als seem to have more difficulty dividing their at¬ Perceptual learning approach An approach to per¬
tention between various stimuli or input channels ception that maintains that the primary determinant of
our perceptual abilities is our previous experience with
(Craik & Simon, 1980). In addition, they seem to
certain environmental stimuli.
have more difficulty filtering or extracting relevant
Ponzo illusion An illusion of size in which the
from irrelevant targets in search or recognition length of a line is perceptually affected by its place in
tasks (Rabbitt, 1977; Wright & Elias, 1979). The the context of surrounding converging lines.
more similar the irrelevant stimuli are to the target Preferential looking A behavioral measure of infant
stimuli, the greater the difficulty all observers have discrimination in which target fixation time is assumed
in detecting targets in a search task. However, el¬ to be positively related to stimulus preference.

derly observers have their performance disrupted at Rooting response An infant reflex consisting of
head turning in the direction of a touch to the face.
levels of difficulty that do not seem to affect youn¬
Saccadic eye movements Sharp, jerky eye move¬
ger observers (Farkas & Hoyer, 1980). Fortunately,
ments that place the images of objects that observers are
when a task does not demand searching irrelevant attending to on the fovea.
stimuli to find a relevant stimulus, the deficits in Schemata Internal models or hypotheses about the
attention in the aged do not appear (Kausler, 1982). external world.
Search Part of selective attention that involves scan¬
ning the visual field for a specific target stimulus.
GLOSSARY Smooth-pursuit eye movement Nonjerky move¬
ments used to track a smoothly moving target, as op¬
The following definitions are specific to this book. posed to saccadic eye movements.
VEP See Visually evoked potential.
Developmental approach An approach to percep¬
tion that presumes that chronological age is the best pre¬ Visual-geometric illusions Simple line drawings in
dictor of perceptual ability. which the actual physical characteristics of certain ele¬
ments differ from the perceived characteristics of those
Dyslexia A form of impaired reading ability charac¬
elements.
terized by confusion of letters that are left-right mirror
Visually evoked potential (VEP) The change in the
images.
electrical activity of the brain produced in response to a
Encoding Entering perceptual information into some
visual stimulus.
form of short-term memory storage to allow recognition
or other processing.
' CHAPTER

Learning and
Experience
EXPERIENCE AND DEVELOPMENT
Restricted and Selective Rearing
Neurophysiological Effects
Perceptual Effects
Human Studies
SENSORY-MOTOR LEARNING
Perceptual Rearrangement
Illusion Decrement
CONTEXT AND MEANING
Eyewitness Testimony
ENVIRONMENTAL AND LIFE HISTORY
DIFFERENCES
Picture Perception
Illusion and Constancy
Speech
Effects of Occupation
Perceptual Set

479
480 Chapter 17/Learning and Experience

A n article in the New York Times spoke of a tea

expert who was called in to determine the
some manner, and now you can see what was for¬
merly invisible. In other words, during the course
components of a blend of tea that an American of the experiment you have learned to see this
company was about to market. A small cup of it pattern.
was poured for him. He sniffed it gently, sipped a
bit, swished it around in his mouth a little, then
looked up. EXPERIENCE AND DEVELOPMENT
“I detect,” he said crisply, ‘‘a rather good
Assam, a run-of-the-mill Darjeeling, a mediocre As an organism develops, its nervous system
Ceylon, and, of course, the tea bag” (Root, 1974). matures, and over the years many changes in phys¬
Although we might be amazed at perfor¬ iology and perceptual ability also come about sim¬
mances such as these, or similar ones of expert ply because of physiological maturation. Of course,
wine tasters, we must realize that this degree of as the months and years roll by, the organism
perceptual discrimination has come about through is also accumulating new experiences with the en¬
years of training and experience. In other words, vironment, and is encountering many chances to
this expert had to learn to taste and identify these learn new perceptual coordinations. It is important
flavors. for us to understand how the natural course of de¬
In some of the previous chapters we have velopment interacts with an individual's life history
mentioned some ways in which our past history, to shape that individual’s perception of the world.
experience, knowledge, and hypotheses affect our Experience can affect the development of the
perception. Most people are willing to admit that individual’s perceptual processes in several differ¬
some aspects of perception may be susceptible to ent ways. We have outlined these in Figure 17-1
the influences of experience, but they are often un¬ (see Aslin, 1985; Gottlieb, 1981).
aware of the magnitude of these effects. In fact, The strongest form of interaction between ex¬
our past can even influence whether we perceive perience and development is induction. Here, the
anything at all in certain circumstances. For exam¬ presence of some sort of relevant experience ac¬
ple, suppose we briefly flash a visual stimulus tually determines both the presence and final level
(such as a word) in front of you. If we have chosen of the ability (Figure 17-1 A). The weakest form of
the duration and intensity of the stimulus carefully, effect we will call maturation. This actually rep¬
you may be unaware of any aspect of the stimulus. resents no effect at all, and the ability might be
If we flash the same stimulus again,\we would ex¬ expected to develop regardless of the individual's
pect that, again, you would see nothing. However, experience or lack of it (17-1B). Another possible
with repeated presentations something about its ap¬ interaction is enhancement. Here the final level of
pearance will begin to change. Soon you will be an ability, which is already present or developing,
able to make out fragments of this stimulus, and is improved because of experiential factors (17-
after a while these fragments will become more 1C). Facilitation increases the rate at which an
complex. Eventually the entire stimulus pattern will ability develops, but not its final level, providing
be identified on every trial (i.e., you can read the earlier acquisition of the skill but not greater profi¬
word), even though the luminance and exposure ciency (17-ID). Finally, maintenance serves to sta¬
durations are the same as for the very first trials bilize, or to keep, an ability that is already present
when you saw and identified nothing (Uhlarik & (17-IE). Of course, different mechanisms might be
Johnson, 1978)! Your prior experience with this expected to produce each of these patterns of in¬
stimulus has changed your perceptual abilities in teraction between development and experience.
 Experience and Development 481

FIGURE 17-1 Various ways in which experience can interact with the development of perceptual
abilities.
482 Chapter 17/Learning and Experience

Restricted and Selective Rearing those cells that are found will not show the usual
degree of orientation and movement selectivity
The most direct method for assessing the relation¬
(Blakemore, 1978; Leventhal & Hirsch, 1980). The
ship between development and experience is to de¬
visual cortex of such animals appears to be very
prive the observer of the opportunity to use a
immature, because of the absence of the usual his¬
particular sensory modality from the moment of
tory of visual experience.
birth. After the individual has fully matured, we
Such neurophysiological disruption caused by
test the perceptual capacities in the deprived mo¬
the absence of visual experience can be found all
dality. If they have developed poorly, we would
along the visual pathways. Some effects are as pe¬
have demonstrated the need for experience in the
ripheral as the retina, but they appear as well at
development of normal functions. This technique is
other places such as the superior colliculus, the lat¬
called restricted rearing. (We previously encoun¬
eral geniculate nucleus, and the visual cortex
tered this procedure in Chapter 10 when we dis¬
(Movshon & van Sluyters, 1981; Riesen & Zilbert,
cussed the development of depth perception.) A
1975). Different aspects of the visual pathways
somewhat more elegant technique involves delib¬
seem to be more or less susceptible to such dam¬
erately altering the pattern of experience that the
age. Thus, we find that X and W cells are rela¬
developing organism is exposed to from birth. For
tively unaffected by dark rearing, whereas Y cells
example, an animal may be exposed to only diffuse
are readily lost if no visual experience is avail¬
light, vertical stripes, the color red, and so forth.
able (Hoffmann & Sherman, 1975; Rothblat &
Such a procedure should selectively bias, rather
Schwartz, 1978). These effects are not irreversible,
than eliminate, certain perceptual abilities, if ex¬
since even animals who have been raised for a year
perience plays a role in their development. This
following birth in total darkness show some recov¬
procedure is known as selective rearing.
ery after several months of exposure to illuminated
surroundings, although recovery is never complete
Neurophysiological Effects
(Cynader, Berman & Hein, 1976).
Recent research indicates that experience may play Much subtler neurophysiological changes
a role in the development of sensory physiological come about through selective-rearing practices. For
structures themselves. For instance, as discussed in instance, in Chapter 3 we stated that cells in the
Chapter 16, the visual pathways and visual cortex visual cortex tend to show ocular dominance. This
of the newborn differ from those of the adult; there means that although most cells in the cortex can be
are fewer responsive cells and these show lesser activated by stimulation of either eye, they tend to
degrees of directional and orientational sensitivity respond more vigorously to one eye than to the
(see also Hickey & Peduzzi, 1987; Norton, 1981a). other. The fact that most cells respond somewhat
Visual experience, in addition to the growth and to each eye’s input probably has to do with the
maturation of the nervous system, seems to be depth cue of binocular disparity (see Chapter 10).
necessary for the development of normal visual Suppose that we rear an animal from birth so that
functioning. This has been demonstrated though it only views the world through one eye. Later we
restricted-rearing studies. test separately the ability of the two eyes to pro¬
Let us consider what happens if we completely duce a response in the visual cortex. We would
deprive an animal of any visual input by rearing it probably find that the majority of the cells are ac¬
in the dark from birth. This animal will have a vi¬ tivated by the experienced eye, and often less than
sual cortex that shows reduced overall responsive¬ 10 percent of the cells can be driven by the de¬
ness, when tested using the electrode implantation prived eye (LeVay, Wiesel & Hubei, 1980).
techniques discussed in Chapter 3. Furthermore, The degree of disruption of normal function-
 Experience and Development 483

ing seems to depend on when the period of depri¬

vation begins. If the animal is deprived of
binocular viewing during the period of 3 weeks to
3 months after birth, large disruptions of the nor¬
mal pattern of binocular response occur. However,
if the monocular viewing period is instituted after
3 months of age, even for periods of up to a year,
virtually no effect is found (Cynader, Timney &
Mitchell, 1980; Held, 1985; Pettigrew, 1978b).
This means that there is a particular time period
during which the visual experience is most required
and most effective. Such an interval is called a
critical period, and it characterizes many aspects
of the interaction between experience and devel¬
opment (Mitchell, 1981). Critical periods may
correspond to periods of maximal growth and de¬
velopment in the nervous system (Aslin, 1985;
Hickey, 1977). Any disruption of normal visual ex¬
FIGURE 17-2 An apparatus for selectively
perience during the critical period, even for periods rearing a kitten so that its only visual experience
as short as 3 days, produces measurable changes in will be with vertical lines.
the responses of cells in the visual system (Free¬
man, Mallach & Hartley, 1981).
Perhaps the most subtle form of selective vi¬ a particular orientation, as we saw in Chapter 3,
sual rearing involves limiting an animal to a world and the particular preferred orientations are rather
containing only contours oriented in one direction. evenly distributed. The left side of Figure 17-3 de¬
Thus, an animal might be exposed to only verti¬ picts this distribution as a set of lines each of which
cally oriented lines from birth. This is accom¬ represents a neuron responding best to that orien¬
plished by either affixing to the animal goggles that tation. However, recording from an animal that has
only contain lines of one orientation, or giving the
animal experience for a few hours each day in an
apparatus similar to that shown in Figure 17-2.
This is simply a large cylinder containing nothing
but vertical stripes and a clear plastic floor on
which the animal stands. Notice that the animal is
wearing a special collar that prevents it from seeing
its own limbs.
What happens in the nervous system after ex¬
posure to this kind of selective rearing and stimu¬
Normally Cat reared with
lation might be called environmental surgery. Such reared cat vertical stripes
surgery drastically alters the response characteris¬
FIGURE 17-3 Distribution of the preferred
tics of neurons in the visual cortex. Normally,
orientation of cortical receptive fields in a
when we insert an electrode into the visual cortex normally reared cat versus that for a cat reared
in order to map receptive fields, we find large num¬ with selective exposure to vertical lines. Each line
bers of cells that respond most strongly to lines in indicates the preferred orientation of 1 cell.
484 Chapter 17/Learning and Experience

never seen horizontal stripes produces quite a dif¬ animals that have been dark-reared for longer peri¬
ferent result. In this animal, virtually no cells are ods (Timney, Mitchel & Griffin, 1978). Although
responsive to horizontally oriented lines, resulting many of the physiological changes appear to be
in a distribution of preferred stimulus orientations permanent, there seems to be enough plasticity in
much like that shown on the right in Figure 17-3 the animal to allow for considerable behavioral re¬
(Hirsch & Spinelli, 1970; Movshon & van Sluy- covery of function after the initial period of depri¬
ters, 1981). It is as if the absence of horizontally vation. Still, there will be measurable deficits in
oriented stimuli in the environment has served as a many visual tasks, including obstacle avoidance,
(figurative) scalpel that has systematically cut off tracking, jumping under visual guidance, and eye
any responding to stimuli other than the vertical blinks to oncoming objects, even after 2 years of
stimuli to which the animal was exposed. Again, normal experience (Mitchell, 1978; Rothblat &
there is a critical period between 3 weeks and 3 Schwartz, 1978).
months, during which this sort of selective stimu¬ Since selective rearing is a more subtle pro¬
lation seems to be most effective (Mitchell, 1981; cedure than restricted rearing, it should not be sur¬
Rothblat & Schwartz, 1978). prising to find that its behavioral effects are often
Blake (1981) has summarized the evidence somewhat indirect and elusive. For instance, the
from such selective rearing studies saying, “In ef¬ most dramatic behavioral effects of rearing animals
fect, the neurophysiologists have compiled a set of with one eye occluded is a reduction in the visual
recipes for creating animals with specific kinds of acuity of the deprived eye (Mitchell, 1981). How¬
neural deficits at sites along the visual pathways” ever, there are a number of interesting visual field
(p. 97). The ingredients that go into these recipes effects as well. Visual field refers to the region of
are particular experiences, or the lack of certain the outside world to which an eye will respond,
normal experiences, with visual stimuli. measured in degrees around the head. For instance,
the top view of Figure 17-4 shows the visual fields
for the right and left eyes of a cat. There is a rather
Perceptual Effects
large region of overlap between the two eyes in the
How do all these unusual environmental experi¬ frontal part of the field. This is the region of bin¬
ences affect what the organism perceives? There is ocular vision, where either or both eyes of the cat
a slight divergence between the physiological and should be able to see an object. Generally, if an
the behavioral data when we answer this question./ interesting stimulus appears in the visual field of a
Consider a kitten reared in total darkness until the cat, it will immediately turn its eyes and head to¬
age of 6 months. When we remove this kitten from ward it, displaying an orienting response (see
darkness, it at first appears to be completely blind; Chapter 15). We can use this response to measure
however, within about 48 hours of exposure to il¬ the effectiveness of stimuli in the visual field of the
luminated surroundings the kitten begins to show cat, and if we cover one eye at a time we can mea¬
some visual responsiveness. Various forms of sen¬ sure each visual field separately.
sory motor coordination begin to appear in a piece¬ Let us first consider an animal that has been
meal fashion, and after a 6-week period of normal completely dark-reared. This animal shows a se¬
experience a great deal of recovery has occurred. vere loss of response in the region of binocular
Direct measures of visual acuity show a gradual overlap. Each eye seems to respond only to objects
improvement. If the animal had been dark-reared on its side of the head, as shown in the middle
for only about 4 months, the acuity gradually view of Figure 17-4. An animal reared with one
would return to that of a normally reared cat; how¬ eye occluded also does not show a binocularly re¬
ever, visual acuity never reaches normal levels for sponsive region of the visual field. The eye that had
 Experience and Development 485

normal visual experience shows a normal visual

field, overlapping well to the opposite side. How¬
ever, the eye that did not receive visual experience
acts as if it only responds to targets that are far to
the side of the head, and excludes all of the visual
field that the normal eye covers, as shown in the
bottom view of Figure 17-4 (Sherman, 1973). A
similar effect was reported for a young man who
was bom with a cataract that prevented any pat¬
terned vision in his left eye (his visual experience
was similar to the monocularly reared animals we
have been discussing). When this cataract was re¬
moved at age 19, the normal eye had its usual vi¬
sual field size, but the patient simply could not
detect any stimuli in a large portion of the region
where the two eyes’ views overlapped (the binoc¬
ular region) with the deprived (cataract) eye. Al¬
though there was some recovery over the next 10
months, the visual field of the deprived eye never
became as large as that of the normal eye (Moran
& Gordon, 1982).
The other form of selective rearing, where an
animal is reared under conditions of exposure to
horizontal or vertical stripes alone, also produces a
behavioral deficit in addition to the change in the
distribution of cells in the cortex with a specific set
of preferred orientations. Here the results are not as
dramatic as we might expect. Animals who have
been reared only with vertical stripes are not blind
to horizontal stripes, rather they have measurably
lower visual acuity for stripes in an orientation
never seen during their rearing (Blasdel, Mitchell,
Muir & Pettigrew, 1977; Hirsch, 1972).

Human Studies

Restricted-rearing studies cannot be conducted with

human observers because of the possibility of pro¬
ducing long-lasting perceptual deficits. However,
some clinical conditions reproduce the circum¬
FIGURE 17-4 Regions of the visual field in which stances needed to study the effects of experience on
a cat will respond to visual stimuli presented to
perception. For example, Senden (1960) collected
the right or left eye are altered by depriving one
or both eyes of visual experience from birth. case reports of individuals who had suffered from
(Based on Sherman, 1973). lifelong blindness due to the presence of cataracts;
486 Chapter 17/Learning and Experience

these individuals later had vision restored through an astigmatism. They were able to show that such
a surgical procedure. After the removal of the cat¬ selective rearing can also alter the distribution of
aracts, these adults were unable to identify familiar preferred orientations of visual cortical neurons,
objects by sight, although they were capable of causing a reduction of the number of cells prefer¬
identifying them if they were allowed to touch the ring the blurred orientation. Severe astigmatism at
objects. For instance, when asked to discriminate an early age in humans results in a permanent loss
between a square and a triangle, these individuals of visual acuity in the direction of the astigmatism.
had to undertake the painstaking procedure of seek¬ This is an acuity loss due to neural changes, be¬
ing out and counting the comers of the figure be¬ cause it remains even after correcting for any opti¬
fore the forms could be distinguished from each cal errors, and is probably the result of selective
other. restriction of exposure to contours in the astigmatic
These newly sighted observers seemed able to direction (Mitchell, 1980).
detect the presence or absence of an object in the A variation of this same selective-rearing ef¬
visual field, but this seemed to be the extent of fect is caused by living in an urbanized environ¬
their abilities. For example, one patient was shown ment. The nature of our carpentered cities means
a watch and was asked whether it was round or that we have frequent exposure to vertical lines
square. When he seemed unable to answer, he was (defining walls, comers, furniture legs, and so
asked whether or not he knew the shape of a square forth) and to horizontal lines (defining floors, ceil¬
or a circle. He was able to position his hands to ings, table edges, and so forth). Proportionally we
form both a square and a circular shape, but he have much less exposure to oblique lines. There¬
could not visually identify the shape of the watch. fore, as inhabitants of such a selectively stimulating
When the watch was placed in his hands, he im¬ environment, it might be expected that we would
mediately recognized it as being round. It appears show reduced acuity for diagonal lines relative to
that his sense of touch, although not more sensitive horizontal and vertical lines. In fact, the human vi¬
than that of a sighted person, had come through sual system is anisotropic, meaning that it often
long experience to be a more reliable source of in¬ reacts differently to stimuli depending on their ori¬
formation about the world than his untrained sense entation. In general, the normal visual system
of vision (see Warren, 1984). shows a slight, but well-defined, preference for
There is an interesting naturally occurring horizontal or vertical stimuli over diagonal stimuli.
analogy to selective rearing of the type used with This is demonstrated in a number of acuity-related
cats in which they see only contours in a single tasks, where resolution acuity and vernier acuity
orientation. This analogy arises from a common vi¬ seem to be poorer for stimuli oriented diagonally
sual problem known as astigmatism. Astigmatism (Bowker & Mandler, 1981; Corwin, Moskowitz-
usually occurs if the cornea of the eye is not per¬ Cook & Green, 1977; Jenkins, 1985; Vogels & Or-
fectly spherical, perhaps being flatter in some ban, 1986). This phenomenon is known as the
places and more curved in others. This deviation oblique effect, and can easily be demonstrated us¬
from perfect sphericity brings contours of some ori¬ ing Demonstration Box 17-1.
entations into sharper focus than those in other Some investigators feel that at least part of the
orientations. Thus, with a vertical astigmatism, oblique effect is caused by genetic factors (Leehey,
horizontal lines will be clear and vertical lines will Moskowitz-Cook, Brill & Held, 1975; Timney &
be blurry, and so forth. The fact that this condition Muir, 1976), but Annis and Frost (1973) provided
can mimic selective-rearing effects was shown by some interesting data that are compatible with se¬
Freeman and Pettigrew (1973), who reared cats lective environmental effects. They compared the
wearing cylindrical lenses that artificially created variations in acuity as a function of the orientation
 Sensory-Motor Learning 487

DEMONSTRATION BOX 17-1. The Oblique Effect

To demonstrate that visual acuity is better for hori¬ It will appear uniform gray at this point, as your
zontal or vertical stimuli than for obliquely oriented resolution acuity fails. Notice, however, that at this
stimuli, prop this book up on a table so that you can distance you still can see that the left circle contains
see the three stimulus patterns. Now slowly walk vertical lines, and the right contains horizontal lines,
backward from the book until you can no longer re¬ thus indicating your greater visual acuity for these
solve clearly the oblique lines in the center circle. orientations.

of lines in a group of students from Queens Uni¬ who are congenitally deaf, tend to have different
versity in Kingston, Ontario, with that observed in brain organizations than do normal-hearing subjects
a group of Cree Indians from James Bay, Quebec. (Neville, 1985; Neville, Schmidt & Kutas, 1983).
The students had all grown up in typical North Parts of the brain normally reserved for auditory
American buildings. The Cree Indians, however, processing now appear to be available for visual
were among the last to be raised in traditional functions.
housing consisting of a cook tent (or meechwop) in
summer, and a winter lodge (or matoocan) during
the rest of the year. Both the insides and outsides SENSORY-MOTOR LEARNING
of these structures consist of a rich array of con¬
tours, with no obvious preponderance of verticals One variable that seems to be essential for the de¬
and horizontals. In addition, the natural environ¬ velopment of normal visual functioning involves
ment of the Cree shows no excesses of verticals not only the eyes but also the entire body. It seems
and horizontals, in contrast to the urbanized envi¬ that normal perceptual development depends on ac¬
ronment of the students. In line with the selective- tive bodily movement under visual guidance. Holst
exposure hypothesis, the students showed the and Mittelsteadt (1950) offered a distinction be¬
expected reduction in acuity for obliquely oriented tween stimulus input that simply acts on a passive
contours, whereas the Cree, without this selective observer, which they called exafference, and stim¬
exposure, did not. ulation that changes as a result of an individual’s
Although we have emphasized the visual mo¬ own movements, called reafference.
dality in our discussions thus far, it is important to Reafference has been suggested as being nec¬
recognize that selective and restricted rearing in essary for the development of accurate visually
other senses also has measurable effects in humans guided spatial behavior (Hein, 1980). An experi¬
and animals. In hearing, for example, it has been ment by Held and Hein (1963) elegantly demon¬
found that individuals who are deaf in one ear, or strates this notion. They reared kittens in the dark
488 Chapter 17/Learning and Experience

until they were 8-12 weeks of age. From that age paw extension (as if to avoid collision) when the
on, the kittens received 3 hours of patterned visual animal is moved quickly toward a surface. In all
exposure in a “carousel” apparatus, shown in Fig¬ three measures, the active animals performed like
ure 17-5. As you can see from the figure, one of normal kittens and the passive animals showed lit¬
the animals is active and can walk around freely. tle evidence of depth perception.
The other animal is passive and is carried around An interesting extension of this work, which
in a gondola that moves in exactly the same direc¬ shows the specificity of experiential effects, was
tion and at exactly the same speed as the move¬ done by Hein, Held, and Gower (1970). They re¬
ments of the active animal. Thus, the moving peated the carousel experiment; however, each an¬
animal experiences changing visual stimuli as a re¬ imal received both active and passive exposure.
sult of its own movements (reafference); the pas¬ One eye was used when the visual exposure was
sive animal experiences the same stimulation, but active, and the other eye was used when the visual
it is not the result of self-generated movements exposure was passive. They reported that when the
(exafference). kittens were tested on the actively exposed eye they
The animals were later tested on a series of seemed to have normal depth perception, whereas
behaviors involving depth perception. These in¬ when tested on the passively exposed eye they
cluded dodging or blinking when presented with a acted as if they did not.
rapidly approaching object and the avoidance of the How much of the development of our visually
deep side of the visual cliff (see Chapter 10). They guided behavior requires practice and exposure?
were also tested for the visual placing response, a Consider the simple tasks of reaching out and pick-

FIGURE 17-5 Kitten carousel for active or passive exposure to visual stimulation. (From r. Held, &. a. Hein,
1963, Journal of Comparative and Physiological Psychology, 56. Copyright 1963 by the American Psychological Association. Reprinted by
permission.) '
 Sensory-Motor Learning 489

ing up an object with one hand. This involves not cats accurately placed their paws, but the unseen
only the accurate assessment of the distance and the limb could not be accurately directed toward the
size of the object, but the ability to guide your limb targets. In another experiment, Held and Bauer
on the basis of the perceptual information. Held (1967) reared a monkey for 34 days without sight
and Hein (1967) reared kittens in the dark until of its limbs. When the animal was first able to see
they were 4 weeks old. After this period, they were its arm, it acted startled and spent much time look¬
allowed 6 hours of free movement each day in a ing at its hand. The animal’s behavior was awk¬
lighted and patterned environment. However, dur¬ ward and inaccurate at first but improved rapidly
ing the time when the cats received their exposure with practice. When the other hand (which had not
to patterned stimuli, they wore lightweight opaque been previously exposed) was tested, it also
collars that prevented them from seeing their bodies showed inaccurate reaching until some experience
or paws while they moved about (see Figure 17-6). had been gained.
The remainder of the time, the kittens were placed There are reports of behaviors in human in¬
in a dark room. After 12 days of such exposure, fants that seem similar to the responses of the de¬
these animals showed normal depth perception, but prived monkey. White (1971) reports that after the
their ability to accurately place their paws by visu¬ first month of life (during which infants are alert
ally directing them toward targets was quite poor. about 5% of the time), infants spend many hours
Nonetheless, after 18 hours of free movement in a watching their hands. Their reaching is quite inac¬
lighted environment, with their paws visible, all di¬ curate at first, but improves steadily. Actually,
rectional confusions seemed to have disappeared. practice can speed up this process of perceptual de¬
Hein and Diamond (1971) conducted a similar velopment. If conditions are arranged so that there
experiment in which each of the cat’s front paws are many objects to reach for and to play with, in¬
was placed in a separate cone, rather than using a fants develop accurate reaching behavior several
collar. One cone was opaque, and the other one weeks earlier than children who have not received
was transparent so that the animal could see its this type of enriched experience. Experience with
limb. When tested using the limb they could see, the sight of actively moving parts of the body
seems to be a necessary condition for the successful
development of visually guided behavior (Hein,
1980).

Perceptual Rearrangement
In 1896, George Stratton reasoned that if some as¬
pects of the perception of space and direction were
learned then it ought to be possible to learn a new
set of spatial percepts. To test this, Stratton used a
technique that altered spatial relations in the visual
world (Stratton, 1897a,b). His technique involved
wearing a set of goggles that optically rotated the
field of view by 180 deg, so that everything ap¬
peared to be upside down. Such a procedure is
FIGURE 17-6 Kitten in collar that prevents view called optical rearrangement (Welch, 1978).
of paws. (From A. Hein, &, R. Held, 1967, Science, 158, 390-92. More recently, Kohler (1962, 1964) elabo¬
Copyright 1967 by the American Association for the Advancement
of Science.) rated on this procedure. Kohler’s observers often
490 Chapter 17/Learning and Experience

wore optically distorting devices for several weeks. place, because when the inverting lenses were re¬
Observers reported that at first the world seemed moved observers experienced a sense of discom¬
very unstable, the visual field appearing to swing fort. The world suddenly appeared to be inverted
as the head was turned. During this stage of the again, and they had difficulty moving about. How¬
experiment observers often had difficulty walking ever, the readaptation to the normal upright world
and needed help to perform very simple tasks. was accomplished within a period of about 1 hour.
However, after about 3 days one observer was able Demonstration Box 17-2 shows how you can ex¬
to ride a bicycle, and after only a few weeks he perience this inverted visual stimulation.
was able to ski. The observers reported that they Most rearrangement studies involve a less
sporadically experienced the world as being up¬ dramatic change of optical input. A common tech¬
right. If they observed common events that have nique is to use a wedge prism, which is a wedge-
definite directional components, such as smoke ris¬ shaped piece of glass that bends, or refracts, light.
ing from a cigarette or water pouring from a The locations of objects viewed through the prism
pitcher, they reported that the world appeared to be seem to be shifted in the direction of the apex (the
upright. This suggests that their ability to adapt to pointed edge of the wedge). If an observer viewed
the optically rearranged visual input was facilitated the world through goggles containing such prisms
by the notion of gravitational direction along with and reached for an object, she would find herself
interaction with familiar events and objects. Kohler missing it. After only a few minutes of practice,
suggested that a real perceptual change had taken however, the observer’s reaching would become

DEMONSTRATION BOX 17-2. Optical

You can experience some of the effects associated

with inverted optical stimulation by holding a mirror
as shown in the accompanying figure. Walk around
and view the world by looking up at the mirror. No¬
tice that the world seems inverted, and also notice
how the world swings as you turn. Now pour some
water from a glass. Does the water pour up or down?
Are you sure?
 Sensory-Motor Learning 491

quite accurate. We would say that she has adapted several degrees to one side. When reaching for an
to the prismatic distortion; in other words, she has object the observer should err in the direction op¬
compensated for the optical distortion. If the ob¬ posite to that of the initial distortion. This is what
server is consciously correcting for the distortion actually does occur. These errors are called after¬
(for instance, saying to herself, “I must reach 10 effects. The occurrence of aftereffects in prism
deg to the right of where the object appears”), adaptation is evidence that some perceptual rear¬
when the goggles were removed she would, of rangement has occurred (Harris, 1980). This pro¬
course, know that the distortion is no longer pres¬ cess is outlined in Figure 17-7.
ent. Being rational, she should then drop this con¬ A number of investigators have attempted to
scious correction and reach for seen stimuli with specify what conditions are necessary for adapta¬
her usual accuracy. However, suppose that some tion to rearranged stimulation (Welch, 1978). Held
perceptual change has occurred. In this case we and Hein (1958) have argued that adaptation de¬
would expect that when the distorting prism is re¬ pends on active movements, as does the develop¬
moved the visual world would appear to be shifted ment of visually guided behavior discussed earlier.

Apparent
target Real target

Initial pointing with

prism goggles

Apparent
target Real target

Aftereffect—pointing behavior
with goggles removed

Adaptation—pointing behavior
after practice with goggles

c
FIGURE 17-7 Prism adaptation and aftereffect.
492 Chapter 17/Learning and Experience

They tested this notion by having an observer view merely watched it rotate. For another group the rod
his hand through a prism under one of three con¬ appeared displaced to the same degree; however, as
ditions. One was a no-movement condition, in the rod swung about it brushed each observer
which the observer viewed only his stationary across the lips, indicating that it was directly in
hand. The second was a passive-movement condi¬ front of the observer rather than off to the side as
tion, in which the observer’s arm was swung back it appeared. Although both groups were passive,
and forth by the experimenter. The third was an the groups receiving the information that their per¬
active-movement condition, in which the observer cept was erroneous (being touched by a stimulus
saw his hand through the prism while he actively that looked like it would pass them by) showed
moved it from side to side. There was considerable perceptual adaptation, whereas the other group did
adaptation to the distortion produced by the prism not. Thus, information indicating how our percepts
under the active-movement condition, whereas in are in error may be sufficient to produce adaptation
the other conditions there was not. These results (Howard, Anstis & Lucia, 1974).
have been verified several times (Pick & Hay, What actually changes during the adaptation
1965). process? This issue is still being debated. Some re¬
Another series of experiments used conditions searchers believe that adaptation simply alters the
similar to the kitten carousel discussed earlier. Ob¬ felt position of various parts of the body (Harris,
servers wearing displacing prism goggles either 1980). This is based on the observation that after
walked around for about 1 hour (active exposure) prism adaptation, when observers are asked to
or were wheeled around in a wheelchair over the point to a straight-ahead position (by feel alone),
same path for about 1 hour (passive exposure). they tend to point off to the side. This indicates
They were then measured to see if any perceptual some proprioceptive or “felt” component in the af¬
change had taken place. Adaptation to the prismatic tereffect (Harris, 1980). Other data indicate that
distortion occurred in the active condition but not this may be only part of the process (Mikaelian,
during the passive exposure condition (Held & 1974; Redding & Wallace, 1976). For example,
Bossom, 1961; Mikaelian & Held, 1964). animals can still adapt to the visual displacement
One important aspect of active movement un¬ when the nerves that provide information about the
der the optically distorted conditions seems to be position of the arm are severed (Bossom & Om-
that it provides observers with some sort of error maya, 1968; Taub & Berman, 1968). The consen¬
feedback, which informs them of the direction and sus is that a change is taking place in both
the extent of the distortion. This information pro¬ proprioceptive and visual perception, which is the
vides a basis for learning a new correlation between result of recalibration of the higher brain centers
the incoming stimuli and the conscious percept. used to interpret perceptual input (Howard, 1982;
The more information we give observers about the Welch, 1978). Some direct evidence for this per¬
nature of their errors, the greater is the adaptation ceptual recalibration comes from an interesting ex¬
to the distortion (Coren, 1966; Welch, 1969, periment by Foley (1970, 1974). She placed wedge
1971). prisms in front of the eyes of observers so that the
Some investigators have suggested that error direction of displacement was different for each
information in the absence of active movement eye. Either one eye saw an upward displacement
is sufficient to produce prism adaptation. Howard, and the other a downward displacement, or one eye
Craske, and Templeton (1965) had observers watch saw a displacement to the right and the other to the
a rotating rod through an optical system that dis¬ left. After several hours of exposure the two eyes
placed it to one side. For some observers the rod were tested separately. The results indicated that
appeared to be displaced to the side, and they each eye had adapted to its own particular distor-
 Sensory-Motor Learning 493

tion. This implies a perceptual recalibration. It decrement, has been demonstrated many times
seems likely that adaptation to optically rearranged (see Coren & Girgus, 1978; Porac & Coren, 1985).
stimuli involves a form of perceptual learning that What is happening in this situation? How does
alters the appearance of visual space, which, like the observer know that the percept is wrong in the
many other forms of learning, is sensitive to what first place, and why does the size of the illusory
the observer is paying attention to (Redding, Clark effect decrease? The answer seems to lie in the pat¬
& Wallace, 1985). However, whether learning to tern of the observer’s eye movements (see Coren,
deal with rearranged spatial stimuli involves the 1986). If we measure the actual pattern of eye
same mechanisms that may have gone into the movements an observer makes over an illusion fig¬
original development of our perception of space is ure, we find that the eyes are directed to move as
not clear. if the distorted percept were actually correct. In
other words, if the eyes were resting on the end of
the line in the perceptually elongated portion of the
Illusion Decrement
Mueller-Lyer figure (Figure 17-8A), an attempt to
There is another form of perceptual learning that is look at the far end of the line would produce an
similar to rearrangement in that it involves learning eye movement that is too long. This eye-movement
to compensate for a perceptual error. It differs from error is in agreement with the percept, which
the situations we have been discussing in that the tends to overestimate the length of the line. A
error is not optical in nature, and the observer usu¬ corrective adjustment in the eye movement must be
ally is not conscious either of the erroneous percep¬ made if the fovea is to come to rest on the exact
tion or of any perceptual change. This situation end of the line. The opposite happens for the
involves visual geometric illusions, which are sim¬ underestimated portion of the Mueller-Lyer figure
ple line drawings that evoke percepts differing in (Figure 17-8B). Here the eye movements are too
size or shape from those expected on the basis of short (again in agreement with the perceptual under¬
physical measurements of the stimuli. We have en¬ estimation of the line length), and a corrective
countered several of these already, in Chapters 1, adjustment must be made. The eye movement pat¬
11, 14, and 16, including the Mueller-Lyer illusion terns over the two portions of the figure are shown
(see Figure 17-8), in which the horizontal line with in Figure 17-8C.
the outward-turned wings appears longer than the In Chapters 15 and 16 we saw instances where
line with the inward-turned wings, despite the fact patterns of eye movements could be used to tell us
that they are physically equal in length. Suppose something about the information-processing abili¬
we present the Mueller-Lyer figure to an observer ties of an observer. The same reasoning can be ap¬
and measure her susceptibility to the line-length plied to the study of eye-movement patterns across
distortion. Next we instruct her to begin moving illusion configurations. As the observer views the
her eyes across the figure, scanning from one end illusory array, eye movements and eye-movement
of the horizontal line to the other on both portions errors provide information about the existence (as
of the figure. We ask her to be as accurate as pos¬ well as the direction and the strength) of the illu¬
sible with her eye movements. At 1-minute inter¬ sory distortion. This error information can be used
vals we stop the scanning process and take by the observer to correct the percept. This point
measurements of illusion magnitude until a total of of view is supported by the fact that an illusion
5 min of viewing time has elapsed. This simple decrement does not occur unless the observer is al¬
process of inspection leads to a 40-percent reduc¬ lowed to scan the figure (Coren, Girgus & Schiano,
tion in the original illusion magnitude (Coren & 1986; Coren & Hoenig, 1972; Festinger, White &
Porac, 1984). This decrease, known as illusion Allyn, 1968).
494 Chapter 17/Learning and Experience

being used to reduce a perceptual error, and the

direction of this change (from greater to lesser il¬
lusion susceptibility) mimics that associated with
perceptual rearrangement studies. The most inter¬
esting aspect of this form of perceptual adjustment,
however, is the fact that nothing about it appears to
be available to consciousness. Unless provided
with a ruler or a direct explanation, the observer
does not consciously know that the original percep¬
tion is in error, nor does she know that illusory
error has been reduced as a result of her active in¬
teractions with the illusion figure! The percept sim¬
ply becomes more accurate with no change in the
observer’s own awareness.
B

CONTEXT AND MEANING

Eye starts here Basically, all percepts are ambiguous. Consider a

target that casts a square image on the retinal sur¬
face. The object the image represents could ac¬
tually be one of an infinite number of different
shapes at any distance or inclination relative to the
observer, as shown in Figure 17-9. Since any ret¬
inal image can be caused by a variety of different
physical targets in the world, it is surprising that
our normal perceptual experiences are generally so
Eye starts here
unambiguous. Actually, what we perceive is the re¬
sult of a decision-making process in which we de¬
duce, on the basis of all available information,
what the stimulus object is. This transactional
viewpoint maintains that any current perceptual ex¬
perience consists of a complex evaluation of the
significance of stimuli reaching our receptors.
FIGURE 17-8 The overestimated (A) and Through our life experience we learn that certain
underestimated (B) portions of the Mueller-Lyer objects or conditions have a high probability of
illusion, and (C) typical eye movements obtained being related to one another. On this basis we de¬
when viewing them.
rive our "best bet” as to what we are viewing. In
a sense, the world we are experiencing is more the
result of perceptual processing than the cause of the
The phenomenon of illusion decrement im¬ perception (Coren, 1983; Ittelson, 1962). The
plies that perceptual learning is taking place. The transactional approach implies that if our expecta¬
information obtained from the eye movements is tions change, or our analysis of the situation
 Context and Meaning 495

FIGURE 17-9 Many different objects at different distances and slants, all of which cast the same square
retinal image. (From Coren &, Girgus, 1978)

changes, then our perceptual experience will also you look at it again. Our ability to perceive these
change (Ames, 1951; Brunswick, 1955). A simple stimuli as objects depends on our prior experience.
example of the effect of context and expectations This was shown by Steinfield (1967), who found
can be seen in Demonstration Box 17-3. that when observers were told a story about an
Most of our percepts are constructed from in¬ ocean cruise they identified Figure 17-IOC as a
complete stimuli. Look at Figure 17-10A. It is steamship in less than 5 sec. Observers who were
clear that this represents a dog, yet it should also told an irrelevant story took six times longer to
be clear that there is no dog present. The figure is identify the figure.
completely constructed in the mind’s eye of the ob¬ Some of the experiences that affect perception
server. The elephant in Figure 17-1 OB will proba¬ may take place outside of conscious awareness. For
bly be somewhat more difficult to identify. The less instance, much earlier in the chapter we spoke of
familiar the object, the more difficult is the identi¬ an experiment in which researchers presented a
fication. Furthermore, you must begin with the ini¬ word for so brief a time period that it could not be
tial hypothesis that there is some object there in the identified. They found that if the word was pre¬
first place, otherwise you may never see any pat¬ sented several times, even though the length of
tern at all (Reynolds, 1985). Flowever, once having time of each presentation was not increased, the
seen (or “constructed”) the figure, the meaningful word was eventually identified (Haber & Hershen-
organization will be apparent immediately when son, 1965; Uhlarik & Johnson, 1978). If the word
496 Chapter 17/Learning and Experience

DEMONSTRATION BOX 17-3. A Context Effect on Perception

Read the accompanying handwritten message. You 15 are identical. In addition, the h in the word phone
probably read it as, “My phone number is area code and the b in the word number are identical, as are the
604, 876-1569. Please call!” If you did, you were d in the word code and the l in the word please. You
being affected by several contextual influences on saw each letter or number within a context, when you
perception. Go back to the message and look care¬ first read the message, and this context determined
fully at the script. You will see that the two pairs of how you interpreted the script character.
characters you read as the word is and the numbers

was not identified on the first presentation, why putations required for identification. The partially
should an observer be able to identify it after re¬ processed stimulus is held in memory, and as the
peated exposures? Dodwell (1971) has suggested information extraction continues, hypotheses are
that very fast presentations of stimuli do not give formed and checked until the stimulus appears to
the brain sufficient time to do the necessary com¬ make sense. At this point, the conscious identifi-

F1GURE 17-10 Some degraded stimuli that may be seen as objects. (Based on Street, 1931)
 Context and Meaning 497

cation response takes place (Doherty & Keeley, saw before. She is funny—big eyes, a big nose,
1972). and a big mouth.” Experimenter: “Look at the
The same perceptual hypotheses, allowing us forehead.” Observer: “She has a small curl in the
to eventually formulate a percept from minimal or middle.”
degraded input, can also modify our perception so Even with more brief presentations, this ob¬
that it no longer accurately represents the stimulus. server still insisted that all that appeared on the
For example, Ross and Schilder (1934) presented forehead was a curl of hair. Third eyes do not oc¬
observers with a series of briefly flashed line draw¬ cur normally, so we apparently correct our percept
ings. Some of the drawings were incomplete or dis¬ on the basis of our expectations—we see extra hair,
torted, such as three-armed people and faces with a not extra eyes. You may see how expectations alter
mouth missing. A look at some of their observers’ our perception in Demonstration Box 17-4.
comments is informative. When presented with the Language as well as expectations may modify
side view of a dog with the left hind leg missing, percepts. This view was advanced by Whorf (1956)
the subject reported: “It’s a dog, a wolf, two ears and Sapir (1939), who suggested that specific lan¬
stand upright, a round mouth, a long tail.” The guage labels for certain types of stimuli increase
experimenter then instructed the observer: “Look the accuracy of perceptual processing. They main¬
at his legs.” Observer: “He has five toes on each tained, for example, that the Eskimo, whose lan¬
leg.” Experimenter: “Look at the hind legs.” Ob¬ guage has a wide variety of different names for
server: “I saw two; the tail goes up.” different kinds of snow, may be able to make better
Despite continued pressure from the experi¬ discrimination among types of snow than those of
menter, the observer continued to correct the per¬ us who speak English and have only the single la¬
cept, filling in the missing leg on the hypothesis bel snow.
that dogs have four legs. An example of how language can affect per¬
These researchers also used a drawing of a ception is seen in a classic experiment conducted
woman’s head facing forward. She had two large by Gottschaldt (1926, 1929), who gave observers
eyes as well as a large third eye on her forehead. from 3 to 520 presentations of a simple target. He
One observer described the picture as “a woman then asked them to find this target in a more com¬
with long hair, black, two eyes, one nose, one plex figure. Such a target is shown in Figure 17-
mouth, two ears.” The stimulus was presented 11, where Figure A is embedded in Figure B. He
again briefly, and the observer was asked if the reported that prolonged experience with the simple
forehead was in order, to which he replied, “Yes.” figure did not make it any easier to find when it
After several other stimuli were presented again, was hidden in the more complex stimulus. Djang
the observer now reported: “The same woman I (1937) repeated this experiment. However, her

DEMONSTRATION BOX 17-4. An Expectancy Effect on Perception

Turn to Color Plate 9 and quickly count the number the aces of spades are printed in red ink, rather than
of aces of spades that you see. Then return to this black. Since you “expect” spades to be black, your
demonstration box. Although you probably only saw recognition process for incongruent or unexpected
three aces of spades, there are actually five. Two of stimuli is impaired.
498 Chapter 17/Learning and Experience

FIGURE 17-11 Figure A may be found embedded in Figure B.

servers did not just look at the simple stimulus; simple line drawings and associated each with a la¬
they were required to draw it. She reports that bel. Observers were then asked to reproduce these
when such active practice is combined with expo¬ drawings. In general, their reproductions were
sure, there was an improvement in later identifica¬ biased in the direction of the verbal label. When
tion of the simple component embedded in the presented with Figure 17-12A and told that it was
more complex figure. Schwartz (1961) then added a broom, observers tended to reproduce patterns
a language component. He had observers learn dis¬ similar to Figure 17-12B. When told that it was a
tinctive verbal labels to attach to each of the simple rifle, observers tended to reproduce patterns simi¬
figures. This modification of the experiment re¬ lar to Figure 17-12C. In this experiment the repro¬
sulted in an improved ability to find and identify ductions occurred only a few moments after the
the named stimuli. stimulus was taken away. Such distortions in our
However, verbal labeling can either help or recollections of what we have seen may have im¬
hinder recognition. Ellis and Muller (1964) trained portant consequences for many behaviors, includ¬
subjects using either a narrow labeling system (1 ing such things as scientific observation and
label for each shape) or a wide labeling system (1 eyewitness testimony (Wells & Loftus, 1984). For
label for 4 shapes). In later tests of recognition for example, eyewitness reports of events that occurred
these forms, the group using the narrow labeling during a crime tend to be remarkably unreliable,
approach performed better than a group that simply even when obtained immediately after the event.
observed the shapes without labeling; the group Observers have a tendency to include details that
who used the wide labeling system had the fewest they could not have seen. Such details are often
correct recognitions. Thus, the wide labeling provided on the basis of the observer’s expectations
seemed to emphasize the learning of similarities or biases (Buckhout, 1976; Loftus, 1979; Yarmey,
between the stimuli, which later resulted in poorer 1979).
discrimination and recognition. Loftus (1974) demonstrated how words used
to question observers about a filmed auto accident
could cause them to distort their perceptual mem¬
Eyewitness Testimony
ory. When witnesses were asked about what hap¬
Language and expectation can also serve to modify pened when one car “smashed into another” as
our reports of what we have seen. Carmichael, Ho¬ opposed to using a more neutral phrase, such as
gan, and Walter (1932) presented observers with “made contact with the other,” they were more
 Environmental and Life History Differences 499

Reproduction when it
Stimulus Reproduction when it
was labeled a "broom”
was labeled a "rifle”

FIGURE 17-12 The effects of labels on later perceptual recall.

likely to report having seen broken glass flying duced later to recall the color of the wallet as
about, even though there was none. We might ar¬ brown (Loftus, 1979). In fact, recent evidence
gue that such distortion is a distortion in memory, seems to indicate that the suggestions of false con¬
rather than in perception. Unfortunately, whenever text that are provided after the actual perceptual
an observer reports what he has seen, he is always event has occurred act only to fill in gaps in the
dealing with events that occurred in the past. For perception, much as the observer’s expectation that
instance, the description of a picture flashed on a dogs have four legs caused the missing leg to be
screen in front of an observer for a fraction of a added perceptually in the experiment discussed ear¬
second is really taken from his memory, since the lier (Yuille, 1984). Structuring the questioning to
image is no longer present by the time the observer fit the actual sequence of events may help to pre¬
begins his report. Thus, any perceptual report must vent such eyewitness errors, perhaps by restoring
have some memory component. the original context that was present during the ac¬
Increasing the time interval between the per¬ tual viewing of the event (Geiselman, Fisher,
ception and the report increases the amount of de¬ MacKinnon & Holland, 1986; Morris & Morris,
viation from what was actually observed. For 1985).
instance, in another experiment Loftus (1974)
showed observers a brief videotape of an automo¬
bile accident and then asked them questions about ENVIRONMENTAL AND LIFE HISTORY
what they had just observed. For one group of ob¬ DIFFERENCES
servers one of the questions was, “How fast was
the white sports car going while traveling along the We have seen earlier what effect the availability,
country road?” For the other group the question or nonavailability, of particular environmental
was, “How fast was the white sports car going stimuli can have on the development of sensory
when it passed the bam while traveling along the systems, and hence on the observer’s later percep¬
country road?” In fact, there was no bam present. tual abilities. There also are aspects of our environ¬
Yet when questioned about the incident a week ment and culture that may teach us different
later, more than 17 percent of the group exposed to perceptual strategies, and may alter the expecta¬
the false suggestion about a bam answered the tions and analyses that we bring to each new per¬
question “Did you see a bam?” by saying “Yes,” ceptual situation. Thus, if we live in the desert or
as opposed to only 3 percent of the group that did on the pampas or the plains, we are exposed to
not get such a suggestion. broad vistas of open space that are never experi¬
When we have clearly perceived something enced by a forest dweller. If we live in a techno¬
the perceptual memory resists any later change. logically advanced country, we are exposed to
Thus, in one experiment subjects viewed a series visual stimuli (such as photographs and television)
of slides showing the theft of a large, bright red that are usually not available to someone dwelling
wallet. Virtually none of the observers could be in¬ in the African bush or the Australian outback. Such
500 Chapter 17/Learning and Experience

differences, especially when experienced over an the picture ought to lead to distorted percepts (Ken¬
entire lifetime, may have dramatic consequences nedy & Ostry, 1976). However, such distortions do
for perceptual processes. not appear (Rosinski & Farber, 1980). We could
enlarge indefinitely this list of discrepancies be¬
tween a real scene and a picture of it. For instance,
Picture Perception
the picture is flat, whereas the real world is three-
In our civilized, urbanized, and media-intensive dimensional; the picture is interpreted correctly
culture, we are inundated with images. Not just the even if its colors are all wrong, or even if there are
images of our immediate environment, but images no colors at all; and so forth. Such considerations
representing environments or objects that are not have led some theorists to conclude that pictures
present. Some of these latter images are in the form may be statements in a sort of visual language that
of changing patterns of color or black and white are created and interpreted according to the agreed-
shown on televisions and in cinemas. There are upon set of conventions in any given culture. Thus,
also photographs in magazines and newspapers, they are not simply representations of reality at all
where we might “see” a baby elephant peaceably (Gombrich, 1972; Goodman, 1968). At the very
grazing a few feet in front of its gigantic mother, least, they must be interpreted as hypotheses shared
all in a 5-cm-square smudge of black ink on a per¬ among individuals growing up with a common her¬
fectly flat surface. If you have some artistic talent, itage (Gregory, 1971). By either of these two the¬
you may be able to represent such a scene with a ories, however, the perception of pictures must be
few strokes of a pen on a sheet of paper, and thus learned in some manner.
be able to let your friends “see for themselves” Before we investigate whether we must learn
what you have seen. This seems like a perfectly to interpret pictures, it is important for us to specify
natural fact of life. that we are really talking about two separate skills.
Pictures, of the sort that we encounter daily, The first is the ability to identify objects depicted
are often viewed as simply “windows” through in a picture, and the second is the ability to inter¬
which we see other worlds (Haber, 1980). Cer¬ pret the three-dimensional arrangement implied in
tainly these images must follow all the same optical the flat image.
laws as the real world. Certainly every observer Hochberg and Brooks (1962) conducted a he¬
must follow these laws to interpret such stimuli in roic experiment, using one of their children as the
the same way that we do. Unfortunately, neither subject. The child was reared to the age of 19
statement is completely “certain.” Pictures do months carefully shielded from any sort of pictorial
contain much information that mimics the optical representation. This meant that the television was
patterns encountered in natural viewing (Gibson, never used in the child’s presence, nor were there
1979; Sedgwick, 1980), but there are many dis¬ magazines or picture books. Even the labels on
crepancies between the pictured and real image. cans and boxes of food were removed or covered.
For instance, the actual sizes of the pictured images When the child was tested after this restricted rear¬
are usually too large or too small, which in turn ing, he had no difficulty identifying pictures of
ruins the geometrical correspondence between the common items. This implies that we need not learn
image and the actual scene (Lumsden, 1980). Fur¬ to interpret patterns or drawings as representations
thermore, even if we could make the geometry of of real world objects.
perspective perfect, it would only be correct for The unlearned nature of picture identification
one single viewing angle, and viewing any pictured seems to be supported by the fact that color photos
image from a vantage point other than the view¬ are interpreted readily when shown to individuals
point adopted by the camera or artist who produced who have lived in cultures where they have never
 Environmental and Life History Differences 501

experienced pictures (Hagen & Jones, 1978). How¬ Chapter 10). Thus, to see a drawing or a photo¬
ever, when black-and-white photos or drawings are graph as representing an arrangement of objects in
used, individuals reared in isolated cultures some¬ three dimensions, rather than as a flat surface with
times have difficulties that are strange to those who different shadings of dark and light, you must at¬
have been reared with the continuous company of tend to some depth cues and ignore others (Pick,
graphic images. Deregowski (1980) has collected a 1987). An observer’s particular perceptual strategy
number of such reports, including one from a Scot¬ may depend on her life history and the relative fre¬
tish missionary working in Malawi (a country in quency with which certain cues are encountered in
southwestern Africa between northern Rhodesia the immediate environment.
and Mozambique) nearly 75 years ago: Hudson (1960, 1962) attempted to separate
cultural factors associated with the use of pictorial
Take a picture in black and white, and the natives
depth information. His technique consisted of using
cannot see it. You may tell the natives: ‘This is a
a series of pictures that depicted certain combina¬
picture of an ox and a dog’; and the people will
look at it and look at you, and that look says that tions of pictorial depth cues. Figure 17-13 shows
they consider you a liar. Perhaps you say again, one picture similar to those used by Hudson; as you
‘Yes, this is a picture of an ox and dog. Look at can see, it depicts a hunting scene containing two
the hom of the ox, and there is his tail!’ And the pictorial depth cues. The first is interposition, in
boy will say, ‘Oh, yes and there is the dog’s nose
which objects closer to the observer block the view
and eyes and ears!’ Then the old people will look
again and clap their hands and say, ‘Oh yes, it is of portions of more distant objects. Since the hun¬
a dog.’ ter and the antelope are covering portions of the
rocks, they appear closer to the observer than the
Clearly, such a report indicates that the indi¬ rocks.
viduals involved did not respond to the photo with The second pictorial depth cue contained in
the immediate spontaneous object identification this drawing is familiar size. We know the relative
characteristic of our viewing of pictures. Still, sizes of familiar objects; therefore, if an object is
when their attention was directed to the relevant depicted as relatively small or large, we will judge
aspects of the pattern they did have an “Aha!” ex¬ its distance from us in a way consistent with our
perience, indicating that the ability to identify the expectations based on its known size. For example,
pattern was there, although they lacked training to an elephant is a very large animal. However, in
direct their attention appropriately. Figure 17-13 the elephant is one of the smaller
Although there may be a general ability to items in the picture. If we are responding to the cue
identify objects depicted in pictures, interpreting of familiar size, we would tend to see the elephant
the implied spatial relationships seems to be more as being the most distant object in this hunting
subject to cultural and educational influences. Iden¬ scene. When something as large as an elephant
tifying depth in a flat image requires a certain casts a smaller image than an antelope, the elephant
amount of selection among the perceptual cues must be farther away, since we know it is physi¬
available. For instance, the photograph might in¬ cally larger than the antelope.
clude such cues for depth as linear perspective, in¬ Hudson used these stimuli because they are
terposition, and texture gradients, among others uniquely constructed to allow for both two-dimen¬
mentioned in Chapter 10. However, there are also sional (no use of pictorial depth) and three-
cues indicating that the picture is flat; there is no dimensional (full use of pictorial depth) types of
binocular disparity between items in the picture, responses. Suppose we asked an observer to de¬
and all the elements in the photo require the same scribe what she saw in this picture. First we would
degree of accommodation and convergence (see expect her to identify correctly all the component
502 Chapter 17/Learning and Experience

FIGURE 17-13 A figure used to test ability to respond to pictorial depth cues. (Based on Hudson, 1962)

objects in the picture. However, suppose we also & Pick, 1972; Pick, 1987). You can explore your
asked her to describe the actions taking place. A own tendencies to use certain depth cues but not
correct three-dimensional response would indicate others by trying Demonstration Box 17-5.
that the hunter was attempting to spear the antelope Culturally determined conventions associated
(which is, of course, nearer to him than the ele¬ with the interpretation of pictures can be shown
phant if you are perceiving pictorial depth). A two- best in situations where the flat, stationary picture
dimensional response would state that the hunter is is supposed to depict not only three-dimensionality
attempting to spear the elephant, which is actually but also motion. For instance, Figure 17-14 depicts
physically closer to the tip of the spear in the pic¬ a scene in which there are three different forms of
ture. Such a response would indicate that the ob¬ motion. From left to right, we see a speeding car,
server had not responded to either the interposition a boy rapidly whipping his head around, and a dog
or the familiar size cues that place the elephant at with a wagging tail. Of course, there is no actual
a greater perceptual distance from the hunter than motion, yet we “read” such motion into the pic¬
the antelope. ture. Within Western cultures, such interpretation
Stimuli similar to these have been used in a of motion in pictorial arrays may appear as early as
number of studies conducted throughout Africa, to 4 years of age (Friedman & Stevenson, 1975).
test observers from a number of tribal and linguistic Non-Western cultures, without pictorial experi¬
groups (Deregowski, 1980). The results indicate ence, however, virtually never “see” movement in
that relatively isolated and uneducated African ob¬ such representations. The likelihood that movement
servers have difficulty seeing pictorial depth within will be seen in such an array increases with edu¬
these pictures, relative to more urbanized Western cation, urbanization, and exposure to pictorial
observers, a fact that has been verified using other materials (Duncan, Gourlay & Hudson, 1973;
types of pictures (Jahoda & McGurk, 1974). The Friedman & Stevenson, 1980).
ability to perceive three-dimensionality in pictures
seems to be improved if more depth cues are added
Illusion and Constancy
(Hagen & Jones, 1978; Killbride & Leibowitz,
1975), or if formal education, involving the use of There are certain facets of the environment that
picture books, drawings, and so forth, has been ex¬ make us more or less responsive to certain patterns
perienced (Killbride & Robbins, 1968; Leibowitz of depth cues appearing in pictures. For instance,
 Environmental and Life History Differences 503

DEMONSTRATION BOX 17-5. Cross-Cultural Differences in Perception

The figure accompanying this box is sometimes called

the “Devil’s tuning fork.” Look at the figure for
about 30 seconds or so; then close the book and try
to draw it from memory. Return to this box when you
have done this.
Most of you probably found this task to be quite
difficult. The source of your difficulty comes from the
fact that your cultural experience with graphic repre¬
sentations has caused you to interpret this two-dimen¬
sional stimulus as a three-dimensional object. Unfor¬
tunately, such an interpretation leads to problems
since the depth cues implied in this figure are ambig¬
uous. It is interesting to note that Africans who have
not received formal education have no difficulty re¬
producing the figure. Since they do not interpret the
figure as three-dimensional, they merely see a pattern
of flat lines, which is easy to reproduce.

the carpentered world hypothesis begins with the tend to use curved furrows. Individuals living in
observation that in the urbanized Western world, such a world would not be expected to rely on lin¬
rooms and buildings are usually rectangular, many ear perspective as heavily as those of us living in a
objects in the environment have right-angled cor¬ more linear environment.
ners, city streets have straight sides, and so forth. In a classic study Segall, Campbell, and Her-
Surrounded by such an environment, we may learn skovits (1966) compared the responsiveness of in¬
to depend more heavily on depth cues based on lin¬ dividuals in carpentered versus noncarpentered
ear perspective than would people who live in more environments to certain types of depth cues. How¬
primitive rural environments (Coren & Girgus, ever, instead of using Hudsonlike pictures as stim¬
1978; Gregory, 1966; Segall, Campbell & Her- ulus materials they chose a more subtle class of
skovits, 1966). For example, rural, isolated Zulus patterns, namely the visual-geometric illusions.
have been described as surrounded by a circular Some of these configurations have already been
culture. They live in round huts with round doors. discussed in Chapter 14, where we pointed out how
They do not plough their land in straight lines but susceptibility to size distortions in some figures,

FIGURE 17-14 Scenes conventionally recognized as depicting motion by Western observers, but not
necessarily by non-Western observers.
504 Chapter 17/Learning and Experience

such as the Mueller-Lyer illusion, may be depen¬ size constancy at greater viewing distances. Turn-
dent on a three-dimensional interpretation of the bull noted one instance when he had taken his
pattern (to refresh your memory refer back to Fig¬ Bambuti guide, Kenge, out of the forest for the first
ure 14-6). As we discussed then, the apparently time in his life. They were crossing over a broad
longer portion of the Mueller-Lyer figure may be plain and happened to spot a herd of buffalo:
interpreted as a comer of a room receding in depth
Kenge looked over the plains and down to where
because the wings of the illusion act as linear per¬
a herd of about a hundred buffalo were grazing
spective cues. The inappropriate application of size some miles away. He asked me what kind of in¬
constancy based on this interpretation of the wings sects they were, and I told him they were buffalo,
of the figure as perspective cues results in our over¬ twice as big as the forest buffalo known to him.
estimation of the size of this segment of the figure He laughed loudly and told me not to tell such
stupid stories. . . . We got into the car and drove
relative to the segment with the inward-turned
down to where the animals were grazing. He
wings. Since pictorial depth information is thought watched them getting larger and larger, and
to play a role in the formation of these illusory though he was as courageous as any Pygmy, he
percepts, these types of configurations are well- moved over and sat close to me and muttered that
suited to an exploration of the carpentered world it was witchcraft. . . . Finally, when he realized
that they were real buffalo he was no longer
hypothesis.
afraid, but what puzzled him still was why they
Segall et al. (1966) gathered data from through¬ had been so small, and whether they really had
out Africa and also from several groups of people been small and suddenly grown larger, or whether
living in Evanston, Illinois. Although there are it had been some kind of trickery. (From C. Turn-
variations within the noncarpentered samples, the bull, American Journal of Psychology, 74. Copy¬
right 1961, The University of Illinois Press.)
average Mueller-Lyer illusion was greater for the
more urban groups. Similar results have been re¬ Turnbull’s description of Kenge’s perceptual
ported for other perspective-related illusions (Coren
impressions suggests that our experience with par¬
& Girgus, 1978; Deregowski, 1980; Killbride ticular stimuli prevalent in our immediate environ¬
& Leibowitz, 1975). ment can result in differences in how we perceive
Although the observed differences in illusion
new stimuli and situations. It seems that we learn
susceptibility for different cultural groups may be
to utilize stimulus information that we encounter
partially caused by factors other than experience
frequently, but fail to learn to utilize stimulus in¬
with a carpentered world (Berry, 1971; Coren &
formation that is rare. This holds for the auditory
Porac, 1978; Pollack & Silvar, 1967), -there is am¬
as well as the visual environment.
ple evidence that the absence of experience with
certain types of depth cues impairs certain other
perceptual functions (such as size constancy) de¬
Speech
pendent on depth perception. One of the most
striking examples of this was provided by the an¬ The most dominant feature in our auditory environ¬
thropologist Turnbull (1961). He observed the be¬ ment is the constant flow of language sounds that
havior of the Bambuti Pygmies, who live in the surrounds us. As discussed in Chapter 12, each
Ituri Forest in the Congo. Since they live in the language uses a small set of word-differentiating
dense rain forest, their vision is generally limited phonemes, which are the functionally characteristic
to short distances, with vistas that only extend for, sounds of that language. Since different languages
at most 30 m. Therefore their life history seems to use different subsets and combinations of these
lack the visual experience needed to learn to use phonemes, experiments on people reared in differ¬
the depth cues responsible for the maintenance of ent linguistic settings offer a unique opportunity to
 Environmental and Life History Differences 505

observe the effects of specific kinds of experience make the discrimination. It seems that at some time
on perception (see Kuhl, 1987). Because some during the first year of exposure to the language,
sounds may be treated as distinctively different in infants begin to selectively respond to certain as¬
some languages, and not in others—for instance, pects of their linguistic environment and to selec¬
the sounds “r” as in rope and “1” as in lope are tively lose their ability to respond to phonemic
different phonemes in English, but not in Japa¬ distinctions not used in their native language (Wer¬
nese—we would expect a bias in the auditory ex¬ ker & Tees, 1984).
perience of individuals brought up surrounded by This is not to say that individuals cannot learn
one or the other of these two languages. Numerous to make certain phonemic distinctions. Evidence
studies have shown that adults who have grown up suggests that learning through exposure and expe¬
with exposure to only one language often have dif¬ rience plays a role in the ability to discriminate
ficulty discriminating certain linguistic contrasts between various linguistic sounds, and some lin¬
characteristic of other languages (Strange & Jen¬ guistic discriminations seem to be learned during
kins, 1978; Werker & Tees, 1984). This type of the childhood years (Eilers, Wilson & Moore,
difficulty may persist even if the adult has learned 1979). However, this appears to be limited to cer¬
the other language and appears to be fluent in it. tain dimensions of the sound. Tees & Werker
For example. Goto (1971) recorded pairs of words (1984) showed that short-term intensive training
that contrasted the “r” and “1” sounds (such as improved the ability of native English speakers to
lead vs. read, or play vs. pray). Several native make certain non-native (Hindi) speech discrimi¬
Japanese speakers, who were bilingual in Japanese nations, although after 5 years of language study
and English, could produce these sounds so that the ability to make these discriminations was al¬
native English-speaking listeners could differentiate ready apparent. An interesting additional finding
them without error. However, this seems to be a pertained to individuals who spent their early years
learned ability to produce rather than to perceive in a setting where Hindi was spoken (perhaps by a
the phonemic difference, since, when asked to lis¬ live-in relative). Even though these individuals had
ten to recordings of pairs of words that contrasted never studied the language, and as adults were un¬
these phonemes, the native Japanese speakers could able to speak, understand, or write more than a few
not do so, even when listening to their own speech words of Hindi, it was found that they could make
productions! the phonemic discriminations that non-natives
The mechanism responsible for our ability to found impossible. Thus, their early experience
discriminate some speech sounds but not others seems to have “tuned” their speech-sound decod¬
is still somewhat mysterious. Surprisingly, infants ing capacity for certain dimensions even though
seem to be bom with the ability to discriminate cer¬ they were not actually speaking the language.
tain sound pairs not used in their native tongue, and In terms of our earlier discussion about the re¬
appear to lose this ability as adults (Trehub, 1976; lationship between experience and development,
Werker & Tees, 1984). A striking example was these results suggest that different aspects of the
provided by Werker, Gilbert, Humphrey, and Tees perception of speechlike sounds follow different
(1981), who presented English-speaking and Hindi¬ courses. Although some auditory discriminations
speaking adults with pairs of sounds that are differ¬ are facilitated through contact with particular
entiated as different phonemes in Hindi, but not in sounds in the linguistic environment, others are lost
English. As we might expect, the adult Hindi through their absence or rarity, thus showing that
speakers could make the discrimination, but the experience is necessary for their maintenance (cf.
adult English speakers could not. The interesting Walley, Pisoni & Aslin, 1981). Overall, this con¬
result, however, is that 6-month-old infants could firms that much of what we perceive and many of
506 Chapter 17/Learning and Experience

the perceptual distinctions we make are strongly in¬ of male office, farm, and factory workers. The
fluenced by the culture and environment in which 0-decibel point on the vertical axis represents the
we were reared. average minimum threshold for an auditory expe¬
rience. The three curves on the graph plot the av¬
erage threshold sound intensity at each of the
Effects of Occupation
frequencies used in the test. As you can see, the
Even within a given culture there is selective ex¬ group of factory workers has a lowered sensitivity
posure to different sets of environmental stimuli. (higher average thresholds). Fortunately, many fac¬
You are exposed to your occupational setting for tory workers have begun to wear earphones while
about one-half of your adult working life, and spe¬ working; these help protect them from the destruc¬
cific sets of occupational experiences can affect tive effects of continual noise exposure.
your perceptual abilities both at the physiological According to Kryter (1985), factory workers
and at higher cognitive levels. One aspect of an are not the only group that should be concerned
occupation that may have physiological effects on about exposure to very loud sounds. For instance,
a sensory system is the magnitude of sound, light, soldiers exposed to the sound of gunfire and airline
or chemical stimulation to which you are exposed. pilots exposed to engine noises have been shown to
Consider, for example, the amount of auditory have hearing deficits. The loss tends to be greatest
input that bombards you in your occupational set¬ in the higher frequency ranges and seems to in¬
ting. Some work environments are relatively quiet crease in severity as the length of the exposure in¬
(such as offices and small stores); others are asso¬ creases. Thus, airline pilots who had 1,000-2,000
ciated with continuous, high-intensity noise (for hr flying time in the noisy planes of the 1960s had
example, factories and mills). Figure 17-15 illus¬ an average audibility threshold of approximately 0
trates the effects of noise on hearing for different dB at a sound frequency of 4000 Hz, whereas more
occupations. The horizontal axis represents the fre¬ experienced pilots, with 10,000-16,000 hr flying
quencies at which hearing was tested in a sample time, had an average audibility threshold of 10 dB
at that sound frequency.
Rice, Ayley, Bartlett, Bedford, Gregory, and
Hallum (1968) have shown that performers of rock
music may also suffer hearing losses. In Figure 17-
16 average audibility thresholds are again plotted
for various test frequencies. The lowest curve rep¬
resents the thresholds of the control group of non¬
performers. Notice that relative to nonperformers
of the same age, the rock performers have elevated
audibility thresholds (lower sensitivity). At 4000
Hz there is an approximately 20-dB difference be¬
tween the thresholds of the performers and the con¬
trols. To give you a reference point, a 20-dB
difference would be roughly equivalent to being
able to hear a normal conversational tone as op¬
posed to a shout. Figure 17-16 also shows the im¬
FIGURE 17-15 The effect of occupation on mediate effects of prolonged exposure to very loud
hearing. (Based on Glorig, Wheeler, Quigle, Grings sounds. The dotted curve on this graph plots the
&. Summerfield, 1970) measured thresholds immediately after 85 min of
 Environmental and Life History Differences 507

often given training to improve the basic visual ca¬

pacities associated with binocular depth perception,
visual acuity, and so forth. There is a strong con¬
viction among students who participate in such pro¬
grams that such visual training is quite effective;
however, actual optometric measures do not show
any beneficial changes (Goodson & Rahe, 1981).
Similarly, the police believe that because of their
training and practical experiences they tend to de¬
velop a superior ability to observe details. Ac¬
tually, much evidence indicates that this is not the
case (see Loftus, 1979; Yarmey, 1979), although
Frequency of test tone (Hz) some recently developed training programs may
improve this ability (Yuille, 1984).
FIGURE 17-16 A comparison of the hearing of
rock musicians to that of nonperformers. (Based
on Rice et al., 1968) Perceptual Set

Experiences we have in an occupational or other

exposure to very loud music; as you can see, the setting may bias our perception and interpretation
threshold at 4000 Hz has risen to 25 dB. of various stimuli. This seems to be because spe¬
In a similar vein, there are occupations that cific past experiences produce a sensitization or
expose the eyes to high-intensity lights (such as predisposition to “see” a situation in a certain
welding arcs and furnace blazes). In the same man¬ way, especially when several alternative perceptual
ner that exposure to prolonged high-intensity sound experiences are possible (as when the stimulus is
can impair hearing, exposure to prolonged high- ambiguous or degraded because of poor viewing
intensity light can permanently impair vision (No- conditions). Technically this is known as a percep¬
ell, 1980). tual set, and it refers specifically to the expectan¬
Some occupations involve specific experiences cies or predispositions an observer brings to the
or training programs that attempt to alter perceptual perceptual situation (Coren, 1984b). In many re¬
abilities. Pilots, especially those in the military, are spects, set can be thought of as another example of

DEMONSTRATION BOX 17-6. Set

Something of the flavor of perceptual set can be ex¬ you pronounce it as if it were organized as the name
perienced by simply reading the set of words below Mac Hines, or did you pronounce it as if it were or¬
out loud: ganized as the more familiar and natural form that
MACBETH MACARTHUR MACWILLIAMS makes the common word machines? If you pro¬
MACNAMARA MACDILLON MACDONALD nounced it as the name, organizing the prefix Mac
MACMASTER MACDOWELL MACHINES into a separate unit, you were demonstrating the ef¬
MACKENZIE fects of perceptual set.
Now look back at the next to the last word. Did
 508 Chapter 17/Learning and Experience

selective attention (as we discussed in Chapter 15), and Schulte were interested in exploring the notion
in which the observer is set to process some but not that police students would be predisposed to inter¬
all incoming information, or to organize it in a spe¬ pret this particular ambiguous situation in terms of
cific manner. To get a better feeling as to how set the violent as opposed to the nonviolent scene.
operates, you might try Demonstration Box 17-6. They compared the performance of advanced police
As an example, let us consider police as ob¬ administration students with two control groups:
servers and eyewitnesses, since this is an occupa¬ beginning police students and university students.
tion in which observation is an important part of In general they found that the advanced police stu¬
the job. Some findings suggest that perceptual set dents interpreted the stereograms as depicting vio¬
may influence the observations of police in certain lence approximately twice as many times as the
situations. In one study, police officers and civil¬ other two groups. Thus, their data provide some
ians were shown films of a street scene over a pe¬
riod of several hours. Their task was to watch for
various wanted people (whose photos were on dis¬
play below the screen) and for certain types of ac¬
tions (normal exchanges of goods vs. theft, and so
forth). The police tended to report more alleged
thefts than the nonpolice, but there was no signifi¬
cant difference between the police and civilians in
their actual detection of people and actions (Clif¬
ford & Bull, 1978).
A more subtle demonstration of this effect of
set was provided by Toch and Schulte (1961), who
studied perception of violence and crime in ambig¬
uous visual scenes. They simultaneously presented
different pictures to each eye in a stereoscope (see
Chapter 10). One eye was shown a violent scene
and the other a nonviolent one, as in the pair of
stimuli in Figure 17-17. If these two views are seen
simultaneously by the two eyes, perceptual confu¬
sion should result. Observers tend to resolve this
ambiguous situation in favor of one scene or the
other; that scene then dominates the percept. Toch

FIGURE 17-18 Although A may appear to be a

relatively random collection of black shapes,
FIGURE 17-17 A stereogram used to test for perhaps depicting some sort of boot, it is actually
occupational influences on the perception of similar to B in that it contains the word FLY
violence. (From H. H. Toch, & R. Schulte, 1961, British Journal depicted in the white areas, only in A the word is
of Psychology, 52, 389-3931
in Chinese calligraphy.
 Glossary 509

evidence that certain occupations, especially those experience is essential if perceptual development is to
requiring intensive training, may set an individual proceed normally.
to interpret ambiguous stimulation in a particular Enhancement An improvement in the final level of
way. an existing ability caused by relevant experience.
We are not singling out the police for partic¬ Exafference Stimulus input that acts on a passive
observer.
ular scrutiny. Perceptual set associated with occu¬
pational training and experience and education can Facilitation An increase in the rate at which an abil¬
ity develops, but not in its final level.
affect all groups of individuals. An example of how
Familiar size A cue to depth based on the known or
our educational background can bias our perception remembered size of objects.
can be seen by looking at Figure 17-18A. Most
Illusion decrement The decrease in the strength of a
Western observers will see a complex pattern of visual illusion with prolonged viewing.
black shapes, which, by some stretching of the Induction A process in development whereby expe¬
imagination, might appear to cohere into some sort rience determines the presence and final level of an
of a boot. Conversely, Figure 17-18B is quite dif¬ ability.
ferent. Here the white spaces clearly shape the Interposition The depth cue based on the blocking
of an object or part of an object from view by another
word FLY while the black spaces serve as the back¬
closer object.
ground. Our familiarity with the English language
Maintenance Preservation of a developed ability by
helps to focus our attention on this region of the
relevant experience.
figure, and we supply the missing contours, subjec¬
Maturation Development of an ability independent
tively, to complete the percept (see Chapter 11). of experience.
Actually, if you were an educated native Chinese, Oblique effect The phenomenon whereby acuity for
you might be more captured by Figure 17-18A, diagonally oriented stimuli is poorer than for horizontally
since it outlines in the white spaces between the or vertically oriented stimuli.
black shapes the calligraphic character for the Perceptual set The expectancies or predispositions
that an observer brings to a perceptual situation.
Chinese word FLY, and Figure 17-18B might ap¬
pear to be merely five meaningless black shapes Reafference Stimulus input that results from an ob¬
server’s own movements.
(see Coren, Porac & Theodor, 1987). Thus, much
Rearrangement An experimental technique that al¬
of what you see is determined by what your expe¬
ters spatial relations in the visual world.
rience, culture, and education set you to see.
Restricted rearing An experimental technique in
which an animal is reared without exposure to a particu¬
lar class of sensory inputs.
GLOSSARY Selective rearing An experimental technique in
which an animal is reared under conditions that bias the
The following definitions are specific to this book. stimulus input it receives toward a particular class of
stimuli (e.g., it sees only vertical stripes).
Aftereffects Errors in hand-eye coordination that
Set See Perceptual set.
follow adaptation to wedge-prism distortion.
Astigmatism A selective visual bias caused by phys¬ Three-dimensional Possessing pictorial depth.
ical distortion of the cornea. Transactional view point Maintains that any per¬
Carpentered world hypothesis States that individ¬ ceptual experience consists of a complex evaluation of
uals living in urban environments characterized by the significance of available stimuli based on expecta¬
straight lines and angles will tend to depend more on tions and experience.
depth cues based on linear perspective than would people Two-dimensional Lacking pictorial depth.
living in more primitive rural environments characterized Visual field The portion of the visual environment to
by curved lines. which an eye will respond, measured in degrees around
Critical period An interval during which sensory the head.
' CHAPTER
18
Individual
Differences
PHYSIOLOGICAL DIFFERENCES
The Effects of Drugs
The Effects of Physical Pathology
GENDER DIFFERENCES
Visual-Spatial Abilities
Physiological Factors
Psychosocial Factors
PERSONALITY AND COGNITIVE STYLE
DIFFERENCES

511
512 Chapter 18/Individual Differences

H ow often have you heard people arguing over

whether a color is green or blue, whether the
limeters. Actually, A and B are directly in line with
each other. The magnitude of this illusion differs
room was too hot or cold, or whether the coffee is among individuals depending on their age, their ed¬
too weak or strong? Such arguments may represent ucation, whether they are male or female, and even
real differences in the perceptions of the individuals how well they do on spatial skills tests, like those
involved. Remember that perception is not simply that form part of many intelligence scales (Coren
a process by which the qualities of the world get & Girgus, 1978; Coren & Porac, 1987; Girgus &
transferred from “out there” to “in here.” Rather, Coren, 1987). Some of these factors are probably
your final conscious experience of a stimulus in¬ not surprising to you, since we have already dis¬
volves many levels of processing. Not only must cussed how age and past experience can affect per¬
the peripheral sensory receptors be stimulated but ception (in Chapters 16 and 17). However, there
stimuli must be interpreted and encoded. As one are many other variables that operate to make each
ancient philosopher said, “The eyes are blind; only person’s perceptual experiences somewhat unique.
the mind sees.” If this premise is true, it is quite pro- These include physiological factors, such as
able that individuals can differ in the way they per- changes in the sensory receptors themselves or the
eive their worlds, since it is certainly true that no two neural apparatus that decodes the sensory informa¬
minds seem to work in exactly the same way. tion. There is also a contribution from an individ¬
There are many factors that can cause individ¬ ual’s cognitive, or perceptual, style, which is ac¬
uals to have different perceptions even when en¬ tually a reflection of personality differences and
countering identical stimuli. For instance, consider different approaches to gathering information from
Figure 18-1, which shows a common visual distor¬ the environment. Even an individual’s gender
tion called the PoggendorfF illusion. For most peo¬ seems to cause differences in the way sensory in¬
ple it appears that if you extend the line marked A formation is processed. All these factors lead to in¬
it will pass below the line marked B by several mil- dividual differences in perception. One of the most
interesting aspects of perception is the considera¬
tion of why the world you perceive may not nec¬
essarily be the same as the world perceived by
others.

PHYSIOLOGICAL DIFFERENCES

Each of us can be seen as a complex physiological

machine, and we certainly have seen how factors
that alter the structure and function of our sensory
apparatus will alter what we perceive. For exam¬
ple, nonfunctional retinal cones will cause color vi¬
sion deficits. Calcium deposits on the bones of the
inner ear will lessen the intensity of sounds from
the environment while making your own voice
seem very loud to you. In addition to such very
FIGUBE 18-1 The PoggendorfF illusion, in which specific factors, there are other physiological fac¬
line A appears as if it would pass below line B if tors that affect the body generally and also affect
extended, even though both are exactly aligned. perception as a side effect.
 Physiological Differences 513

The Effects of Drugs duced responsiveness to sweet (sucrose), acid

(acetic acid), salt (sodium chloride), and bitter
A number of drugs can cause marked changes in (quinine) stimuli. The strongest reduction of taste
sensory capacities. For instance, smokers are con¬ response was for the bitter stimuli. Alcohol may
tinually ingesting a number of active chemicals. also effect other sensory systems, and recent stud¬
The most important of these is the poison nicotine; ies suggest that these effects may be cumulative.
next in importance is the gas carbon monoxide. One study tested the auditory system and found that
Since these chemicals enter the body predominantly chronic alcoholics show delays in the neural re¬
via the mouth, it is not surprising that the major sponse to auditory signals when compared to non¬
sensory effects of smoking tobacco are on the sense alcoholics (Begleiter, Porjesz & Chou, 1981).
of taste. Absolute taste thresholds are higher for Studies of the effect of alcohol on visual abilities
smokers than for nonsmokers, with smokers being have produced similar findings. Research with
especially insensitive to bitter tastes such as quinine chronic alcoholics has demonstrated differences in
(Kaplan & Glanville, 1964; Sinnot & Rauth, color vision, with alcoholics displaying a higher in¬
1937). However, you may be surprised to find that cidence of color vision deficiencies that nonalco¬
smoking also affects vision. Most of these effects holic observers (Granger & Ikeda, 1968; Reynolds,
are probably due to the inhalation of carbon mon¬ 1979).
oxide, a substance that has been shown to produce You do not have to be an alcoholic or a habit¬
alterations in some visual tasks, particularly those ual drinker to show the effects of alcohol on visual
involving sustained visual attention (Gliner, Hor¬ perception. In some studies of nonalcoholic observ¬
vath & Mihevic, 1983). In addition, smoking tends ers there are indications that higher doses of alco¬
to reduce the ability to make light-intensity- dis¬ hol decrease an individual’s ability to follow a
criminations, especially under scotopic illumination moving target with the eyes (Flom, Brown, Adams
conditions (Rhee, Kim & Kim, 1965). This might & Jones, 1976; Levy, Lipton & Holzman, 1981)
explain why smokers tend to have more nighttime and also decrease the ability of the eye to accom¬
driving accidents than nonsmokers. There is also modate or change the focus of the lens (Miller, Pi-
evidence that ingestion of nicotine slows the rate of gion & Martin, 1985). However, even small
recovery from certain visual aftereffects (such as amounts of alcohol can decrease an observer’s abil¬
the afterimages associated with viewing bright tar¬ ity to detect the onset of a moving stimulus
gets, or the fatigue effects of prolonged staring at (MacArthur and Sekuler, 1982).
certain types of patterns, as we discussed in Chap¬ The overall sensitivity of the visual and audi¬
ters 4 and 5). It has been hypothesized that nicotine tory systems is also affected by many depressant
slows recovery from aftereffects because it in¬ drugs. A popular technique for measuring visual
creases neural inhibition within the visual system responsiveness is the critical flicker fusion fre¬
(Amure, 1978). quency task (usually abbreviated CFF). This task
One general theme that characterizes the data requires an observer to view a flickering light. As
describing the effects of drugs on perception is that the flicker rate is increased the observer will even¬
drugs that depress neural activity—such as seda¬ tually no longer see the successive on-and-off cy¬
tives, barbiturates, tranquilizers, or alcohol—also cles, but rather will see them fused into a steady,
decrease sensory acuity. For example, Helekant continuous light. The flicker speed that results in
(1965) measured the effect of alcohol on taste sen¬ the perceptual shift from an apparently flickering to
sitivity by recording directly from the chorda tym- an apparently steady light is the CFF. The more
pani nerve of a cat. This nerve conveys taste sensitive the eye is to changes in illumination level,
information from most of the tongue. Alcohol re¬ the faster the rate that the light must be cycled on
514 Chapter 18/Individual Differences

and off to cause the perception of flicker to disap¬ 8

pear. There is a similar task used in hearing called

the auditory flutter fusion (AFF) task. In measur¬
ing the AFF a tone, rather than a light, is switched
on and off repeatedly. The AFF is the rate that the
tone must be cycled on and off for the listener to
hear the signal as a continuous sound. Depressant
drugs, such as alcohol and tranquilizers, tend to
lower both the CFF and AFF, thus indicating that
the visual and auditory systems are acting slug¬
gishly and with less sensitivity under the influence
of such drugs (Besser, 1966; Holland, 1960).
Drugs that increase the arousal level of the ob¬ FIGURE 18-2 The Necker cube, in which the
server, such as stimulants like caffeine or amphet¬ face with corner A sometimes appears nearer
amines (and even some of the B vitamins), may in than the face with corner B, and sometimes
reverses so that the face with corner B appears
some instances improve the sensitivity of the ob¬
nearer.
server. These effects, however, do not seem to be
as widespread or as reproducible as those obtained
with depressant drugs. We do find that amphet¬
amines and caffeine tend to increase the respon¬ entation from time to time. Sometimes the face
siveness of both the visual and auditory systems with the comer labeled A seems closer than the face
when sensitivity is measured using CFF and AFF. with the comer labeled B, and at other times B
Such drugs seem also to increase the sensitivity of seems closer. Most people have a fairly constant
the olfactory system (Turner, 1968). rate of reversal for this figure. Many factors can
Another technique used to monitor drug ef¬ affect the reversal rate, such as the way you attend
fects involves reversible figures, such as the one to it (Reisberg & O’Shaughnessy, 1984; Wallace &
shown as Figure 18-2. When you look steadily at Priebe, 1985). Depressive drugs, such as some
this figure (which is called the Necker cube) you tranquilizers, tend to slow the reversal rate (Phillip-
will notice that it tends to reverse its apparent ori- son & Harris, 1984). Demonstration Box 18-1 pro-

DEMQNSTRATION BOX 18-1. The Effects of Stimulants on Figure Reversals

For this demonstration you will need a friend and a keeping you awake for the rest of the chapter). The
watch. Have your friend monitor you while you look effects take about 15 minutes to appear; after this in¬
at Figure 18-2 for 1 minute. Call out each time the terval repeat the viewing process. Look at Figure 18-
figure reverses its orientation and have your friend 2 again for 1 minute while your friend records the
keep count. Next, drink a cup of coffee. Don’t use number of reversals. You should find that the stimu¬
decaffeinated coffee since we want you to receive a lant has increased the number of perceptual shifts that
dose of caffeine. Since caffeine is a stimulant it you experience.
should increase your visual responsiveness (as well as
 Physiological Differences 515

vides a procedure for testing the effects of a ably does not account for the results, since on those
stimulant on this phenomenon. trials when the stimuli were reported, the reaction
The hallucinogenic and psychoactive drugs— times were as quick as when observers were not
including LSD, mescaline, psilocybin, and mari¬ under the influence of the drug (Moskowitz,
juana—are often reported to have profound percep¬ Sharma & McGlothlin, 1972). Intake of marijuana
tual effects. For instance, Aldous Huxley (1963) has also been shown to increase the interstimulus
described his visual experiences after taking mes¬ interval at which visual masking occurs (visual
caline saying: “First and most important is the ex¬ masking is discussed in Chapter 11), indicating that
perience of light. . . . All colors are intensified to it acts somewhat like a sedative and decreases the
a pitch far beyond anything seen in the normal speed of visual information processing (Braff, Sil-
state, and at the same time the mind’s capacity for verton, Saccuzzo & Janowsky, 1981). As with al¬
recognizing fine distinctions of tone and hue is no¬ cohol, prolonged use of marijuana seems to have a
tably heightened.” Unfortunately, although some cumulative effect. This shows up particularly in
aspects of the subjective experience seem to be color discrimination, which has been shown to be
heightened, actual measurements do not always in¬ poorer in habitual marijuana users. These effects
dicate increased sensory sensitivity. For instance, are particularly marked in the blue region (Adams,
Hartman and Hollister (1963) found that LSD, Brown, Haegerstrom-Portnoy & Flom, 1976).
mescaline, and psilocybin all reduced the accuracy There are also some more complex sensory ef¬
of color discriminations. Carlson (1958) showed fects that have been reported after the smoking of
that LSD reduced visual sensitivity in a threshold cannabis. Some observers experience changes in
task, and there are reports of blurred vision (Hoffer depth perception and distortions in the perception
& Osmond, 1967) and slower than normal dark ad¬ of size (Tart, 1971). In addition, there is a report
aptation for observers under the influence of LSD that the autokinetic effect, which is the illusory
(Ostfeld, 1961). Susceptibility to at least one visual movement of a stationary light viewed in total
geometric illusion, the Mueller-Lyer, increases un¬ darkness (illustrated in Demonstration Box 13-8),
der the influence of LSD (Edwards & Cohen, may become exaggerated. This last observation has
1961). However, LSD does seem to improve audi¬ led one group of experimenters to caution against
tory acuity and seems to enhance the CFF (Hoffer night driving while under the influence of mari¬
& Osmond, 1967; Williams, 1979). juana (Sharma & Moskowitz, 1972). Yet mari¬
Similarly, when observers are asked to de¬ juana, even at relatively high intake levels, does
scribe their experiences after ingesting moderate not seem to impair eye-movement facility, since
doses of marijuana (cannabis), they often report neither saccadic eye movements nor the ability to
improved visual clarity and acuity. Unfortunately, pursue a moving visual stimulus with the eyes is
the experimental results indicate that as with LSD affected by its ingestion (Flom, Brown, Adams &
the actual perceptual effects involve losses in sen¬ Jones, 1976).
sitivity. For instance, in a vigilance task where ob¬ Although contact with hallucinogenic drugs
servers were asked to fixate a target and report involves a departure from everyday behavior for
stimuli appearing in the periphery of vision, those most people, many of the stimulants (such as caf¬
who had smoked marijuana produced fewer accu¬ feine) and depressants (such as tobacco and alco¬
rate reports. Such an effect could be due to a nar¬ hol) that alter perception are used commonly.
rowing of attention induced by the drug, or it might Everyday drugs, including antihistamines and as¬
occur because the observers could not be bothered pirin, can cause the perceptual responses of indi¬
to press the switch. This second explanation prob¬ viduals to differ. For instance, aspirin may cause
516 Chapter 18/Individual Differences

dimness of vision or ringing in the ears (Goodman Greek a meaning “not” and gnosis meaning “in¬
& Gilman, 1965). Thus, an individual who has just tuitive knowledge.” People suffering from agno¬
had a cup of coffee or a martini, or who has tried sias seem to perceive but are not capable of
to alleviate a headache, may differ from other in¬ understanding the information presented to them.
dividuals in perceptual responses because of the ac¬ Freud (1953) noticed a form of perceptual dis¬
tions of the ingested drugs. turbance that he called visual object agnosia.
Some of his patients were unable to identify famil¬
iar objects, although there seemed to be no psycho-
The Effects of Physical Pathology
pathological disturbance or readily detectable
There are many pathological conditions that affect elementary damage to the visual apparatus. More
perception. The most obvious of these are maladies recent work (Luria, 1973) has suggested that ag¬
that directly damage a particular receptor organ. nosias might arise from lesions in the secondary
Glaucoma, which causes a pressure increase inside visual areas of the cortex. These lesions do not
the eye, can produce blindness if left untreated, and cause blindness nor do they seem to diminish visual
otosclerosis, which causes the bones of the middle acuity. Rather, they make it difficult for a person
ear to become immobile, will impair hearing. to combine parts of an object so as to identify it.
There are, however, some pathological conditions For example, Luria describes a patient who was
that cause disturbances in very special and complex given a picture of a pair of eyeglasses. The patient
aspects of perception, rather than simply causing a examined the picture carefully in a manner indicat¬
loss of sensitivity to a given stimulus dimension. ing that he was confused and did not know exactly
Such effects are often caused by severe toxic con¬ what it represented. He then started to guess.
ditions, such as carbon monoxide poisoning, as “There is a circle . . . and another circle ... a
well as diseases or injuries that damage or reduce crossbar . . . why, it must be a bicycle?”
the functioning in some parts of the brain (Critch- Such patients also have problems in separating
ley, 1964; Davidoff, 1975; Luria, 1973). These can the parts of the figure from the overall context.
affect such complex functions as the ability to iden¬ Thus, if the patient is shown a drawing of a clock,
tify objects or to place them in space, and they may such as is shown in Figure 18-3A, he can usually
also affect the ability to distribute attention. In gen¬ identify it correctly. However, if the clock is sim¬
eral, such a problem is called an agnosia, from the ply crossed out with a couple of lines, as in Figure

FIGURE 18-3 (A) A figure identified as a clock. (B) A figure no longer identifiable to a visual agnosic.
 Physiological Differences 517

18-3B, the patient can no longer identify what the daily, they give a visual list of most of the details,
picture represents. Such a patient may identify a as opposed to an overall integration of the parts
telephone, with a dial, as a clock, or perceive a into a whole figure. A drawing typical of such a
sofa, upholstered in brown fabric, as a trunk. Such patient is shown as the copy in Figure 18-5.
difficulties seem to be even more pronounced when Evidence from several physiological experi¬
the stimuli are presented for less than 500 msec. ments implies that this defect is caused by distur¬
What sort of underlying mechanisms are in¬ bances in the temporal region of the cortex. It also
volved in these perceptual disturbances? As long seems to be specific to the way in which attention
ago as 1909, the Hungarian neurologist Balint is distributed to the visual targets (Butters, Barton
made some observations that suggest an attentional & Brody, 1970; Gerbrandt, Spinelli & Pribram,
mechanism. He found that his patients had a defi¬ 1970). Luria (1973) claimed that injections of caf¬
nite decrease in attention span, being able to see feine (to stimulate the appropriate region of the
only one object at a time, regardless of its size. For cortex) can reduce some of the symptoms, thus al¬
instance, such a person could not place a dot in the lowing the patient to be able to attend to two or
center of a circle since this would require paying three objects in the visual field simultaneously. Un¬
attention to both the circle and the dot simulta¬ fortunately, this improvement lasts only as long as
neously. This type of patient is said to be suffering the drug is active.
from simultagnosia. Thus, if the patient were Some of the agnosias are quite general in
shown a series of overlapping objects, such as scope and may involve more than one sensory mo¬
those in Figure 18-4, she might report a single ob¬ dality. Patients with diseases of the parietal lobe of
ject, for example, the hammer, and deny that she the brain may show a spatial agnosia. They have
can see any of the others (Williams, 1970). If such difficulty negotiating their way through the world.
individuals are asked to copy a simple drawing, They make wrong turns even in familiar surround¬
such as the one shown as the specimen in Figure ings, do not easily recognize landmarks, and can
18-5, they depict only its individual parts. Essen- become lost in their own homes. This problem does
not appear to be caused by a defect in a single sen¬
sory modality. These patients seem to be just as
impaired using their tactile or kinesthetic senses
as their visual (Heaton, 1968; Weinstein, Cole,
Mitchell & Lyerly, 1964). Such patients often
show a unilateral neglect of space. For example, if
asked to draw symmetrical objects, they will usu¬
ally produce some sort of imperfection on one side.
Thus, an individual with left-sided spatial agnosia
would reproduce the specimen in Figure 18-6 as
the copy shown.
Although some of these perceptual effects are
quite general, others are very specific. For in¬
stance, there is a rare disorder called prosopagno¬
sia. In this type of agnosia the patient has difficulty
perceiving and identifying human faces. In extreme
cases the patient may not even know his or her own
face in a mirror. Prosopagnosia is of interest be¬
FIGURE 18-4 A test figure for simultagnosia. cause the human face is a very important stimulus.
518 Chapter 18/Individual Differences

Specimen Copy

FIGURE 18-5 A target figure to be copied, and a reproduction typical of a person suffering from visual
object agnosia.

Your mother’s face was probably one of the first A related disorder is call autotopagnosia,
visual forms to which you attended as an infant. which is the distorted perception of body image and
When children draw, the face is usually the first body parts. For example, one patient, when asked
part of the body to be depicted. So we find that to point to her ear, looked around for it and replied
facial perception is one of the aspects of form per¬ that she must have lost it (Pick, 1922). Finger ag¬
ception that is lost only in cases of serious brain nosias and finger-naming difficulties are the most
injury, usually involving the right temporal lobe, widely known forms of specific autotopagnosias,
and often returns much earlier than other cognitive and these are thought to be associated with lesions
functions when recovery from brain damage occurs in the left parietal portion of the brain (Pizorrolo,
(Heaton, 1968). 1978).

Specimen Copy

FIGURE 18-6 A target figure to be copied, and a reproduction typical of a person with unilateral spatial
agnosia.
 Physiological Differences 519

Although we have concentrated on the visual centers involved in the interpretation of stimuli.
sense in this discussion, similar difficulties are Thus, when we find agnosias, we tend often to find
found in speech and sound perception. These are damage of particular brain sites. However, these
usually grouped under the overall heading of are usually not the primary receiving areas of the
aphasia (from a meaning “not” and phasis mean¬ cortex for that particular sensory modality. Visual
ing “utterance”). Aphasia sufferers have an inabil¬ agnosias often are associated with damage to the
ity to name common objects and often fail to recall more forward portions of the occipital cortex, gen¬
the meanings of words designating common objects erally Areas 18 and 19, which are the secondary
(Luria, 1972; Tsvetkova, 1972). In addition, there visual areas, and to the temporal lobes, which are
are specific auditory agnosias, resulting from dam¬ tertiary visual processing areas and seem to be as¬
age to the auditory pathways, that lead to the selec¬ sociated with complex visual analysis, as we saw
tive loss of the perception of words, called pure in Chapter 3. Similarly for the other sensory mo¬
word deafness, the perception of nonlinguistic dalities, the various forms of agnosia are associated
sounds, called sound agnosia, and the perception with secondary and tertiary areas of the cortex
of music, called sensory amusia (Pizorrolo, 1978). rather than the primary receiving areas. A map of
Thus, it should be clear that there are a large num¬ areas typically found to be damaged when an indi¬
ber of different forms of agnosia, or higher percep¬ vidual demonstrates various agnosias is shown as
tual disruption. A number of these are listed and Figure 18-7.
named in Table 18-1. Specifiable physiological differences are not
Most agnosias seem to have been caused by the only source of individual differences in percep¬
physiological damage, usually of the higher brain tion, however. In the next sections, we will

Table 18-1. Some of the More Common Forms of Agnosia That Manifest Themselves as Complex
Perceptual Deficits

Type of Agnosia Sensory Modality Perceptual Deficit

Object agnosia Visual Inability to name, recognize, or use objects

Color agnosia Visual Inability to associate colors with objects
Drawing agnosia Visual Inability to recognize drawn stimuli
Spatial agnosia Visual Deficits in stereoscopic vision and ability to relate objects in space
Simultagnosia Visual Inability to attend to more than one visual object at a time
Prosopagnosia Visual Inability to recognize faces
Unilateral neglect Visual Apparent deficit in processing stimuli on one side
Amusia Auditory Inability to recognize melodies, often accompanied by inability to
reproduce rhythm or tempo
Sound agnosia Auditory Inability to identify the meaning of nonverbal sounds (i.e., bells,
dog bark)
Sensory aphasia Auditory Inability to comprehend speech, although verbal production unim¬
(Wernicke’s aphasia) paired (as opposed to motor or Broca’s aphasia, which affects
production but not comprehension)
Astereagnosia Somatosensory Inability to recognize objects by touch
Autotopagnosia Somatosensory Inability to name or localize body parts
Asomatagnosia Somatosensory Inability to recognize bodily states
Asymbolia for pain Somatosensory Inability to localize or properly react to pain
520 Chapter 18/Individual Differences

FIGURE 18-7 Damage to particular portions of the brain is often associated with the indicated agnosias.

sider factors affecting perception that may or may varies over the course of the menstrual cycle,
not have a physiological basis, or that may be the reaching its peak at midcycle when estrogen (one
result of a combination of physiological and exper¬ of the major female hormones) levels are at their
iential processes. Unlike drug and specific sensory highest (Mair, Bouffard, Engen & Morton, 1978;
damage, the causes of gender and personality dif¬ Parlee, 1983). Women whose ovaries are less ac¬
ferences are harder to specify, although both appear tive than normal have impaired smell sensitivity,
to exist. but this defect can be remedied by the administra¬
tion of estrogen. Conversely, doses of androgen
(such as the male hormone testosterone) make the
GENDER DIFFERENCES sense of smell less sensitive (Schneider, Costiloe,
Howard & Wolf, 1958).
The gender of an individual (whether a person is Some hormonal effects on perception are quite
male or female) may partially determine what is subtle. For example, there is a gender difference
perceived in any given stimulus situation. Gender, in taste preference rather than in taste sensitivity;
of course, carries with it many physiological impli¬ females prefer the sweet taste more than males
cations. One of the most important of these is the do, and this preference varies with the menstrual
massive chemical differences between the bodies of cycle (Aaron, 1975). This is true for rats as well as
men and women due to the presence of specific humans. When the ovaries of female rats are re¬
male or female hormones. Hormones are carried in moved, their preference for the sweet taste dimin¬
the blood, which infuses and supplies all our sen¬ ishes; therapeutic doses of estrogen restore the
sory receptors, and are found in many parts of the preference (Zucker, Wade & Ziegler, 1972). This
brain. Thus, it would not be surprising to find that may also help us to understand why women using
men and women might differ in certain sensory and contraceptive pills (which contain estrogen as a
perceptual capacities. component) often complain that they have a ten¬
Examples of gender differences in perception dency to overeat sweets and gain weight.
are found in both taste and olfaction. For instance, Male-female differences in sensory sensitivity
women, on average, have more acute senses of are found in other modalities as well (McGuiness,
smell than men (Money, 1965). This difference 1976a). For example, women usually show greater
seems to be directly attributable to hormonal influ¬ touch sensitivity than men (Ippolitov, 1973; Wein¬
ences since the acuity of a female’s sense of smell stein & Sersen, 1961). They also show superior
 Gender Differences 521

hearing sensitivity, especially at higher frequencies produce consistent sex differences favoring males
and in older people (Corso, 1959; McGuinness, (Halpem, 1986). One of these involves disembed-
1972; Royster, Royster & Thomas, 1980). Women ding, or the ability to disentangle a target object
generally are more sensitive than men to pain pro¬ from a surrounding, and often confusing, context.
duced by electric shock, and women’s pain thresh¬ For example, in Figure 18-8A, you see a figure
olds also seem to vary over the menstrual cycle marked Target that is hidden, or embedded, in the
(Goolkasian, 1980; Tedford, Warren & Flynn, more complex figure beside it. The observer’s task
1977). is to find the simple shape as quickly as possible.
Gender differences in vision seem to be more Such tasks are usually called the embedded figures
complex. Males generally appear to have much test or the hidden figures test. A different spatial
better visual acuity under photopic conditions task involves the ability to recognize targets when
(Burg, 1966; Roberts, 1964), whereas females have they have been rotated. An example of this mental
lower absolute thresholds under scotopic conditions rotation task is shown in Figure 18-8B. The ob¬
(McGuinness & Lewis, 1976). This difference server has to recognize the shape marked Target
seems to be present from childhood (see Brabyn & from among the three figures next to it. It is often
McGuinness, 1979). There is also some evidence difficult to recognize which shape is exactly the
that the visual acuity of women varies with their same as the target, since the correct shape has been
menstrual cycle (Parlee, 1983; Scher, Pionk & Pur¬ rotated into a different spatial orientation.
cell, 1981), being poorest just prior to and during Both disembedding and mental rotation tasks
menstruation. The hormone progesterone (another produce performance differences that favor males
predominately female hormone) is often prescribed (Halpem, 1986; Wilson, DeFries, McCleam, Van-
for women who suffer from severe anxiety or denberg, Johnson & Rashad, 1975). These differ¬
depression during menstruation. Progesterone re¬ ences are often quite sizable, amounting to 16
lieves these symptoms and also restores visual acu¬ percent or more, depending on the tests involved
ity to its normal level in most patients (Dalton, (Sanders, Soares & D’Aquila, 1982). Males either
1964). When acuity is measured at different spatial are more accurate in their responses or show
frequencies (measurements of contrast sensitivity, greater speed when completing such tasks (B lough
see Chapter 4), females show greater sensitivity in & Slavin, 1987; Harris, 1981; Lohman, 1986). In
the low spatial frequency ranges and males are addition, McGlone (1981) has shown that females
more sensitive at the high spatial frequencies (Bra¬ approach these tasks differently than males. Fe¬
byn & McGuinness, 1979). Also, there is some males appear to make more rotational hand move¬
suggestion that females dark-adapt more rapidly ments while completing cognitive rotations; in
than males do (McGuinness, 1976b). other words, they more frequently need concrete
aids or verbal strategies to successfully complete
the task (Clarkson-Smith & Halpem, 1983). Dem¬
Visual-Spatial Abilities
onstration Box 18-2 provides an opportunity for
An interesting and complex gender difference con¬ you to test this sex-related difference in spatial
cerns visual-spatial abilities. These are tasks that ability for yourself.
involve nonverbal cognitive manipulations of ob¬
jects, and may include the ability to visualize how
Physiological Factors
objects will appear when they are rotated, to detect
the orientations of and relationships between differ¬ Disembedding a figure, or recognizing it when it
ent stimuli, and to correctly perceive complex vi¬ has been rotated in space, is a complex task that
sual patterns (McGee, 1979). Such tasks seem to would seem to involve many learned skills, and
522 Chapter 18/Individual Differences

Complex figure in which

target is embedded

Target

FIGURE 18-8 (A) An embedded figures test in which the target is found in the more complex array. (B) A
mental rotation test in which the target is found as one of the three test figures, but in a different
orientation.

would seem to be affected by a familiarity with sponsible for these effects. Dawson (1967) used a
such things as maps and blueprints, which might series of these tests on a number of West African
involve the use of similar skills. So it is somewhat males who suffered from a disease that results in
surprising to find that there is evidence suggesting estrogen levels higher than those usually found in
that some of the same physiological factors distin¬ males. When tested on a series of spatial tasks,
guishing males from females might be partially re¬ these males showed reduced spatial ability relative
 Gender Differences 523

DEMONSTRATION BOX 18-2. Gender Differences in Mental Rotation

For this demonstration you will need a stopwatch or those 5 as quickly as possible. Start your watch, show
a wristwatch that allows you to read seconds (either them the figure, and time how long it takes for them
with a sweep hand or digitally). You will also need a to find the 5 correct ones. If they get any wrong, tell
couple of male and female friends. Test them one at them, but keep the time going until all 5 are found.
a time. First show them what is meant by a mental The correct answers are on the bottom of page 528.
rotation task by using Figure 18-8B. If they have dif¬ You should notice that, on average, females will
ficulty, point out that only Stimulus 3 can be rotated take longer at this task than males. Another interest¬
to be identical to the target, whereas the other 2 are ing observation should be that females are more likely
differently shaped figures. Next, tell your observers to perceive the task as being difficult, as indicated by
that they will see another target figure and a set of 12 comments like “I can never do this sort of thing” or
test figures. Five of the test figures are identical in “I’m terrible at this,” and so forth.
shape to the target figures, and their task is to pick out
524 Chapter 18/Individual Differences

to a sample of nonaffected males. Similar effects cause many aspects of perception are subject to
were found in certain South American tribes, where learning influences, as we discussed in Chapter 17.
the males habitually chew the leaves of the coca Research has shown that even at an early age males
shrub, thus releasing cocaine, which when ingested and females may differ in the types of tasks and
decreases the secretion of the male hormone testos¬ activities they engage in, and also in the tools, im¬
terone. Such males tend to show typical signs of plements, or utensils they use in their everyday ac¬
feminization (including enlarged breasts, widened tivities (see Harris, 1981). These factors can also
hips, softened skin texture, and the like), and also influence some aspects of perception. Demonstra¬
show reduced spatial abilities, similar to those of tion Box 18-3 provides an object-identification task
females. Male hormones influence these spatial that occasionally produces different responses from
abilities in the opposite way. Thus, males who pro¬ males and females. The task shows differences that
duce little or are insensitive to androgen (e.g., tes¬ are most likely to have an experiential basis rather
tosterone) also show reduced spatial abilities, than the kind of physiological basis we discussed
whereas females with high androgen (e.g., andro- above.
stenedione) levels show greater spatial abilities Perhaps the strongest data in favor of psycho¬
(Hier & Crowley, 1982; Masica, Money, Ehrhardt social factors contributing to gender differences in
& Lewis, 1969; Peterson, 1976). Hormonal effects spatial skills comes from studies in which specific
in these complex spatial abilities have also been training was given to boys and girls using toys and
implicated by the finding that the spatial abilities of games that have a spatial component and that are
pregnant women (who have higher than usual lev¬ typically preferred by boys (such as blocks, tinker
els of estrogen) differ from those who are not preg¬ toys, or paper cut in geometric shapes). Individuals
nant (Woodfield, 1984). who received such training tended to do better on
There are other factors consistent with phys¬ spatial skills tests, such as the embedded figures
iological determination of spatial ability. For in¬ task, suggesting that there is a learned component
stance, the rate at which individuals mature seems for spatial ability (Smith, Frazier, Ward & Webb,
to predict spatial performance. In this case, “ma¬ 1983; Sprafkin, Serbin, Denier & Conner, 1983).
turing” means showing their secondary sexual However, improvement does not occur with train¬
characteristics. Typically, late maturing individuals ing for all of the spatial skills that usually show
are better on such spatial tasks than early maturing gender differences (Thomas, Jamison & Hammel,
individuals (Petersen & Crockett, 1985; Waber, 1973).
1976, 1977). This is consistent with the usually ob¬ It seems likely that differences in spatial skills
served gender differences since males tend to ma¬ are due to the interaction between biological and
ture later than females. psychosocial factors (Halpem, 1986). Such inter¬
actions may explain why identification behavior
differs for individuals who are more strongly sex-
Psychosocial Factors
typed, that is, who identify themselves to be “a
typical male or female,” versus those who are
We have been dealing with some of the physiolog¬ “androgynous,” showing a mixture of typical male
ical variables that seem to produce different pat¬ and female behaviors (Bern, 1981). They may also
terns of perceptual abilities in males and females. help to explain why male homosexuals resemble
Of course, there will be many factors relating to heterosexual females more than they resemble het¬
experience, life history, and cultural influences that erosexual males in terms of their performance on
will also affect perceptual behaviors. This is be¬ spatial tasks (Sanders & Ross-Field, 1986).
 Personality and Cognitive Style Differences 525

DEMONSTRATION BOX 18-3. Sex Differences and Object Identification

Look at the three accompanying figures and de¬

cide what each looks like. Do this before reading any
further.
Responses to patterns similar to these show differ¬
ences depending on the sex of the observer. Most
males view the top figure as a brush or a centipede,
whereas females tend to view it as a comb or teeth.
Most men view the middle figure as a target, whereas
women tend to view it as a dinner plate (but both
respond equally with “ring” and “tire”). Most men
see the bottom figure as a head, whereas women tend
to view it as a cup.

PERSONALITY AND COGNITIVE individual’s personality. People with different per¬

STYLE DIFFERENCES sonalities tend to behave differently in many social
situations and tend to respond differently to infor¬
There are myriad nonperceptual ways in which in¬ mation of various sorts. Do they also perceive the
dividuals differ from one another. Some are out¬ world differently?
going and sociable; others are withdrawn and prefer There have been many attempts to link indi¬
to be alone. Some are careful and methodical in vidual differences in personality to individual dif¬
everything; others are haphazard and unsystematic. ferences in perception. Often the perceptual re¬
The total of all these behavior traits composes the sponses themselves are used to classify individuals
526 Chapter 18/Individual Differences

as belonging to one personality type or another. personality theory. Characteristic of this approach
There is a lot of evidence that individuals differ in is the work of Eysenck (1967). He divided individ¬
their ability to disembed figures from one another uals into two groups on the basis of whether they
in tasks such as the one we discussed above and were outgoing and sociable (extrovert) or more
illustrated in Figure 18-8A (e.g., T. B. Ward, withdrawn and self-contained (introvert). Eysenck
1985). Some investigators have suggested that these found that he could classify individuals along this
tests not only separate individuals according to dimension on the basis of a simple questionnaire,
their spatial abilities but also separate individuals and he speculated on some physiological differ¬
according to underlying personality type. Observers ences that might account for the differences in per¬
who have difficulty with this task are called field sonality traits. He suggested that extroverts have a
dependent. They have been classified by person¬ neural system that is slower to respond and more
ality tests as being socially dependent, eager to weakly aroused by stimuli than that of introverts.
make a good impression, conforming, and sensitive In addition, they generate neural inhibition more
to their social surroundings (Konstadt & Forman, quickly. If this physiological speculation is correct,
1965; Linton & Graham, 1959; Ruble & Naka¬ then introverts should be more perceptually sensi¬
mura, 1972). Individuals who have little difficulty tive than extroverts.
with such perceptual disembedding tasks are called Several studies have investigated the effect of
field independent people. They have been charac¬ introversion-extroversion on perception. Introverts
terized by the same tests as being self-reliant, in- do seem to have more sensitive perceptual systems
ner-directed, and individualistic (Alexander & as predicted by the theory. They show lower aver¬
Gudeman, 1965; Crutchfield, Woodworth & Al¬ age thresholds for vision (Siddle, Morish, White &
brecht, 1958; Klein, 1970). Mangen, 1969), hearing (Stelmack & Campbell,
Witkin has been one of the major proponents 1974), touch (Coles, Gale & Kline, 1971), and
of this approach. He and his associates look upon pain (Halsam, 1967). In addition, introverts are
both the personality and perceptual effects as ex¬ better at tasks requiring sustained attention or vigi¬
amples of an individual’s cognitive style (Witkin lance (Harkins & Green, 1975).
& Berry, 1975). They maintain that perceptual, When studying the effects of personality fac¬
cognitive, personality, and social interactions all tors on perception, it is important to be sure that
are affected by the same set of processes that de¬ we are measuring perceptual sensitivity rather than
termine how a person approaches the world. In ef¬ simply detecting differences in how observers re¬
fect, cognitive style is part of what we call in spond. It could be the case that introverts simply
everyday language “life-style,” affecting not only say “Yes, I detected the stimulus” more often than
our habitual interpersonal and task-oriented behav¬ extroverts. Signal detection theory (discussed in
iors but also the way we process information, and Chapter 2) allows us to separate these possibilities.
in effect, the way we perceive the world. Thus, by When Stelmack and Campbell (1974) analyzed
measuring how you normally respond in complex their data from this viewpoint, they found that in¬
perceptual situations, we can predict to some extent troverts have more sensitive hearing than extro¬
how you will approach many other, nonperceptual verts, even though extroverts are more biased
aspects of your life. toward saying yes.
Some investigators use perception as the start¬ Another way to ascertain sensitivity indepen¬
ing point and move from there into predictions dent of the observer’s response bias is to use direct
about personality; others have attempted to go in physiological measurements. One technique is
the opposite direction, predicting individual differ¬ called evoked response recording. An electrode is
ences in perception from prior considerations of placed on an observer’s head over the region of the
 Personality and Cognitive Style Differences 527

cortex receiving the primary sensory information ences in processing of sensory information, it is not
for the sense modality being tested. Another elec¬ surprising that there are dramatic perceptual effects
trode, elsewhere on the body, serves as a reference associated with certain severe personality disorders.
electrode. Any changes in the electrical activity of The perceptual responses that differentiate schizo¬
this brain region can be picked up by sensitive re¬ phrenic from nonschizophrenic observers is one
cording devices, and such activity presumably area that has received a large amount of research
means that the sensory information has, at least, attention. Schizophrenia (from the Greek for
been registered in the brain. In this way, Stelmack, “split mind”) is the most frequent diagnosis of a
Achom, and Michaud (1977) demonstrated that in¬ severe or psychotic personality disorder. It is usu¬
troverts seem to have greater auditory sensitivity ally characterized by a withdrawal (or “splitting
than extroverts. Unfortunately, not all researchers off”) from the environment, reduced levels of
have been able to verify these findings (Campbell, emotional response, a reduction in abstract think¬
Baribeau-Braun & Braun, 1981). This may mean ing, and a general diminishing of daily activity. In
that nonsensory factors, such as motivation or dis¬ other words, schizophrenia is a disorder that affects
tribution of attention, or even the sort of cognitive all aspects of the sufferer’s social and cognitive
style we discussed above, rather than direct neuro¬ life. Studies of the perceptual responses of schizo¬
logical differences, may account for the differences phrenics have shown that they differ from control
between introverts and extroverts on sensory tasks. groups in their performance on time estimation
It is surprising nonetheless, that the answers (Wahl & Sieg, 1980), attentional tasks (Cegalis &
to a few questions about how a person interacts Deptula, 1981), and even on the perception of vi¬
with other individuals can be used to predict how sual aftereffects (Tress & Kugler, 1979). Several
one person’s perceptual responses may differ from studies have shown also that schizophrenics display
those of another. Demonstration Box 18-4 allows eye-movement patterns that differ from those of
you to estimate your own degree of introversion control groups; they perform poorly when they are
and extroversion, and to test a typical perceptual asked to track a moving target with their eyes (Ia-
preference for yourself. cono, Peloquin, Lumry, Valentine & Tuason,
Since personality factors that differ among 1982; Levin, Lipton & Holzman, 1981). Since
those in the general population are related to differ¬ poor eye-tracking behavior is also found in the

DEMONSTRATION BOX 18-4. Introversion-Extroversion and Taste Perception

It is easy to determine your own standing on intro¬ If you answered all the questions “Yes,” you are
version versus extroversion by answering the follow¬ rather extroverted; if you answered them all “No,”
ing questions with a “Yes” or a “No.” you are rather introverted.
Have some friends and/or relatives answer these
Do you often wish for more excitement in life?
questions, but add one additional item to the list.
Do you often say things without stopping to think?
Do you like going out a lot? Do you like spicy foods?
Do other people think of you as being lively?
What answer do you expect extroverts versus intro¬
Do you like interacting with people?
verts to give? What sensory data would lead you to
expect that answer?
528 Chapter 18/Individual Differences

close relatives of schizophrenics (who are not af¬ Disembedding The ability to disentangle a target ob¬
fected with the disorder), it has been suggested that ject from a surrounding, and often confusing, context.
eye-movement behavior may be a genetic marker Embedded figures test A task used to determine
spatial abilities, in which a subject is asked to find a
for the disorder (Iacono et al., 1982). This suggests
simple shape hidden in a more complex figure.
that perceptual behavior can be used as an indicator
Evoked response Overall electrical response of the
of the presence of underlying processes that could brain to presentation of a stimulus.
promote personality disorders. Extrovert An outgoing and sociable person.
Overall, who you are, the kind of person you Field dependent Descriptive of individuals exhibit¬
are, and the life history you have had all affect ing difficulty with embedded-figures tasks.
what you perceive in any stimulus situation. Since Field independent Descriptive of individuals exhib¬
you differ along many dimensions from those iting little difficulty with embedded-figures tasks.
around you, your perception of the world has a Hidden figures test See Embedded figures test.
unique flavor. What you perceive in any situation Introvert A withdrawn and self-contained person.
is not necessarily the same as what is perceived by Mental rotation task A task in which observers are
the person next to you. asked to recognize a visual target in different spatial ori¬
entations (rotations).
Necker cube The drawing in Figure 18-2, in which
GLOSSARY the three-dimensional interpretation alternates between
two equally compelling possibilities.
The following definitions are specific to this book.
Poggendorff illusion An illusion of direction that
Agnosia A pathological condition in which an shows both age and gender differences.
individual can no longer attach meaning to a sensory Prosopagnosia A perceptual disorder in which an
impression. individual cannot identify human faces.
Aphasia A disorder characterized by difficulties in¬ Schizophrenia A psychotic disorder characterized
volving speech and sound perception. by withdrawal from the environment, reduced levels of
Auditory flutter fusion (AFF) The rate of interrup¬ emotional response, a reduction in abstract thinking, and
tion of a continuous tone at which an observer first hears a general diminishing of daily activity.
the tone as continuous. Simultagnosia An attentional disorder in which an
Autotopagnosia The distorted perception of body individual cannot pay attention to more than one stimulus
image and body parts. at a time.
Cognitive style The overall personality and percep¬ Spatial agnosia A perceptual disorder in which
tual predispositions that are characteristics of a particular individuals cannot accurately localize objects for
individual. themselves.
Critical flicker fusion frequency (CFF) The mini¬ Visual object agnosia An inability to identify famil¬
mum rate of a flickering light at which the light is per¬ iar objects in the absence of psychopathology or organic
ceived as continuous. damage to the visual apparatus.

Answers to Demonstration Box 18-2: 3, 6, 7, 9, 11.

APPENDIX

Primer of
Neurophysiology
NEURONS AND THE NERVOUS
SYSTEM
THE NATURE OF NEURAL ACTIVITY
TECHNIQUES TO MEASURE NEURAL
FUNCTION
530 Appendix/Primer of Neurophysiology

T here are several places in the book where our

discussions assume that you know some neu¬
Sensory endings

rophysiology. You need to know some terminology

for parts of the nervous system, how a neuron
functions, and how we investigate neural activity
in sensory systems. This appendix provides you
with that information in a fairly condensed form.
More details are available from any of several basic
texts in biopsychology (e.g., Carlson, 1988; Kalat,
1984).

NEURONS AND THE NERVOUS

SYSTEM

The human nervous system contains approximately

10 to 14 billion neurons. Figure A-l shows some
aspects of several types of neurons. A sensory
neuron conducts information from sensory recep¬
tors (sometimes part of the sensory neuron itself)
toward the brain, an interneuron conducts infor¬
mation between other neurons, and a motor neu¬
ron conducts nerve impulses outward to the
muscles. Each of these neurons is a separate cell,
generally composed of three distinct parts: a cell
body, an axon, and dendrites. The cell body con¬
tains the nucleus (which contains the genetic ma¬
terial) and a large variety of other molecules that
govern the functioning of the neuron. The dendrites
are branching structures that make contact with in¬
coming nerve fibers from other neurons, and the
axons are usually long fibers that conduct nerve im¬
pulses toward other neurons (or muscle fibers). Ax¬
ons typically terminate near dendrites of other
neurons. Many, but not all, neurons have axons
that are covered by protective and nutritive cells
called glial cells. Some glial cells (including a va¬
riety called Schwann cells) form the myelin
sheath around the axon, which, as you will see
below, helps to increase the speed at which an ac-

FIGURE A-l Neurons of various types and their

Motor endings-
important parts and connections with each other.
 The Nature of Neural Activity 531

tion potential travels along a neuron. The myelin negatively charged proteins that are bound to the
sheath is interrupted about every millimeter by inside of the cell. The ions are not distributed
nodes of Ranvier, which are the gaps between equally on both sides of the membrane. Of the neg¬
successive glial cells. Most sensory and motor ative ions, chloride and bicarbonate are many times
nerves are myelinated (have a myelin sheath); a no¬ more common outside. Of the positive ions, so¬
table exception are the small C neurons that termi¬ dium is 14 times more common outside, and potas¬
nate in the skin and respond best to noxious sium 28 times more common inside, mainly
stimulation (see Chapter 9). because of the sodium-potassium pump, a bio¬
When many axons gather together into a path¬ chemical process that ejects 3 sodium ions for
way to carry information from one part of the body every 2 potassium ions it allows in. Also, potas¬
to the other, that pathway is called a nerve. Sen¬ sium passes more easily through the membrane
sory information usually is carried by nerves to the than does sodium.
central nervous system (CNS), which consists of When a neuron is stimulated, either by a phys¬
the brain and the spinal cord. In the central nervous ical stimulus or another neuron, the difference in
system, a pathway is no longer called a nerve but electrical potential across the cell membrane either
rather a tract, although the terms fasciculus or pe¬ becomes less negative by moving toward 0 mV
duncle are used sometimes for certain pathways. A (depolarization) or more negative by moving far¬
bundle of nerve fibers ascending the spinal cord is ther away from 0 mV (hyperpolarization). This
quite often referred to as a lemniscus. In addition happens because the stimulation causes changes in
to pathways in the central nervous system (white the permeability of the cell membrane to various
matter), there are distinct regions containing gray ions. Typically, when sodium is blocked from its
matter (the cell bodies of many neurons grouped usual (slow) flow across the membrane the negative
together). The islands of gray matter are referred to charge inside the neuron increases (hyperpolariza¬
as nuclei. Much of the sensory information pro¬ tion). Conversely, if sodium ions are allowed to
cessing, and complex channeling of information, flow more rapidly into the cell the negative charge
takes place in nuclei of the brain and spinal cord. decreases (depolarization).
There are several different ways such stimu¬
lation occurs. Different sensory cells use different
THE NATURE OF NEURAL ACTIVITY methods to change the environmental stimulation
into a depolarization or a hyperpolarization. We
Information is passed along neurons, and from one discuss some of these processes in the chapters de¬
to another, by electrochemical changes in the neu¬ voted to particular sensory systems, since they are
ron. When unstimulated, the inside of a neuron is rather specialized. For neurons stimulated by other
electrically negative with respect to the outside, neurons, however, the changes in permeability are
with a resting potential of about —70 millivolts fairly standard. They are caused by release of a
(mV). This charge is due to the action of millions transmitter substance across the synapse separat¬
of ions (atoms that have gained or lost an electron, ing the two cells. Figure A-2 shows the most im¬
and hence are electrically charged). The most com¬ portant parts of the synapse. The transmitter
mon ions involved in neural action are sodium substance is stored in the synaptic vesicle located
(Na + ), potassium (K + ), chloride (Cl-), and bicar¬ in the synaptic knob near the presynaptic mem¬
bonate (HC03-). The negative electrical charge of brane, and is released across the synaptic cleft in
the neuron is the result of a dynamic chemical amounts determined by the amount of activity in
equilibrium involving the flow of various ions the incoming neuron. Transmitter substance comes
across the cell membrane, plus the existence of in two forms, excitatory transmitter, which
532 Appendix/Primer of Neurophysiology

the initial small depolarization is suddenly followed

t by a much larger and more rapid depolarization
This way to cell body
when sodium ions flow into the axon. This rapid
depolarization occurs within about 1 msec and is
quickly reversed and followed by a period of hy¬
perpolarization as potassium ions are pumped out.
After this sudden swing in electrical charge, the
potential returns to the resting level of —70 mV.
During the time of hyperpolarization following a
spike, called the refractory period, the neuron is
much more difficult to excite.
These changes in electrical potentials in the
axon can be recorded by various electrical record¬
ing devices, and one such recording is shown as
Figure A-3. Increased stimulation does not change
the size of the spike, but rather increases the num¬
ber of responses per unit time. You might imagine
FIGURE A-2 A typical synapse. Transmitter
substances are stored in the synaptic vesicles and that each neural response is a bark from a dog and
released across the synaptic cleft. the excitement of the dog is measured by the speed
at which he is barking. Thus, the frequency of re¬
sponses shows the level of stimulation. Because of
depolarizes the postsynaptic membrane, and in¬ the refractory period, the maximum rate at which
hibitory transmitter, which hyperpolarizes the spikes can occur is about 1,000 spikes per second.
postsynaptic membrane. Both forms are found in Most of the information carried to the central ner-
numerous places in the central nervous system and
there are several different varieties of each type.
Transmitter substances are active only a short time
after release, being quickly neutralized by enzymes
that are always present in the synaptic cleft.
Usually, the stronger the excitatory stimulus to
a neuron is, the greater the change in potential dif¬
ference it causes. Such changes are called graded
potentials. The neural responses in several sensory
systems, such as the parts of the retina that respond
first (namely photoreceptors, bipolars, horizontal
cells), involve only a continuous graded change in
electrical potential. For retinal amacrine and gan¬
glion cells, however, and for virtually all nonsen-
sory neurons, this is not the case. Instead, if a
depolarization reaches some critical level, a more —i—i—i—i_i_i i ■ i i

dramatic and rapid change in the electrical state of 1 23456789 10 11

Time (msec)
the neuron follows that is referred to as the action
potential or spike potential. In the spike potential, FIGURE A-3 A typical spike, or action potential.
 Techniques to Measure Neural Function 533

vous system is in the form of a series of nerve TECHNIQUES TO MEASURE NEURAL

spikes. FUNCTION
The change in ion flow causes the spike poten¬
tial to move along the axon of the neuron until it Neural responses (mainly spikes) are measured us¬
stimulates the release of transmitter substance ing microelectrodes, which consist of tiny glass
where the neuron makes synapses with dendrites or tubes (the tip might be .01 mm in size, or smaller)
cell bodies of other neurons. The spike moves more filled with salt water. The electrode is inserted into
quickly in thicker axons than in thin ones, with a the cell body or the axon, and the potential differ¬
maximum speed of about 10 m/sec. However, even ence between that electrode and another reference
the thickest axons require help in getting conduc¬ electrode outside the cell is amplified and then re¬
tion speeds up to levels where they are useful for corded by a computer, which converts the contin¬
quick sensing and muscular action. This help is uous voltages to digital numbers. The continuous
provided by the myelin sheath worn by many neu¬ signal is also usually displayed on an oscilloscope,
rons (see Figure A-l). In such neurons, spikes which is a sensitive voltmeter that displays voltage
jump electrically from one node of Ranvier to the changes over time. Typical spike responses are
next at the speed of electrical conduction, which is shown in Figure A-4, which shows two neurons
about 300 million m/sec, whereas the slower bio¬ responding at different rates.
chemical processes occur only in the dendrites, The use of microelectrode techniques to record
over the cell body, and in the nodes themselves. the activity of single neurons has produced some of
This allows net conduction speeds of up to 120 m/ the most exciting data in the field of sensory phys¬
sec in myelinated cells. Remarkably, these conduc¬ iology. Generally, the procedure involves the ap¬
tion speeds measured with modem methods agree plication of a muscle relaxant combined with local
very well with those estimated by Helmholtz in anesthetics to reduce the discomfort from the
1850 using reaction-time techniques. restraining device used to hold the animal. A

Relatively Relatively
active cell inactive cell
Each represents

Time (msec)

FIGURE A-4 Oscilloscope records of spikes generated by two neurons penetrated by recording
microelectrodes.
534 Appendix/Primer of Neurophysiology

stereotaxic instrument is used to allow the re¬ nected to that are responding to the stimulus. A
searcher to accurately place electrodes in the brain. decrease in firing rate indicates that the stimulus or
Figure A-5 shows a cat with its head in a stereo¬ other neurons are inhibiting the neuron from which
taxic instrument. Note that the cat is viewing a recordings are being made. Typically, the rec¬
screen on which visual stimuli may be presented. ords of neural activity along with the conditions
The electrode is attached through a set of amplifiers under which they occurred are analyzed by the
to the computer, the oscilloscope, and also often to computer for patterns that indicate their functional
a speaker. The loudspeaker transforms the ampli¬ significance.
fied neural response into a series of pops or clicks,
each click caused by a single spike. Researchers
can then listen to the neural response, keeping their
eyes free to tend to other matters. An increase in GLOSSARY
the rate of clicking means an increase in the fre¬
quency of cell firing, and a decrease means a re¬ The following definitions are specific to this book.
duction. An increase in firing rate when a stimulus Action potential The large depolarization of a neu¬
is applied means that the neuron is being excited ron, 1 msec in duration, that occurs when a graded de¬
by the stimulus or whatever other neurons it is con¬ polarization exceeds a certain threshold.

Audio
Amplifier Oscilloscope speaker

FIGURE A-5 Setup to record neural activity from the visual system of the cat.
 Glossary 535

Axon The long, slender part of a neuron that con¬ Postsynaptic membrane The cell membrane of a
ducts membrane potential changes away from the cell dendrite that is receiving transmitter substance across a
body and makes synapses with other neurons’ dendrites synapse.
or with muscle fibers. Presynaptic membrane The cell membrane of a
Cell body The part of a neuron that contains the synaptic knob of a neuron that is sending transmitter sub¬
nucleus. stance across a synapse.
Central nervous system (CNS) The brain and the Refractory period A period of time shortly after a
spinal cord. spike during which a neuron is hyperpolarized and diffi¬
Dendrites The branching part of a neuron that makes cult to depolarize above threshold to generate another
synapses with other neurons’ axons or with sensory spike.
receptors. Resting potential The usual electrical state of a neu¬
Depolarization A change in electrical potential ron, typically with the inside about -70 mV with re¬
across the cell membrane of a neuron in the direction of spect to the outside.
lesser negativity. Schwann cells One type of glial cells that form the
Excitatory transmitter A substance released across myelin sheath around axons.
a synapse that causes the postsynaptic membrane to Sensory neuron A neuron that carries information
depolarize. from a sensory receptor (often the neuron itself) toward
Glial cells Cells that protect and feed neurons; some the brain or spinal cord.
types form the myelin sheath as well. Sodium-potassium pump A mechanism that ejects
Graded potential A change in electrical potential 3 sodium ions from inside the neuron for every 2 potas¬
across the cell membrane of a neuron that changes sium ions it lets in; helps maintain the resting potential
in magnitude as the stimulating event changes in at —70 mV.
magnitude. Spike potential See Action potential.
Hyperpolarization A change in electrical potential Stereotaxic instrument A device to allow electrodes
across the cell membrane of a neuron in the direction of to be placed precisely in neurons of experimental
greater negativity. animals.
Inhibitory transmitter A substance released across Synapse The place where two neurons almost touch
a synapse that causes the postsynaptic membrane to and across which one releases transmitter substance to
hyperpolarize. stimulate the other.
Interneuron A neuron that conducts information Synaptic cleft A small space separating a synaptic
from one neuron to another. knob of an axon of one neuron, or sensory receptor, from
Lemniscus A bundle of nerve fibers ascending the the cell membrane of a dendrite of another neuron, or a
spinal cord. muscle fiber; transmitter substances travel across the syn¬
aptic cleft.
Motor neuron A neuron that conducts spikes out¬
ward from the central nervous system to the muscles. Synaptic knob One of the knoblike endings of an
axon; there may be many thousands at the end of each
Myelin sheath A covering of glial cells over the
axon.
axon of a neuron that allows electrical conduction be¬
tween nodes of Ranvier, and provides an overall increase Synaptic vesicle A small reservoir in a synaptic
in the speed of information transmission. knob that contains transmitter substance and releases the
transmitter across the synaptic cleft when stimulated by
Nerve A bundle of axons that carries neural infor¬
a potential change in the cell membrane, usually a spike.
mation from one part of the body to another.
Tract The most common name for a nerve in the
Nodes of Ranvier Interruptions approximately every
central nervous system.
millimeter in the myelin sheath covering an axon.
Transmitter substance A substance released across
Nuclei Groups of cell bodies of neurons found in the
the synaptic cleft when a neuron experiences a change in
spinal cord and brain.
electrical potential, most usually a spike.
Oscilloscope A sensitive voltmeter that displays
voltage changes over time.
- * ■
 References
Aantaa, E. (1970). Light-induced and spontaneous variations in zoid window. Psychological Monographs, 65, (14, Whole
the amplitude of the electro-oculogram. Acta Otolaryngolo- No. 324).
gica, (Suppl. 267). Ames, A., Jr. (1955). The nature of our perception, apprehen¬
Aaron, M. (1975). Effect of the menstrual cycle on subjective sions and behavior. Princeton, NJ: Princeton University
ratings of sweetness. Perceptual and Motor Skills, 40, 974. Press.
Abbs, J. H., & Sussman, H. M. (1971). Neurophysiological Amoore, J. E. (1969). A plan to identify most of the primary
feature detectors and speech perception: Discussion of theo¬ odors. In C. Pfaffman (Ed.), Olfaction and taste III (pp.
retical implications. Journal of Speech and Hearing Re¬ 158-171). New York: Rockefeller University Press.
search, 14, 23-36. Amoore, J. E. (1970). Molecular basis of odor. Springfield, IL:
Abramov, I., Gordon, J., Henderson, A., Hainline, L., Dob¬ Thomas.
son, V., & La Brossiere, E. (1982). The retina of the new¬ Amoore, J. E. (1975). Four primary odor modalities of man:
born human infant. Science, 217, 265-267. Experimental evidence and possible significance. In D. A.
Adam, N., Rosner, B. S., Hosick, E. C., & Clark, D. L. Denton & J. P. Coghlan (Eds.), Olfaction and taste V (pp.
(1971). Effect of anesthetic drugs on time production and 283-289). New York: Academic Press.
alpha rhythm. Perception & Psychophysics, 10, 133-136. Amoore, J. E., Pelosi, P., & Forrester, L. J. (1977). Specific
Adams, A. S., Brown, B., Haegerstrom-Portnoy, G., & Flom, anosmias to 5a-androst-16 en-3 one and w-pentadecaloc-
M. C. (1976). Evidence for acute effects of alcohol and tone: The urinous and musky odors. Chemical Senses and
marijuana on color discrimination. Perception & Psycho¬ Flavor, 5, 401-425.
physics, 20, 119—124. Amure, B. O. (1978). Nicotine and decay of the McCullough
Adams, J. A. (1977). Feedback theory of how joint receptors effect. Vision Research, 18, 1449-1451.
regulate the timing and positioning of a limb. Psychological Andersen, G. J. (1986). Perception of self-motion: Psychophys¬
Review, 84, 503-523. ical and computational approaches. Psychological Bulletin,
Adams, R. D. (1977). Intervening stimulus effects on category 99, 52-65.
judgments of duration. Perception & Psychophysics, 21, Andersen, G. J., & Braunstein, M. L. (1985). Induced self-
527-534. motion in central vision. Journal of Experimental Psychol¬
Akil, H., & Watson, S. J. (1980). The role of endogenous opi¬ ogy: Human Perception and Performance, 11, 122-132.
ates in pain control. In H. W. Kosterlitz & L. Y. Terenius Anderson, N. H. (1970). Functional measurement and psycho¬
(Eds.), Pain and Society (pp. 201-222). Weinheim: Verlag physical judgment. Psychological Review, 77, 153-170.
Chemie. Anderson, N. H. (1975). On the role of context effects in psy¬
Alexander, J. B., & Gudeman, H. E. (1965). Personal and in¬ chophysical judgment. Psychological Review, 82, 462-482.
terpersonal measures of field dependence. Perceptual and Anderson, N. S., & Fitts, P. M. (1958). Amount of information
Motor Skills, 20, 70-86. gained during brief exposures of numerals and colors. Jour¬
Alexander, K. R., & Shansky, M. S. (1976). Influence of hue, nal of Experimental Psychology, 56, 362-369.
value, and chroma on the perceived heaviness of colours. Andrews, B. W., & Pollen, D. A. (1979). Relationship between
Perception & Psychophysics, 19, 72-74. spatial frequency selectivity and receptive field profile of
Algom, D., Raphaeli, N., & Cohen-Raz, L. (1986). Integration simple cells. Journal of Physiology, 287, 163-176.
of noxious stimulation across separate somatosensory com¬ Annis, R. C., & Frost, B. (1973). Human visual ecology and
munications systems: A functional theory of pain. Journal orientation anisotropies in acuity. Science, 182, 729-731.
of Experimental Psychology: Human Perception and Perfor¬ Anstis, S. M. (1978). Apparent movement. In R. Held, H. W.
mance, 12, 92-102. Leibowitz, & H. L. Teuber (Eds.), Handbook of sensory
Allison, A. C. (1953). The structure of the olfactory bulb and physiology (pp. 655-673). New York: Springer-Verlag.
its relation to the olfactory pathways in the rabbit and the Anstis, S. M., & Mather, G. (1985). Effects of luminance and
rat. Journal of Comparative Neurology, 98, 309-348. contrast on direction of ambiguous motion. Perception, 14,
Allman, J., Miezin, F., & McGuinness, E. (1985). Direction- 167-179.
and velocity-specific responses from beyond the classical re¬ Anstis, S. M., Rogers, B., & Henry, J. (1978). Interactions
ceptive field in the middle temporal visual area (MT). Per¬ between simultaneous contrast and colored afterimages. Vi¬
ception, 14, 105—126. sion Research, 18, 899-911.
Allport, D. A., Antonis, B., & Reynolds, P. (1972). On the Antes, J. R. (1974). The time course of picture viewing. Jour¬
division of attention: A disproof of the single channel hy¬ nal of Experimental Psychology, 103, 62-70.
pothesis. Quarterly Journal of Psychology, 24, 225-235. Antes, J. R., & Penland, J. (1981). Picture context effects on
Ames, A., Jr. (1951). Visual perception and the rotating trape¬ eye movement patterns. In D. Fisher, R. Monty, & J. Send-

537
538 References

ers (Eds.), Eye movements: Cognition and visual perception ception. Psychological Review, 61, 183—193.
(pp. 157-170). New Jersey: Lawrence Erlbaum. Attneave, F. (1955). Symmetry, information, and memory for
Arend, L. E., & Goldstein, R. (1987). Lightness models, gra¬ patterns. American Journal of Psychology, 68, 209-222.
dient illusions, and curl. Perception & Psychophysics, 42, Aubert, H. (1886). Die bewegungsempfindung. Archiv fuer die
65-80. Gesamte Physiologie des Menschen and der Tiere, 39, 347-
Arlin, M. (1986). The effects of quantity, complexity, and at- 370.
tentional demand on children’s time perception. Perception Augenstine, L. G. (1962). A model of how humans process
& Psychophysics, 40, 177-182. information. Biometrics, 18, 420-421.
Arvidson, K., & Friberg, U. (1980). Human taste response and Avant, L. L. (1965). Vision in the ganzfeld. Psychological Bul¬
taste bud number in fungiform papillae. Science, 209, 807- letin, 64, 246-258.
808. Avant, L. L., & Lyman, P. J. (1975). Stimulus familiarity mod¬
Aschoff, J. (1979). Circadian rhythms: General features and en¬ ifies perceived duration in prerecognition visual processing.
docrinological aspects. In D. T. Krieger (Ed.), Endocrine Journal of Experimental Psychology: Human Perception and
rhythms. New York: Raven. Performance, 1, 205-213.
Aschoff, J. (1981). Handbook of behavioral neurobiology (Vol. Avant, L. L., Lyman, P. J., & Antes, J. R. (1975). Effects of
4). New York: Plenum. stimulus familiarity upon judged visual duration. Perception
Aschoff, J. (1984). Circadian timing. Annals of the New York & Psychophysics, 17, 253-262.
Academy of Sciences, 423, 442-468.
Ashby, F. G., & Perrin, N. A. (1988). Toward a unified theory
of similarity and recognition. Psychological Review, 95, Baddeley, A. D. (1966). Time estimation at reduced body tem¬
124-150. perature. American Journal of Psychology, 79, 475-479.
Aslin, R. N. (1981a). Development of smooth pursuit in human Baird, J. C. (1975). Psychophysical study of numbers. IV. Gen¬
infants. In D. Fisher, R. Monty, & J. Senders (Eds.), Eye eralized preferred state theory. Psychological Research,
movements: Cognition and visual perception. New Jersey: 38, 175-187.
Lawrence Erlbaum. Baird, J. C., Green, D. M., & Luce, R. D. (1980). Variability
Aslin, R. N. (1981b). Experiential influences and sensitive pe¬ and sequential effect in cross modality matching of area and
riods in perceptual development: A unified model. In R. N. loudness. Journal of Experimental Psychology: Human Per¬
Aslin, J. R. Albers, & M. R. Petersen (Eds.), Development ception & Psychophysics, 6, 277-289.
of perception (pp. 45-93). New York: Academic Press. Baird, J. C., Lewis, C., & Romer, D. (1970). Relative frequen¬
Aslin, R. N. (1985). Effects of experience on sensory and per¬ cies of numerical responses in ratio estimation. Perception
ceptual development: Implications for infant cognition. In J. <& Psychophysics, 8, 358-362.
Mehler & R. Fox (Eds.), Neonate cognition: Beyond the Baird, J. C., & Noma, E. (1978). Fundamentals of scaling and
blooming buzzing confusion (pp. 157-184). Hillsdale, NJ: psychophysics. New York: Wiley.
Lawrence Erlbaum. Baker, C. L., & Braddick, O. J. (1985). Temporal properties
Aslin, R. N. (1987). Motor aspects of visual development in of the short-range process in apparent motion. Perception,
infancy. In P. Salapatek & L. Cohen (Eds.), Handbook of 14, 181-192.
infant perception: Vol. 1. From sensation to perception (pp. Balint, R. (1909). Seelenlahmung des “Schauens,” optische
43-113). Orlando, FL: Academic Press. Ataxie, raumliche Stoning der Aufmerksamkeit. Mon-
Aslin, R. N., & Dumais, S. (1980). Binocular vision in infants: atsschr. Psychiatr. Neurol. 25, 51-81.
A review and a theoretical framework. In H. Reese & L. Ball, W., & Vurpillot, E. (1976). La perception du mouvement
Lipsett (Eds.), Advances in child development and behavior: en profondur chez le nourrisson. L'Annee Psychologique,
(Vol. 15, pp. 54-95). New York: Academic Press. 67, 393-400.
Aslin, R. N., & Smith, L. B. (1988). Perceptual development. Balogh, R. D.. & Porter. R. H. (1986). Olfactory preferences
Annual Review of Psychology, 39, 435-473. resulting from mere exposure in human neonates. Infant Be¬
Atkinson, J. (1979). Development of optokinetic nystagmus in havior and Development, 9, 395-401.
the human infant and monkey infant: An analogue to devel¬ Banks, M. S., & Dannemiller, J. L. (1987). Infant visual psy¬
opment in kittens. In R. D. Freeman (Ed.), Developmental chophysics. In P. Salapatek & L. Cohen (Eds.), Handbook
neurobiology of vision (pp. 277-288). New York: Plenum. of infant perception: Vol. 1. From sensation to perception
Atkinson, J., & Braddick, O. (1981). Acuity, contrast sensitiv¬ (pp. 115-184). Orlando, FL: Academic Press.
ity, and accommodation in infancy. In R. Aslin, J. Alberts, Banks, M. S., & Salapatek, P. (1983). Infant visual perception.
& M. Petersen (Eds.), Development of perception: Psycho- In M. M. Haith & J. J. Campos (Eds.), Handbook of child
biological perspectives: Vol. 2. The visual system (pp. 243- psychology. New York: Wiley.
278). New York: Academic Press. Bannatyne, A. (1971). Language, reading, and learning dis¬
Atkinson, J., Braddick, O., & French, J. (1979). Contrast sen¬ abilities. Springfield, IL: Thomas.
sitivity of the human neonate measured by the visual evoked Barbeito, R. (1981). Sighting dominance: An explanation based
potential. Investigative Ophthalmology and Visual Sciences, on the processing of visual direction in tests of sighting
18, 210-213. dominance. Vision Research, 21, 855-860.
Attneave, F. (1954). Some informational aspects of visual per¬ Barclay, C. D., Cutting, J. E., & Kozlowski, L. T. (1978).
 References 539

Temporal and spatial factors in gait perception that influence Benedetti, F. (1985). Processing of tactile spatial information
gender recognition. Perception & Psychophysics, 23, 145— with crossed fingers. Journal of Experimental Psychology:
152. Human Perception and Performance, 11, 517-525.
Barlow, H. B. (1985). The role of single neurons in the psy¬ Bentham, J. van. (1985). Semantics of time. In J. A. Michon
chology of perception. Quarterly Journal of Experimental & J. L. Jackson (Eds.), In time, mind, and behavior (pp.
Psychology, 37A, 121-145. 266-278). Berlin: Springer-Verlag.
Bartleson, C. J. (1960). Memory colors of familiar objects. Berbaum, K., Bever, T., & Chung, C. S. (1983). Light source
Journal of the Optical Society of America, 50, 73-77. position in the perception of object shape. Perception, 12,
Bartoshuk, L. M. (1974). Taste illusions: Some demonstrations. 411-416.
Annals of the New York Academy of Sciences, 237, 279- Berbaum, K., Bever, T., & Chung, C. S. (1984). Extending
285. the perception of shape from known to unknown shading.
Bartoshuk, L. M. (1978). Gustatory system. In R. B. Masterton Perception, 13, 479-488.
(Ed.), Handbook of behavioral neurobiology: Vol. I. Sen¬ Berbaum, K., & Lenel, J. C. (1983). Objects in the path of
sory integration (pp. 503-567). New York: Plenum. apparent motion. American Journal of Psychology, 96, 491-
Bartoshuk, L. M. (1979). Bitter taste of saccharine related to SOl.
the genetic ability to taste the bitter substance 6-n-propy- Berbaum, K., Tharp, D., & Mroczek, K. (1983). Depth percep¬
thiouracil. Science, 205, 934-935. tion of surfaces in pictures: Looking for conventions of de¬
Bashford, J. A. & Warren, R. M. (1987). Multiple phonemic piction in Pandora’s box. Perception, 12, 5-20.
restorations follow the rules for auditory induction. Percep¬ Berg, K. M., & Smith, M. C. (1983). Behavioral thresholds for
tion & Psychophysics, 42, 114-121. tones during infancy. Journal of Experimental Child Psy¬
Batteau, D. W. (1967). The role of the pinna in human locali¬ chology, 35, 409-425.
zation. Proceedings of the Royal Society of London, Series Bergeijk, W. A. van. (1967). The evolution of vertebrate hear¬
B, 168, 158-180. ing. In W. D. Neff (Ed.), Contributions to sensory physiol¬
Beauchamp, G. K., & Cowart, B. J. (1985). Congenital and ogy (Vol. 2, pp. 1-49). New York: Academic Press.
experiential factors in the development of human flavor pref¬ Berger, G. O. (1896). Uber den Einfluss der Reizstarke auf die
erences. Appetite, 6, 357-372. Dauer einfacher psychischer Vorgange mit besonderer Ruck-
Beck, J. (1965). Apparent spatial position and the perception of sicht auf Lichtreize. Philosophische Studien, (Wundt), 3,
lightness. Journal of Experimental Psychology, 69, 170- 38-93.
179. Berkley, M. A. (1982). Neural substrates of the visual percep¬
Beck, J. (1966). Effects of orientation and of shape similarity tion of movement. In A. H. Wertheim, W. A. Wagenaar, &
on perceptual grouping. Perception & Psychophysics, 1, H. W. Leibowitz (Eds.), Tutorials on motion perception
300-302. (pp. 201-229). New York: Plenum.
Beck, J. (1982). Textural segmentation. In J. Beck (Ed.), Or¬ Berlin, B., & Kay, P. (1969). Basic color terms. Berkeley: Uni¬
ganization and Representation in Perception (pp. 285-318). versity of California Press.
Hillsdale, NJ: Lawrence Erlbaum. Bernstein, L. R., & Green, D. M. (1987). Detection of simple
Beck, J., & Schwartz, T. (1979). Vemer acuity with dot test and complex changes of spectral shape. Journal of the
objects. Vision Research, 19, 313-319. Acoustical Society of America, 82, 1587-1592.
Beck, N. C., & Siegel, L. J. (1980). Preparation for childbirth Berry, J. W. (1971). Mueller-Lyer susceptibility: culture, ecol¬
and contemporary research on pain, anxiety, and stress re¬ ogy, race? International Journal of Psychology, 7, 193—
duction: A review and critique. Psychosomatic Medicine, 196.
42, 429-447. Besser, G. (1966). Centrally acting drugs and auditory flutter.
Begleiter, H., Porjesz, B., Chou, C. L. (1981). Auditory brain¬ In A. Herxheimer (Ed.), Proceedings of the Symposium on
stem potentials in chronic alcoholics. Science, 211, 1064- Drugs and Sensory Functions (pp. 199-200). London:
1066. Churchill.
Beidler, L. M., & Smallman, R. L. (1965). Renewal of cells Bhatia, B. (1975). Minimum separable as function of speed of
within taste buds. Journal of Cell Biology, 27, 263-272. a moving object. Vision Research, 15, 23-33.
Bekesy, G. von. (1947). A new audiometer. Acta Oto-laryngo- Biederman, I. (1987). Recognition-by-components: A theory of
logica, 35, 411-422. human image understanding. Psychological Review, 94,
Bekesy, G. von. (1959). Synchronism of neural discharges and 115-147.
their demultiplication in pitch perception on the skin and in Biederman, I., Glass, A. L., & Stacey, E. W., Jr. (1973).
learning. Journal of the Acoustical Society of America, 31, Searching for objects in real-world scenes. Journal of Ex¬
338-349. perimental Psychology, 97, 22-27.
Bekesy, G. von. (1960). Experiments in hearing. New York: Billings, B. L., & Stokinger, T. E. (1977). Investigation of sev¬
McGraw-Hill. eral aspects of low-frequency (200 Hz) central masking.
Bekesy, G. von. (1967). Sensory inhibition. Princeton, NJ: Journal of the Acoustical Society of America, 61, 1260-
Princeton University Press. 1263.
Bern, S. L. (1981). Gender schema theory: A cognitive account Birch, E. E., Shimojo, S., & Held, R. (1985). Preferential-
of sex typing. Psychological Review, 88, 354-364. looking assessment of fusion and stereopsis in infants aged
540 References

1-6 months. Investigative Ophthalmology & Visual Science, Bonnet, C. (1984). Discrimination of velocities and mechanisms
26, 366-370. of motion perception. Perception, 13, 275-282.
Birren, J., Woods, A., & Williams, M. (1980). Behavioral Borg, G., Diamant, H., Oakley, B., Strom, L., & Zotterman,
slowing with age: Causes, organization, and consequences. Y. (1967). A comparative study of neural and psychophysi¬
In L. Poon (Ed.), Aging in the 1980s (pp. 293-308). Wash¬ cal responses to gustatory stimuli. In T. Hayashi (Ed.), Ol¬
ington, D.C.: American Psychological Association. faction and taste II. (pp. 253-264). Oxford: Pergamon.
Bishop, P. O. (1981). Binocular vision. In R. A. Moses (Ed.), Bomstein, M. H. (1973). Color vision and color naming: A psy-
Adler’s physiology of the eye: Clinical applications (7th chophysiological hypothesis of cultural difference. Psycho¬
ed.). (pp. 575-649). St. Louis: Mosby. logical Bulletin, 80, 257-285.
Blake, R. (1981). Strategies for assessing visual deficits in ani¬ Bomstein, M. H. (1975). The influence of visual perception on
mals with selective neural deficits. In R. N. Aslin, J. R. culture. American Anthropologist, 77, 774-798.
Alberts, & M. R. Petersen (Eds.), Development of percep¬ Bomstein, M. H. (1977). Developmental pseudocyanapsia: On¬
tion: Vol. 2. The visual system (pp. 95-110). New York: togenetic change in human color vision. American Journal
Academic Press. of Optometry & Physiological Optics, 54, 464-469.
Blakemore, C. (1978). Maturation and modification in the de¬ Bomstein, M. H. (1981). Two kinds of perceptual organization
veloping visual system. In R. Held, M. W. Leibowitz, & near the beginning of life. In W. Collins (Ed.), Aspects of
H. L. Teuber (Eds.), Handbook of sensory physiology: Vol. the development of competence. Hillsdale, NJ: Lawrence
8. Perception (pp. 377-436). New York: Springer-Verlag. Erlbaum.
Blakemore, C., & van Sluyters, R. C. (1975). Innate and envi¬ Bomstein, M. H. (1985). Infant into adult: Unity to diversity in
ronmental factors in the development of the kitten’s visual the development of visual categorization. In J. Mehler & R.
cortex. Journal of Physiology, 248, 633-716. Fox (Eds.), Neonate cognition: Beyond the blooming buzz¬
Blakeslee, A. F., & Salmon, T. H. (1935). Genetics of sensory ing confusion (pp. 115-138). Hillsdale, NJ: Lawrence Erl¬
thresholds: Individual taste reactions for different sub¬ baum.
stances. Proceedings of the National Academy of Sciences of Bomstein, M. H., Kessen, W., & Weiskopf, S. (1976). Color
the U.S.A., 21, 84-90. vision and hue categorization in young human infants. Jour¬
Blarney, P. J., Dowell, R. C., Brown, A. M., Clark, G. M., nal of Experimental Psychology: Human Perception and
& Seligman, P. M. (1987). Vowel and consonant recogni¬ Performance, 115-129.
tion of cochlear implant patients using formant-estimating Bomstein, M. H., & Monroe, M. D. (1978). Color-naming evi¬
speech processors. Journal of the Acoustical Society of dence for tritan vision in the fovea. American Journal of
America, 82, 48-57. Optometry & Physiological Optics, 55, 627-630.
Blasdel, G. G., Mitchell, D. E., Muir, D. W., & Pettigrew, Bossom, J., & Ommaya, A. K. (1968). Visuo-motor adaptation
J. D. (1977). A combined physiological and behavioral (to prismatic transformation of the retinal image) in monkeys
study of the effect of early visual experience with contours with bilateral dorsal rhizotomy. Brain, 91, 161-172.
of a single orientation. Journal of Physiology, 265, 615— Botte, M. C., Baruch, C., & Scharf, B. (1986). Loudness re¬
636. duction and adaptation induced by a contralateral tone. Jour¬
Blazynski, C., & Ostroy, S. E. (1981). Dual pathways in the nal of the Acoustical Society of America, 80, 73-81.
photolysis of rhodopsin: Studies using a direct chemical Botte, M. C., Canavet, G., & Scharf, B. (1982). Loudness ad¬
method. Vision Research, 21, 833-841. aptation induced by an intermittent tone. Journal of the
Bliss, J. C., Katcher, M. H., Rogers, C. H., & Shepard, Acoustical Society of America, 72, 727-739.
R. P. (1970). Optical-to-tactile image conversion for the Botwinick, J. (1984). Aging and behavior: A comprehensive in¬
blind. IEEE Transactions on Man-Machine Systems, 11, tegration of research findings, (3rd ed.). New York: Sprin¬
58-65. ger.
Block, R. A. (1974). Memory and the experience of duration in Bowen, R. W. (1981). Latencies for chromatic and achromatic
retrospect. Memory and Cognition, 2, 153-160. visual mechanisms. Vision Research, 2, 1457-1466.
Block, R. A. (1978). Remembered duration: Effects of event Bowker, D. O., & Mandler, M. B. (1981). Apparent contrast
and sequence complexity. Memory and Cognition, 6, 320- of suprathreshold gratings varies with stimulus orientation.
326. Perception & Psychophysics, 29, 585-588.
Block, R. A., George, E. J., & Reed, M. A. (1980). A watched Bowmaker, J. K., & Dartnall, H. J. A. (1980). Visual pigments
pot sometimes boils: A study of duration experience. Acta of rods and cones in a human retina. Journal of Physiology,
Psychologica, 46, 81-94. 298, 501-511.
Blough, P. M., & Slavin, K. (1987). Reaction time assessments Boyd, I. A., & Roberts, T. D. M. (1953). Proprioceptive dis¬
of gender differences in visual-spatial performance. Percep¬ charges from the stretch receptors in the knee-joint of the
tion & Psychophysics, 41, 276-281. cat. Journal of Physiology (London), 122, 38-58.
Boer, L., & Keuss, P. (1982). Global precedence as a post per¬ Boynton, R. M. (1971). Color vision. In J. W. King & L. A.
ceptual effect: An analysis of speed accuracy trade off func¬ Riggs (Eds.), Woodworth and Schlossberg's experimental
tions. Perception & Psychophysics, 31, 358-366. psychology (3rd ed.). (pp. 315-368). New York: Holt,
Bolton, T. L. (1894). Rhythm. American Journal of Psychol¬ Rinehart & Winston.
ogy, 6, 145-238. Boynton, R. M. (1978). Ten years of research with the
 References 541

minimally distinct border. In J. C. Armington, J. Kraus- (Ed.), Allsemaie chirurgie der gehirnkrankheiten. Stuttgart:
kopf, & B. R. Wooten (Eds.), Visual psychophysics and F. Enke.
physiology: A volume dedicated to Lorrin Riggs (pp. 193— Brooks, R. A. (1981). Symbolic reasoning among 3-D models
207). New York: Academic Press. and 2-D images. Artificial Intelligence, 17, 205-244.
Boynton, R. M. (1979). Human color vision. New York: Holt, Brou, P., Sciancia, T. R., Linden, L., & Lettvin, J. Y. (1986).
Rinehart & Winston. The colors of things. Scientific American, 255 (3), 84-91.
Boynton, R. M. (1988). Color vision. Annual Review of Psy¬ Brown, A. C., Beeler, W. J., Kloka, A. C., & Fields, R. W.
chology, 39, 69-100. (1985). Spatial summation of pre-pain and pain in human
Boynton, R. M., & Gordon, J. (1965). Bezold-Brucke hue shift teeth. Pain, 21, 1-16.
measured by color-naming technique. Journal of the Optical Brown, B. (1972). Resolution thresholds for moving targets at
Society of America, 55, 78-86. the fovea and in the peripheral retina. Vision Research, 12,
Braddick, O. J. (1980). Low-level and high-level processes in 293-304.
apparent motion. Philosophical Transactions of the Royal Brown, E. L., & Deffenbacher, K. (1979). Perception and the
Society of London, Series B, 290, 137-151. senses. New York: Oxford University Press.
Braddick, O., & Atkinson, J. (1979). Accommodation and acu¬ Brown, P. E. (1972). Use of acupuncture in major surgery.
ity in the human infant. In R. D. Freeman (Ed.), Develop¬ Lancet, 1, 1328-1330.
mental neurobiology of vision (pp. 289-300). New York: Brown, P. K., & Wald, G. (1964). Visual pigments in single
Plenum. rods and cones of the human retina. Science, 144, 45-52.
Braddick, O., Atkinson, J., Julesz, B., Kropfl, W., Bodis- Brown, R. E., & MacDonald, D. W. (Eds.). (1985a). Social
Wollner, I., Raab, E. (1980). Cortical binocularity in in¬ odours in mammals: Vol. 1. Oxford: Clarendon.
fants. Nature, 288, 363-385. Brown, R. E., & MacDonald, D. W. (Eds.). (1985b). Social
Bradley, A., & Skottun, B. C. (1987). Effects of contrast and odours in mammals: Vol. 2. Oxford: Clarendon.
spatial frequency on vernier acuity. Vision Research, 27, Brown, S. W. (1985). Time perception and attention: The ef¬
1817-1824. fects of prospective versus retrospective paradigms and task
Bradley, R. M., & Stem, 1. B. (1967). The development of the demands on perceived duration. Perception & Psychophys¬
human taste bud during the foetal period. Journal of Anat¬ ics, 38, 115-124.
omy, 101, 743-752. Brown, T. S. (1975). General biology of sensory systems. In
Braff, D. L., Silverton, L., Saccuzzo, D. P., & Janowsky, B. Scharf (Ed.), Experimental sensory psychology (pp. 69-
D. S. (1981). Impaired speed of visual information process¬ 111). Glenview. IL: Scott-Foresman.
ing in marihuana intoxication. American Journal of Psychia¬ Brown, W. (1910). The judgment of difference. University of
try, 138 (5), 613-617. California, Berkeley, Publications in Psychology, 1, 1-71.
Braine, L. G., Plastow, E., & Greene, S. I. (1987). Judgments Brownell, W. E., Bader, C. R., Bertrand, D., & de Ribau-
of shape orientation: A matter of contrasts. Perception & pierre, Y. (1985). Evoked mechanical responses of isolated
Psychophysics, 41, 335-344. cochlear outer hair cells. Science, 227, 194-196.
Brazelton, T., Scholl, M., & Robey, J. (1966). Visual re¬ Bruce, C., Desimone, R., & Gross, C. G. (1981). Visual prop¬
sponses in the newborn. Pediatrics, 37, 284-290. erties of neurons in a polysensory area in superior temporal
Bregman, A. S. (1978). Auditory streaming: Competition sulcus of the macaque. Journal of Neurophysiology, 46,
among alternative organizations. Perception & Psychophys¬ 369-384.
ics, 23, 391-398. Bruner, J. S., Postman, L., & Rodrigues, J. (1951). Expecta¬
Briand, K. A., & Klein, R. M. (1987). Is Posner’s “beam” the tions and the perception of color. American Journal of Psy¬
same as Treisman’s “glue”?: On the relation between visual chology, 64, 216-227.
orienting and feature integration theory. Journal of Experi¬ Brunswick, E. (1952). The conceptual framework of psychol¬
mental Psychology: Human Perception and Performance, ogy. International Encyclopedia of Unified Science, 1, No.
13, 228-241. 10.
Bridges, C. D. B. (1986). Biochemistry of vision—A perspec¬ Brunswick, E. (1955). Representative design and probabilistic
tive. Vision Research, 26, 1317-1337. theory in a functional psychology. Psychological Review,
Brillat-Savarin, J. A. (1971). The physiology of taste: Or med¬ 62, 193-217.
itations on transcendental gastronomy. (M. F. K. Fisher Brunswick, E. (1956). Perception as a representative design of
Trans.). New York: Knopf. (Original work published 1825) psychological experiments. Berkeley: University of Califor¬
Broadbent, D. (1958). Perception and communication. Oxford: nia Press.
Pergamon. Brussell, E. M., & Festinger, L. (1973). The Gelb effect:
Broadbent, D. E., & Gregory, M. (1963). Vigilance considered Brightness contrast plus attention. American Journal of Psy¬
as a statistical decision. British Journal of Psychology, 54, chology, 86, 225-235.
309-323. Bryden, M. P. (1982). Laterality: Functional symmetry in the
Broadbent, D. E., & Gregory, M. (1965). Effects of noise and intact brain. New York: Academic Press.
of signal rate upon vigilance analyzed by means of decision Buckhout, R. (1976). Eyewitness testimony. In R. Held & W.
theory. Human Factors, 7, 155-162. Richards (Eds.), Recent progress in perception (pp. 205-
Brodmann, K. (1914). Physiologie des gehimg. In F. Krause 213). San Francisco: Freeman.
542 References

Bundesen, C., Larsen, A., & Farrell, J. E. (1983). Visual ap¬ Cain, W. S., & Johnson, F., Jr. (1978). Lability of odor pleas¬
parent movement: Transformations of size and orientation. antness: Influence of mere exposure. Perception, 7, 459-
Perception, 12, 549-568. 465.
Burg, A. (1966). Visual acuity as measured by dynamic and Calis, G., & Leeuwenberg, E. (1981). Grounding the figure.
static tests: A comparative evaluation. Journal of Applied Journal of Experimental Psychology: Human Perception and
Psychology, 50, 460-466. Performance, 7, 1386-1397.
Bums, E. M. (1981). Circularity in relative pitch judgments for Callaghan, T. C., Lasaga, M. L., & Gamer, W. R. (1986).
inharmonic complex tones: The Shepard demonstration re¬ Visual texture segregation based on orientation and hue.
visited, again. Perception & Psychophysics, 30, 467-472. Perception & Psychophysics, 39, 32-38.
Butler, D. L., & Kring, A. M. (1987). Integration of features Campbell, F. W., & Maffei, L. (1981). The influence of spatial
in depictions as a function of size. Perception & Psycho¬ frequency and contrast on the perception of moving patterns.
physics, 41, 159-164. Vision Research, 21, 713-721.
Butler, R. A. (1987). An analysis of the monaural displacement Campbell, F. W., & Robson, J. G. (1968). Application of Four¬
of sound in space. Perception & Psychophysics, 41, 1-7. ier analysis to the visibility of gratings. Journal of Physiol¬
Butler, R. A., Levy, E. T., & Neff, W. D. (1980). Apparent ogy, 197, 551-566.
distance of sounds recorded in echoic and anechoic cham¬ Campbell, K. B., Baribeau-Braun, J., & Braun, C. (1981).
bers. Journal of Experimental Psychology: Human Percep¬ Neuroanatomical and physiological foundations of extraver¬
tion and Physiology, 6, 745-750. sion. Psychophysiology, 18, 263—267.
Butters, N., Barton, M., & Brody, B. A. (1970). Right parietal Canevet, G., Heilman, R., & Scharf, B. (1986). Group esti¬
lobe and cross-model associations. Cortex, 6, 19-46. mation of loudness in sound fields. Acustica, 60, 277-282.
Butterworth, G. (1981). The origins of auditory-visual percep¬ Cannon, M. W., Jr. (1983). Contrast sensitivity: Psychophysi¬
tion and visual proprioception in human development. In cal and evoked potential methods compared. Vision Re¬
R. D. Walk & H. L. Pick, Jr. (Eds.), Intersensory percep¬ search, 23, 87-95.
tion and sensory integration (pp. 37-70). New York: Carey, S. (1981). The development of face perception. In G.
Plenum. Davies, H. Ellis, & J. Shepherd (Eds.), Perceiving and Re¬
membering Faces (pp. 9-38). London: Academic Press.
Carey, S., & Diamond, R. (1977). From piecemeal to configu¬
Cacace, A. T., & Margolis, R. H. (1985). On the loudness of rational representation of faces. Science, 195, 312-314.
complex stimuli and its relationship to cochlear excitation. Carey, S., Diamond, R., & Woods, B. (1980). The develop¬
Journal of the Acoustical Society of America, 78, 1568— ment of face recognition: A maturational component. Devel¬
1573. opmental Psychology, 16, 257-269.
Caelli, T. (1982). On discriminating visual textures and images. Carlson, C. R. (1983). A simple model for vernier acuity. In¬
Perception & Psychophysics, 31, 149-159. vestigations in Ophthalmology and Visual Science Supple¬
Caelli, T. (1984). On the specification of coding principles for ment, 24, 276.
visual image processing. In P. C. Dodwell & T. Caelli Carlson, N. R. (1988). Foundations of physiological psychol¬
(Eds.), Figural Synthesis (pp. 153-184). Hillsdale, NJ: ogy. Boston: Allyn & Bacon.
Lawrence Erlbaum. Carlson, V. R. (1958). Effect of lysergic acid diethylamide
Cagen, R. H., & Rhein, L. D. (1980). Biochemical basis of (LSD-25) on the absolute visual threshold. Journal of Com¬
recognition of taste and olfactory stimuli. In H. van der parative and Physiological Psychology, 51, 528-531.
Starre (Ed.), Olfaction and taste VII (pp. 35-44). London: Carlson, V. R. (1977). Instructions and perceptual constancy
IRL Press. judgments. In W. Epstein (Ed.), Stability and constancy in
Cahoon, D., & Edmonds, E. M. (1980). The watched pot still visual perception: Mechanisms and processes (pp. 217-
won’t boil: Expectancy as a variable in estimating the pas¬ 254). New York: Wiley.
sage of time. Bulletin of the Psychonomic Society, 16, 115- Carmichael, L., Hogan, H. P., & Walter, A. A. (1932). An
116. experimental study of the effect of language on the repro¬
Cain, D. P., & Bindra, D. (1972). Response of amygdala single duction of visually perceived forms. Journal of Experimen¬
units to odors in the rat. Experimental Neurology, 35, 98- tal Psychology, 15, 73-86.
110. Caron, A., Caron, R., Caldwell, R., & Weiss, S. (1973). Infant
Cain, W. S. (1969). Odor intensity: Differences in the exponent perception of the structural properties of the face. Develop¬
of the psychophysical function. Perception & Psychophys¬ mental Psychology, 9, 385-399.
ics, 6, 349-354. Carpenter, D. L., & Dugan, M. P. (1983). Motion parallax in¬
Cain, W. S. (1977). Differential sensitivity for smell: “Noise” formation for direction of rotation in depth: Order and direc¬
at the nose. Science, 195, 796-798. tion components. Perception, 12, 559-569.
Cain, W. S. (1979). To know with the nose: Keys to odor iden¬ Carson, C. R. (1983). A simple model for vernier acuity. Inves¬
tification. Science, 203, 467-470. tigative Ophthalmology and Visual Science, 276 (Suppl.
Cain, W. S., & Engen, T. (1969). Olfactory adaptation and the 24).
scaling of odor intensity. In C. Pfaffman (Ed.), Olfaction Carter, R. R., & Hirsch, J. (1955). An experimental compari¬
and taste III (pp. 127-141). New York: Rockefeller Univer¬ son of several psychological scales of weight. American
sity Press. Journal of Psychology, 68, 645-49.
 References 543

Casey, K. L. (1978). Neural mechanisms of pain. In E. C. Car- Clark, H. H., & Clark, E. V. (1977). Psychology and lan¬
terette & M. P. Friedman (Eds.), Handbook of perception: guage: An introduction to psycholinguistics. New York:
Vol. 6. Feeling and hurting (pp. 183-230). New York: Ac¬ Harcourt, Brace & Jovanovich.
ademic Press. Clark, J. C., & Whitehurst, G. S. (1974). Asymmetrical stim¬
Casey, K. L., & Morrow, T. J. (1983). Ventral posterior tha¬ ulus control and the mirror-image problem. Journal of Ex¬
lamic neurons differentially responsive to noxious stimula¬ perimental Child Psychology, 17, 147-166.
tion of the awake monkey. Science, 221, 675-677. Clark, W. C., & Yang, J. C. (1974). Acupunctural analgesia?
Cattell, J. M. (1886). The influence of the intensity of the stim¬ Evaluation by signal detection theory. Science, 184, 1096-
ulus on the length of the reaction time. Brain, 9, 512-514. 1098.
Cavanaugh, P. (1984). Image transforms in the visual system. Clarkson-Smith, L., & Halpem, D. F. (1983). Can age-related
In P. C. Dodwell & T. Caelli (Eds.), Figural Synthesis (pp. deficits in spatial memory be attenuated through the use of
185-218). Hillsdale, NJ: Lawrence Erlbaum. verbal coding? Experimental Aging Research, 9, 179-184.
Cegalis, J. A., & Deptula, D. (1981). Attention in schizophre¬ Clifford, B. R. & Bull, R. (1978). The psychology of person
nia: Signal detection in the visual periphery. Journal of Ner¬ identification. London: Routledge & Kegan Paul.
vous and Mental Health Diseases, 169, 751-760. Clifton, R. K., Morrongiello, B. A., & Dowd, J. M. (1984). A
Cerella, J., Poon, L., & Williams, D. (1980). Age and the com¬ developmental look at an auditory illusion: The precedence
plexity hypothesis. In L. Poon (Ed.), Aging in the 1980s effect. Developmental Biology, 17, 519-536.
(pp. 332-345). Washington, D.C.: American Psychological Cogan, R., & Spinnato, J. A. (1986). Pain and discomfort
Association. thresholds in late pregnancy. Pain, 27, 63-68.
Cemoch, J. M., & Porter, R. H. (1985). Recognition of mater¬ Cohen, K. (1981). The development of strategies of visual
nal axillary odors by infants. Child Development, 56, 1593— search. In D. Fisher, R. Monty, & J. Senders (Eds.), Eye
1598. movements: Cognition and visual perception. Hillsdale, NJ:
Chapman, C. R. (1978). The hurtful world: Pathological pain Lawrence Erlbaum.
and its control. In E. C. Carterette & M. P. Friedman Cohen, W. (1958). Color perception in the chromatic Ganzfeld.
(Eds.), Handbook of perception: Vol. 6B. Feelings and hurt¬ American Journal of Psychology, 71, 390-394.
ing (pp. 264-301). New York: Academic Press. Cole, R. A., Rudnicky, A. I., Zue, V. W., & Reddy, D. R.
Chapman, C. R., Casey, K. L., Dubner, R., Foley, K. M., (1980). Speech as patterns on paper. In R. A. Cole (Ed.),
Gracely, R. H., & Reading, A. E. (1985). Pain measure¬ Perception and production of fluent speech (pp. 3-50).
ment: An overview. Pain, 22, 1—31. Hillsdale, NJ: Lawrence Erlbaum.
Cheeseman, J., & Merikle, P. M. (1984). Priming with and Coles, M. G., Gale, A., & Kline, P. (1971). Personality and
without awareness. Perception & Psychophysics, 36, 387- habituation of the orienting reaction: Tonic and response
395. measures of electrodermal activity. Psychophysiology, 8,
Cheng, P. W. (1985). Restructuring versus automaticity: Alter¬ 54-63.
native accounts of skill acquisition. Psychological Review, Collings, V. B. (1974). Human taste response as a function of
92, 414-423. locus of stimulation on the tongue and soft palate. Percep¬
Cheng, T. O. (1973). Acupuncture anesthesia. Science, 179, tion & Psychophysics, 16, 169-174.
521. Comfort, A. (1971). Likelihood of human pheromones. Nature,
Cherry, E. C. (1953). Some experiments on the recognition of 230, 432-433.
speech, with one and with two ears. Journal of the Acousti¬ Condon, W. S., & Sander, L. W. (1974). Neonate movement
cal Society of America, 25, 975-979. is synchronized with adult speech: Interactional participation
Chevrier, J., & Delorme, A. (1983). Depth perception in Pan¬ and language acquisition. Science, 183, 99-101.
dora’s box and size illusion: Evolution with age. Perception, Cooke, N. M., Breen, T. J., & Schvaneveldt, R. W. (1987). Is
12, 177-185. consistent mapping necessary for high-speed search? Journal
Chocolle, R. (1940). Variations des temps de reaction auditifs of Experimental Psychology: Learning, Memory, and Cog¬
en fonction de l’intensite a diverses frequences. Annee Psy- nition, 13, 223-229.
chologique, 41, 65-124. Cooper, B. Y., Vierck, C. J., Jr., & Yeomans, D. C. (1986).
Chocolle, R. (1962). Les effets des interactions interaurales Selective reduction of second pain sensations by systemic
dans l’audition. Journale de Psychologie, 3, 255, 282. morphine in humans. Pain, 24, 93-116.
Cholewiak, R. W., & Craig, J. C. (1984). Vibrotactile pattern Coren, S. (1966). Adaptation to prismatic displacement as a
recognition and discrimination at several body sites. Percep¬ function of the amount of available information. Psycho-
tion & Psychophysics, 35, 503-514. nomic Science, 4, 407-408.
Chomsky, N., & Miller, G. A. (1963). Introduction to the for¬ Coren, S. (1969). Brightness contrast as a function of figure-
mal analysis of natural languages. In R. D. Luce, R. Bush, ground relations. Journal of Experimental Psychology, 80,
& E. Galanter (Eds.), Handbook of mathematical psychol¬ 517-524.
ogy (Vol. 2, pp. 269-321). New York: Wiley. Coren, S. (1972). Subjective contours and apparent depth. Psy¬
Choudhurt, B. P., & Crossey, A. D. (1981). Slow-movement chological Review, 79, 359-367.
sensitivity in the human field of vision. Physiology and Be¬ Coren, S. (1983). Set. In R. J. Corsini (Ed.), The encyclopedia
havior, 26, 125—128. of psychology. New York: Wiley.
544 References

Coren, S. (1984a). Lens model (Brunswick). In R. J. Corsini Coren, S., & Porac, C. (1984). Structural and cognitive com¬
(Ed.), The encyclopedia of psychology (Vol. 2, pp. 302- ponents in the Mueller-Lyer illusion assessed via cyclopean
303). New York: Wiley. presentation. Perception & Psychophysics, 35, 313-318.
Coren S. (1984b). Set. In R. J. Corsini (Ed.), The encyclopedia Coren, S., & Porac, C. (1987). Individual differences in visual-
of psychology (Vol. 3, pp. 296-298). New York: Wiley. geometric illusions: Predictions from measures of spatial
Coren, S. (1986). An efferent component in the visual percep¬ cognitive abilities. Perception & Psychophysics, 41, 211 —
tion of direction and extent. Psychological Review, 93, 391— 219.
410. Coren, S., Porac, C., & Duncan, P. (1981). Lateral preference
Coren, S., Bradley, D. R., Hoenig, P., & Girgus, J. S. (1975). in pre-school children and young adults. Child Development,
The effect of smooth tracking and saccadic eye movements 52, 443-450.
on the perception of size: The shrinking circle illusion. Vi¬ Coren, S., Porac, C., & Theodor, L. H. (1987). Set and sub¬
sion Research, 15, 49-55. jective contour. In S. Petry & G. E. Meyer (Eds.), The per¬
Coren, S. & Girgus, J. S. (1972a). Density of human lens pig¬ ception of illusory contours (pp. 237-245). New York:
mentation: In vivo measures over an extended age range. Springer-Verlag.
Vision Research, 12, 343-346. Cormack, R. H. (1984). Stereoscopic depth perception at far
Coren, S., & Girgus, J. S. (1972b). Differentiation and decre¬ viewing distances. Perception & Psychophysics, 35, 423-
ment in the Mueller-Lyer illusion. Perception & Psycho¬ 428.
physics, 12, 466-470. Comsweet, T. N. (1956). Determination of the stimuli for in¬
Coren, S., & Girgus, J. S. (1977). Illusions and constancies. In voluntary drifts and saccadic eye movements. Journal of the
W. Epstein (Ed.), Stability and constancy in visual percep¬ Optical Society of America, 46, 987-993.
tion: Mechanisms and processes (pp. 255-284). New York: Comsweet, T. N. (1970). Visual Perception. New York: Aca¬
Wiley. demic Press.
Coren, S., & Girgus, J. S. (1978). Seeing is deceiving: The Comsweet, T. N. (1985). Prentice Award Lecture: A simple
psychology of visual illusions. Hillsdale, NJ: Lawrence Erl- retinal mechanism that has complex and profound effects on
baum. perception. American Journal of Optometry and Physiologi¬
Coren, S., & Girgus, J. S. (1980). Principles of perceptual or¬ cal Optics, 62, 427-438.
ganization and spatial distortion: The gestalt illusions. Jour¬ Correia, M. J., & Guedry, F. E. (1978). The vestibular system:
nal of Experimental Psychology: Human Perception & Per¬ Basic biophysical and physiological mechanisms. In R. B.
formance, 6, 404-412. Masterton (Ed.), Handbook of sensory neurobiology: Vol. 1.
Coren, S., Girgus, J. S., & Schiano, D. (1986). Is adaptation Sensory integration. New York: Plenum.
of orientation-specific cortical cells a possible explanation of Corso, J. F. (1959). Age and sex differences in thresholds.
illusion decrement? Bulletin of the Psychonomic Society, 24, Journal of the Acoustical Society of America, 31, 498-509.
207-210. Corso, J. F. (1981). Aging sensory systems and perception.
Coren, S., & Hakstian, A. R. (1987). Visual screening without New York: Praeger.
the use of technical equipment: Preliminary development of Corwin, T. R., Moskowitz-Cook, A., & Green, M. A. (1977).
a behaviorally validated questionnaire. Applied Optics, 26, The oblique effect in a vernier acuity situation. Perception
1468-1472. and Psychophysics, 21, 445-449.
Coren, S., & Hakstian, A. R. (1988). Color vision screening Costanzo, R. M. & Gardner, E. P. (1981). Multiple-joint neu¬
without the use of technical equipment: Scale development rons in somatosensory cortex of awake monkeys. Brain Re¬
and cross-validation. Perception & Psychophysics, 43, 115— search, 24, 321-333.
120. Costanzo, R. M., & Graziadei, P. P. C. (1987). Development
Coren, S., & Hoenig, P. (1972). Eye movements and decrement and plasticity of the olfactory system. In T. E. Finger &
in the Oppel-Kundt illusion. Perception & Psychophysics, W. L. Silver (Eds.), Neurobiology of taste and smell (pp.
12, 224-225. 230-250). New York: Wiley.
Coren, S., & Keith, B. (1970). Bezold-Brucke effect: Pigment Cowey, A. (1981). Why are there so many visual areas? In
or neural locus? Journal of the Optical Society of America, F. O. Schmitt, F. G. Worden, G. Adelman, & S. G. Dennis
60, 559-562. (Eds.), The organization of the cerebral cortex, (pp. 395-
Coren, S., & Komoda, M. K. (1973). Apparent lightness as a 413). Cambridge, MA: MIT Press.
function of perceived direction of incident illumination. Cowley, J. J., Johnson, A. L., & Brooksbank, B. W. L.
American Journal of Psychology, 86, 345-349. (1977). The effect of two odorous compounds on perfor¬
Coren, S., & Porac, C. (1978). Iris pigmentation and visual- mance in an assessment-of-people test. Psychoneuroendocri¬
geometric illusions. Perception, 7, 473-478. nology, 2, 159-172.
Coren, S., & Porac, C. (1983a). The creation and reversal of Craig, J. C. (1977). Vibrotactile pattern perception: Extraordi¬
the Mueller-Lyer illusion through attentional manipulation. nary observers. Science, 196, 450-452.
Perception, 12, 49-54. Craig, J. C. (1978). Vibrotactile pattern recognition and mask¬
Coren, S., & Porac, C. (1983b). Subjective contours and appar¬ ing. In G. Gordon (Ed.), Active touch: The mechanism of
ent depth: A direct test. Perception & Psychophysics, 33, recognition of objects by manipulation (pp. 229-242). Ox¬
197-200. ford: Pergamon.
 References 545

Craig, J. C. (1981). Tactile letter recognition: Pattern duration in cognitive functioning. In L. Poon (Ed.), Aging in the
and modes of pattern generation. Perception & Psychophys¬ 1980s (pp. 327-331). Washington, D.C.: American Psycho¬
ics, 30, 540-546. logical Association.
Craig, J. C. (1983a). The role of onset in the perception of Curcio, C. A., Sloan, K. R., Packer, O., Hendrickson, A. E.,
sequentially presented vibrotactile patterns. Perception & & Kalina, R. E. (1987). Distribution of cones in human and
Psychophysics, 34, 421-432. monkey retina: Individual variability and radial asymmetry.
Craig, J. C. (1983b). Some factors affecting tactile pattern rec¬ Science, 236, 579-582.
ognition. International Journal of Neuroscience, 19, 47-58. Cutler, W. B., Preti, G. Krieger, A., Huggins, G. R., Garcia,
Craig, J. C., & Evans, P. M. (1987). Vibrotactile masking and C. R., & Lawley, H. J. (1986). Human axillary secretions
the persistence of tactual features. Perception & Psycho¬ influence women’s menstrual cycles: The role of donor ex¬
physics, 42, 309-317. tract from men. Hormones and Behavior, 20, 463-473.
Craig, K. D. (1978). Social modeling influences on pain. In Cutting, J. E. (1976). Auditory and linguistic processes in
R. A. Stembach (Ed.), The psychology of pain. New York: speech perception: Inferences from six fusions in dichotic
Raven. listening. Psychological Review, 83, 114-140.
Craig, K. D., Best, H., & Ward, L. M. (1975). Social model¬ Cutting, J. E. (1978). Generation of synthetic male and female
ling influences on psychophysical judgments of electrical walkers through manipulation of a biomechanical invariant.
stimulation. Journal of Abnormal Psychology, 84, 366-373. Perception, 7, 393-405.
Craig, K. D., & Coren, S. (1975). Signal detection analysis of Cutting, J. E. (1986). Perception with an eye for motion. Cam¬
social modelling influences on pain expressions. Journal of bridge, MA: MIT Press.
Psychosomatic Research, 19, 105-112. Cutting, J. E. (1987). Perception and information. Annual Re¬
Craig, K. D., & Prkachin, K. M. (1978). Social modelling in¬ view of Psychology, 38, 61-90.
fluences on sensory decision theory and psychophysiological Cutting, J. E., & Kozlowski, L. T. (1977). Recognizing friends
indexes of pain. Journal of Personality and Social Psychol¬ by their walk: Gait perception without familiarity cues. Bul¬
ogy, 36, 805-815. letin of the Psychonomic Society, 9, 353-356.
Craik, F., & Simon, E. (1980). The roles of attention and depth Cutting, J. E., & Proffitt, D. R. (1981). Gait perception as an
of processing in understanding age differences in memory. example of how we may perceive events. In R. Walk &
In L. Poon, J. Fozard, L. Cermak, & L. Thompson (Eds.), H. L. Pick, Jr. (Eds.), Intersensory perception and sensory
New directions in memory and aging: Proceedings of the integration (pp. 249-273). New York: Plenum.
George A. Talland memorial conference (pp. 95-112). Cutting, J. E., Proffitt, D. R., & Kozlowski, L. T. (1978). A
Hillsdale NJ: Lawrence Erlbaum. biomechanical invariant for gait perception. Journal of Ex¬
Crane, H. D. (1982). IHC-TM connect-disconnect in relation to perimental Psychology: Human Perception and Perfor¬
sensitization and masking of a HF-tone burst by a LF tone. mance, 4, 357-372.
IV. Journal of the Acoustical Society of America, 71, 1183— Cynader, M., Berman, N., & Hein, A. (1976). Recovery of
1193. function in cat visual cortex following prolonged depriva¬
Cratty, B. (1979). Perceptual and motor development in infants tion. Experimental Brain Research, 25, 139-156.
and children. Englewood Cliffs, NJ: Prentice-Hall. Cynader, M., Timney, B. N., & Mitchell, D. E. (1980). Period
Creed, R. S., Denny-Brown, D., Eccles, J. C., Liddell, of susceptibility of kitten visual cortex to the effects of mon¬
E. G. T., & Sherrington, C. S. (1932). Reflex activity of the ocular deprivation extends beyond 6 months of age. Brain
spinal cord. Reprint. London & New York: Oxford Univer¬ Research, 191, 545-550.
sity Press, Clarendon, 1972.
Critchley, M. (1964). The problem of visual agnosia. Journal Daan, S., Beersma, D. G. M., & Borbely, A. A. (1984). Tim¬
of Neurological Science, I, 274. ing of human sleep: Recovery process gated by a circadian
Crook, C. (1987). Taste and olefaction. In P. Salapatek & L. pacemaker. American Journal of Physiology, 246, 161-178.
Cohen. (Eds.), Handbook of infant perception: Vol.2. From Dacey, D. M. (1988). Dopamine-accumulating retinal neurons
perception to cognition (pp. 237-264). Orlando, FL: Aca¬ revealed by in vitro flourescence display a unique morphol¬
demic Press. ogy. Science, 240, 1196—1198.
Crossman, E. R. F. W. (1953). Entropy and choice time: The Dallenbach, K. M. (1927). The temperature spots and end-or¬
effect of frequency unbalance on choice response. Quarterly gans. American Journal of Psychology, 39, 402-427.
Journal of Experimental Psychology, 5, 41-51. Dallenbach, K. M. (1939). Pain: History and present status.
Crutchfield, R. S, Woodworth, D. G., & Albrecht, R. E. American Journal of Psychology, 52, 331-347.
(1958). Perceptual performance and the effective person. Dallos, P. (1978). Biophysics of the cochlea. In E. C. Friedman
(WADC-TN-58-60). Lackland Air Force Base, TX: Wright & M. P. Carterette (Eds.), Handbook of Perception: Vol. 4.
Air Development Center. (NTIS No. AD-151-039). Hearing (pp. 125-162).
Cuddy, L. L., Cohen, A. J., & Mewhort, D. J. K. (1981). Dallos, P. (1981). Cochlear physiology. Annual Review of Psy¬
Perception of structure in short melodic sequences. Journal chology, 32, 153-190.
of Experimental Psychology: Human Perception and Perfor¬ Dallos, P., Santos-Sacchi, J., & Flock, A. (1982). Intracellular
mance, 7, 869-883. recordings from cochlear outer hair cells. Science, 18, 582-
Cunningham, W. (1980). Speed, age and qualitative differences 584.
546 References

Dalton, K. (1964). The premenstrual syndrome. Springfield, IL: Deutsch, D. (1975). Two channel listening to musical scales.
Thomas. Journal of the Acoustical Society of America, 57, 1156—
Dalziel, C. C., & Egan, D. J. (1982). Crystalline lens thickness 1160.
changes as observed by pachometry. American Journal of Deutsch, D. (1978). The psychology of music. In E. C. Carter-
Optometry & Physiological Optics, 59, 442-447. ette & M. P. Friedman (Eds.), Handbook of perception:
Daniels, J. D., Pettigrew, J. D., & Norman, J. L. (1978). De¬ Vol. 10. Perceptual Ecology (pp. 191-224). New York: Ac¬
velopment of single-neuron responses in kittens’ lateral ge¬ ademic Press.
niculate nucleus. Journal of Neurophysiology, 41, 1373- Deutsch, D. (Ed.). (1982). The psychology of music. New
1393. York: Academic Press.
Darian-Smith, I., Sugitani, M., Heywood, J., Karita, K., & Deutsch, D. (1986). A musical paradox. Music Perception, 3,
Goodwin, A. (1982). Touching textured surfaces: Cells in 275-280.
somatosensory cortex respond both to finger movement and Deutsch, D., & Feroe, J. (1981). The internal representation of
to surface features. Science, 218, 906-909. pitch sequences in tonal music. Psychological Review, 88,
Dark, V., Johnston, W., Myles-Worsley, M., & Farah, M. 503-522.
(1985). Levels of selection and capacity limits. Journal of Deutsch, J. A., & Deutsch, D. (1963). Attention: Some theo¬
Experimental Psychology: General, 114, 472-497. retical considerations. Psychological Review, 70, 80-90.
Dartnall, H. M. A. (1957). The visual pigments. London: Me¬ DeValois, R. L., Albrecht, D. G., & Thorell, L. G. (1982).
thuen. Spatial frequency selectivity of cells in the macaque visual
Davidoff, J. B. (1975). Differences in visual perception: The cortex. Vision Research, 22, 545-559.
individual eye. New York: Academic Press. DeValois, R. L., & DeValois, K. K. (1975). Neural coding of
Dawson, J. L. (1967). Cultural and physiological influences color. In E. C. Carterette & M. P. Friedman (Eds.), Hand¬
upon spatial processes in West Africa. I. International Jour¬ book of perception: Vol. V. Seeing (pp. 117-168). New
nal of Psychology, 2, 115-128. York: Academic Press.
Day, M. (1975). Developmental trends in visual scanning. In DeValois, R. L., & DeValois, K. K. (1980). Spatial vision.
H. W. Reese (Ed.), Advances in child development & be¬ Annual Review of Psychology, 31, 309-341.
havior: Vol. 10 (pp. 154-193). New York: Academic Press. DeValois, R. L., & DeValois, K. K. (1987). Spatial vision.
Day, M. C., & Stone, C. A. (1980). Childrens’ use of percep¬ New York: Oxford University Press.
tual set. Journal of Experimental Child Psychology, 29, DeValois, R. L., & Jacobs, G. H. (1984). Neural mechanisms
428-445. of color vision. In I. Darian-Smith (Ed.), Handbook of phys¬
Day, R. H., Stuart, G. W., & Dickinson, R. G. (1980). Size iology: Vol. III. The nervous system (pp. 425-465). Balti¬
constancy does not fail below half a degree. Perception & more: Williams & Wilkins.
Psychophysics, 28, 263-265. DeValois, R. L., Yund, E. W., & Hepler, N. (1982). The ori¬
Day, R. S. (1968) Fusion in dichotic listening. Unpublished entation and direction selectivity of cells in macaque visual
doctoral dissertation, Stanford University. cortex. Visual Research, 22, 531-544.
Day, R. S. (1970). Temporal order judgments in speech: Are Devaney, K. O., & Johnson, H. A. (1980). Neuron loss in the
individuals language-bound or stimulus-bound? Haskins aging visual cortex of man. Journal of Gerontology, 35,
Laboratories Status Report, SR-2II22, 71-87. 836-841.
Delk, J. L., & Fillenbaum, S. (1965). Differences in perceived De Vries, H., & Stuiver, M. (1961). The absolute sensitivity of
color as a function of characteristic color. American Journal the human sense of smell. In W. A. Rosenblith (Ed.), Com¬
of Psychology, 78, 290-293. munication processes (pp. 159-167). New York: MIT
Delorme, A., & Martin, C. (1986). Roles of retinal periphery Press.
and depth periphery in linear vection and visual control of De Vries, J. V. (1968). Perspective. New York: Dover. (Orig¬
standing in humans. Canadian Journal of Psychology, 40, inal work published 1604)
176-187. Diamant, H., Funakoshi, M., Strom, L., & Zotterman, Y.
De Monasterio, F. M. (1978). Center and surround mechanisms (1963). Electrophysiological studies on human taste nerves.
of opponent-color X and Y ganglion cells of retina of ma¬ In Y. Zotterman (Ed.), Olfaction and taste (pp. 191-203).
caques. Journal of Neurophysiology, 41, 1418-1434. Oxford: Pergamon.
Deregowski, J. (1980). Illusions, patterns and pictures: A Diamant, H., & Zotterman, Y. (1969). A comparative study on
cross-cultural perspective. London: Academic Press. the neural and psychophysical response to taste stimuli. In
Derrington, A. M., & Fuchs, A. F. (1981). The development C. Pfaffman (Ed.), Olfaction and taste III (pp. 428-435).
of spatial-frequency selectivity in kitten striate cortex. Jour¬ New York: Rockefeller University Press.
nal of Physiology, 316, 1-10. Dichgans, J., & Brandt, T. (1978). Visual-vestibular interac¬
Desimone, R., Albright, T. D., Gross, C. G., & Bruce, C. tion: Effects on self-motion perception and postural control.
(1980). Responses of inferior temporal neurons to complex In R. Held, H. W. Leibowitz, & H. L. Teuber (Eds.),
visual stimuli. Society of Neurosciences: Abstracts, 6, 581. Handbook of sensory physiology: Vol. 7. Perception (pp.
Desimone, R., & Gross, C. G. (1979). Visual areas in the tem¬ 755-804). New York: Springer-Verlag.
poral cortex of the macaque. Brain Research, 178, 363- Dickhaus, H., Pauser, G., & Zimmerman, M. (1985). Tonic
380. descending inhibition affects intensity coding of nociceptive
 References 547

responses of spinal dorsal hom neurons in the cat. Pain, 23, (1978). Communication of gender from human axillary
145-158. odors: Relationship to perceived intensity and hedonicity.
Diehl, R. L. (1981). Feature detectors for speech: A Critical Behavioral Biology, 23, 373-380.
reappraisal. Psychological Bulletin, 89, 1-18. Dowling, W. J., & Carterette, E. C. (Eds.). (1987). The under¬
Dineen, I. T., & Meyer, W. J. (1980). Developmental changes standing of melody and rhythm. Special Issue of Perception
in visual orienting behavior to featural versus structural in¬ & Psychophysics, 41, 482-656.
formation in the human infant. Developmental Psychology, Dowling, W. J., & Harwood, D. L. (1986). Music cognition.
13, 123-130. Orlando, FL: Academic Press.
Ditchbum, R. W. (1973). Eye movements and perception. Ox¬ Droscher, V. B. (1971). The magic of the senses: New discov¬
ford: Clarendon. eries in animal perception. New York: Harper.
DiZio, P. A., & Lackner, J. R. (1986). Perceived orientation, Drum, B. (1980). Relation of brightness to threshold for light-
motion, and configuration of the body during viewing of an adapted and dark-adapted rods and cones: Effects of retinal
off-vertical rotating surface. Perception & Psychophysics, eccentricity and target size. Perception, 9, 633-650.
39, 39-46. Drum, B. (1981). Brightness interactions between rods and
Djang, S. (1937). The role of past experience in the visual ap¬ cones. Perception & Psychophysics, 29, 505-510.
prehension of masked forms. Journal of Experimental Psy¬ Duncan H. F., Gourlay, N., & Hudson, W. (1973). A study of
chology, 20, 29-59. pictorial perception among the Bantu and white primary
Dobson, V. (1976). Spectral sensitivity of the 2-month-old in¬ school children in South Africa. Johannesberg: Witwaters-
fant as measured by the visual evoked cortical potential. Vi¬ rand University Press.
sion Research, 16, 367-374. Duncker, K. (1929). Uber induzierte Bewegung (ein Beitrag zur
Dobson, V., & Teller, D. Y. (1978). Visual acuity in human Theorie optisch warigenommener Bewegung). Psycholo-
infants: A review and comparison of behavioral and electro- gische Forschung, 2, 180-259.
physiological studies. Vision Research, 18, 1469-1483. Durlach, N. I., & Braida, L. D. (1969). Intensity perception. I.
Dodwell, P. C. (1971). On perceptual clarity. Psychological Re¬ Preliminary theory of intensity resolution. Journal of the
view, 78, 275-279. Acoustical Society of America, 46, 372-383.
Dodwell, P. C., Humphrey, G. K., & Muir, D. W. (1987).
Shape and pattern perception. In P. Salapatek & L. Cohen Easterbrook, J. A. (1959). The effect of emotion on cue utili¬
(Eds.), Handbook of infant perception: Vol. 2. From per¬ zation and the organization of behavior. Psychological Re¬
ception to cognition (pp. 1-80). Orlando. FL: Academic view, 66, 183-201.
Press. Easton, R. D., & Basala, M. (1982). Perceptual dominance dur¬
Doetsch, G. S., Ganchrow, J. J., Nelson, L. M., & Erickson, ing lipreading. Perception & Psychophysics, 32, 562-570.
R. P. (1969). Information processing in the taste system of Edwards, A., & Cohen, S. (1961). Visual illusions, tactile sen¬
the rat. In C. Pfaffman (Ed.), Olfaction and taste 111 (pp. sibility and reaction time under LSD-25. Psychopharmacol-
492-511). New York: Rockefeller University Press. ogia, 2, 297-303.
Doherty, M. E., & Keeley, S. M. (1972). On the identification Efron, R. (1967). The duration of the present. Annals of the
of repeatedly presented visual stimuli. Psychological Bulle¬ New York Academy of Sciences, 138, 713-729.
tin, 78, 142-154. Efron, R. (1973). Conservation of temporal information by per¬
Donaldson, I. M. L., & Long, A. C. (1980). Interactions be¬ ceptual systems. Perception & Psychophysics, 14, 518-530.
tween extraocular proprioceptive and visual signals in the Egan, J. P. (1975). Signal detection theory and ROC-analysis.
superior colliculus of the cat. Journal of Physiology, 298, New York: Academic Press.
85-110. Egeth, H. E., Virgi, R. A., & Garbart, H. (1984). Searching
Donchin, E. (1981). Surprise! .... Surprise? Psychophysiol¬ for conjunctively defined targets. Journal of Experimental
ogy, 18, 493-513. Psychology: Human Perception and Performance, 10, 32-
Doner, J., Lappin, J. S. & Perfetto, G. (1984). Detection of 39.
three-dimensional structure in moving optical patterns. Jour¬ Eichengreen, J. M., Coren, S., & Nachmias, J. (1966). Visual-
nal of Experimental Psychology: Human Perception and cliff preference by infant rats: Effects of rearing and test
Performance, 10, 1-11. conditions. Science, 151, 830-831.
Doty, R. L. (1985). The primates III: Humans. In R. E. Brown Eijkman, E. G. J., Jongsma, H. J., & Vincent, J. (1981). Two
& D. W. MacDonald (Eds.), Social odours in mammals: dimensional filtering oriented line detectors and figural as¬
Vol. 2. (pp. 804-832). Oxford: Clarendon. pects as determinants of visual illusions. Perception & Psy¬
Doty, R. L., Applebaum, S., Zusho, H., & Settle, R. G. chophysics, 29, 352-358.
(1985). Sex differences in odor identification ability: A Eilers, R., Wilson, W., & Moore, T. (1979). Speech perception
cross-cultural analysis. Neuropsychologia, 23, 667-672. in the language innocent and the language wise: A study in
Doty, R. L., Green, P. A., Ram, C., & Yankell, S. L. (1982). the perception of voice onset time. Journal of Child Lan¬
Communication of gender from human breath odors: Rela¬ guage, 6, 1-18.
tionship to perceived intensity and pleasantness. Hormones Eimas, P. D., & Corbit. J. D. (1973). Selective adaptation of
and Behavior, 16, 13-22. linguistic feature detectors. Cognitive Psychology, 4, 99-
Doty, R. L., Kligman, A., Leyden, J., & Omdorff, M. M. 109.
548 References

Eimas, P. D., & Miller, J. D. (1980). Contextual effects in Epstein, W., & Hanson, S. (1977). Discrimination of unique
infant speech perception. Science, 209, 1140-1141 motion-path length. Perception & Psychophysics, 22, 152—
Eimas, P. D., Siqueland, E. R., Jusczyk, P., & Vigorito, J. 158.
(1971). Speech perception in infants. Science, 171, 303- Epstein, W., Hatfield, G., & Muise, G. (1977). Perceived
306. shape at a slant as a function of processing time and pro¬
Einstein, A., & Besso, M. (1972). Correspondence 1903-1955. cessing load. Journal of Experimental Psychology: Human
Paris: Hermann. Perception and Performance, 3, 473-483.
Eland, J. M., & Anderson, J. E. (1977). The experience of pain Epstein, W., & Lovitts, B. E., (1985). Automatic and atten-
in children. In A. Jacox (Ed.), Pain: A sourcebook for tional components in perception of shape-at-a-slant. Journal
nurses and other professionals. Boston: Little, Brown. of Experimental Psychology: Human Perception and Perfor¬
Elkind, D. (1978). The child’s reality: Three developmental mance, 11, 355-366.
themes. Hillsdale, NJ: Lawrence Erlbaum. Epstein, W., & Park, J. N. (1964). Shape constancy: Functional
Ellingson, R. (1968). Clinical applications of evoked potential relationships and theoretical formulations. Psychological
techniques in infants and children. Electroencephalography Bulletin, 62, 180-196.
and Clinical Neurophysiology, 24, 293. Erickson, R. P. (1963). Sensory neural patterns and gustation.
Ellingson, R., Lathrop, G., Nelson, G., & Donahy, T. (1972). In Y. Zotterman (Ed.), Olfaction and taste (pp. 205-213).
Visual evoked potentials of infants. Reveue d'Electrocence- Oxford: Pergamon.
phalographie et de Neurophysiologie Clinique, 2, 395-400. Erickson, R. P. (1985). Definitions: A matter of taste. In D. W.
Ellis, H. (1905). Sexual selection in man. New York: Davis. Pfaff (Ed.), Taste, olfaction, and the central nervous system
Ellis, H. C., & Muller, D. G. (1964). Transfer in perceptual (pp. 129-150). New York: Rockefeller University Press.
learning following stimulus predifferentiation. Journal of Erickson, R. P, & Covey, E. (1980). On the singularity of taste
Experimental Psychology, 68, 388-395. sensations: What is a taste primary? Physiology & Behavior.
Emmerson, P. G., & Ross, H. E. (1986). The effect of bright¬ 25, 527-533.
ness on colour recognition under water. Ergonomics, 29, Erickson. R. P., & Schiffman, S. S. (1975). The chemical sen¬
1647-1658. ses: A systematic approach. In M. S. Gazzaniga & C.
Emmerton, J. (1983). Pattern discrimination in the near-ultravi¬ Blakemore (Eds.), Handbook of psychobiology (pp. 393-
olet by pigeons. Perception & Psychophysics, 34, 555-559. 425). New York: Academic Press.
Engen, T. (1982). Perception of odors. New York: Academic Eriksen, C. W., & Collins, J. F. (1968). Sensory traces versus
Press. the psychological movement in the temporal organization of
Engen, T. (1987). Remembering odors and their names. Amer¬ form. Journal of Experimental Psychology, 77, 376-382.
ican Scientist, 75, 497-503. Eriksen, C. W., & Hake, H. W. (1955). Absolute judgments as
Engen, T., Lipsitt, L. P., & Kaye. H. (1963). Olfactory re¬ a function of stimulus range and number of stimulus and
sponses and adaptation in the human neonate. Journal of responses categories. Journal of Experimental Psychology,
Comparative and Physiological Psychology, 56, 73-77. 49, 323-332.
Engen, T., & Tulunary, U. (1956). Some sources of error in Eriksen, C. W., & Hoffman, J. E. (1972). Some characteristics
half-heaviness judgments. Journal of Experimental Psychol¬ of selective attention in visual perception determined by vo¬
ogy, 54, 208-212. cal reaction time. Perception & Psychophysics, 11, 169—
Enns, J. T. (1986). Seeing textons in context. Perception & 171.
Psychophysics, 39, 143-147. Eriksen, C. W., & Murphy, T. D. (1987). Movement of atten-
Enns, J. T. (1987). A developmental look at pattern symmetry tional focus across the visual field: A critical look at the
in perception and memory. Developmental Psychology, 23, evidence. Perception & Psychophysics, 42, 299-305.
839-850. Eriksen, C. W., & St. James, J. D. (1986). Visual attention
Enns, J. T., & Cameron, S. (1987). Selective attention in young within and around the field of focal attention: A zoom lens
children: The relations between visual search, filtering and model. Perception & Psychophysics, 40, 225-240.
priming. Journal of Experimental Child Psychology, 44, Eriksen, C. W., & Yeh, Y. (1985). Allocation of attention in
38-63. the visual field. Journal of Experimental Psychology: Hu¬
Enns, J. T., & Girgus, J. S. (1985). Perceptual grouping and man Perception and Performance, 11, 583—597.
spatial distortion: A developmental study. Developmental Ernst, M., Lee, M. H. M., Dworkin, B., & Zaretsky, H. H.
Psychology, 21, 241-246. (1986). Pain perception decrement produced through re¬
Epstein, W. (1973). The process of taking into account in visual peated stimulation. Pain, 26, 221-231.
perception. Perception, 2, 267-285. Erulkar, S. C. (1972). Comparative aspects of spatial localiza¬
Epstein, W., & Baratz, S. S. (1964). Relative size in isolation tion of sound. Physiological Review, 52, 237-360.
as a stimulus for relative perceived distance. Journal of Ex¬ Essen, D. C. van. (1979). Visual areas of the mammalian ce¬
perimental Psychology, 67, 507-513. rebral cortex. Annual Review of Neurosciences, 2, 227-263.
Epstein, W., & Broota, K. D. (1986). Automatic and atten- Essen, D. C. van. (1984). Functional organization of primate
tional components in perception of size-at-a-distance. Per¬ visual cortex. In A. Peters & E. G. Jones (Eds.), Cerebral
ception & Psychophysics, 40, 256-262. Cortex: Vol. 3 (pp. 259-329). New York: Plenum.
 References 549

Evans, C. R., & Wells, A. M. (1967). Fragmentation phenom¬ Fender, D. H. (1971). Time delays in the human eye-tracking
ena associated with binocular stabilization. British Journal system. In P. Bach-y-Rita, C. C. Collins, & J. E. Hyde
of Physiological Optics, 24, 45-50. (Eds.), The control of eye movements (pp. 539-543). New
Evans, E. F. (1975). Cochlear nerve and cochlear nucleus. In York: Academic Press.
W. D. Keidel & W. D. Neff (Eds.), Handbook of sensory Festinger, L., Allyn, M. R., & White, C. W. (1971). The per¬
physiology: Vol. 2. Auditory system: Physiology (CNS). Be¬ ception of color with achromatic stimulation. Vision Re¬
havioral Studies. Psychoacoustics (pp. 1-108). New York: search, 11, 591-612.
Springer-Verlag. Festinger, L., Coren, S., & Rivers, G. (1970). The effect of
Evans, P. M. (1987). Vibrotactile masking: Temporal integra¬ attention on brightness contrast and assimilation. American
tion, persistence, and strengths of representation. Perception Journal of Psychology, 83, 189-207.
& Psychophysics, 42, 515-525. Festinger, L., & Easton, M. (1974). Inference about the efferent
Evans, P. M., & Craig, J. C. (1986). Temporal integration and system based on a perceptual illusion produced by eye
vibrotactile backward masking. Journal of Experimental movements. Psychological Review, 81, 44-58.
Psychology: Human Perception and Performance, 12, 160- Festinger, L., White, C. W., & Allyn, M. R. (1968). Eye
168. movements and decrement in the Mueller-Lyer illusion. Per¬
Eysenck. H. J. (1967). The biological basis of personality. ception & Psychophysics, 3, 376-382.
Springfield, IL: Thomas. Filer, R., & Meals, D. (1949). The effect of motivating condi¬
tions on the estimation of time. Journal of Experimental
Falk, J., & Bindra, D. (1954). Judgment of time as a function Psychology, 39, 327-331.
of serial position and stress. Journal of Experimental Psy¬ Filsinger, E. E., & Fabes, R. A. (1985). Odor communication,
chology, 47, 279-282. pheromones, and human families. Journal of Marriage and
Falmagne, J. C. (1974). Foundations of Fechnerian psychophys¬ the Family, 47, 349-360.
ics. In D. Krantz. R. C. Atkinson, R. D. Luce, & P. Suppes Findlay, J. (1981). Local and global influences on saccadic eye
(Eds.), Contemporary developments in mathematical psy¬ movements. In D. Fisher. R. Monty, & J. Senders (Eds.),
chology (Vol. 2, pp. 121-159). San Francisco: Freeman. Eye movements: Cognition and visual perception (pp. 171 —
Falmagne, J. C. (1985). Elements of psychophysical theory. 179). Hillsdale, NJ: Lawrence Erlbaum.
New York: Oxford University Press. Fisher, D. L., Duffy, S. A., Young, C., & Pollatsek, A.
Fant, G. (1967). Auditory patterns of speech. In W. Wathen- (1988). Understanding the central processing limit in consis¬
Dunn (Ed.), Models for the perception of speech and visual tent-mapping visual search tasks. Journal of Experimental
form (pp. 111-125). Cambridge, MA: MIT Press. Psychology: Human Perception and Performance, 14, 253-
Fantz, R. L. (1961). The origin of form perception. Scientific 266.
American, 204 , 66-72. Fisher, R. (1967). The biological fabric of time. In Interdisci¬
Fantz, R. L. (1965). Ontogeny of perception. In A. M. Schrier, plinary perspectives of time. Annals of the New York Acad¬
H. F. Harlow, & F. Stollnitz (Eds.), Behavior of nonhuman emy of Sciences, 138, 451-465.
primates (Vol. 2, pp. 365-403). New York: Academic Fisk, A. D., & Schneider, W. (1981). Control and automatic
Press. processing during tasks requiring sustained attention: A new
Fantz, R. L., & Miranda, S. B. (1977). Visual processing in approach to vigilance. Human Factors, 23, 737-750.
the newborn preterm, and mentally high-risk infant. In L. Fitzpatrick, V., Pasnak, R., & Tyer, Z. E. (1982). The effect
Gluck (Ed.), Intrauterine asphyxia and the developing fetal of familiar size at familiar distances. Perception, 11, 85-
brain (pp. 453-471). Chicago: Year Book Medical Publish¬ 91.
ers. Flandrin, J. M., & Jeannerod, M. (1981). Effects of unilateral
Fantz, R. L., & Yeh, J. (1979). Configurational selectives: Crit¬ superior colliculus ablation on oculomotor and vestibulo-
ical for development of visual perception and attention. Ca¬ ocular responses in the cat. Experimental Brain Research,
nadian Journal of Psychology, 33, 277-287. 42, 73-80.
Farkas, M., & Hoyer, W. (1980). Processing consequences of Flannery, R., & Butler, R. A. (1981). Spectral cues provided
perceptual grouping in selective attention. Journal of Ger¬ in the pinna for monaural localization in the horizontal
ontology, 35, 27-216. plane. Perception & Psychophysics, 29, 438-444.
Farrell, J. E. (1983). Visual transformations underlying appar¬ Flin, R. H. (1980). Age effects in children’s memory for unfa¬
ent movement. Perception & Psychophysics, 33, 85-92. miliar faces. Developmental Psychology, 16, 373-374.
Faurion, A., Saito, S., & MacLeod, P. (1980). Sweet taste in¬ Flock, H., & Freedberg, E. (1970). Perceived angle of inci¬
volves several distinct receptor mechanisms. Chemical Sen¬ dence and achromatic surface color. Perception & Psycho¬
ses, 5, 107-121. physics, 8, 251-256.
Favreau, O. E., & Cavanagh, P. (1981). Color and luminance: Flock, H. R., & Nusinowitz, S. (1984). Visual structures for
Independent frequency shifts. Science, 212, 831-832. achromatic color perceptions. Perception & Psychophysics,
Fechner, G. T. (1966). Elements of psychophysics. (H. E. 36, 111-130.
Alder, Trans.). New York: Holt, Rinehart & Winston. Flom, M. C., Brown, B. Adams, A. J., & Jones, R. T. (1976).
(Original work published 1860) Alcohol and marihuana effects on ocular tracking. American
 550 References

Journal of Optometry and Physiological Optics, 53, 764- Fisher, R. Monty, & J. Senders (Eds.), Eye movements:
773. Cognition and visual perception (pp. 137—155). Hillsdale,
Florentine, M. (1986). Level discrimination of tones as a func¬ NJ: Lawrence Erlbaum.
tion of duration. Journal of the Acoustical Society of Amer¬ Friedman, R. B. (1980). Identity without form: Abstract repre¬
ica, 79, 792-798. sentations of letters. Perception & Psychophysics, 28, 53-
Fodor, J. (1983). The modularity of mind. Cambridge, MA: 60.
MIT Press. Friedman, S. L., & Stevenson, M. (1975). Developmental
Foley, J. E. (1970). Prism adaptation with opposed base orien¬ changes in the understanding of implied motion in two-di¬
tation: The weighting of direction information from the two mensional pictures. Child Development, 46, 773-778.
eyes. Perception & Psychophysics, 8, 23-25. Friedman, S. L., & Stevenson, M. (1980). Perception of move¬
Foley, J. E. (1974). Factors governing interocular transfer of ments in pictures. In M. Hagen (Ed.), Perception of pic¬
prism adaptation. Psychological Review, 81, 183-186. tures: Vol. 1. Alberti’s Window: The projective model of
Foulke, E., & Sticht, T. (1969). Review of research on the in¬ pictural information (pp. 225-255). New York: Academic
telligibility and comprehension of accelerated speech. Psy¬ Press.
chological Bulletin, 72 , 50-62. Frisby, J. P. (1980). Seeing: Illusion, brain and mind. Oxford:
Fox, R., Aslin, R. N., Shea, S. L., & Dumais, S. T. (1980). Oxford University Press.
Stereopsis in infants. Science, 207, 323-324. Fuld, K., Wooten, B. R., & Whalen, J. J. (1981). The elemen¬
Fox, R., & McDaniel, C. (1982). The perception of biological tal hues of short-wave and extraspectral lights. Perception &
motion by human infants. Science, 486-487. Psychophysics, 29, 317-322.
Fozard, J., Wolf, E., Bell, B., McFarland, R., & Podolsky, S. Fuller, C. A., Lydic, R., Sulzman, F. M., Albers, H. E., Tep-
(1977). Visual perception and communication. In. J. Birren per, B., & Moore-Ede, M. C. (1981). Circadian rhythm of
& K. Schaie (Eds.), Handbook of the psychology of aging. body temperature persists after suprachiamatic lesions in the
New York: Van Nostrand Reinhold. squirrel monkey. American Journal of Physiology, 241,
Fraisse, P. (1963). The psychology of time. New York: Harper R385-R391.
& Row. Fulton. J. F., & Bailey, P. (1929). Tumors in the region of the
Frank, M. (1975). Response patterns of rat glossopharyngeal third ventricle: Their diagnosis and relation to pathological
taste neurons. In D. A. Denton & P. Coghlan (Eds.), Olfac¬ sleep. Journal of Nen'ous and Mental Disorders, 69, 1-25,
tion and taste V (pp. 59-64). New York: Academic Press. 145-164, 261-277.
Frank, M. E. (1985). On the neural code for sweet and salty Funakoshi, M., Kasahara, Y., Yamamoto, T., & Kawamura,
taste. In D. W. Pfaff (Ed.), Taste, olfaction, and the central Y. (1972). Taste coding and central perception. In D.
nervous system (pp. 107-128). New York: Rockefeller Uni¬ Schneider (Ed.), Olfaction and taste IV (pp. 336-342).
versity Press. Stuttgart: Wissenshaftliche Verlagsgesellschaft MBH.
Frankenhauser, M. (1959). The estimation of time. Stockholm: Funkenstein, H. H., Nelson, P. G., Winter, P. L., Wolberg.
Almqvist & Wiksell. Z. , & Newman, J. D. (1971). Unit responses in auditory
Freeman, R., Mallach, R., & Hartley, S. (1981). Responsivity cortex of awake squirrel monkeys to vocal stimulation. In
of normal kitten striate cortex deteriorates after brief binoc¬ M. B. Saschs (Ed.), Physiology of the auditory system (pp.
ular deprivation. Journal of Neurophysiology, 45, 1074- 307-326). Baltimore: National Educational Consultants.
1084.
Freeman, R., & Pettigrew, J. (1973). Alteration of visual cortex
from environmental asymmetries. Nature, 246, 359-360. Gaddes, W. H. (1976). Prevalence estimates and the need for
Fregnac, V., & Imbert, M. (1978). Early development of visual definition of learning disabilities. In R. M. Knights & D. J.
cortical cells in normal and dark reared kittens: Relation¬ Bakker (Eds.), The neuropsychology of learning disorders.
ships between orientation selectivity and ocular dominance. Baltimore: University Park Press.
Journal of Physiology, 278, 27-44. Gaddes, W. H. (1980). Learning disabilities and brain function:
Freud, S. (1953). An aphasia. London: Imago. A neurophysiological approach. New York: Springer-Ver-
Frey, M. von., & Goldman, A. (1915). Der zeitliche Verlauf lag.
det Einstellung bei den Druckempfindungen. Zeitschrift feur Galanter, E. (1962). Contemporary psychophysics. In R.
Biologie, 65, 183-202. Brown. E. Galanter, E. Hess, & G. Mandler (Eds.), New
Frey, M. von., & Kiesow, F. (1899). Uber die function der directions in psychology (pp. 87-157). New York: Holt,
tastkorperchen. Zeitschrift feur Psychologie, 20, 126-163. Rinehart & Winston.
Freyd, J., & Tversky, B. (1984). Force of symmetry in form Ganchrow, J. R., Steiner, J. E., & Daher, M. (1983). Neonatal
perception. American Journal of Psychology, 97, 109-126. facial expressions in response to different qualities and in¬
Friedman, A. (1979). Framing pictures: The role of knowledge tensities of gustatory stimuli. Infant Behavior and Develop¬
in automatized encoding and memory for gist. Journal of ment, 6, 189-200.
Experimental Psychology: General, 108, 316-355. Gardner, E. B., & Costanzo, R. H. (1981). Properties of kin¬
Friedman. A., & Liebelt, L. (1981). On the time course of esthetic neurons in somatosensory cortex of awake monkeys.
viewing pictures with a view towards remembering. In D. Brain Research, 214, 301-319.
 References 551

Gardner. E. P. (1983). Cortical neuronal mechanisms underly¬ Gibson, J. J. (1950). Perception of the visual world. Boston:
ing the perception of motion across the skin. In C. von Eu¬ Houghton Mifflin.
ler, O. Franzen, U. Lindblom, & D. Ottoson (Eds.), Soma¬ Gibson, J. J. (1966). The senses considered as perceptual sys¬
tosensory mechanisms (pp. 93-112). New York: Plenum. tems. Boston: Houghton Mifflin.
Gardner, R. J. (1979). Lipophilicity and the perception of bit¬ Gibson, J. J. (1979). The ecological approach to visual percep¬
terness. Chemical Senses and Flavor, 4, 275-286. tion. Boston: Houghton Mifflin.
Gamer, W. R. (1953). An informational analysis of absolute Gibson, R. H., & Tomko, D. L. (1972). The relation between
judgments of loudness. Journal of Experimental Psychology, category and magnitude estimates of tactile intensity. Per¬
46, 373-380. ception & Psychophysics, 12, 135-138.
Gamer, W. R. (1962). Uncertainty and structure as psycholog¬ Gilbert, A. N., & Wysocki, C. J. (1987). The smell survey:
ical concepts. New York: Wiley. Results. National Geographic, 172, 514-525.
Gamer, W. R. (1974). The processing of information and struc¬ Gilchrist, A. L. (1980). When does perceived lightness depend
ture. Potomac, MD: Lawrence Erlbaum. on perceived spatial arrangement? Perception & Psycho¬
Gamer, W. R. (1978). Aspects of a stimulus: Features, dimen¬ physics, 28, 527-538.
sions and configurations. En E. H. Rosch & B. B. Lloyd Gilchrist, A. L., Delman, S., & Jacobsen, A. (1983). The clas¬
(Eds.), Cognition and categorization (pp. 99-139). Hills¬ sification and integration of edges as critical to the percep¬
dale, NJ: Lawrence Erlbaum. tion of reflectance and illumination. Perception & Psycho¬
Gamer, W. R., & Clement. D. E. (1963). Goodness of pattern physics, 33, 425-436.
and pattern uncertainty. Journal of Verbal Learning and Gilchrist, A. L., & Jacobsen, A. (1984). Perception of lightness
Verbal Behavior, 2, 446-452. and illumination in a world of one reflectance. Perception,
Gamer, W. R., & Hake, H. W. (1951). The amount of infor¬ 13, 5-19.
mation in absolute judgments. Psychological Review, 58, Gilinsky, A. S. (1955). The effect of attitude upon the percep¬
446-459. tion of size. American Journal of Psychology, 68, 173-192.
Geiselman, R. E., Fisher, R. P., MacKinnon, D. P., & Hol¬ Gillam. B. (1980). Geometrical illusions. Scientific American,
land, H. L. (1986). Enhancement of eyewitness memory 242, 102-111.
with the cognitive interview. American Journal of Psychol¬ Gintzler, A. R. (1980). Endorphin-mediated increases in pain
ogy, 99, 385-401. threshold during pregnancy. Science, 210, 193-195.
Gelb, A. (1929). Die “Farbenkonstanz” der sehdinge. Hand- Girgus, J. S., & Coren, S. (1987). The interaction between
buch der normalen und pathologische Phvsiologie, 12, 549- stimulus variations and age trends in the Poggendorff illu¬
678. sion. Perception & Psychophysics. 41, 60-66.
Geldard, F. A. (1972). The human senses (2nd ed.). New York: Glass, A. L., & Holyoak, K. J. (1986). Cognition. New York:
Wiley. Random House.
Gent, J. F. (1979). An exponential model for adaptation in Gliner, J. A., Horvath, S. M., & Mihevic, P. M. (1983). Car¬
taste. Sensory Processes, 3, 303-316. bon monoxide and human performance in a single and dual
Gerbrandt, L. K., Spinelli, D. N., & Pribram, K. H. (1970). task methodology. Aviation, Space and Environmental Med¬
Interaction of visual attention and temporal cortex stimula¬ icine, 54, 714-717.
tion on electrical activity evoked in striate cortex. Electro¬ Glorig, A., Wheeler, D., Quigle, R., Grings, W., & Summer-
encephalography and Clinical Neurology, 29, 146. field, A. (1970). 1954 Wisconsin State Fair hearing survey:
Gescheider, G. A.. & Verrillo, R. T. (1982). Contralateral en¬ Statistical treatment of clinical and audiometric data. Cited
hancement and suppression of vibrotactile sensation. Per¬ in D. D. Kryter, The effects of noise on man (p. 116). New
ception & Psychophysics, 32, 69-74. York: Academic Press.
Gesteland, R. C. (1986). Speculations on receptor cells as ana¬ Glucksberg, S., & Cowen, G. N., Jr. (1970). Memory for non-
lyzers and filters. Experientia, 42, 287-291. attended auditory material. Cognitive Psychology, 1, 149—
Gesteland, R. C., Lettvin, J. Y., Pitts, W. H., & Rojas, A. 156.
(1963). Odor specificities of the frog’s olfactory receptors. Gogel, W. C., & DaSilva, J. A. (1987a). A two-process theory
In Y. Zotterman (Ed.), Olfaction and taste (pp. 19-34). Ox¬ of the response to size and distance. Perception & Psycho¬
ford: Pergamon. physics, 41, 220-238.
Getchell, T. V., & Getchell, M. L. (1987). Peripheral mecha¬ Gogel, W. C., & DaSilva, J. A. (1987b). Familiar size and the
nisms of olfaction: Biochemistry and neurophysiology. In theory of off-sized perceptions. Perception & Psychophys¬
T. E. Finger & W. L. Silver (Eds.), Neurobiology of taste ics, 41, 318-328.
and smell (pp. 91-124). New York: Wiley. Gogel, W. C., Gregg, J. M., & Wainwright, A. (1961). Con¬
Giachetti, I., & MacLeod, P. (1975). Cortical neuron responses vergence as a cue to absolute distance. (Report No. 467,
to odours in the rat. In D. A. Denton & J. P. Coghlan pp. 1-16). Fort Knox, KY: U.S. Army Medical Research
(Eds.), Olfaction and taste V (pp. 303-307). New York: Laboratory.
Academic Press. Gogel, W. C., & Koslow, M. (1972). The adjacency principle
Gibbons, B. (1986). The intimate sense of smell. National Geo¬ and induced movement. Perception & Psychophysics, 11,
graphic, 170, 324-361. 309-324.
552 References

Goldfoot, D. A. (1981). Olfaction, sexual behavior and the Gouras, P., & Zrenner, E. (1981). Color coding in primate ret¬
pheromone hypothesis in the rhesus monkey: A critique. ina. Vision Research, 21, 1591-1598.
American Zoologist, 21, 153-164. Graham, C. H. (1965). Visual space perception. In C. H. Gra¬
Goldfoot, D. A., Essock-Vitale, S. M., Asa, C. S., Thornton, ham (Ed.), Vision and visual perception (pp. 504-547).
J. E., & Leshner, A. I. (1978). Anosmia in male rhesus New York: Wiley.
monkeys does not alter copulatory activity with cycling fe¬ Graham, C. H., & Hsia, Y. (1958). Color defect and color the¬
males. Science, 199, 1095-1096. ory. Science, 127, 657-682.
Goldfoot, D. A., Kravetz, M. A., Goy, R. W., & Freeman, S. Graham, N. (1980). Spatial-frequency channels in human vi¬
K. (1976). Lack of effect of vaginal lavages and aliphatic sion: Detecting edges without edge detectors. In C. S. Harris
acids on ejaculatory responses in Rhesus monkeys: Behav¬ (Ed.), Visual coding and adaptability (pp. 215-262). New
ioral and chemical analysis. Hormones and Behavior, 7, 1 — York: Lawrence Erlbaum.
27. Graham, N. (1981). Psychophysics of spatial-frequency chan¬
Goldstein, E. B. (1980). Sensation and Perception. Belmont, nels. In M. Kubovy & J. Pomerantz (Eds.), Perceptual or¬
CA: Wadsworth. ganization (pp. 1-26). Hillsdale, NJ: Lawrence Erlbaum.
Goldstein, J. L. (1973). An optimum processor theory for the Granger, G. W., & Ikeda, H. (1968). Drugs and visual thresh¬
central formation of the pitch of complex tones. Journal of olds. In A. Herxheimer (Ed.), Drugs and sensory functions
the Acoustical Society of America, 54, 1496-1516. (pp. 299-344). London: Churchill.
Goldstone, S., Boardman, W. K., & Lhamon, W. T. (1958). Granrud, C. E. (1986). Binocular vision and spatial perception
Effect of quinal barbitone dextro-amphetamine, and placebo in 4- and 5-month-old infants. Journal of Experimental Psy¬
on apparent time. British Journal of Psychology, 49, 324- chology: Human Perception and Performance, 12, 36-49.
328. Granrud, C. E., Haake. R. J., & Yonas, A. (1985). Infants’
Gombrich, E. H. (1972). The mask and the face: The perception sensitivity to familiar size: The effect of memory on spatial
of physiognomic likeness in life and in art. In E. H. Gom¬ perception. Perception & Psychophysics, 37, 459-466.
brich, J. Hochberg, & M. Black (Eds.), Art, perception and Granrud, C. E., Yonas, A., & Pettersen, L. (1984). A compar¬
reality (pp. 1-46). Baltimore: Johns Hopkins Press. ison of monocular and binocular depth perception in 5- and
Goodman, L., & Gilman, A. (Eds.), (1965). The pharmacolog¬ 7-month-old infants. Journal of Experimental Child Psy¬
ical basis of therapeutics. New York: Macmillan. chology, 38, 19-32.
Goodman, N. (1968). Languages of art. New York: Bobbs- Gravetter, F., & Lockhead, G. R., (1973). Criterial range as a
Merrill. frame of reference for stimulus judgment. Psychological Re¬
Goodson, R., & Rahe, A. (1981). Visual training effects on view, 80, 203-216.
normal vision. American Journal of Optometry and Physio¬ Green, B. G. (1984). Thermal perception on lingual and labial
logical Optics, 58, 787-791. skin. Perception & Psychophysics, 36, 209-220.
Goodwin, M., Gooding, K. M., & Regnier, F. (1979). Sex Green, B. G. (1985). Heat pain thresholds in the oral-facial re¬
pheromone in the dog. Science, 203, 559-561. gion. Perception & Psychophysics, 38, 110-114.
Goolkasian, P. (1980). Cyclic changes in pain perception: A Green, B. G.H1986). Oral perception of the temperature of liq¬
ROC analysis. Perception & Psychophysics, 27, 499-504. uids. Perception <£ Psychophysics, 39, 19-24.
Goto, H. (1971). Auditory perception by normal Japanese adults Green, B. G. (1987). The effect of cooling on the vibrotactile
of the sounds “L” or “R”. Neuropsychologia, 9, 317-323. sensitivity of the tongue. Perception & Psychophysics, 42,
Gottlieb, G. (1981). Roles of early experience in species-spe¬ 423-430.
cific perceptual development. In R. Aslin, J. Alberts, & M. Green, D. G., & Powers, M. K. (1982). Mechanisms of light
Petersen (Eds.), Development of perception: Psychobiologi- adaptation in rat retina. Vision Research, 22, 209-216.
cal perspectives: Vol. 1. Audition, somatic perception and Green, D. M. (1976). An introduction to hearing. New York:
the chemcial senses (pp. 5-44). New York: Academic Academic Press.
Press. Green, D. M. (1987). Profile analysis: Auditory intensity dis¬
Gottlieb, M. D., Kietzman, M. 1., & Bemhaus, I. J. (1985). crimination. New York: Oxford University Press.
Two-pulse measures of temporal integration in the fovea and Green, D. M., Nachmias, J., Kearny, J. K., & Jeffress, L. A.
peripheral retina. Perception & Psychophysics, 37, OS- (1979). Intensity discrimination with gated and continuous
OS. sinusoids. Journal of the Acoustical Society of America, 66.
Gottschaldt, K. (1926). Uber den Einfluss der Erfahrung auf die 1051-1056.
Wahmehmung von Figuren. I. Psychologische Forschung, Green, D. M., & Swets, J. A. (1966). Signal detection theory
8, 261-317. and psychophysics. Reprint. New York: Krieger, 1974.
Gottschaldt, K. (1929). Uber den Einfluss der Erfahrung auf die Green, D. W., Hammond, E. J., and Supramaniam, S. (1983).
Wahmehmung von Figuren. II. Psychologische Forschung, Letters and shapes: Developmental changes in search strate¬
12, 1-87. gies. British Journal of Psychology, 74, 11-16.
Gottschaldt, K. M., & Vahle-Hinz, C. (1981). Merkle cell re¬ Greenberg, M. J. (1981). The dependence of odor intensity on
ceptors: Structure and transducer function. Science, 214, the hydrophobic properties of molecules. In H. R. Moskow-
183-185. itz & C. B. Warren (Eds.), Odor quality and chemical struc-
 References 553

ture (pp. 177-194). Washington, D.C.: American Chemical Haber, R. N. (1980). Perceiving space from pictures: A theo¬
Society.
retical analysis. In M. Hagen (Ed.), Perception of Pictures:
Greene, H. A., & Madden, D. J. (1987). Adult age differences Vol. 1. Alberti’s Window: The projective model of pictorial
in visual acuity, stereopsis, and contrast sensitivity. Ameri¬ information (pp. 3-31). New York: Academic Press.
can Journal of Optometry and Physiological Optics, 64, Haber, R. N., & Hershenson, M. (1965). The effects of re¬
749-753. peated brief exposures on the growth of a percept. Journal
Gregory, R. L. (1966). Eye and brain. New York: World Uni¬ of Experimental Psychology, 69, 40-46.
versity Library. Hagen, M., & Jones, R. (1978). Cultural effects on pictorial
Gregory, R. L. (1971). Concepts and mechanisms of percep¬ perception: How many words is one picture really worth? In
tion. London: Duckworth. R. Walk & H. Pick (Eds.), Perception and Experience (pp.
Gregory, R. L. (1978). Eye and brain. (3rd ed.). New York: 171-212). New York: Plenum.
McGraw-Hill. Hahn, PI. (1934). Die Adaptation des Geschmackssinnes. Zeit-
Grice, G. R., Nullmeyer, R., & Schnizlein, J. M. (1979). Var¬ schraft fuer Sinnesphysiologie, 65, 105-145.
iable criterion analysis of brightness effects in simple reac¬ Hainline, L. (1978). Developmental changes in the scanning of
tion time. Journal of Experimental Psychology: Human Per¬ face and nonface patterns by infants. Journal of Experimen¬
formance and Perception, 5, 303-314. tal Child Psychology, 25, 90-115.
Groos, G., & Daan, S. (1985). The use of the biological clocks Haith, M. M., Bergman, T., & Moore, M. J. (1977). Eye con¬
in time perception. In J. A. Michon & J. L. Jackson (Eds.), tact and face scanning in early infancy. Science, 198, 853-
Time, mind and behavior (pp. 65-74). Berlin: Springer-Ver- 855.
lag. Hall, J. W. Ill, & Peters, R. W. (1982). Change in the pitch of
Groos, G., & Meijer, J. H. (1985). The effects of illumination a complex tone following its association with a second com¬
on suprachiasmatic nucleous electrical discharge. Annals of plex tone. Journal of the Acoustical Society of America, 71,
the New York Academy of Sciences. 142-146.
Gross, C. G., Rocha-Miranda, E. C., & Bender, D. B. (1972). Hall, M. J., Bartoshuk, L. M., Cain, W. S., & Stevens, J. C.
Visual properties of neurons in inferotemporal cortex of the (1975). PTC taste blindness and the taste of caffeine. Nature
macaque. Journal of Neurophysiology, 35, 96-111. (London), 253, 442-443.
Grossberg, J. M., & Grant, B. F. (1978). Clinical psychophys¬ Hall, W. G., & Oppenheim, R. W. (1987). Developmental psy¬
ics. Psychological Bulletin, 85, 1154-1176. chobiology: Prenatal, perinatal and early postnatal aspects of
Grossberg, S. (1987). Cortical dynamics of three-dimensional behavioral development. Annual Review of Psychology, 38,
form, color, and brightness perception: I. Monocular theory. 91-128.
Perception & Psychophysics, 41, 87-116. Halpem, B. P., & Meiselman, H. L. (1980). Taste psychophys¬
Grunau, R. V. E., & Craig, K. D. (1987). Pain expression in ics based on a simulation of human drinking. Chemical Sen¬
neonates: Facial action and cry. Pain, 28, 395-410. ses, 5, 279-294.
Grzegorczyk, P. B., Jones, S. W., & Mistretta, C. M. (1979). Halpem, D. F. (1981). The determinants of illusory-contour
Age-related differences in salt taste acuity. Journal of Ger¬ perception. Perception, 10, 191-213.
ontology, 34, 834-940. Halpem, D. F. (1986). Sex differences in cognitive abilities.
Guitton, D., Crommelink, M., & Roucoux, A. (1980). Stimu¬ Hillsdale, NJ: Lawrence Erlbaum.
lation of the superior colliculus in the alert cat: Eye move¬ Halpem, D. L., Blake, R., & Hillenbrand, J. (1986). Psycho¬
ment and neck EMG activity evoked when the head is re¬ acoustics of a chilling sound. Perception & Psychophysics,
strained. Experimental Brain Research, 39, 63-74. 39, 77-80.
Gulick, W. L. (1971). Hearing: Physiology and psychophysics. Halsam, D. (1967). Individual differences in pain threshold and
London & New York: Oxford University Press. level of arousal. British Journal of Psychology, 58, 139—
Gumsey, R., & Browse, R. A. (1987). Micropattem properties 142.
and presentation conditions influencing visual texture dis¬ Hamalainen, H., & Jarvilehto, T. (1981). Peripheral neural ba¬
crimination. Perception & Psychophysics, 41, 239-252. sis of tactile sensations in man: I. Effect of frequency and
Guzman, A. (1971). Analysis of curved line drawings using probe area on sensations elicited by single mechanical pulses
context and global information. Machine Intelligence, 6, on hairy and glabrous skin of the hand. Brain Research,
325-375. Edinburgh: Edinburgh University Press. 219, 1-12.
Gwiazda, J., Brill, S., Mohindra, I., & Held, R. (1980). Pref¬ Hamasaki, D. J., & Sutija, V. G. (1979). Development of X-
erential looking acuity in infants from 2 to 58 weeks of age. and Y-cells in kittens. Experimental Brain Research, 35, 9—
American Journal of Optometry and Physiological Optics, 23.
57, 428-432. Handel, S., & Gamer, W. R. (1965). The structure of visual
pattern associates and pattern goodness. Perception & Psy¬
chophysics, 1, 33-38.
Haaf, R. (1977). Visual responses to complex facelike patterns Handel, S., & Oshinsky, J. S. (1981). The meter of syncopated
by 15 and 20 week old infants. Developmental Psychology, auditory polyrhythms. Perception & Psychophysics, 30, 1-
38, 893-899. 9.
554 References

Hanna, T. E., von Gierke, S. M., & Green, D. M. (1986). Hatfield, G., & Epstein, W. (1985). The status of the minimum
Detection and intensity discrimination of a sinusoid. Journal principle in the theoretical analysis of visual perception.
of the Acoustical Society of America, 80, 1335-1340. Psychological Bulletin, 97, 155—186.
Hardie, R. C., & Kirschfe’ld, K. (1983). Ultraviolet sensitivity He, L. (1987). Involvement of endogenous opioid peptides in
of fly photoreceptors R7 and R8: Evidence for a sensitizing acupuncture analgesia. Pain, 31, 99-122.
function. Biophysics of Structure and Mechanism, 9, 171 — He, L., Lu, R., Zhuang, S., Zhang, X., Pan, X. (1985). Pos¬
180. sible involvement of opioid peptides of caudate nucleus in
Hardy, J. D., Wolff, H. G., & Goodell, H. (1943). The pain acupuncture analgesia. Pain, 23, 83-93.
threshold in man. Research Publications Association for Re¬ Head, H. (1920). Studies in neurology. London & New York:
search in Nervous and Mental Disease, 23, 1-15. Oxford University Press.
Hardy, J. D., Woiff, H. G., & Goodell, H. (1947). Studies on Heaton, J. M. (1968). The eye: Phenomenology and psychology
pain: Discrimination of differences in intensity of a pain of function and disorder. London: Tavistock.
stimulus as a basis of a scale of pain intensity. Journal of Hecht, S., & Mandelbaum, M. (1938). Rod-cone dark adapta¬
Clinical Investigation, 26, 1152-1158. tion and vitamin A. Science, 88, 219-221.
Hardy, J. D., Stolwijk, J. A. J., & Hoffman, D. (1968). Pain Hecht, S., Shlaer, S., & Pirenne, M. H. (1942). Energy quanta
following step increase in skin temperature. In D. R. Ken- and vision. Journal of General Physiology, 25, 819-840.
shalo (Ed.), The skin senses (pp. 444-457). Springfield, IL: Heckenmueller, E. G. (1965). Stabilization of the retinal image:
Thomas. A review of method, effects, and theory. Psychological Bul¬
Harkins, S., & Green, R. G. (1975). Discriminability and cri¬ letin, 63, 157-169.
terion differences between extraverts and introverts during Heggelund, P. (1981a). Receptive field organization of simple
vigilance. Journal of Research in Personality, 9, 335-340. cells in cat striate cortex. Experimental Brain Research, 42,
Harkins, S. W., & Chapman, C. R. (1977). The perception of 89-98.
induced dental pain in young and elderly women. Journal of Heggelund, P. (1981b). Receptive field organization of complex
Gerontology, 32, 428-435. cells in cat striate cortex. Experimental Brain Research, 42,
Hannon, L. D. (1973). The recognition of faces. Scientific 99-107.
American, 229, 70-82. Hein, A. (1980). The development of visually guided behavior.
Harmon, L. D., & Julesz, B. (1973). Masking in visual recog¬ In C. Harris (Ed.), Visual coding and adaptability (pp. 51-
nition: Effects of two dimensional filtered noise. Science, 68). Hillsdale, NJ: Lawrence Erlbaum.
180, 1194-1197. Hein, A., & Held, R. (1967). Dissociation of the visual placing
Harper, R. S. (1953). The perceptual modification of coloured response into elicited and guided components. Science, 158,
figures. American Journal of Psychology, 66, 86-89. 390-392.
Harris, C. S. (1980). Insight or out of sight?: Two examples of Hein, A., Held, R., & Gower, E. C. (1970). Development and
perceptual plasticity in the human adult. In C. S. Harris segmentation of visually controlled movement by selective
(Ed.), Visual coding and adaptability (pp. 95-149). Hills¬ exposure during rearing. Journal of Comparative and Phys¬
dale, NJ: Lawrence Erlbaum. iological Psychology, 73, 181-187.
Harris, L. J. (1981). Sex related variations in spatial skill. In Hein, S., & Diamond, R. M. (1971). Contrasting development
L. S. Liben, A. H. Patterson, & N. Newcombe (Eds.), Spa¬ of visually triggered and guided movements in kittens with
tial representation and behavior across the lifespan: Theory respect to interocular and interline equivalence. Journal of
and application (pp. 83-128). New York: Academic Press. Comparative and Physiological Psychology, 76, 219-224.
Harris, P., & MacFarlane, A. (1974). The growth of the effec¬ Held, R. (1985). Binocular vision—behavioral and neuronal de¬
tive visual field from birth to seven weeks. Journal of Ex¬ velopment. In J. Mehler & R. Fox (Eds.), Neonate cogni¬
perimental Child Psychology, 18, 340-348. tion: Beyond the blooming buzzing confusion (pp. 37-44).
Harter, M., & Suitt, C. (1970). Visually-evoked cortical re¬ Hillsdale, NJ: Lawrence Erlbaum.
sponses and pattern vision in the infant: A longitudinal Held, R., & Bauer, J. A. (1967). Visually guided reaching in
study. Psychonomic Science, 18, 235-237. infant monkeys after restricted rearing. Science, 155, 718-
Hartline, H. K. (1940). The receptive fields of optic nerve fi¬ 720.
bers. American Journal of Physiology, 130, 690-699. Held, R., & Bossom, J. (1961). Neonatal deprivation and adult
Hartline, H. K., & Ratliff, F. (1957). Inhibitory interaction of rearrangement: Complementary techniques for analyzing
receptor units in the eye of Limulus. Journal of General plastic sensory-motor coordinations. Journal of Comparative
Physiology, 40, 357-376. and Physiological Psychology, 54, 33-37.
Hartman, A., & Hollister, L. (1963). Effect of mescaline, lys¬ Held, R., Dichgans, J., & Bauer, J. (1975). Characteristics of
ergic acid diethylamide and psilocybin on color perception. moving visual senses influencing spatial orientation. Vision
Psychopharmacologia, 4, 441-451. Research, 15, 357-365.
Harvey, L. O., Jr., & Leibowitz, H. (1967). Effects of expo¬ Held, R., & Hein, A. (1958). Adaptation of disarranged hand-
sure duration, cue reduction, and temporary monocularity on eye coordination contingent upon re-afferent stimulation.
size matching at short distances. Journal of the Optical So¬ Perceptual and Motor Skills, 8, 87-90.
ciety of America, 57, 249-253. Held, R., & Hein, A. (1963). Movement-produced stimulation
 References 555

in the development of visually guided behavior. Journal of light and shade cues of depth. Journal of Comparative and
Comparative and Physiological Psychology, 56, 872-876. Physiological Psychology, 43, 112-122.
Held, R., & Hein, A. (1967). On the modifiability of form per¬ Hick, W. E. (1952). On the rate of gain of information. Quar¬
ception. In W. Wathen-Dunn (Ed.), Models for the percep¬ terly Journal of Experimental Psychology, 4, 11-26.
tion of speech and visual form (pp. 296-304). Cambridge, Hickey, T. L. (1977). Postnatal development of the human lat¬
MA: MIT Press. eral geniculate nucleus: Relationship to a critical period for
Hellekant, G. (1965). Electrophysiological investigation of the the visual system. Science, 198, 836-838.
gustatory effect of ethyl alcohol: The summated response of Hickey, T. L., & Peduzzi, J. E. (1987). Structure and devel¬
the chorda tympani in the cat, dog and rat. Acta Physiolo- opment of the visual system. In P. Salapatek & L. Cohen
gica Scandinavica, 64, 392-397. (Eds.), Handbook of infant perception: Vol. 1. From sensa¬
Hellstrom, A. (1979). Time errors and differential sensation tion to perception (pp. 1-43). Orlando, FL: Academic
weighting. Journal of Experimental Psychology: Human Press.
Perception and Performance, 5, 460-477. Hicks, R. E., Miller, G. W., Gaes, G., & Bierman, K. (1977).
Hellstrom, A. (1985). The time-order error and its relatives: Concurrent processing demands and the experience of time
Mirrors of cognitive processes in comparing. Psychological in passing. American Journal of Psychology, 90, 431-446.
Bulletin, 97, 35-61. Hicks, R. E., Miller, G. W., & Kinsboume, M. (1976). Pro¬
Helmholtz, H. E. F. von. (1930). The sensations of tone (A. J. spective and retrospective judgments of time as a function
Ellis, Trans.). New York: Longmans, Green. (Original work of amount of information processed. American Journal of
published 1863) Psychology, 89, 719-730.
Helmholtz, H. E. F. von. (1962). Treatise on physiological op¬ Hier, D. B., & Crowley, W. F., Jr. (1982). Spatial ability in
tics. (J. P. C. Southall, Ed. and Trans.). New York: Dover. androgen-deficient men. New England Journal of Medicine,
(Original work published 1909) 306, 1202-1205.
Helson, H. (1959). Adaptation level theory. In S. Koch (Ed.), Higashiyama, A. (1985). The effects of familiar size on judg¬
Psychology: A study of a science (Vol. 1, pp. 565-621). ments of size and distance; an interaction of viewing attitude
New York: McGraw-Hill. with spatial cues. Perception & Psychophysics, 35, 305-
Helson, H. (1964). Adaptation level theory: An experimental 312.
and systematic approach to behavior. New York: Harper. Hirsch, H. V. (1972). Visual perception in cats after environ¬
Henmon, V. A. C. (1906). The time of perception as a measure mental surgery. Experimental Brain Research, 15, 409-
of differences in sensations. Archives of Philosophy, Psy¬ 423.
chology and Scientific Methods, No. 8. Hirsch, H. V., & Spinelli, D. N. (1970). Visual experience
Henn, V., Cohen, B., & Young, L. (1980). Visual-vestibular modifies distribution of horizontally and vertically oriented
interaction in motion perception and the generation of nys¬ receptive fields in cats. Science, 168, 869-871.
tagmus. Neurosciences Research Program Bulletin, 18, Hirsh, I. J. (1948). The influence of interaural phase on inter-
459-651. aural summation and inhibition. Journal of the Acoustical
Henn, V., Young, L. R., & Finley, C. (1974). Vestibular nu¬ Society of America, 20, 536-544.
cleus units in alert monkeys are also influenced by moving Hoagland, H. (1933). The physiological control of judgment of
visual field. Brain Research, 71, 144-149. duration: Evidence for a chemical clock. Journal of General
Henning, H. (1915). DerGeruch. I. Zietschrift fur Psychologie, Psychology, 9, 267-287.
73, 161-257. Hochberg, J. (1971). Perception. II. Space and movement. In
Henning, H. (1916). Die Qualitatenreihe des Geschmaks. Zeit- J. W. Kling & L. A. Riggs (Eds.), Woodworth and Schloss-
schrift fur Psychologie, 74, 203-219. berg’s experimental psychology. (3rd ed.). (pp. 475-550).
Hensel, H. (1981). Thermoreception and temperature regula¬ New York: Holt, Rinehart & Winston.
tion. New York: Academic Press. Hochberg, J. (1972). Nativism and empiricism in perception. In
Hering, E. (1964). Outlines of a theory of the light sense. L. Postman (Ed.), Psychology in the making (pp. 255-330).
(L. M. Hurvich & D. Jameson, Trans.). Cambridge, New York: Knopf.
MA: Harvard University Press. (Original work published Hochberg, J. (1974). Higher-order stimuli and interresponse
1878) coupling in the perception of the visual world. In R. B.
Hernandez-Peon, R. (1964). Psychiatric implications of neuro¬ Macleod & H. L. Pick (Eds.), Perception: Essays in honor
physiological research. Bulletin of the Meninger Clinic, 28, of James J. Gibson (pp. 17-39). Ithaca: Cornell University
165-185. Press.
Hershberger, W. (1987). Saccadic eye movements and the per¬ Hochberg, J. (1981). On cognition in perception: Perceptual
ception of visual direction. Perception & Psychophysics, 41, coupling and unconscious inference. Cognition, 10, 127—
35-44. 134.
Hershenson, M. (1982). Moon illusion and spiral aftereffect: Il¬ Hochberg, J. (1982). How big is a stimulus? In J. Beck (Ed.),
lusions due to the loom-zoom system? Journal of Experi¬ Organization and representation in perception (pp. 191 —
mental Psychology: General, 111, 423-440. 218). Hillsdale, NJ: Lawrence Erlbaum.
Hess, E. H. (1950). Development of the chick’s response to Hochberg, J., & Beck, J. (1954). Apparent spatial arrangement
556 References

and perceived brightness. Journal of Experimental Psychol¬ quency detection. Perception & Psychophysics, 35, 256—
ogy, 47, 263-266. 264.
Hochberg, J., & Brooks, V. (1960). The psychophysics of Howell, P., Cross, I., & West, R. (Eds.), (1985). Musical
form: Reversible-perspective drawings of spatial objects. structure and cognition. London: Academic Press.
American Journal of Psychology, 73, 337-354. Hoyer, W., & Plude, D. (1980). Attentional and perceptual pro¬
Hochberg, J., & Brooks, V. (1962). Pictorial recognition as an cesses in the study of cognitive aging. In L. Poon (Ed.),
unlearned ability. A study of one child’s performance. Aging in the 1980s (pp. 227-238). Washington, D.C.:
American Journal of Psychology, 75, 624-628. American Psychological Association.
Hockey, G. R. (1970). Effect of loud noise on attentional selec¬ Hubbell, W. L., & Bownds, M. D. (1979). Visual transduction
tivity. Quarterly Journal of Experimental Psychology, 22, in vertebrate photoreceptors. Annual Review of Neurosci¬
28-36. ences, 2, 17-34.
Hoffer, A., & Osmond, H. (1967). The hallucinogens. New Hubei, D. H., & Wiesel, T. N. (1962). Receptive fields, bin¬
York: Academic Press. ocular interaction and functional architecture in the cat’s
Hoffman, J. E. (1980). Interaction between global and local lev¬ visual cortex. Journal of Physiology (London), 160, 106-
els of form. Journal of Experimental Psychology: Human 154.
Perception and Performance, 6, 222-234. Hubei, D. H., & Wiesel, T. N. (1968). Receptive fields and
Hoffman, J. E., Nelson, B., & Houck, M. R. (1983). The role functional architecture of monkey striate cortex. Journal of
of attentional resources in automatic detection. Cognitive Physiology (London), 195, 215-243.
Psychology, 51, 379-410. Hubei, D. H., & Wiesel, T. N. (1979). Brain mechanisms of
Hoffmann, K. P. (1979). Optokinetic nystagmus and single cell vision. Scientific American, 82, 84-97.
responses in the nucleus tractus opticus after early monocu¬ Hudson, W. (1960). Pictorial depth perception in subcultural
lar deprivation in the cat. In. R. D. Freeman (Ed.), Devel¬ groups in Africa. Journal of Social Psychology, 52, 183—
opmental neurobiology of vision (pp. 63-72). New York: 208.
Plenum. Hudson, W. (1962). Pictorial perception and educational adap¬
Hoffman, K. P., & Sherman, S. (1975). Effects of early bin¬ tation in Africa. Psychologia, Africana, 9, 226-239.
ocular deprivation on visual input to cat superior colliculus. Hudspeth, A. J. (1985). The cellular basis of hearing: The bio¬
Journal of Neurophysiology, 38, 1049-1059. physics of hair cells. Science, 230, 745-752.
Holender, D. (1986). Semantic activation without conscious Hughes, H. C. (1986). Asymmetric interference between com¬
identification in dichotic listening, parafoveal vision, and vis¬ ponents of suprathreshold compound gratings. Perception &
ual masking: A survey and appraisal. The Behavioral and Psychophysics, 40, 241-250.
Brain Sciences, 9, 1-23. Hughes, H. C., Layton, W. M.. Baird, J. C., & Lester, L. S.
Holland, H. (1960). Drugs and personality. XII. A comparison (1984). Global precedence in visual pattern recognition.
of several drugs by the flicker-fusion method. Journal of Perception & Psychophysics, 35, 361—371.
Mental Science, 106, 858-861. Hughes, H. C., & Zimba, L. D. (1985). Spatial maps of di¬
Holst, E. von, & Mittelstadt, H. (1950). Das Reafferenzprincip rected visual attention. Journal of Experimental Psychology:
(wechselwirkungen zeischen zentral Nervensystem und Per¬ Human Perception and Performance, 11, 409-430.
ipherie). Naturwissenschaften, 37, 464-476. Humphrey, G. K., Humphrey, D. E., Muir, D. W., & Dod-
Holway, A. F., & Boring, E. G. (1941). Determinants of ap¬ well, P. C. (1986). Pattern perception in infants: Effects of
parent visual size with distance variant. American Journal of structure and transformation. Journal of Experimental Child
Psychology, 54, 21-37. Psychology, 41, 128-148.
Honda, H. (1984). Functional between-hand differences and Humphreys, G. W. (1983). Reference frames and shape percep¬
outflow eye position information. Quarterly Journal of Ex¬ tion. Cognitive Psychology, 15, 151-196.
perimental Psychology, 36A, 75-88. Humphreys, G. W. (1984). Shape constancy: The effects of
Houtsma, A. J. M., & Goldstein, J. L. (1972). The central or¬ changing shape orientation and the effects of changing the
igin of the pitch of complex tones: Evidence from musical position of focal features. Perception & Psychophysics, 36,
interval recognition. Journal of the Acoustical Society of 50-64.
America, 51, 520-529. Hunzelmann, N., & Spillman, L. (1984). Movement adaptation
Howard, I. P. (1982). Human visual orientation. Chichester: in the peripheral retina. Vision Research, 24, 1765-1769.
Wiley. Hurlbert, A. C., & Poggio, T. A. (1988). Synthesizing a color
Howard, I. P., Anstis, T., & Lucia, H. C. (1974). The relative algorithm from examples. Science, 239, 482-485.
lability of mobile and stationary components in a visual-mo¬ Hurvich, L. M. (1981). Color vision. Sunderland, MA: Sinauer
tor adaptation task. Quarterly Journal of Experimental Psy¬ Associates.
chology, 26, 293-300. Hurvich, L. M., & Jameson, D. (1974). Opponent processes as
Howard, 1. P., Craske, B., & Templeton, W. B. (1965). Vis- a model of neural organization. American Psychologist, 29,
uomotor adaptation to discordant exafferent stimulation. 88-102.
Journal of Experimental Psychology, 70, 189-191. Hutz, C. S., & Bechtoldt, H. P. (1980). The development of
Howard, J. H., O’Toole, A. J., Parasuraman, R., & Bennett, binocular discrimination in infants. Bulletin of the Psycho-
K. B. (1984). Pattern-directed attention in uncertain fre¬ nomic Society, 16, 83-86.
 References 557

Huxley, A. (1963). The doors of perception and heaven and Jenkins, B. (1985). Orientational anisotropy in the human visual
hell. New York: Harper. system. Perception & Psychophysics, 37, 125-134.
Hyman, A., Mentyer, T., & Calderone, L. (1979). The contri¬ Jennings, J. A. M., & Charman, W. N. (1981). Off-axis image
bution of olfaction to taste discrimination. Bulletin of the quality in the human eye. Vision Research, 21, 445-455.
Psychonomic Society, 13, 359-362. Jensen, D., & Engel, R. (1971). Statistical procedures for relat¬
ing dichotomous responses to maturation and EEG measure¬
Iacono, W. G., Peloguin, L. J., Lumry, A. E., Valentine, ments. Electroencephalography and Clinical Neurophysiol¬
R. H., & Tuason, V. B. (1982). Eye tracking in patients ogy, 30, 437-443.
with unipolar and bipolar affective disorders in remission. Jesteadt, W. (1980). An adaptive procedure for subjective judg¬
Journal of Abnormal Psychology, 91, 35-44. ments. Perception c£ Psychophysics, 28, 85-88.
Imbert, M. (1985). Physiological underpinnings of perceptual Jesteadt, W., Bacon, S. P., & Lehman, J. R. (1982). Forward
development. In J. Mehler & R. Fox (Eds.), Neonate cog¬ masking as a function of frequency, masker level, and signal
nition: Beyond the blooming buzzing confusion (pp. 69-88). delay. Journal of the Acoustical Society of America, 71,
Hillsdale, NJ: Lawrence Erlbaum. 950-962.
Ingram, R. M. & Barr, A. (1979). Changes in refraction be¬ Jesteadt, W., & Wier, C. C. (1977). Comparison of monaural
tween the ages of 1 and 3Vi years. British Journal of Oph¬ and binaural discrimination of intensity and frequency. Jour¬
thalmology, 63, 39-342. nal of the Acoustical Society of America, 61, 1599-1603.
Intraub, H. (1985). Visual dissociation: An illusory conjunction Jesteadt, W., Wier, C. C., & Green, D. M. (1977). Intensity
of pictures and forms. Journal of Experimental Psychology: discrimination as a function of frequency and sensation
Human Perception and Performance, 11, 431-442. level. Journal of the Acoustical Society of America, 61,
Ippolitov, F. W. (1973). Interanalyser differences in the sensi¬ 169-177.
tivity-strength parameter for vision, hearing and cutaneous Johansson, G. (1976a). Visual motion perception. In R. Held &
modalities. In V. D. Nebylitsyn & J. A. Gray (Eds.), Bio¬ W. Richards (Eds.), Recent progress in perception: Read¬
logical bases of individual behavior (pp. 43-61). New ings from Scientific American (pp. 67-75). San Francisco:
York: Academic Press. Freeman.
Ittelson, W. H. (1951). Size as a cue to distance: Static locali¬ Johansson, G. (1976b). Spatio-temporal differentiation and in¬
zation. American Journal of Psychology, 64, 54-67. tegration in visual motion perception. Psychological Re¬
Ittelson, W. H. (1960). Visual space perception. Berlin & New search, 38, 379-393.
York: Springer-Verlag. Johansson, G., von Hofsten, C., & Jansson, G. (1980). Event
Ittelson, W. H. (1962). Perception and transactional psychol¬ perception. Annual Review of Psychology, 31, 27-63.
ogy. In S. Koch (Ed.), Psychology: A study of a science Johnson, C. H., & Hastings, J. W. (1986). The elusive mech¬
(Vol. 4, pp. 660-704). New York: McGraw-Hill. anism of the circadian clock. American Scientist, 74(1), 29-
36.
Jacobs, G. H. (1976). Color vision. Annual Review of Psychol¬ Johnson, D. H. (1980). The relationship between spike rate and
ogy, 27, 63-89. synchrony in responses of auditory-nerve fibers to single
Jacobs, G. H. (1986). Cones and opponency. Vision Research, tones. Journal of the Acoustical Society of America, 68,
26, 1533-1541. 1115-1122.
Jahoda, G., & McGurk, H. (1974). Pictorial depth perception: Johnson, M. A. (1986). Color vision in the peripheral retina.
A developmental study. British Journal of Psychology, 65, American Journal of Optometry & Physiological Optics, 63,
141-149. 97-103.
James, W. (1890). The principles of psychology. New York: Jones, L. A. (1986). Perception of force and weight: Theory
Holt, Rinehart & Winston. and research. Psychological Bulletin, 100, 29-42.
Jameson, D., & Hurvich, L. M. (1959). Note on factors influ¬ Jonides, J. (1980). Towards a model of the mind’s eye’s move¬
encing the relation between stereoscopic acuity and obser¬ ments. Canadian Journal of Psychology, 34, 103-112.
vation distance. Journal of the Optical Society of America, Jonides, J , & Mack, R. (1984). On the cost and benefit of cost
49, 639. and benefit. Psychological Bulletin, 96, 29-44.
Jameson, D., & Hurvich, L. M. (1964). Theory of brightness Jonides, J., & Yantis, S. (1988). Uniqueness of abrupt visual
and color contrast in human vision. Vision Research, 4, onset in capturing attention. Perception & Psychophysics,
135-154. 43, 346-355.
Jarvis, J. R. (1977). On Fechner-Benham subjective colour. Vi¬ Julesz,. B. (1971). Foundations of cyclopean perception. Chi¬
sion Research, 17, 445-451. cago: University of Chicago Press.
Javel, E. (1981). Suppression of auditory nerve responses I: Julesz, B. (1978). Perceptual limits of texture discrimination
Temporal analysis, intensity effects and suppression con¬ and their implications to figure-ground separation. In E.
tours. Journal of the Acoustical Society of America, 69, Leeuwenberg & H. Buffart (Eds.), Formal Theories of Per¬
1735-1745. ception (pp. 205-216). New York: Wiley.
Jen, D. H., & Steele, C. R. (1987). Electrokinetic model of Julesz, B. (1980). Spatial nonlinearities in the instantaneous
cochlear hair cell motility. Journal of the Acoustical Society perception of textures with identical power spectra. In C.
of America, 82, 1667—1678. Longuet-Higgins & N. S. Sutherland (Eds.), The psychology
558 References

of vision. Philosophical transactions of the Royal Society, Kauer, J. S. (1987). Coding in the olfactory system. In T. E.
London, 290, 83-94. Finger & W. L. Silver (Eds.), Neurobiology of taste and
Julesz, B. (1981). Textons, the elements of texture perception smell (pp. 205-232). New York: Wiley.
and their interactions. Nature, 290, 91-97. Kaufman, L. (1974). Sight and mind: An introduction to visual
Julesz, B., & Bergen, J. R. (1983). Textons, the fundamental perception. London & New York: Oxford University Press.
elements in preattentive vision and perception of textures. Kaufman, L., & Rock, I. (1962). The moon illusion. Scientific
The Bell System Technical Journal, 62, 1619-1645. American, 207, 120-130.
Julesz, B., & Schumer, R. A. (1981). Early visual perception. Kaufmann, R., Maland, J., & Yonas, A. (1981). Sensitivity of
Annual Review of Psychology, 32, 575-627. 5- and 7-month-old infants to pictorial depth information.
Jusczyk, P. W. (1986). Toward a model of the development of Journal of Experimental Child Psychology, 32, 162-168.
speech perception. In J. S. Perkell & D. H. Klatt (Eds.), Kausler, D. H. (1982). Experimental psychology and human ag¬
Invariance and Variability in Speech Processes (pp. 1-19). ing. New York: Wiley.
Hillsdale, NJ: Lawrence Erlbaum. Kawabata, N. (1986). Attention and depth perception. Percep¬
tion, 15, 563-572.
Kaas, J. H. (1983). The organization of somatosensory cortex Kaye, M., Mitchell, D. E., & Cynader, M. (1982). Depth per¬
in primates and other mammals. In C. von Euler, O. Fran- ception, eye dominance and cortical binocularity of dark-
zen, U. Lindblom, & D. Ottoson (Eds.), Somatosensory reared cats. Developmental Brain Research, 2, 37-53.
mechanisms (pp. 51-60). New York: Plenum. Keithley, E. M., & Schreiber, R. C. (1987). Frequency map of
Kaess, D. W. (1980). Instructions and decision times of size- the spiral ganglion in the cat. Journal of the Acoustical So¬
constancy responses. Perception & Psychophysics, 27, 477- ciety of America, 81, 1036-1042.
482. Keller, H. (1931). The story of my life. New York: Doubleday.
Kahneman, D. (1966). Time-intensity reciprocity in acuity as a Kellman, P. J. (1984). Perception of three-dimensional form by
function of luminance and figure-ground contrast. Vision Re¬ human infants. Perception & Psychophysics, 36, 353-358.
search, 6, 207-215. Kellman, P. J., & Spelke, E. S. (1983). Perception of partly
Kahneman, D. (1968). Method, findings, and theory in studies occluded objects in infancy. Cognitive Psychology, 15, 483-
of visual masking. Psychological Bulletin, 70, 404-425. 524.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs, Kendrick, K. M., & Baldwin, B. A. (1987). Cells in the tem¬
NJ: Prentice-Hall. poral cortex of conscious sheep can respond preferentially to
Kahneman, D., Norman, J., & Kubovy, M. (1967). Critical the sight of faces. Science, 236, 448-450.
duration for the resolution of form: Centrally or peripherally Kennedy, J., & Ostry, D. (1976). Approaches to picture percep¬
determined? Journal of Experimental Psychology, 73, 323- tion: Perceptual experience and ecological optics. Canadian
327. Journal of Psychology, 30, 90-98.
Kahneman, D., & Treisman, A. (1984). Changing views of at¬ Kennedy, J. M., & Domander, R. (1985). Shape and contour:
tention and automaticity. In R. Parasuraman & D. R. Davies The points of maximum change are least useful for recogni¬
(Eds.), Varieties of attention, (pp. 29-61). Orlando, FL: tion. Perception, 14, 367-370.
Academic Press. Kenshalo, D. R., & Isensee, O. (1983). Responses of primate
Kaiser, P. K., & Boynton, R. M. (1985). Role of the blue SI cortical neurons to noxious stimuli. Journal of Neuro¬
mechanism in wavelength discrimination. Vision Research, physiology, 50, 1479-1496.
25, 523-529. Kenshalo, D. R., Nafe, J. P., & Brooks, B. (1961). Variations
Kalat, J. W. (1984). Biological psychology. (2nd ed.). Bel¬ in thermal sensitivity. Science, 134, 104-105.
mont, CA: Wadsworth. Kenshalo, D. R., & Scott, H. A., Jr. (1966). Temporal course
Kaneko, A., Nishimura, Y., Tachibana, M., Tauchi, M., & of thermal adaptation. Science, 151, 1095-1096.
Shimai, K. (1981). Physiological and morphological studies Keren, G., & Baggen, S. (1981). Recognition models of al¬
of signal pathways in the carp retina. Vision Research, 21, phanumeric characters. Perception <& Psychophysics, 29,
1519-1526. 234-245.
Kaplan, A., & Glanville, E. (1964). Taste thresholds for bitter¬ Kessen, W., Salapatek, P., & Haith, M. M. (1972). The visual
ness and cigarette smoking. Nature (London), 202, 1366. response of the human newborn to linear contour. Journal
Karmel, B. Z., & Maisel, E. B. (1975). A neuronal activity of Experimental Child Psychology, 13, 9-20.
model for infant visual attention. In L. B. Cohen & P. Sa- Keveme, E. B. (1978). Olfactory cues in mammalian behavior.
lapatek (Eds.), Infant perception: From sensation to cogni¬ In J. E. Hutchinson (Ed.), Biological determinants of sexual
tion: Vol. I. Basic visual processes (pp. 78-133). New behavior (pp. 727-763). Chichester: Wiley.
York: Academic Press. Khanna, S. N., & Leonard, D. G. B. (1982). Basilar membrane
Kasamatsu, T. (1976). Visual cortical neurons influenced by the tuning in the cat cochlea. Science, 215, 305-306.
oculomotor input: Characterization of their receptive field Kiang, K. Y. S., Rho, J. M., Northrop, C. C., Liberman, M.
properties. Brain Research, 113, 271-292. C., & Ryugo. D. K. (1982). Hair-cell innervation by spiral
Kauer, J. S. (1980). Some spatial characteristics of central in¬ ganglion cells in adult cats. Science, 217, 175-177.
formation processing in the vertebrate olfactory pathway. In Kilbride, P. E., Hutman. L. P., Fishman, M., & Read, J. S.
H. van der Starre (Ed.)., Olfaction and taste VII (pp. 227- (1986). Foveal cone pigment density difference in the aging
236). London: 1RL Press. human eye. Vision Research, 26, 321-325.
 References 559

Killbride, P. L., & Leibowitz, H. W. (1975). Factors affecting Kobler, J. B., Isbey, S. F., & Casseday, J. H. (1987). Auditory
the magnitude of the Ponzo illusion among the Baganda. pathways to the frontal cortex of the mustache bat, Pteron-
Perception & Psychophysics, 17, 543-548. otus pamelli. Science, 236. 824-826.
Killbride, P. L., & Robbins, M. (1968). Linear perspective pic¬ Kohler, I. (1962). Experiments with goggles. Scientific Ameri¬
torial depth perception and education among the Baganda. can, 206, 62-86.
Perception and Motor Skills, 27, 601-602. Kohler, I. (1964). The formation and transformation of the per¬
Kimura, K., & Beidler, L. M. (1961). Microelectrode study of ceptual world. Psychological Issues, 3 (Whole No. 4).
taste receptors of rat and hamster. Journal of Cellular and Kohler, W. (1923). Zur Theories des Sukzessivvergleichs und
Comparative Physiology, 58, 131-140. der Zeitfehler. Psychologische Forschung, 4, 115-175.
Kinchla, R. A., Solis-Macias, V., & Hoffman, J. E. (1983). Kolb, B., & Whishaw , I. O. (1985). Fundamentals of human
Attending to different levels of structure in a visual image. neuropsychology (2nd ed.). New York: Freeman.
Perception & Psychophysics, 33, 1-10. Kolb, H., Nelson, R., & Mariani, A. (1981). Amacrine cells,
Kinchla, R. A., & Wolfe, J. (1979). The order of visual pro¬ bipolar cells and ganglion cells of the cat retina: A Golgi
cessing: “Top-down”, “bottom-up”, or “middle-out.” study. Vision Research, 21, 1081-1114.
Perception & Psychophysics, 25, 225-231. Kolers, P. A., & Brewster, J. M. (1985). Rhythms and re¬
Kirk-Smith, M. D., & Booth, D. A. (1980). Effects of andros- sponses. Journal of Experimental Psychology: Human Per¬
tenone on choice of location in other’s presence. In H. van ception and Performance, 11, 150-167.
der Starre (Ed.), Olfaction and taste VII (pp. 397-400). Kolers, P. A., & Green, M. (1984). Color logic of apparent
London: IRL Press. motion. Perception, 13, 249-254.
Kirk-Smith. M. D.. Booth, D. A., Caroll, D., & Davies, P. Kolers, P. A., & von Grunau. M. (1976). Shape and color in
(1978). Human social attitudes affected by androstenol. Re¬ apparent motion. Vision Research, 16, 329-335.
search communications in psychology, psychiatry and be¬ Komoda, M. K., Festinger, L., Phillips, L. J., Duckman, R.
havior, 3, 379-384. H., & Young, R. A. (1973). Some observations concerning
Kitzes, L. M., Gibson, M. M., Rose, J. E., & Hind, J. E. saccadic eye movement. Vision Research, 13, 1009-1020.
(1978). Initial discharge latency and threshold considerations Konstadt, N., & Forman, E. (1965). Field dependence and ex¬
for some neurons in cochlear nucleus complex of the cat. ternal directedness. Journal of Personality and Social Psy¬
Journal of Neurophysiology, 41, 1165-1182. chology, 1, 490-493.
Klatt, D. H. (1980). Speech perception: A model of acoustic- Kombrot, D. E. (1984). Mechanisms for categorization: Deci¬
phonetic analysis and lexical access. In R. Cole (Ed.), Per¬ sion criteria and the form of the psychophysical function.
ception and production of fluent speech (pp. 243-288). British Journal of Mathematical and Statistical Psychology,
Hillsdale, NJ: Lawrence Erlbaum. 37, 184-198.
Klatzky, R. L., Lederman, S. J., & Metzger, V. A. (1985). Kosterlitz, H. W., & McKnight, A. T. (1981). Opioid peptides
Identifying objects by touch: An “expert system.” Percep¬ and sensory function. In D. Ottoson (Ed.), Progress in sen¬
tion & Psychophysics, 37, 299-302. sory physiology (Vol. L, pp. 31-95). Heidelberg: Springer-
Klatzky, R. L., Lederman. S. J., & Reed, C. (1987). There’s Verlag.
more to touch than meets the eye: The salience of object Kozlowski, L. T., & Cutting, J. E. (1977). Recognizing the sex
attributes for haptics with and without vision. Journal of Ex¬ of a walker from a dynamic point-light display. Perception
perimental Psychology: General, 116, 356-369. & Psychophysics, 21, 575-580.
Klein, G. S. (1970). Perception, motives and personality. New Krauskopf, J., & Reeves, A. (1980). Measurement of the effect
York: Knopf. of photon noise on detection. Vision Research, 20, 193—
Klein, S. A., & Levi, D. M. (1985). Hyperacuity threshold of 196.
1.0 second: Theoretical predictions and empirical validation. Kremenitzer, J. P., Vaughan, H. G., Kurtzberg, D., & Dowl¬
Journal of the Optical Society of America, A2, 1170-1190. ing, K. (1979). Smooth-pursuit eye movements in the new¬
Kluender, K. R., Diehl, R. L., & Killeen, P. R. (1987). Japa¬ born infant. Child Development, 50, 442-448.
nese quail can learn phonetic categories. Science, 237, Kries, J. von. (1895). Uber die Natur gewisser mit den psy-
1195-1197. chischen Vorgangen verknupfter Ghimzustande. Zeitschrift
Kluver, H., & Bucy, P. C. (1937). “Psychic blindness” and fur Psychologie, 8, 1-33.
other symptoms following bilateral temporal lobectomy in Krueger, L. E. (1982). A word superiority effect with print and
Rhesus monkeys. American Journal of Physiology, 119, braille characters. Perception & Psychophysics, 31, 345-
352-353. 352.
Klymenko, V., & Weisstein, N. (1986). Spatial frequency dif¬ Kruger, J. (1981). The difference between x- and y-type re¬
ferences can determine figure-ground organization. Journal sponses in ganglion cells of the cat’s retina. Vision Re¬
of Experimental Psychology: Human Perception and Perfor¬ search, 21, 1685-1687.
mance, 12, 324-330. Kruger, J., & Gouras, P. (1980). Spectral selectivity of cells
Knudsen, E. I., & Konishi, M. (1978a). A neural map of au¬ and its dependence on slit length in monkeys’ visual cortex.
ditory space in the owl. Science, 200, 795-797. Journal of Neurophysiology, 43, 1055-1069.
Knudsen, E. I., & Konishi, M. (1978b). Center-surround orga¬ Krumhansl, C. L., Bharucha, J. J., & Kessler, E. J. (1982).
nization of auditory receptive fields in the owl. Science, Perceived harmonic structure of chords in three related
202, 778-780. musical keys. Journal of Experimental Psychology: Human
560 References

Perception and Performance, 8, 24-36. sponses to noxious stimulation and comparison with re¬
Krumhansl, C. L., & Kessler, E. J. (1982). Tracing the dy¬ sponses to non-noxious stimulation. Experimental Brain
namic changes in perceived tonal organization in a spatial Research, 49, 35—45.
representation of musical keys. Psychological Review, 89, Land, E. H. (1977). The retinex theory of color vision. Scien¬
334-368. tific American, 237(6), 108-128.
Krumhansl, C. L., & Shepard, R. N. (1979). Quantification of Landolt, E. (1889). Tableua d’optotypes pour la determination
the hierarchy of tonal functions within a diatonic context. de l’acuite visuelle. Societe Francois d’Ophthalmologie, 1,
Journal of Experimental Psychology: Human Perception & 385ff.
Performance, 5, 579-594. Lappin, J. S., & Preble, L. D. (1975). A demonstration of
Kryter, K. D. (1970). The effects of noise on man. New York: shape constancy. Perception & Psychophysics, 17, 439-
Academic Press. 444.
Kryter, K. D. (1985). The effects of noise on man. (2nd ed.). Lauter, J. L., Herscovitch, P., Formby, C., & Raichle, M. E.
Orlando, FL: Academic Press. (1985). Tonotopic organization in human auditory cortex re¬
Kuffler, S. W. (1953). Discharge patterns and functional orga¬ vealed by positron emission tomography. Hearing Research,
nization of mammalian retina. Journal of Neurophysiology, 20, 199-205.
16, 37-68. Lawless, H. T., & Stevens, D. A. (1988). Responses by hu¬
Kuhl, P. K. (1987). Perception of speech and sound in early mans to oral chemical irritants as a function of locus of stim¬
infancy. In P. Salapatek & L. Cohen (Eds.), Handbook of ulation. Perception & Psychophysics, 43, 72-78.
infant perception: Vol. 2. From perception to cognition (pp. Lea, S. E. G. (1984). Instinct, environment, and behavior. Lon¬
275-382). Orlando, FL: Academic Press. don: Methuen.
Kuhl, P. K., & Meltzoff, A. N. (1982). The bimodal perception Leehey, S. C., Moskowitz-Cook, A., Brill, S., & Held, R.
of speech in infancy. Science, 218, 1138-1141. (1975). Orientational anisotropy in infant vision. Science,
Kuhl, P. K., & Padden, D. M. (1983). Enhanced discriminabil- 190, 900-902.
ity at the phonetic boundaries for the place feature in ma¬ Lefton, L. A. (1973). Metacontrast: A review. Perception &
caques. Journal of the Acoustical Society of America, 73, Psychophysics, 13, 161-171.
1003-1010. Lehmkuhle, S., & Fox, R. (1980). Effect of depth separation of
Kupchella, C. (1976). Sights and sounds. Indianapolis: Bobbs- metacontrast masking. Journal of Experimental Psychology:
Merrill. Human Perception & Performance, 6, 605-621.
Kurtz, D., & Butter, C. M. (1980). Impairments in visual dis¬ Lehmkuhle, S., Kratz, K. E., Mangel. S. C., & Sherman, S.
crimination performance and gaze shifts in monkeys with M. (1980). Spatial and temporal sensitivity of x- and y-cells
superior colliculus lesions. Brain Research, 196, 109-124. in dorsal lateral geniculate nucleus of the cat. Journal of
Kuyk, T., Veres, J. G., Ill, Lahey, M. A., & Clark, D. J. Neurophysiology, 43, 520-541.
(1986). The ability of protan color defectives to perform Lehmkuhle, S., Kratz, K. E., & Sherman, S. M. (1982). Spa¬
color-dependent air traffic control tasks. American Journal tial and temporal sensitivity of normal and amblyopic cats.
of Optometry & Physiological Optics, 63, 582-586. Journal of Neurophysiology, 48, 372-387.
Leibowitz, H. W., & Bourne, L. E. (1956). Time and intensity
as determiners of perceived shape. Journal of Experimental
Psychology, 51, 277-281.
LaBerge, D. (1983). Spatial extent of attention to letters and Leibowitz, H. W., & Harvey, L. O., Jr. (1969). Effect of in¬
words. Journal of Experimental Psychology: Human Per¬ structions, environment, and type of test object on matched
ception and Performance, 9, 371-379. size. Journal of Experimental Psychology, 81, 36-43.
Ladefoged, P. (1975). A course in phonetics. New York: Har- Leibowitz, H. W., & Moore, D. (1966). Role of changes in
court Brace Jovanovich. accommodation and convergence in the perception of size.
Lakowski, R. (1962). Is the deterioration of colour discrimina¬ Journal of the Optical Society of America, 56, 1120-1123.
tion with age due to lens or retinal changes? Farbe, 11, 69- Leibowitz, H. W., & Owens, D. A. (1977). Nighttime acci¬
86. dents and selective visual degradation. Science, 197 422-
Lakowski, R., Aspinall, P. A., & Kinnear, P. R. (1972). As¬ 423.
sociation between colour vision losses and diabetes mellitus. Leibowitz, H. W., & Pick, H. (1972). Cross-cultural and edu¬
Ophthalmic Research, 4, 145-159. cational aspects of the Ponzo perspective illusion. Percep¬
Lakowski, R., & Drance, S. M. (1979). Acquired dyschroma- tion & Psychophysics, 12, 430-432.
topsias: The earliest functional losses in glaucoma. Docu- Leibowitz, H. W., Post, R. B., Brandt, T., & Dichgans, J.
menta Ophthalmologica, Proceedings Series 19, 159-165. (1982). Implications of recent developments in dynamic spa¬
Lakowski, R., & Morton, B. A. (1977). The effect of oral con¬ tial orientation and visual resolution for vehicle guidance. In
traceptives on colour vision in diabetic women. Canadian A. H. Wertheim, W. A. Wagenaar. & H. W. Leibowitz
Journal of Ophthalmology, 12, 89-97. (Eds.), Tutorials on motion perception (pp. 231-260). New
Laming, D. (1986). Sensory Analysis. Orlando, FL: Academic York: Plenum.
Press. Leibowitz, H. W., Post, R. B., & Ginsburg, A. (1980). The
Lamour, Y., Wilier, J. C., & Guilbaud, G. (1983). Rat soma¬ role of fine detail in visually controlled behavior. Investiga¬
tosensory (SmI) cortex. I. Characteristics of neuronal re¬ tive Ophthalmology and Visual Science, 19, 846-848.
 References 561

Leibowitz, H. W., Shupert, C. L., Post, R. B., & Dichgans, J. smooth apparent motion. Journal of the Optical Society of
(1983). Autokinetic drifts and gaze deviation. Perception & America, 53, 302-306.
Psychophysics, 33, 455-459. Lie, 1. (1980). Visual detection and resolution as a function of
Leibowitz, H. W., Wilcox, S. B., & Post, R. B. (1978). The retinal locus. Vision Research, 20, 967-974.
effect of refractive error on size constancy and shape con¬ Liebeskind, J. C., & Melzack, R. (1987). The International
stancy. Perception, 7, 557-562. Pain Foundation: Meeting a need for education in pain man¬
Leinonen, L. (1983). Integration of somatosensory events in the agement. Pain, 30, 1.
posterior parietal cortex of the monkey. In C. von Euler, O. Lim, D. J. (1980). Cochlear anatomy related to cochlear micro¬
Franzen, U. Lindblom, & D. Ottoson (Eds.), Somatosensory mechanics: A review. Journal of the Acoustical Society of
mechanisms (pp. 113-124). New York: Plenum. America, 67, 1686-1695.
Lennie, P. (1980). Parallel visual pathways: A review. Vision Lindsay, P. H., & Norman, D. A. (1977). Human information
Research, 20, 561-594. processing (2nd ed.). New York: Academic Press.
LePage, E. L. (1987a). Frequency-dependent self-induced bias Linton, H., & Graham, E. (1959). Personality correlates of per-
of the basilar membrane and its potential for controlling sen¬ suasibility. In I. Janis (Eds.), Personality andpersuasibility.
sitivity and tuning in the mammalian cochlea. Journal of the New Haven, CT: Yale University Press.
Acoustical Society of America, 82, 139-154. Lipsett, L. P. (1977). Taste in human neonates: Its effect on
LePage, E. L. (1987b). A spatial template for the shape of tun¬ sucking and heart rate. In J. M. Weiffenbach (Ed.), Taste
ing curves in the mammalian cochlea. Journal of the Acoust¬ and development: The ontogeny of sweet preference (pp.
ical Society of America, 82, 155-164. 125-140). Washington, D.C.: U. S. Government Printing
Lerman, S. (1984). Biophysical aspects of corneal and lenticular Office.
transparency. Current Eye Research, 3, 3-14. Lisker, L., & Abramson, A. (1970). The voicing dimension:
LeVay, S., Wiesel, T. N., & Hubei, D. H. (1980). The devel¬ Some experiments in comparative phonetics. Proceedings of
opment of ocular dominance columns in normal and visually the 6th International Congress of Phonetic Sciences, 563-
deprived monkeys. Journal of Comparative Neurology, 191, 567.
1-51. Livingstone, M., & Hubei, D. (1988). Segregation of form,
Leventhal, A., & Hirsch, H. (1980). Receptive-field properties color, movement, and depth: Anatomy, physiology, and per¬
of different classes of neurons in visual cortex of normal and ception. Science, 240, 740-749.
dark-reared cats. Journal of Neurophysiology, 43, 1111 — Lockhead, G. R. (1966). Effects of dimensional redundancy on
1132. visual discrimination. Journal of Experimental Psychology,
Levin, S., Lipton, R. B., & Holtzman, P. S. (1981). Pursuit 72, 95-104.
eye movements in psychopathology: Effects of target char¬ Lockhead, G. R. (1970). Identification and the form of multi¬
acteristics. Biological Psychiatry, 16, 255-267. dimensional discrimination space. Journal of Experimental
Levine, M. W., & Shefner, J. M. (1981). Fundamentals of sen¬ Psychology, 85, 1-10.
sation and perception. Reading, MA: Addison-Wesley. Lockhead, G. R. (1972). Processing dimensional stimuli: A
Levy, D. L., Lipton, R. B., & Holzman, P. S. (1981). Smooth note. Psychological Review, 79, 410-419.
pursuit eye movements: Effects of alcohol and chloral hy¬ Lockhead, G. R. (1979). Holistic versus analytic process mod¬
drate. Journal of Psychiatric Research, 16, 1—11. els: A reply. Journal of Experimental Psychology: Human
Lewis, J. W., Terman, G. W., Shavit, Y.. Nelson, L. R., & Perception & Performance, 5, 746-755.
Liebeskind, J. C. (1984). Neural, neurochemical, and hor¬ Lockhead, G. R., & Byrd, R. (1981). Practically perfect pitch.
monal bases of stress-induced analgesia. In L. Kruger & J. Journal of the Acoustical Society of America, 70, 387-389.
C. Liebeskind (Eds.), Neural mechanisms of pain (pp. 277- Lockhead, G. R., & King, M. C. (1977). Classifying integral
288). New York: Raven. stimuli. Journal of Experimental Psychology: Human Per¬
Liberman, A. M., (1982). On finding that speech is special. ception and Performance, 3, 436-443.
American Psychologist, 37, 148-167. Loewenstein, W. R. (1960). Biological transducers. Scientific
Liberman, A. M., Cooper, F. S., Shankweiler, D. P., & Stud- American, 203, 98-108.
dert-Kennedy, M. (1967). Perception of the speech code. Loftus, E. (1974). Reconstructing memory: The incredible eye
Psychological Review, 74, 431-461. witness. Psychology Today, 8, 116-119.
Liberman, A. M., Harris, K. S., Hoffman, H. A., & Griffith, Loftus, E. (1979). Eyewitness testimony. Cambridge, MA: Har¬
B. C. (1957). The discrimination of sounds within and vard University Press.
across phoneme boundaries. Journal of Experimental Psy¬ Loftus, G., & Mackworth, N. (1978). Cognitive determinants
chology, 54, 358-368. of fixation location during picture viewing. Journal of Ex¬
Liberman, A. M., & Mattingly, I. G. (1985). The motor theory perimental Psychology: Human Perception and Perfor¬
of speech perception revised. Cognition, 21, 1-36. mance, 4, 565-572.
Liberman, M. C. (1982). Single-neuron labeling in the cat au¬ Lohman, D. F. (1986). The effect of speed-accuracy tradeoff on
ditory nerve. Science, 216, 1239-1241. sex differences in mental rotation. Perception & Psycho¬
Lichte, W. H., & Borresen, C. R. (1967). Influence of instruc¬ physics, 39, 427-436.
tions on degree of shape constancy. Journal of Experimental Long, G. R., & Cullen, J. K., Jr. (1985). Intensity difference
Psychology, 74, 538-542. limens at high frequencies. Journal of the Acoustical Society
Lichtenstein, M. (1963). Spatio-temporal factors in cessation of of America, 78, 507-513.
562 References

Loomis, J. M. (1981) Tactile pattern perception. Perception, MacLeod, D. I. (1978). Visual sensitivity. Annual Review of
10, 5-27. Psychology, 29, 613-645.
Loop, M. S. (1984). Effect of duration on detection by the chro¬ MacLeod, P. (1971). An experimental approach to the periph¬
matic and achromatic systems. Perception & Psychophysics, eral mechanisms of olfactory discrimination. In G. Ohloff &
36, 65-67. A. F. Thomas (Eds.), Gustation and olfaction (pp. 28-44).
Lowenstein, O., & Sand, A. (1940). The mechanism of the New York: Academic Press.
semicircular canal: A study of the responses of single-fiber MacLeod, R. (1947). The effects of “artificial penumbra” on
preparations to angular accelerations and to rotation at con¬ the brightness of included areas. In A. Michotte (Ed.), Mis¬
stant speed. Proceedings of the Royal Society of London, cellanea psychologica (pp. 1-22). Paris: Librairie Philoso-
Series B, 129, 256-275. phique.
Luce, R. D., Green, D. M., & Weber, D. L. (1976). Attention MacNichol, E. F., Jr. (1986). A unifying presentation of pho¬
bands in absolute identification. Perception & Psychophys¬ topigment spectra. Vision Research, 29, 543-546.
ics, 20, 49-54. Madden, T. M., & Burt, G. S. (1981). Inappropriate constancy
Luce, R. D., & Narens, L. (1987). Measurement scales on the scaling theory and the Mueller-Lyer illusion. Perceptual and
continuum. Science, 236, 1527-1532. Motor Skills, 52, 211-218.
Lumsden, E. (1980). Problems of magnification and minifica- Mair, R. G., Bouffard, J. A., Engen, T., & Morton, T. (1978).
tion: An explanation of the distortions of distance, slant, Olfactory sensitivity during the menstrual cycle. Sensory
shape, and velocity. In M. Hagen (Ed.), Perception of pic¬ Process, 2, 90-98.
tures: Vol. 1. Alberti’s window: The projective model of pic¬ Mandler, G. (1980). Recognizing: The judgment of previous oc¬
torial information (pp. 91-135). New York: Academic currence. Psychological Review, 87, 252-271.
Press. Manning, K. A., Riggs, L. A., & Komenda. J. K. (1983). Re¬
Luria, A. R. (1972). Memory disturbances in local brain le¬ flex eyeblinks and visual suppression. Perception & Psycho¬
sions. Neuropsychologia, 9, 367-375. physics, 34, 250-256.
Luria, A. R. (1973). The working brain. London: Penguin. Marks, L. E. (1968). Stimulus range, number of categories, and
Lynn, P. A., & Sayers, B. M. A. (1970). Cochlear innervation, form of the category scale. American Journal of Psychology,
signal processing, and their relation to auditory time-inten¬ 81, 467-479.
sity effects. Journal of the Acoustical Society of America, Marks, L. E. (1974). On scales of sensation: Prolegomena to
47, 523-533. any future psychophysics that will be able to come forth as
science. Perception & Psychophysics, 16, 358-376.
MacArthur, R. O., & Sekuler, R. (1982). Alcohol and motion Marks, L. E. (1979a). Summation of vibrotactile intensity: An
perception. Perception <6 Psychophysics, 31, 502-505. analogy to auditory critical bands? Sensory Processes, 3.
MacDonald, D. W., & Brown, R. E. (1985). Introduction: The 188-203.
pheromone concept in mammalian chemical communication. Marks, L. E. (1979b). A theory of loudness and loudness judg¬
In R. E. Brown & D. W. MacDonald (Eds.), Social odours ments. Psychological Review, 86, 256-285.
in mammals (Vol. 1, pp. 1-18). Oxford: Clarendon Press. Marks, L. E., Szczesiul, R., & Ohlott, P. (1986). On the cross-
MacFarlane, A. (1975). Olfaction in the development of social modal perception of intensity. Journal of Experimental Psy¬
preferences in the human neonate. In Ciba Foundation Sym¬ chology: Human Perception and Performance, 12, 517-534.
posium 33: The human neonate in parent-infant interaction Marks, W. B.. Dobelle, W. H., & MacNichol, E. F. (1964).
(pp. 103-177). Amsterdam: Elsevier. Visual pigments of single primate cones. Science, 143,
Mach, E. (1959). The analysis of sensations and the relation of 1181-1183.
the physical to the psychical. New York: Dover. (Originally Marley, A. A. J., & Cook, V. T. (1984). A fixed rehearsal
published 1886) capacity interpretation of limits on absolute identification
Mack, A., & Herman, E. (1972). A new illusion: The under¬ performance. British Journal of Mathematical and Statistical
estimation of a distance during pursuit eye movements. Per¬ Psychology, 37, 136-151.
ception & Psychophysics, 12, 471-473. Marr, D. (1974). The computation of lightness by the primate
Mack, A., Heuer, F., Fendrich, R., Vilardi, K., & Chambers, retina. Vision Research, 14, 1377-1388.
D. (1985). Induced motion and oculomotor capture. Journal Marr, D. (1982). Vision. San Francisco: Freeman.
of Experimental Psychology: Human Perception and Perfor¬ Marr, D., & Poggio, T. (1979). A computational theory of hu¬
mance, 11, 329-345. man stero vision. Proceedings of the Royal Society (London)
MacKain, K., Studdert-Kennedy, M., Spieker, S., & Stem, D. Series B(204), 301-328.
(1983). Infant intermodal speech perception is a left hemi¬ Marshall, D. A., & Moulton, D. G. (1981). Olfactory sensitiv¬
sphere function. Science, 219, 1347-1349. ity to a-ionine in humans and dogs. Chemical Senses, 6,
Mackworth, N. H. (1948). The breakdown of vigilance during 53-61.
prolonged visual search. Quarterly Journal of Experimental Marslen-Wilson, W. D. (1980). Speech understanding as a psy¬
Psychology, 1, 6-21. chological process. In J. C. Simon (Ed.), Spoken language
Mackworth, N. H., & Bruner, J. S. (1970). How adults and generation and understanding (pp. 39-67). Dordrecht: Rei-
children search and recognize pictures. Human Develop¬ del.
ment, 13, 149-177. Martin, D. K., & Holden, B. A. (1982). A new method for
 References 563

measuring the diameter of the in vivo human cornea. Amer¬ McBurney, D. H., & Bartoshuk, L. M. (1972). Water taste in
ican Journal of Optometry and Physiological Optics, 59, mammals. In D. Schneider (Ed.), Olfaction and taste IV
436-441. (pp. 329-335). Wissenshafliche Verlagsgesellschaft MBH.
Martin, M. (1979). Local and global processing: The role of McBumey, D. H., & Gent, J. F. (1979). On the nature of taste
sparsity. Memory and Cognition, 7, 476-484. qualities. Psychological Bulletin, 86, 151-167.
Masica, D. N., Money, J., Ehrhardt, A. A., & Lewis, V. G. McBumey, D. H., Levine, J. M., & Cavanaugh, P. H. (1977).
(1969). IQ, fetal sex hormones and cognitive patterns: Stud¬ Psychophysical and social ratings of human body odor. Per¬
ies in testicular feminizing syndrome of androgen insensitiv¬ sonality and Social Psychology Bulletin, 3, 135-138.
ity. Johns Hopkins Medical Journal, 124, 34. McCall, R. B. (1979). Individual differences in the pattern of
Masland, R. H. (1986). The functional architecture of the ret¬ habituation at five and ten months of age. Developmental
ina. Scientific American, 255, 102-111. Psychology, 15, 559-569.
Massaro, D. W. (1987). Speech perception by ear and eye: A McClain, L. (1983). Interval estimation: Effect of processing
paradigm for psychological inquiry. Hillsdale, NJ: Law¬ demands on prospective and retrospective reports. Percep¬
rence Erlbaum. tion & Psychophysics, 34, 185-189.
Mather, J. A., & Fisk, J. D. (1985). Orienting to targets by McClelland, J. L., & Elman, J. L. (1986). The TRACE model
looking and pointing: Parallels and interactions in ocular and of speech perception. Cognitive Psychology, 18, 1-86.
manual performance. Quarterly Journal of Experimental McClintock, M. K. (1971). Menstrual synchrony and suppres¬
Psychology, 37A, 315-338. sion. Nature (London), 229, 244-245.
Matin, L. (1982). Visual localization and eye movements. In McColgin, F. H. (1960). Movement threshold in peripheral vi¬
A. H. Wertheim, W. A. Wagenaar, & H. W. Leibowitz sion. Journal of the Optical Society of America, 50, 774-
(Eds.), Tutorials on motion perception (pp. 101-156). New 779.
York: Plenum. McCollough, C. (1965). Color adaptation of edge detectors in
Matin. L., & MacKinnon, G. E. (1964). Autokinetic move¬ the human visual system. Science, 149, 1115-1116.
ment: Selective manipulation of directional components by McCready, D. (1986). Moon illusions redescribed. Perception
image stabilization. Science, 143, 147-148. & Psychophysics, 39, 64-72.
Matthews, B. H. C. (1933). Nerve endings in mammalian mus¬ McFarland, R. A., Domey, R. G., Warren, A. B., & Ward,
cle. Journal of Physiology (London), 78, 1-53. D. C. (1960). Dark-adaptation as a function of age. I. A.
Maunsell, J. H. R., & van Essen, D. C. (1983). Functional statistical analysis. Journal of Gerontology, 15, 149-154.
properties of neurons in middle temporal visual area of the McGee, M. G. (1979). Human spatial abilities: Psychometric
macaque monkey. I. Selectivity for stimulus direction, studies and environmental, genetic, hormonal, and neurolog¬
speed, and orientation. Journal of Neurophysiology, 49, ical influences. Psychological Bulletin, 86, 889-918.
1127-1147. McGlone, J. (1981). Sexual variations in behavior during spatial
Maurer, D. (1975). Infant visual perception: Methods of study. and verbal tasks. Canadian Journal of Psychology, 35, 277-
In L. B. Cohen & P. Salapatek (Eds.), Infant perception: 282.
From sensation to cognition, basic visual processes (Vol. 1, McGuinness. D. (1972). Hearing: Individual differences in per¬
pp. 1-77). New York: Academic Press. ceiving. Perception, 1, 465-473.
Maurer, D., & Barrera, M. (1981). Infant’s perception of natu¬ McGuinness, D. (1976a). Away from a unisex psychology: In¬
ral and distorted arrangements of a schematic face. Child dividual differences in visual sensory and perceptual pro¬
Development, 52, 196-202. cesses. Perception, 5, 279-294.
Maurer. D., & Lewis, T. L. (1979). A physiological explana¬ McGuinness, D. (1976b). Sex differences in the organization of
tion of infants’ early visual development. Canadian Journal perception and cognition. In B. Lloyd & U. Archer (Eds.),
of Psychology, 33, 232-251. Exploring sex differences (pp. 123-156). New York: Aca¬
Maurer, D., & Martello. M. (1980). The discrimination of ori¬ demic Press.
entation by young infants. Vision Research, 20, 201-204. McGuinness, D., & Lewis, I. (1976). Sex differences in visual
Maurer, D., & Salapatek, P. (1976). Development changes in persistence: Experiments on the Ganzfeld and the after-im¬
the scanning of faces by young infants. Child Development, age. Perception, 5, 295-301.
47, 523-527. McGurk, H., & MacDonald, J. (1976). Hearing lips and seeing
Maxwell, J. C. (1873). Treatise on electricity and magnetism. voices. Nature, 264, 746-748.
Oxford: Clarendon Press. Meer, H. C. van der. (1979). Interrelation of the effects of bin¬
Mayer, D. J., & Watkins, L. R. (1984). Multiple endogenous ocular disparity and perspective cues on judgments of depth
opiate and nonopiate analgesia systems. In L. Kruger & and height. Perception & Psychophysics, 26, 481-488.
J. C. Liebeskind (Eds.), Neural mechanisms of pain (pp. Meiselman, H. L., Bose, H. E., & Nykvist, W. F. (1972).
253-276). New York: Raven. Magnitude production and magnitude estimation of taste in¬
Mayhew, J. E. W., & Frisby, J. P. (1980). The computation of tensity. Perception & Psychophysics, 12, 249-252.
binocular edges. Perception, 9, 69-86. Melzack, R., & Casey, K. L. (1968). Sensory, motivational,
McBumey, D. H. (1969). Effects of adaptation on human taste and central control determinants of pain. In D. R. Kenshalo
function. In C. Pfaffman (Ed.), Olfaction and taste III (pp. (Ed.), The skin senses (pp. 423-443). Springfield, IL:
407-419). New York: Rockefeller University Press. Thomas.
564 References

Melzack, R., & Wall, P. D. (1965). Pain mechanisms: A new chology, 77, 257-263.
theory. Science, 150, 971-979. Miles, F. A., & Fuller, J. E. (1975). Visual tracking and the
Melzack, R., & Wall, P. D. (1982). The challenge of pain. primate flocculus. Science, 189, 1000-1002.
Harmondsworth: Penguin. Milewski, A. E. (1976). Infant’s discrimination of internal and
Melzack, R., Wall, P. D., & Ty, T. C. (1982). Acute pain in external pattern elements. Journal of Experimental Child
an emergency clinic: Latency of onset and descriptor pat¬ Psychology, 22, 229-246.
terns related to different injuries. Pain, 14, 33-43. Mill, J. (1829). Analysis of the phenomena of the human mind.
Mergner, R., Anastasopoulos, D., Becker, W., & Deecke, L. London.
(1981). Discrimination between trunk and head rotation: A Millan, M. J. (1986). Multiple opioid systems and pain. Pain,
study comparing neuronal data from the cat with human psy¬ 27, 303-347.
chophysics. Acta Psychologica, 48, 291-302. Millar, J. M., & Whitaker, H. A. (1983). The right hemi¬
Merkel, J. (1885). Die zeitlichen Verhaltnisse der Willenstatig- sphere’s contribution to language: A review of the evidence
keit. Philosophische Studien (Wundt), 2, 73-127. from brain-damaged subjects. In S. Sagalowitz (Ed.), Lan¬
Mershon, D. H., & Bowers, J. N. (1979). Absolute and relative guage function and brain organization (pp. 87-114). New
cues for the auditory perception of egocentric distance. Per¬ York: Academic Press.
ception, 8, 311-322. Miller, D. L., Moore, R. K., & Wooten, B. R. (1984). When
Mershon, D. H., Desaulniers, D. H., & Amerson, T. L., Jr. push comes to pull: Impressions of visual direction. Percep¬
(1980). Visual capture in auditory distance perception: Prox¬ tion & Psychophysics, 36, 396-397.
imity image effect reconsidered. Journal of Auditory Re¬ Miller, G. A. (1947). Sensitivity to changes in the intensity of
search, 20, 129-136. white noise and its relation to masking and loudness. Jour¬
Mershon, D. H., Desaulniers, D. H., Kiefer, S. A., & Amer¬ nal of the Acoustical Society of America, 19, 609-619.
son, T. L., Jr. (1981). Perceived loudness and visually de¬ Miller, G. A. (1956). The magical number seven, plus or minus
termined auditory distance. Perception, 10, 531-543. two: Some limits on our capacity for processing information.
Mershon, D. H., & Gogel, W. C. (1970). Effect of stereoscopic Psychological Review, 63, 81-97.
cues on perceived whiteness. American Journal of Psychol¬ Miller, G. W., Hicks, R. E., & Willette, M. (1978). Effects of
ogy, 83, 55-67. concurrent verbal rehearsal and temporal set upon judgments
Mershon, D. H., & King, L. E. (1975). Intensity and reverber¬ of temporal duration. Acta Psychologica, 42, 173-179.
ation as factors in the auditory perception of egocentric dis¬ Miller, J. (1982). Divided attention: Evidence for coactivation
tance. Perception & Psychophysics, 18, 409-415. with redundant signals. Cognitive Psychology, 14, 247-279.
Metzler, D. E., & Harris, C. M. (1978). Shapes of spectral Miller, J. L., & Liberman, A. M. (1979). Some effects of later-
bands of visual pigments. Vision Research, 18, 1417-1420. occurring information on the perception of stop consonants
Michael, C. R. (1981). Columnar organization of color cells in and semivowel. Perception & Psychophysics, 25, 457-465.
monkey’s striate cortex. Journal of Neurophysiology, 46, Miller, N. D. (1965). Visual recovery from brief exposures to
587-604. high luminance. Journal of the Optical Society of America,
Michael, C. R. (1985). Laminar segregation of color cells in the 55, 1661-1669.
monkey’s striate cortex. Vision Research, 25, 415-423. Miller, R. J., Pigion, R. G., & Martin, K. D. (1985). The ef¬
Michael, R. P., & Keveme, E. B. (1968). Pheromones in the fects of ingested alcohol on accommodation. Perception &
communication of sexual status in primates. Nature, 218, Psychophysics, 37, 407-414.
746-749. Mills, A. W. (1958). On the minimum audible angle. Journal
Michael, R. P., Keveme, E. B., & Bonsall, R. W. (1971). of the Acoustical Society of America, 30, 127-246.
Pheromones: Isolation of male sex attractants from a female Mills, A. W. (1960). Lateralization of high-frequency tones.
primate. Science, 172, 964-966. Journal of the Acoustical Society of America, 32, 132-134.
Michael, S., & Sherrick, M. F. (1986). Perception of induced Milne, J. & Milne, M. (1967). The senses of animals and men.
visual motion: Effects of relative position, shape and size of New York: Atheneum.
the surround. Canadian Journal of Psychology, 40, 122— Mitchell, D. (1978). Effect of early visual experience on the
125. development of certain perceptual abilities in animals and
Michaels, C. F., & Carello, C. (1981). Direct perception. En¬ man. In R. Walk & H. Pick (Eds.), Perception and experi¬
glewood Cliffs, NJ: Prentice-Hall. ence, New York: Plenum.
Michell, J. (1986). Measurement scales and statistics: A clash Mitchell, D. (1980). The influence of early visual experience on
of paradigms. Psychological Bulletin, 100, 398-407. visual perception. In C. Harris (Ed.), Visual coding and
Michon, J. (1985). The compleat time experiencer. In J. A. adaptability (pp. 1-50). Hillsdale, NJ: Lawrence Erlbaum.
Michon & J. L. Jackson (Eds.), Time, mind and behavior Mitchell, D. (1981). Sensitive periods in visual development. In
(pp. 20-52). Berlin: Springer-Verlag. R. Aslin, J. Alberts, & M. Petersen (Eds.), Development of
Mikaelian, H. (1974). Adaptation to displaced hearing: A non- perception (pp. 1-43). New York: Academic Press.
proprioceptive change. Journal of Experimental Psychology, Monahan, J. S., & Lockhead, G. R. (1977). Identification of
103, 326-330. integral stimuli. Journal of Experimental Psychology: Gen¬
Mikaelian, H., & Held, R. (1964). Two types of adaptation to eral, 106, 94-110.
an optically-rotated visual field. American Journal of Psy¬ Moncrieff, R. W. (1956). Olfactory adaptation and colour like-
 References 565

ness. Journal of Physiology (London), 133, 301-316. Murphy, C., & Cain, W. S. (1980). Taste and olfaction: Inde¬
Money, J. (1965). Psychosexual differentiation. In J. Money pendence vs. interaction. Physiology and Behavior, 24,
(Ed.), Sex research: New developments (pp. 3-23). New 601-605.
York: Holt. Murphy, T. D., & Eriksen, C. W. (1987). Temporal changes in
Montellese, S., Sharpe, L. T., & Brown, J. L. (1979). Changes the distribution of attention in the visual field in response to
in critical duration during dark-adaptation. Vision Research, precues. Perception & Psychophysics, 42, 576-586.
19, 1147-1153. Mustillo, P. (1985). Binocular mechanisms mediating crossed
Montgomery, J. C., & MacDonald, J. A. (1987). Sensory tun¬ and uncrossed stereopsis. Psychological Bulletin, 97, 187—
ing of lateral line receptors in Antarctic fish to the move¬ 201.
ments of planktonic prey. Science, 235, 195-196. Myers, A. K. (1982). Psychophysical scaling and scales of
Mooney. R. D., Dubin, M. W., & Russoff, A. C. (1979). In- physical stimulus measurement. Psychological Bulletin, 92,
temeuron circuits in the lateral geniculate nucleus of mon- 203-214.
ocularly deprived cats. Journal of Comparative Neurology,
187(3), 533-544. Nagy, A. L. (1980). Short-flash Bezold-Brucke hue shifts. Vi¬
Moore, B. (1977). Introduction to the psychology of hearing. sion Research, 20, 361-368.
Baltimore: University Park Press. Naka, Ken-Ichi (1982). The cells horizontal cells talk to. Vision
Moore, L. M., Nielson, C. R., & Mistretta, C. M. (1982). Su¬ Research, 22, 653-660.
crose taste thresholds: Age-related differences. Journal of Narens, L., & Luce, R. D. (1986). Measurement: The theory
Gerontology, 37, 64-69. of numerical assignments. Psychological Bulletin, 99, 166—
Moran, J., & Gordon, B. (1982). Long term visual deprivation 180.
in a human. Vision Research, 22, 27-36. Nathans, J., Piantanida, T. P., Eddy, R. L., Shows, T. B., &
Moray, N. (1959). Attention in dichotic listening: Affective Hogness, D. S. (1986). Molecular genetics of inherited
cues and the influence of instructions. Quarterly Journal of variation in human color vision. Science, 232, 203-210.
Experimental Psychology, 11, 56-60. Navon, D. (1977). Forest before trees: The precedence of global
Moray, N. (1969). Attention: Selective processes in vision and features in visual perception. Cognitive Psychology, 9, 353-
hearing. London: Hutchinson Educational Ltd. 383.
Morris, V., & Morris, P. E. (1985). The influence of question Navon, D., & Gopher, D. (1979). On the economy of the hu¬
order on eyewitness accuracy. British Journal of Psychol¬ man-processing system. Psychological Review, 86, 214—
ogy, 76, 365-371. 255.
Morrison. J. D., & Whiteside, T. C. D. (1984). Binocular cues Neff, D. L.. & Green, D. M. (1987). Masking produced by
in the perception of distance of a point source of light. Per¬ spectral uncertainty with multicomponent maskers. Percep¬
ception, 13, 555-566. tion & Psychophysics, 41, 409-415.
Moskowitz, H., Sharma, S., & McGlothlin, W. (1972). Effect Neisser, U. (1967). Cognitive psychology. New York: Apple-
of marijuana upon peripheral vision as a function of the in¬ ton.
formation processing demands in central vision. Perceptual Neisser, U. (1976). Cognition and reality: Principles and impli¬
and Motor Skills, 35, 875. cations of cognitive psychology. San Francisco: Freeman.
Moskowitz-Cook, A. (1979). The development of photopic Neisser, U., & Becklin, R. (1975). Selective looking: Attending
spectral sensitivity in human infants. Vision Research, 9, to visually specified events. Cognitive Psychology, 7, 480-
113-114. 494.
Mountcastle, V. B., Poggio, G. F.. & Werner, G. (1963). The Nelson, R., Kolb, H., Robinson, M. M., & Mariani, A. P.
relation of thalamic cell response to peripheral stimuli varied (1981). Neural circuity of the cat retina: Cone pathways to
over an intensive continuum. Journal of Neurophysiology, ganglion cells. Vision Research, 21, 1527-1537.
26, 807-834. Neuweiler, G., Bruns, V., & Schuller, G. (1980). Ears adapted
Mountcastle, V. B., & Powell, T. P. S. (1959). Central nervous for the detection of motion; or how echolocating bats have
mechanisms subserving position sense and kinesthesis. Bul¬ exploited the capacities of the mammalian auditory system.
letin of the Johns Hopkins Hospital, 105, 173-200. Journal of the Acoustical Society of America, 68, 741-753.
Movshon, J. A., & van Sluyters, R. C. (1981). Visual neural Neville. H. J. (1985). Effects of early sensory and language
development. Annual Review of Psychology, 32, All-522. experience on the development of the human brain. In J.
Mozel, M. M.. Smith, B., Smith, P., Sullivan, R., & Swender, Mehler & R. Fox (Eds.), Neonate cognition: Beyond the
P. (1969). Nasal chemoreception in flavor identification. Ar¬ blooming buzzing confusion (pp. 349-364). Hillsdale, NJ:
chives of Otolaryngology, 90, 367-373. Lawrence Erlbaum.
Muir, D., & Field, j. (1979). Newborn infants orient to sounds. Neville, H. J., Schmidt, A., & Kutas, M. (1983). Altered vi¬
Child Development, 50, 431-436. sual evoked potentials in congenitally deaf adults. Brain Re¬
Mulligan, R. M., & Schiffman, H. R. (1979). Temporal expe¬ search, 266, 127-132.
rience as a function of organization in memory. Bulletin of Newhall, S. M., Bumham, R. W., & Clark, J. R. (1957). Com¬
the Psychonomic Society, 14, 417-420. parison of successive with simultaneous color matching.
Munsell, A. H. (1915). Atlas of the Munsell color system. Mal- Journal of the Optical Society of America, 47, 43-56.
din, MA: Wadsworth, Howland. Newhall, S. M., Nickerson, D., & Judd, D. B. (1943). Final
 566 References

report of the O.S.A. subcommittee on spacing of the Mun- visual corticotectal systems in the cat. Journal of Neuro¬
sell colors. Journal of the Optical Society of America, 33, physiology, 52, 1226-1245.
385-418. Ohala, J. J. (1986). Phonological evidence for top-down pro¬
Newland, J. (1972). Children's knowledge of left and right. cessing in speech perception. In J. S. Perkell & D. H. Klatt
Unpublished master’s thesis, University of Auckland. Cited (Eds.), Invariance and variability in speech processes (pp.
in M. C. Corballis & J. L. Beale (1976). The psychology of 386-397). Hillsdale, NJ: Lawrence Erlbaum.
left and right (p. 167). Hillsdale, NJ: Lawrence Erlbaum. Oldfield, S. R., & Parker, S. P. A. (1984). Acuity of sound
Noble, W., & Gates, A. (1985). Accuracy, latency, and lis¬ localisation: A topography of auditory space. II. Pinna cues
tener-search behavior in localization in the horizontal and absent. Perception, 13, 601-617.
vertical planes. Journal of the Acoustical Society of Amer¬ Oldfield, S. R., & Parker, S. P. A. (1986). Acuity of sound
ica, 78, 2005-2012. localisation: A topography of auditory space. III. Monaural
Noda, H., Freeman, R. B., & Creutzfeldt, O. D. (1972). Neu¬ hearing conditions. Perception, 15, 67-81
ronal correlates of eye movements in the cat visual cortex. Olsho, L. W. (1984). Infant frequency discrimination. Infant
Science, 175, 661-664. Behavior and Development, 7, 27-35.
Noell, W. (1980). Possible mechanisms of photoreceptor dam¬ Olson, R., & Attneave, F. (1970). What variables produce sim¬
age by light in mammalian eyes. Vision Research, 20, ilarity grouping? American Journal of Psychology, 83, 1-
1163-1172. 21.
Norman, D. A. (1968). Toward a theory of memory and atten¬ Olzak, L. (1986). Widely separated spatial frequencies: Mech¬
tion. Psychological Review, 75, 522-536. anism interactions. Vision Research, 26, 1143-1154.
Norman, D. A. (1969). Memory while shadowing. Quarterly O’Mahony, M. (1979). Salt taste adaptation: The psychophysi¬
Journal of Experimental Psychology, 21, 85-93. cal effects of adapting solutions and residual stimuli from
Norman, D. A. (1976). Memory and attention. (2nd ed.). New prior tastings on the taste of sodium chloride. Perception, 8,
York: Wiley. 441-476.
Norman, D. A., Rumelhart, D. E., and the LNR Research O’Mahony, M., & Heintz, C. (1981). Direct magnitude esti¬
Group. (1975). Explorations in Cognition. San Francisco: mation of salt taste intensity with continuous correction for
Freeman. salivary adaptation. Chemical Senses, 6, 101-112.
Norton, T. T. (1981a). Development of the visual system and Ono, H. (1969). Apparent distance as a function of familiar
visually guided behavior. In R. Aslin, J. Alberts, & M. Pe¬ size. Journal of Experimental Psychology, 79, 109-115.
tersen (Eds.), Development of perception: Psychobiological Ono, H., & Comerford, T. (1977). Stereoscopic depth con¬
perspectives: Vol. 2. The visual system (pp. 113-156). New stancy. In W. Epstein (Ed.), Stability and constancy in vi¬
York: Academic Press. sual perception: Mechanisms and processes. New York:
Norton, T. T. (1981b). Geniculate and extrageniculate visual Wiley.
systems in the tree shrew. In A. R. Morrison & P. L. Strick Ono, H., & Weber, E. U. (1981). Nonveridical visual direction
(Eds.), Changing concepts of the nervous system (pp. 377- produced by monocular viewing. Journal of Experimental
410). New York: Academic Press. Psychology: Human Perception & Performance, 7, 937—
Norwich, K. H. (1981). The magical number seven: Making a 947.
“bit” of “sense”. Perception & Psychophysics, 29, 409- Ono, M. E., Rivest, J., & Ono, H. (1986). Depth perception as
422. a function of motion parallax and absolute-distance infor¬
Norwich, K. H. (1983). To perceive is to doubt: The relativity mation. Journal of Experimental Psychology: Human Per¬
of perception. Journal of Theoretical Biology, 102, 175— ception and Performance, 12, 331-337.
190. Orban, G. A. (1984). Neuronal operations in the visual cortex.
Norwich, K. H. (1984). The psychophysics of taste from the Berlin: Springer-Verlag.
entropy of the stimulus. Perception & Psychophysics, 35, Orban, G. A., Kennedy, H., & Maes, H. (1981a). Response to
269-278. movement of neurons in areas 17 and 18 of the cat: Velocity
Norwich, K. H. (1987). On the theory of Weber fractions. Per¬ sensitivity. Journal of Neurophysiology, 45, 1043-1058.
ception & Psychophysics, 42, 286-298. Orban, G. A., Kennedy, H., & Maes, H. (1981b). Response to
Nusbaum, H. C., & Schwab, E. C. (1986). The role of atten¬ movement of neurons in areas 17 and 18 of the cat: Direc¬
tion and active processing in speech perception. In E. C. tion sensitivity. Journal of Neurophysiology, 45, 1059-
Schwab & H. C. Nusbaum (Eds.), Pattern recognition by 1073.
humans and machines: Vol. 1. Speech Perception (pp. 113- Omstein, R. E. (1969). On the experience of time. London:
157). Orlando, FL: Academic Press. Penguin.
Osaka. N. (1981). Brightness exponent as a function of flash
Oakley, B. (1985). Taste responses of human chorda tympani duration and retinal eccentricity. Perception & Psychophys¬
nerve. Chemical Senses, 10, 469-481. ics, 30, 144-148.
O’Connell, R. J., & Mozell, M. M. (1969). Quantitative stim¬ Osterberg, G. (1935). Topography of the layer of rods and
ulation of frog olfactory receptors. Journal of Neurophysi¬ cones in the human retina. Acta Ophthalmologica,
ology, 32, 51-63. 6(Suppl.).
Ogasawara, K., McHaftie, J. G., & Stein, B. E. (1984). Two Ostfeld, A. (1961). Effects of LSD-25 and JB318 on tests of
 References 567

visual and perceptual functions in man. Federation Proceed¬ quality effects in previewed displays. Journal of Experimen¬
ings, Federation of American Societies for Experimental Bi¬ tal Psychology: Human Perception and Performance, 10,
ology, 20, 876-883. 429-448.
Ottoson, D. (1956). Analysis of the electrical activity of the Pashler, H. (1987). Detecting conjunctions of color and form:
olfactory epithelium. Acta Physiologica Scandinavica, Reassessing the serial search hypothesis. Perception & Psy¬
5J(Suppl. 122), 1-83. chophysics, 41, 191-201.
Owens, M. E. (1984). Pain in infancy: Conceptual and meth¬ Pasnak, R., Tyer, Z. A., & Allen, J. A. (1985). Effect of dis¬
odological issues. Pain, 20, 213-220. tance instructions on size judgments. American Journal of
Owsley, C. (1983). The role of motion in infants’ perception of Psychology, 98, 297-304.
solid shape. Perception, 12, 707-717. Pastore, R. E., Schmeckler, M. A., Rosenblum, L., & Szczes-
Oyama, T. (1968). A behavioristic analysis of Steven’s magni¬ iul, R. (1983). Duplex perception with musical stimuli. Per¬
tude estimation method. Perception & Psychophysics, 3, ception & Psychophysics, 33, 469-474.
317-320. Patterson, R. D. (1969). Noise masking of a change in residue
Oyama, T. (1986). The effect of stimulus organization on nu- pitch. Journal of the Acoustical Society of America, 45,
merosity discrimination. Japanese Psychological Research, 1520-1524.
28, 77-86. Paulus, K., & Haas, E. M. (1980). The influence of solvent
viscosity on the threshold values of primary tastes. Chemical
Senses, 5, 23-32.
Palmer, A. R., Winter, I. M., & Darwin, C. J. (1986). The Paulus, K., & Reisch, A. M. (1980). The influence of temper¬
representation of steady-state vowel sounds in the temporal ature on the threshold values of primary tastes. Chemical
discharge pattern of the guinea pig cochlear nerve and pri¬ Senses, 5, 11—21.
marylike cochlear nucleus neurons. Journal of the Acousti¬ Peeples, D. R., & Teller, D. Y. (1978). White-adapted pho-
cal Society of America, 79, 100-113. topic spectral sensitivity in human infants. Vision Research,
Palmer, C., & Krumhansl, C. L. (1987). Independent temporal 18, 39-53.
and pitch structures in determination of musical phrases. Peichl, L., & Wassle, H. (1979). Size, scatter and coverage of
Journal of Experimental Psychology: Human Perception and ganglion-cell receptive-field centers in the cat retina. Journal
Performance, 13, 116—126. of Physiology (London), 291, 117.
Palmer, J. (1986). Mechanisms of displacement discrimination Pelosi, P., & Pisanelli, A. M. (1981). Specific anosmia to 1,8-
with and without perceived movement. Journal of Experi¬ cineole: The camphor primary odor. Chemical Senses, 6,
mental Psychology: Human Perception and Performance, 87-93.
12, 411-421. Penfield, W., & Rasmussen, T. (1950). The cerebral cortex of
Palmer, S. E. (1975a). The effects of contextual scenes on the man. New York: Macmillan.
identification of objects. Memory and Cognition, 3, 519— Perkell, J. S., & Klatt, D. H. (Eds.). (1986). Invariance and
526. variability in speech processes. Hillsdale, NJ: Lawrence Erl-
Palmer, S. E. (1975b). Visual perception and world knowledge: baum.
notes on a model of sensory-cognitive interaction. In D. A. Perl, E. R. (1984). Characterization of nociceptors and their ac¬
Norman & D. E. Rumelhart (Eds.), Explorations in cogni¬ tivation of neurons in the superficial dorsal horn: First steps
tion (pp. 297-307). San Francisco: Freeman. for the sensation of pain. In L. Kruger & J. C. Liebeskind
Paquet, L., & Merikle, P. M. (1984). Global precedence: The (Eds.), Neural mechanisms of pain (pp. 23-52). New York:
effect of exposure duration. Canadian Journal of Psychol¬ Raven.
ogy, 38, 45—53. Peterson, A. C. (1976). Physical androgyny and cognitive func¬
Parasuraman, R. (1984). Sustained attention in detection and tioning in adolescence. Developmental Psychology, 12,
discrimination. In R. Parasuraman & D. R. Davies (Eds.), 524-533.
Varieties of attention, (pp. 243—271). Orlando, FL: Aca¬ Peterson, A. C., & Crockett, L. (1985, August). Factors influ¬
demic Press. encing sex differences in spatial ability during adolescence.
Parasuraman, R., & Mouloua, M. (1987). Interaction of signal In S. L. Willis (Chair.), Sex differences in spatial ability
discriminability and task type in vigilance decrement. Per¬ across the lifespan. Symposium conducted at the Ninety-
ception & Psychophysics, 41, 17-22. third Annual Convention of the American Psychological As¬
Parducci, A. (1965). Category judgment: A range-frequency sociation, Los Angeles.
model. Psychological Review, 72, 407-418. Petrig, B., Julesz, B., Kropfl, W., Baumgartner, G., & Anli-
Parker, D. E. (1980). The vestibular apparatus. Scientific Amer¬ ker, M. (1981). Development of stereopsis and cortical bin-
ican, 243, 118-135. ocularity in human infants: Electrophysiological evidence.
Parks, T. E. (1965). Post-retinal visual storage. American Jour¬ Science, 213, 1402-2405.
nal of Psychology, 78, 145-147. Petry, S., & Meyer, G. E. (1987). The perception of illusory
Parlee, M. B. (1983). Menstrual rhythms in sensory processes: contours. New York: Springer-Verlag.
A review of fluctuations in vision, olfaction, audition, taste Pettigrew, J. (1978a). Stereoscopic visual processing. Nature,
and touch. Psychological Bulletin, 93, 539-548. 273, 9-11.
Pashler, H. (1984). Evidence against late selection: Stimulus Pettigrew, J. (1978b). The paradox of the critical period for
568 References

striate cortex. In C. W. Cotman (Ed.), Neuronal plasticity L. M. (1985). Speech perception, word recognition and the
(pp. 311-330). New York: Raven. structure of the lexicon. Speech Communication, 4, 75-95.
Pfaff, D. (1968). Effects of temperature and time of day on Pizorrolo, F. J. (1978). The neuropsychology of developmental
judgment. Journal of Experimental Psychology, 76, 419— reading disorders. New York: Praeger.
422. Plateau, M. H. (1872). Sur la mesure des sensations physiques,
Pfaffman, C. (1955). Gustatory nerve impulses in rat, cat, and et sur la loi qui lie l'intensite de la cause excitante. Bulletin
rabbit. Journal of Neurophysiology, 18, 429-440. de TAcademie Royale de Belgique, 33, 376-388.
Pfaffman, C. (1974). Specificity of the sweet receptors of the Podgomy, P., & Shepard, R. N. (1983). Distribution of visual
squirrel monkey. Chemical Senses and Flavor, 1, 61-67. attention over space. Journal of Experimental Psychology:
Pfaffman, C., Bartoshuk, L., & McBumey, D. H. (1971). Human Perception and Performance, 9, 380-393.
Taste psychophysics. Handbook of Sensory Physiology, 1, Poggio, G. F., & Mountcastle, V. B. (1960). A study of the
75-101. functional contributions of the lemniscal and spinothalamic
Pfaffman, C., Frank, M., & Norgren, R. (1979). Neural mech¬ systems to somatic sensibility: Central nervous mechanisms
anisms and behavioral aspects of taste. Annual Review of in pain. Bulletin of the Johns Hopkins Hospital, 106, 266-
Psychology, 30, 283-325. 316.
Pfeiffer, R. R. (1966). Classification of response patterns of Pokomy, J., & Smith, V. C. (1986). Eye disease and color
spike discharges for units in the cochlear nucleus: Tone- defects. Vision Research, 26, 1573-1584.
burst stimulation. Experimental Brain Research, 1, 220- Pollack, I. (1952). The information of elementary auditory dis¬
235. plays. Journal of the Acoustical Society of America, 24,
Phillips, C. G., Zeki, S., & Barlow, H. B. (1984). Localization 745-749.
of function in the cerebral cortex. Brain, 107, 328-360. Pollack, I. (1953). The information of elementary auditory dis¬
Phillips, D. P., & Brugge, J. F. (1985). Progress in neurophys¬ plays. II. Journal of the Acoustical Society of America, 25,
iology of sound localization. Annual Review of Psychology, 765-769.
36, 245-274. Pollack, I. (1978). Decoupling of auditory pitch and stimulus
Phillipson, O. T., & Harris, J. P. (1984). Effects of chloro- frequency: The Shepard demonstration revisited. Journal of
promazine and promazine on the perception of some multi¬ the Acoustical Society of America, 63, 202-206.
stable visual figures. Quarterly Journal of Experimental Psy¬ Pollack, I., & Pickett, J. M. (1964). Intelligibility of excerpts
chology, 36A, 291-308. from fluent speech: Auditory vs. structural context. Journal
Piaget, J. (1969). The mechanisms of perception. (G. N. Sea- of Verbal Learning and Verbal Behavior, 3, 79-84.
grine, Trans.). New York: Oxford University Press. Pollack, R. H., & Silvar, S. D. (1967). Magnitude of the Muel-
Pick, H. L. (1987). Information and the effects of early percep¬ ler-Lyer illusion in children as a function of pigmentation of
tual experience. In N. Eisenberg (Ed.), Contemporary topics fundus oculi. Psychonomic Science, 8, 83-84.
in developmental psychology (pp. 59-76). New York: Pollen, D. A., Lee, J. R., & Taylor J. H. (1971). How does
Wiley. the visual cortex begin the reconstruction of the visual
Pick, H. L., Jr., & Hay, J. C. (1965). A passive test of the world? Science, 173, 74-77.
Held reafference hypothesis. Perceptual and Motor Skills, Poltrock, S. E., Lansman, M., & Hunt, E. (1982). Automatic
20, 1070-1072. and controlled attention processes in auditory target detec¬
Pick, H. L., Jr., & Pick, A. D. (1970). Sensory and perceptual tion. Journal of Experimental Psychology: Human Percep¬
development. In P. H. Mussen (Ed.), Carmichael’s manual tion and Performance, 8, 37-45.
of child development (pp. 773-848). New York: Wiley. Pomerantz, J. R. (1983). Global and local precedence: Selective
Pierce, J. R. (1983). The science of musical sound. New York: attention in form and motion perception. Journal of Experi¬
Scientific American Books. mental Psychology: General, 112, 511-535.
Piggins, D. J., Kingham, J. R., & Holmes, S. M. (1972). Col¬ Pomerantz, J. R. (1986). Visual form perception: An overview.
our, colour saturation and pattern induced by intermittent il¬ In E. C. Schwab & H. C. Nusbaum (Eds.), Pattern recog¬
lumination: An initial study. British Journal of Physiological nition by humans and machines: Vol. 2. Visual Perception
Optics, 27, 120-125. (pp. 1-30). Orlando, FL: Academic Press.
Pinkers, A., & Marre, M. (1983). Basic phenomena in acquired Pomerantz, J., Goldberg, D., Golder, P., & Tetewsky, S.
colour vision deficiency. Documenta Ophthalmologica, 55, (1981). Subjective contours can facilitate performance in a
251-271. reaction-time task. Perception & Psychophysics, 29, 605-
Pisoni, D. B. (1973). Auditory and phonetic codes in the dis¬ 611.
crimination of consonants and vowels. Perception & Psy¬ Pomerantz, J. R.. Sager, L. C., & Stoever, R. J. (1977). Per¬
chophysics, 13, 253-260. ception of wholes and of their component parts: Some con-
Pisoni, D. B., & Luce, P. A. (1986). Speech perception: Re¬ figural superiority effects. Journal of Experimental Psychol¬
search, theory, and the principal issues. In E. C. Schwab & ogy: Human Perception and Performance, 1, 422-435.
H. C. Nusbaum (Eds.), Pattern recognition by humans and Pons, T. P., Garraghty P. E., Friedman, D. P., & Mishkin, M.
machines: Vol. 1. Speech perception (pp. 1-50). Orlando, (1987). Physiological evidence for serial processing in so¬
FL: Academic Press. matosensory cortex. Science, 237, 417-420.
Pisoni, D. B., Nusbaum, H. C., Luce, P. A., & Slowiaczek, Poppel, E. (1978). Time perception. In R. Held, H. W. Lei-
 References 569

bowitz, & H. L. Teuber (Eds.), Handbook of Sensory movements with and without saccadic correction. Vision Re¬
Physiology: Vol. VIII. Perception (pp. 713-729). New search, 9, 295-303.
York: Springer-Verlag.
Popper, R., Parker, S., & Galanter, E. (1986). Dual loudness Quinn, P. C., Wooten, B, R., & Ludman, E. J. (1985). Ach¬
scales in individual subjects. Journal of Experimental Psy¬ romatic color categories. Perception & Psychophysics, 37,
chology: Human Perception and Performance, 12, 61-69. 198-204.
Porac, C., & Coren, S. (1976). The dominant eye. Psychologi¬
cal Bulletin, 83, 880-897. Rabbitt, P. (1977). Changes in problem solving ability in old
Porac, C., & Coren, S. (1981). Life-span age trends in the per¬ age. In J. Birren & K. Schaie (Eds.), Handbook of the psy¬
ception of the Mueller-Lyer: An additional evidence for the chology of aging. New York: Van Nostrand Reinhold.
existence of two illusions. Canadian Journal of Psychology, Rabbitt, P. (1984). The control of attention in visual search. In
35, 58-62. R. Parasuraman & D. R. Davies (Eds.), Varieties of atten¬
Porac, C., & Coren, S. (1985). Transfer of illusion decrement: tion (pp. 273-291). Orlando, FL: Academic Press.
The effects of global versus local figural variations. Percep¬ Rabin, M. D., & Cain, W. S. (1984). Odor recognition: Famil¬
tion & Psychophysics, 37, 515-522. iarity, identifiability, and encoding consistency. Journal of
Porac, C., & Coren. S., (1986). Sighting dominance and ego¬ Experimental Psychology: Learning, Memory, and Cogni¬
centric localization. Vision Research, 26, 1709-1713. tion, 10, 316-325.
Porter, R. H., Balogh, R. D., Cemoch, J. M., & Franchi, C. Rabin, M. D.. & Cain, W. S. (1986). Determinants of mea¬
(1986). Recognition of kin through characteristic body sured olfactory sensitivity. Perception & Psychophysics, 39,
odors. Chemical Senses, 11, 389-395. 281-286.
Porter, R. H.. & Moore, J. D. (1981). Human kin recognition Rakerd, B., & Hartmann, W. M. (1985). Localization of sound
by olfactory cues. Physiology & Behavior, 27, 493-495. in rooms, II: The effects of a single reflecting surface. Jour¬
Posner, M. I. (1978). Chronometric exploration of mind. Hills¬ nal of the Acoustical Society of America, 78, 524-533.
dale, NJ: Lawrence Erlbaum. Ramachandran, V. S. (1986). Capture of stereopsis and appar¬
Posner, M. I. (1980). Orienting of attention. Quarterly Journal ent motion by illusory contours. Perception & Psychophys¬
of Experimental Psychology, 32, 3-25. ics, 39, 361-373.
Post, B., & Leibowitz, H. W. (1985). A revised analysis of the Ramachandran, V. S. (1988). Perceiving shape from shading.
role of efference in motion perception. Perception, 14, 631- Scientific American, 259, 76-83.
643. Rand, T. C. (1974). Dichotic release from masking for speech.
Postman, L., & Egan, J. P. (1949). Experimental psychology. Journal of the Acoustical Society of America, 55, 678-680.
New York: Harper & Row. Randsom-PIogg, A., & Spillman, L. (1980). Perceptive field
Poulton, E. C. (1979). Models for biases in judging sensory size in fovea of the light and dark adapted. Vision Research,
magnitudes. Psychological Bulletin, 86, 777-803. 20, 221-228.
Poulton, E. C., Edwards, R. S., & Fowler, T. J. (1980). Elim¬ Ratliff, F. (1965). Mach bands: Quantitative studies on neural
inating subjective biases in judging the loudness of a 1-KHz networks in the retina. San Francisco: Holden-Day.
tone. Perception & Psychophysics, 27, 93—103. Rayleigh, Lord (1907). On our perception of sound direction.
Powers, M. K., Schneck, M., & Teller, D. Y. (1981). Spectral Philosophical Magazine, 13(6), 214—232.
sensitivity of human infants at absolute visual threshold. Vi- Raymond, J. E., Shapiro, K. L. & Rose, D. J. (1984). Opto¬
sion Research, 21, 1005-1016. kinetic backgrounds affect perceived velocity during ocular
Poynter, W. D., & Holma, D. (1985). Duration judgment and tracking. Perception & Psychophysics, 36, 221-224.
the experience of change. Perception and Psychophysics, Reason, J. (1984). Lapses of attention in everyday life. In R.
33, 548-560. Parasuraman & D. R. Davies (Eds.), Varieties of attention
Preti, G., Cutler, W. B., Garcia, C. R., Huggins, G. R., & (pp. 515-549). Orlando, FL: Academic Press.
Lawley, H. J. (1986). Human axillary secretions influence Redding, G. M., Clark, S. E., & Wallace, B. (1985). Attention
women’s menstrual cycles: The role of donor extract of fe¬ and prism adaptation. Cognitive Psychology, 17, 1-25.
males. Hormones and Behavior, 20, 474-482. Redding, G. M., & Wallace, B. (1976). Components of dis¬
Price, J. L. (1987). The central and accessory olfactory sys¬ placement adaptation in acquisition and decay as a function
tems. In T. E. Finger & W. L. Silver (Eds.), Neurobiology of hand and hall exposure. Perception & Psychophysics, 20,
of taste and smell (pp. 179-204). New York: Wiley. 453-459.
Prinzmetal, W. (1981). Principles of feature integration in vi¬ Reddy, D. R. (1976). Speech recognition by machine: A re¬
sual perception. Perception & Psychophysics, 30, 330-340. view. Proceedings of the IEEE, 64, 501-531.
Prinzmetal, W., & Millis-Wright, M. (1984). Cognitive and lin¬ Reed, C. F. (1984). Terrestrial passage theory of moon illusion.
guistic factors affect visual feature integration. Cognitive Journal of Experimental Psychology: General, 113, 489-500.
Psychology, 16, 305-340. Regan, D. M., & Beverly, K. I. (1982). How do we avoid
Pritchard, R. M., Heron, W., & Hebb, D. O. (1960). Visual confounding the direction we are looking and the direction
perception approached by the method of stabilized images. we are moving? Science, 215, 194-196.
Canadian Journal of Psychology, 14, 67-77. Rehn, T. (1978). Perceived odor intensity as a function of air¬
Puckett, J. deW., & Steinman, R. M. (1969). Tracking eye flow through the nose. Sensory Processes, 2, 198-205.
570 References

Reisberg, D., & O’Shaughnessy, M. (1984). Diverting subjects’ tation in speech perception. Perception & Psychophysics,
concentration slows figural reversals. Perception, 13, 461 — 30, 309-314.
468. Robertson, P. W. (1967). Color words and colour vision. Biol¬
Remez, R. E., Rubin, P. E., Pisoni, D. B., & Carrell, T. D. ogy and Human Affairs, 33, 28-33.
(1981). Speech perception without traditional speech cues. Robinson, D. W., & Dadson, R. S. (1956). A redetermination
Science, 212, 947-950. of the equal-loudness relations for pure tones. British Jour¬
Remington, R., & Pierce, L. (1984). Moving attention: Evi¬ nal of Applied Physics, 7, 166-181.
dence for time-invariant shifts of visual selective attention. Robson, J. G. (1980). Neural images: The physiological basis
Perception & Psychophysics, 35, 393-399. of spatial vision. In C. S. Harris (Ed.), Visual coding and
Repp, B. H. (1987). The sound of two hands clapping: An ex¬ adaptability (pp. 177-214.). Hillsdale, NJ: Lawrence Erl¬
ploratory study. Journal of the Acoustical Society of Amer¬ baum.
ica, 81, 1100-1109. Rock, I. (1973). Orientation and form. New York: Academic
Restle, F. (1971). Visual illusions. In M. H. Appley (Ed.), Ad¬ Press.
aptation-level theory (pp. 55-69). New York: Academic Rock, I. (1975). An introduction to perception. New York:
Press. Macmillan.
Restle, F. (1978). Assimilation predicted by adaptation level Rock, I. (1983). The logic of perception. Cambridge, MA: MIT
theory with variable weights. In J. S. Castellan & F. Restle Press.
(Eds.), Cognitive theory: Vol. 3 (pp. 75-92). Hillsdale, NJ: Rock, I., & Guttman, D. (1981). The effect of inattention on
Lawrence Erlbaum. form perception. Journal of Experimental Psychology: Hu¬
Reynolds, D. C. (1979). A visual profile of the alcoholic driver. man Perception and Performance, 7, 275-285.
American Journal of Optometry and Physiological Optics, Rock, I., & Halper, F. (1969). Form perception without a
56, 241-251. retinal image. American Journal of Psychology, 82, 425-
Reynolds. R. I. (1985). The role of object-hypotheses in the 440.
organization of fragmented figures. Perception, 14, 49-52. Rockland, K. S., & Pandya, P. N. (1981). Cortical connections
Rhee, K., Kim, D., & Kim, Y. (1965). The effects of smoking of the occipital lobe in the rhesus monkey: Interconnections
on night vision. 14th Pacific Medical Conference (Profes¬ between areas 17, 18, 19 and the superior temporal sulcus.
sional papers). Brain Research, 212, 249-270.
Rhodes, G. (1987). Auditory attention and the representation of Rodieck, R. W. (1973). The vertebrate retina: Principles of
spatial information. Perception & Psychophysics, 42, 1-14. structure and function. San Francisco: Freeman.
Rice, C. G., Ayley, J. B., Bartlett, B., Bedford, W., Gregory, Rodieck, R. W. (1979). Visual pathways. Annual Review of
W., & Hallum, G. (1968). A pilot study on the effects of Neurosciences, 2, 193-225,
pop group music on hearing. Cited in K. D. Kryter (1970). Roelofs, C. O. (1935). Optische lokalisation. Archiv fur Augen-
The effects of noise on man (p. 203). New York: Academic heilkunde 109, 395-415.
Press. Rogel, M. J. (1978). A critical evaluation of the possibility of
Richards, W. (1977). Lessons in constancy from neurophysiol¬ higher primate reproductive and sexual pheromones. Psy¬
ogy. In W. W. Epstein (Ed.), Stability and constancy in chological Bulletin, 85, 810-830.
visual perception: Mechanisms and processes (pp. 421 — Rohrbaugh, J. W. (1984). The orienting reflex: Performance
436.), New York: Wiley. and central nervous system manifestations. In R. Parasura-
Riesen, A., & Zilbert, D. (1975). Behavioral consequences of man & D. R. Davies (Eds.), Varieties of attention, (pp.
variations in early sensory environments. In A. Riesen 323-373). Orlando, FL: Academic Press.
(Ed.), The developmental neuropsychology of sensory dep¬ Rollman, G. B., & Harris, G. (1987). The detectability, discri-
rivation (pp. 211-252). New York: Academic Press. minability, and perceived magnitude of painful electric
Rieser, J., Yonas, A., & Wikner, K. (1976). Radial localization shock. Perception & Psychophysics, 42, 257-268.
of odors by human newborns. Child Development, 47, 856- Romani, G. L., Williamson, S. J., & Kaufman, L. (1982).
859. Tonotopic organization of the human auditory cortex. Sci¬
Riesz, R. R. (1928). Differential intensity sensitivity of the ear ence, 216, 1339-1340.
for pure tones. Physical Review, 31, 867-875. Romano, P. E., Romano, J. A., & Puklin, J. E. (1975). Ster¬
Riggs, L. A., Ratliff, F., Comsweet, J. C., & Comsweet, T. N. eoactivity development in children with normal single vi¬
(1953). The disappearance of steadily fixated visual test ob¬ sion. American Journal of Ophthalmology, 79, 966-971.
jects. Journal of the Optical Society of America, 43, 495- Root, W. (1974, December 22). Of wine and noses. New York
SOl. Times Magazine, p. 14 et seq.
Rijnsdorp, A., Daan, S., & Dijkstra, C. (1981). Hunting in the Rose, J. E., Brugge, J. F., Anderson, D. J., & Hind, J. E.
kestrel (Falco tinnunculus) and the adaptive significance of (1967). Phase-locked response to low frequency tones in sin¬
daily habits. Oecologia, 50, 391-406. gle auditory nerve fibers of the squirrel monkey. Journal of
Roberts, J. (1964). Binocular visual acuity of adults. Washing¬ Neurophysiology, 30, 769-793.
ton, D.C.: U.S. Department of Health, Education and Wel¬ Rose, J. E., Galambos, R., & Hughes, J. (1959). Microelec¬
fare. trode studies of the cochlear nuclei of the cat. Johns Hopkins
Roberts, M. & Summerfield, A. Q. (1981). Audio-visual adap¬ Hospital Bulletin, 14, 211-251.
 References 571

Rose, J. E., Galambos, R., & Hughes, J. (1960). Organization Rusak, B., & Zucker, I. (1979). Neural regulation of circadian
of frequency sensitive neurons in the cochlear nuclear com¬ rhythms. Physiological Review, 59, 449-526.
plex of the cat. In G. L. Rasmussen & W. F. Windle (Eds.), Rushton, W. A. H. (1962). Visual pigments in man. Scientific
Neural mechanisms of the auditory and vestibular systems American, 205, 120-132.
(pp. 116-136). Springfield, IL: Thomas. Rushton, W. A. H. (1965). Cone pigment dynamics in the deu-
Rosinski, R., & Farber, J. (1980). Compensation for viewing teranope. Journal of Physiology (London), 176, 38-45.
point in the perception of pictured space. In M. Hagen Russell, M.. J. (1976). Human olfactory communication. Na¬
(Ed.), Perception of pictures: Vol. 7. Alberti’s Window: The ture (London), 260, 520-522.
projective model of pictorial information. New York: Aca¬ Russoff, A. C. (1979). Development of ganglion cells in the
demic Press. retina of the cat. In R. D. Freeman (Ed.), Developmental
Ross. H. (1975, June 19). Mist, murk and visual perception. neurobiology of vision (pp. 19-30). New York: Plenum.
New Scientist, 658-660. Russoff, A. C., & Dubin, M. W. (1977). Development of re¬
Ross, H. E., Brodie, E., & Benson, A. (1984). Mass discrimi¬ ceptive-field properties of retinal ganglion cells in kittens.
nation during prolonged weightlessness. Science, 225, 219— Journal of Neurophysiology, 40, 1188-1198.
221.
Ross, H. E., & Reschke, M. F. (1982). Mass estimation and Sachs, M. B., & Kiang, N. Y. S. (1968). Two-tone inhibition
discrimination during brief periods of zero gravity. Percep¬ in auditory nerve fibers. Journal of the Acoustical Society of
tion & Psychophysics, 31, 429-436. America, 43, 1120-1128.
Ross, N., & Schilder, P. (1934). Tachistoscopic experiments on Salapatek, P. (1975). Pattern perception in early infancy. In
the perception of the human figure. Journal of General Psy¬ L. B. Cohen & P. Salapatek (Eds.), Infant perception: From
chology, 10, 152-172. sensation to cognition (Vol. 1, pp. 133-248). New York:
Rothblat, L., & Schwartz, M. (1978). Altered early environ¬ Academic Press.
ment: Effects on the brain and visual behavior. In R. Walk Salapatek, P., & Kessen, W. (1973). Prolonged investigation of
& H. Pick (Eds.), Perception and experience (pp. 7-36). a plane geometric triangle by the human newborn. Journal
New York: Plenum. of Experimental Child Psychology, 15, 22-29.
Royster, L. H., Royster, J. D., & Thomas, W. G. (1980). Rep¬ Samuel, A. G. (1981). Phonemic restoration: Insights from a
resentative hearing levels by race and sex in North Carolina new methodology. Journal of Experimental Psychology:
industry. Journal of Acoustical Society of America, 68, 551— General, 110, 474-494.
566. Samuel, A. G. (1986). Red herring detectors and speech percep¬
Rozin, P. (1978). The use of characteristic flavorings in human tion: In defense of selective adaptation. Cognitive Psychol¬
culinary practice. In C. M. Apt (Ed.), Flavor: Its Chemical, ogy, 18, 452-499.
Behavioral, and Commercial Aspects (pp. 101-127). Boul¬ Sanders, B., Soares, M. P., & D’Aquila, J. M. (1982). Sex
der, CO: Westview Press. difference on one test of spatial visualization: A nontrivial
Rozin, P. (1982). “Taste-smell confusions” and the duality of difference. Child Development, 53, 1106-1110.
the olfactory sense. Perception & Psychophysics, 31, 397- Sanders, G., & Ross-Field, L. (1986). Sexual orientation and
401. visuo-spatial ability. Brain and Cognition, 5, 280-290.
Rozin, P., Ebert, L., & Schull, J. (1982). Some like it hot: A Sanocki, T. (1987). Visual knowledge underlying letter percep¬
temporal analysis of hedonic responses to chili pepper. Ap¬ tion: Font-specific, schematic tuning. Journal of Experimen¬
petite, 3, 13-22. tal Psychology: Human Perception and Performance, 13,
Rubin, E. (1915). Synoplevede figuren. Copenhagen: Glyden- 267-278.
dalske. Sapir, E. (1939). Language. New York: Harcourt, Brace &
Rubin, E. (1921). Visuell wahrgenommene figuren. Copen¬ World.
hagen: Glydendalske. Sawusch, J. R. (1986). Auditory and phonetic coding of speech.
Ruble, D. N., & Nakamura, C. Y. (1972). Task orientation In E. C. Schwab & H. C. Nusbaum (Eds.), Pattern recog¬
versus social orientation in young children and their atten¬ nition by humans and machines: Vol. 1. Speech Perception
tion to relevant social cues. Child Development, 43, 471 — (pp. 51-88). Orlando, FL: Academic Press.
480. Schaie, K. W., & Geiwitz, J. (1982). Adult development and
Ruggieri, V., Cei, A., Ceridono, D., & Bergerone, C. (1980). aging. Boston: Little, Brown.
Dimensional approach to the study of sighting dominance. Scharf, B. (1964). Partial masking. Acustica, 14, 16-23.
Perceptual and Motor Skills, 51, 247-251. Scharf, B. (1975). Audition. In B. Scharf (Ed.), Experimental
Runeson, S., & Frykholm, G. (1983). Kinematic specifications sensory psychology (pp. 112-149.) Glenview, IL: Scott,
of dynamics as an informational basis for person-and-action Foresman.
perception: Expectation, gender recognition, and deceptive Scharf, B. (1978). Loudness. In E. C. Carterette & M. P. Fried¬
intention. Journal of Experimental Psychology: General, man (Eds.), Handbook of Perception: Vol. IV. Hearing.
112, 585-615. New York: Academic Press.
Rusak, B., & Groos, G. (1982). Suprachiasmatic stimulation Scharf, B., Quigley, S., Aoki, C., Peachey, N., & Reeves, A.
phase shifts rodent circadian rhythms. Science, 215, 1407— (1987). Focused auditory attention and frequency selectivity.
1409. Perception & Psychophysics, 42, 215-223.
572 References

Schenkel, K. D. (1967). Die beidohrigen Mithorschoellen von tomatic and control processing and attention. In R. Parasur-
Impulsen. Acustica, 18, 38-46. aman & D. R. Davies (Eds.), Varieties of attention, (pp. 1-
Scher, D., Pionk, M., & Purcell, D. G. (1981). Visual sensitiv¬ 27). Orlando, FL: Academic Press.
ity fluctuations during the menstrual cycle under dark and Schneider, W., & Shiffrin, R. M. (1977). Controlled and auto¬
light adaptation. Bulletin of the Psychonomic Society, 18, matic human information processing. I. Detection, search
159-160. and attention. Psychological Review, 84, 1-66.
Schiffman, S. S. (1974). Physiochemical correlates of olfactory Schouten, M. E. H. (1980). The case against a speech mode of
quality. Science, 185, 112-117. perception. Acta Psychologica, 44, 71-98.
Schiffman, S. S. (1977). Food recognition by the elderly. Jour¬ Schubert, E. D. (1978). History of research on hearing. In
nal of Gerontology, 32, 586-592. E. C. Carterette & M. P. Friedman (Eds.), Handbook of
Schiffman, S. S., & Dackis, C. (1975). Taste of nutrients: perception: Vol. IV. Hearing (pp. 41—80). New York: Aca¬
Amino acids, vitamins, and fatty acids. Perception & Psy¬ demic Press.
chophysics, 17, 140-146. Schull, J., Kaplan, H., & O’Brien, C. P. (1981). Naloxone can
Schiffman, S. S., & Erickson, R. P. (1971). A theoretical re¬ alter experimental pain and mood in humans. Physiological
view: A psychophysical model for gustatory quality. Physi¬ Psychology, 9, 245-250.
ology & Behavior, 1, 617-633. Schulman, P. H. (1979). Eye movements do not cause induced
Schiffman, S. S., McElroy, A. E., & Erickson, R. F. (1980). motion. Perception & Psychophysics, 26, 381-383.
The range of taste quality of sodium salts. Physiology & Schwartz, C. B. (1961). Visual discrimination of camouflaged
Behavior, 24, 217-224. figures. Unpublished doctoral dissertation, University of
Schiffman, S. S., & Pasternak, M. (1979). Decreased discrim¬ California at Berkeley.
ination of food odors in the elderly. Journal of Gerontology, Schwartz, S. H., & Loop, M. S. (1984). Effect of duration on
84, 73-79. detection by the chromatic and achromatic systems. Percep¬
Schiffman, S. S., Reilly, D. A., & Clark, T. B., III. (1979). tion & Psychophysics, 36, 65-67.
Qualitative differences among sweeteners. Physiology & Be¬ Scott, T. R. (1987). Coding in the gustatory system. In T. E.
havior, 23, 1-9. Finger & W. L. Silver (Eds.), Neurobiology of taste and
Schiller, P. H. (1986). The central visual system. Vision Re¬ smell (pp. 355-378). New York: Wiley.
search, 26, 1351-1386. Scott, T. R., & Erickson, R. P. (1971). Synaptic processing of
Schindler, R. A., & Merzenich, M. M. (Eds.), (1985). Coch¬ taste-quality information in the thalamus of the rat. Journal
lear Implants. New York: Raven. of Neurophysiology, 34, 868-884.
Schleidt, M., Hold, B., & Attili, G. (1981). A cross-cultural Sedgwick, H. (1980). The geometry of spatial layout in pictorial
study on the attitude towards personal odors. Journal of representation. In M. Hagen (Ed.), Perception of pictures:
Chemical Ecology, 7, 19-31. Vol. 1. Alberti’s window: The projective model of pictorial
Schmiedt, R. A., Zwislocki, J. J., & Hamemik, R. P. (1980). information (pp. 33-90). New York: Academic Press.
Effects of hair cell lesions on responses of cochlear nerve Segall, M. H., Campbell, D. T., & Herskovits, M. J. (1966).
fibers. I. Lesions, tuning curves, two-tone inhibition, and The influence of culture on visual perception. Indianapolis:
responses to trapezoidal-wave patterns. Journal of Neuro¬ Bobbs-Merrill.
physiology, 43, 1367-1389. Sekuler, R. (1975). Visual motion perception. In E. C. Carter¬
Schnapf, J. L., & Baylor, D. A. (1987). How photoreceptor ette & M. P. Friedman (Eds.), Handbook of perception:
cells respond to light. Scientific American, 256, 40-47. Vol. 5 (pp. 387-433). New York: Academic Press.
Schneider, B. A., & Bissett, R. J. (1981). The dimensions of Sekuler, R., Ball, K., Tynan, P., & Machmer, J. (1982). Psy¬
tonal experience: A nonmetric scaling approach. Perception chophysics of motion perception. In A. H. Wertheim,
& Psychophysics, 30, 39-48. W. A. Wagenaar, & H. W. Leibowitz (Eds.), Tutorials on
Schneider, B. A., & Parker, S. (1987). Can we measure sen¬ motion perception (pp. 81-100). New York: Plenum.
sory intensity? In M. Teghtsoonian & R. Teghtsoonian Sekuler, R., & Blake, R. (1985). Perception. New York:
(Eds.), Fechner day 87 (pp. 29-33). Durham, NH: Interna¬ Knopf.
tional Society for Psychophysics. Sekuler, R., & Ganz, L. (1963). A new aftereffect of seen
Schneider, D. (1969). Insect olfaction: Deciphering system for movement with a stabilized retinal image. Science, 139,
chemical messages. Science, 163, 1031-1037. 419-420.
Schneider, G. E. (1969). Two visual systems. Science, 163, Sekuler, R., & Hutman, L. P. (1980). Spatial vision and aging:
895-902. I. Contrast sensitivity. Journal of Gerontology, 35, 692-
Schneider, R., Costiloe, J., Howard, R., & Wolf, S. (1958). 699.
Olfactory perception thresholds in hypogonadal women: Selfridge, O. G. (1959). Pandemonium: A paradigm for learn¬
Changes accompanying administration of androgen and es¬ ing. In D. V. Blake & A. M. Uttley (Eds.), Proceedings of
trogen. Journal of Clinical Endocrinology, 18, 379-390. the symposium on the mechanization of thought processes
Schneider, S. L., Hughes, B., Epstein, W., & Bach-y-Rita, P. (pp. 511-529). London: HM Stationery Office.
(1986). The detection of length and orientation changes in Semple, M. N., & Kitzes, L. M. (1987). Binaural processing
dynamic vibrotactile patterns. Perception & Psychophysics, of sound pressure level in the inferior colliculus. Journal of
40, 290-300. Neurophysiology, 57, 1130-1147.
Schneider, W., Dumais, S. T., & Shiffrin, R. M. (1984). Au¬ Senden, M. von. (1960). Space and sight: The perception of
 References 573

space and shape in congenitally blind patients before and Shimojo, S., & Richards, W. (1986). “Seeing” shapes that are
after operation. London: Methuen. almost totally occluded: A new look at Park’s camel. Per¬
Serpell, R. (1971). Discrimination of orientation by Zambian ception & Psychophysics, 39, 418-426.
children. Journal of Comparative Physiology, 75, 312. Shockey, L., & Reddy, R. (1974, August). Quantitative analy¬
Shaffer, H. L. (1975). Multiple attention in continuous verbal sis of speech perception: Results from transcription of con¬
tasks. In P. M. A. Rabbitt & S. Domic (Eds.), Attention nected speech from unfamiliar languages. Paper presented
and Performance V. London: Academic Press. at the Speech Communications Seminar. Stockholm.
Shaffer, L. H. (1985). Timing in action. In J. A. Michon & Shower, E. G., & Biddulph, R. (1931). Differential pitch sen¬
J. L. Jackson (Eds.), Time, mind and behavior (pp. 226- sitivity of the ear. Journal of the Acoustical Society of Amer¬
242). Berlin: Springer-Verlag. ica, 3, 275-287.
Shallice, T., & Vickers, D. (1964). Theories and experiments Shulman, G. L., Remington, R. W., & McLean, J. P. (1979).
on discrimination times. Ergonomics. 7, 37-49. Moving attention through visual space. Journal of Experi¬
Shannon, C. E., & Weaver, W. (1949). The mathematical the¬ mental Psychology: Human Perception and Performance, 5,
ory of communication. Urbana: University of Illinois Press. 522-526/
Shapley, R., & Enroth-Cugell, C. (1984). Visual adaptation and Shulman, G. L., Sheehy, J. B., & Wilson, J. (1986). Gradients
retinal gain controls. Progress in Retina! Research, 3, 263- of spatial attention. Acta Psychologica, 61, 167-181.
346. Shulman, G. L., Wilson, J., & Sheehy, J. B. (1985). Spatial
Shapley, R., & Reid, R. C. (1985). Contrast and assimilation determinants of the distribution of attention. Perception &
in the perception of brightness. Proceedings of the National Psychophysics, 37, 59-65.
Academy of Sciences of the U.S.A., 82, 5983-5986. Siddle, D. A., Morish, R. B., White, K. D., & Mangen, G. L.
Sharma, S., & Moskowitz, H. (1972). Effect of marijuana on (1969). Relation of visual sensitivity to extraversion. Jour¬
the visual autokinetic phenomenon. Perceptual and Motor nal of Experimental Research in Personality, 3, 264-267.
Skills, 35, 891. Sidman, M., & Kirk, B. (1974). Letter reversals in naming,
Shea, S. L., Fox, R., Aslin, R. N., & Dumais, S. T. (1980). writing, and matching to sample. Child Development, 45,
Assessment of stereopsis in human infants. Investigative 616-625.
Ophthalmology, 19, 1400-1404. Silver, W. L. (1987). The common chemical sense. In T. E.
Shebilske, W. (1977). Visuomotor coordination in visual direc¬ Finger & W. L. Silver (Eds.), Neurobiology of Taste and
tion and position constancies. In W. Epstein (Ed.), Stability Smell (pp. 65-87). New York: Wiley.
and constancy in visual perception: Mechanisms and pro¬ Simmons, F. B., Epley, J. M., Lummis, R. C., Guttman, N.,
cesses (pp. 23-70). New York: Wiley. Frishkopf, L. S., Harmon, L. D., & Zwicker, E. (1965).
Shebilske, W. L. (1976). Extraretinal information in corrective Auditory nerve: Electrical stimulation in man. Science, 148,
saccades and inflow vs. outflow theories of visual direction 104-106.
constancy. Vision Research, 16, 621-628. Sinclair, D. C., & Stokes, B. A. R. (1964). The production and
Sheedy, J. E., Bailey, I. L., Buri, M., & Bass, E. (1986). characteristics of “second pain.” Brain, 87, 609-618.
Binocular vs. monocular task performance. American Jour¬ Sinnot, J., & Rauth, J. (1937). Effect of smoking on taste
nal of Optometry & Physiological Optics, 63, 839-846. thresholds. Journal of General Psychology, 17, 155-162.
Shepard, R. N. (1964). Circularity in judgments of relative Sivian, L. S., & White, S. D. (1933). On minimum audible
pitch. Journal of the Acoustical Society of America, 36, sound fields. Journal of the Acoustical Society of America,
2346-2353. 4, 288-321.
Shepard, R. N. (1982). Geometrical approximations to the Skowbo, D. (1984). Are McCollough effects conditioned re¬
structure of musical pitch. Psychological Review, 89, 305- sponses? Psychological Bulletin, 96, 215-226.
333. Sloboda, J. A. (1985). The musical mind: The cognitive psy¬
Shepard, R. N., & Zare, S. L. (1983). Path-guided apparent chology of music. Oxford: Oxford University Press.
motion. Science, 220, 632-634. Small, L. H., & Bond, Z. S. (1986). Distortions and deletions:
Sherman, S. (1973). Visual field defects in monocularly and Word-initial consonant specificity in fluent speech. Percep¬
binocularly deprived cats. Brain Research, 49, 25-45. tion & Psychophysics, 40, 20-26.
Sherman, S. M. (1985). Parallel W-, X- and Y-cell pathways in Smith, A., & Over, R. (1979). Motor aftereffect with subjective
the cat: A model for visual function. In D. Rose & V. G. contours. Perception & Psychophysics, 25, 95-98.
Dobson (Eds.)., Models of the visual cortex (pp. 71-84). Smith, D. V. (1985). Brainstem processing of gustatory infor¬
Chichester: Wiley. mation. In D. W. Pfaff (Ed.), Taste, olfaction, and the cen¬
Sherrington, C. S. (1906). Integrative action of the nervous sys¬ tral nervous system (pp. 151-177). New York: Rockefeller
tem. New Haven, CT: Yale University Press. University Press.
Shiffrin, R. M., & Schneider, W. (1977). Controlled and auto¬ Smith, J., Hausfeld, S., Power, R. P., & Gorta, A. (1982).
matic human information processing. II. Perceptual learn¬ Ambiguous musical figures and auditory streaming. Percep¬
ing, automatic attending and a general theory. Psychological tion & Psychophysics, 32, 454-464.
Review, 84, 127—190. Smith, W. S., Frazier, N. I., Ward, S., & Webb, F. (1983).
Shimojo, S., & Held, R. (1987). Vernier acuity is less than Early adolescent girls’ and boys’ learning of a spatial visu¬
grating acuity in 2- and 3-month olds. Vision Research, 27, alization skill-replications. Journal of Education, 67, 239-
77-86. 243.
574 References

Snellen, H. (1862). Probebuchstaben zur bestimmung der seh- Stelmack, R. M., & Campbell, K. B. (1974). Extraversion and
scharfe. Utrecht: Weijer. auditory sensitivity to high and low frequency. Perceptual
Snyder, S. H. (1977). Opiate receptors and internal opiates. Sci¬ and Motor Skills, 38, 875-879.
entific American, 236, 44-56. Stephan, F, K., & Nunez, A. A. (1977). Elimination of circa¬
So, Y. T., & Shapley, R. (1981). Spatial tuning of cells in and dian rhythms in drinking activity, sleep, and temperature by
around lateral geniculate nucleus of the cat: X and y cells isolation of suprachiasmatic nuclei. Behavioral Biology, 20,
and perigeniculate intemeurons. Journal of Neurophysiol¬ 1-16.
ogy, 45, 107-120. Stephens, P. R., & Young, J. Z. (1982). The stacocyst of
Sokolov, E. N. (1975). The neuronal mechanisms of the orient¬ the squid Loligo. Journal of Zoology, (London), 197, 241-
ing reflex. In E. N. Sokolov & O. S. Vinogradova (Eds.), 266.
Neuronal mechanisms of the orienting reflex (pp. 217-238). Stem, J. A., Oster, P. J., & Newport, K. (1980). Reaction time
New York: Wiley. measures, hemispheric specialization, and age. In L. Poon
Southwick, E., & Schiffman, S. S. (1980). Odor quality of pyr- (Ed.), Aging in the 1980's (pp. 309-326). Washington,
idyl ketones. Chemical Senses, 5, 343-357. D.C.: American Psychological Association.
Sperling, H. G. (1986). Spectral sensitivity, intense spectral Stembach, R. A. (1963). Congenital insensitivity to pain: A re¬
light studies and the color receptor mosaic of primates. Vi¬ view. Psychological Bulletin, 60, 252-264.
sion Research, 26, 1557-1571. Stembach, R. A., & Tursky, B. (1964). On the psychophysical
Sperry, R. W. (1943). Effect of 180 degree rotation of the ret¬ power function in electric shock. Psychonomic Science, 1,
inal field on visuomotor coordination. Journal of Experimen¬ 247-248.
tal Zoology, 92, 263-277. Sternberg, S. (1975). Memory scanning: New findings and cur¬
Spoendlin, H. H. (1978). The afferent innervation of the coch¬ rent controversies. Quarterly Journal of Experimental Psy¬
lea. In R. F. Naunton & C. Femandey (Eds.), Evoked elec¬ chology, 27, 1-32.
trical activity in the auditory nervous system (pp. 21-42). Stevens, D. A., & Lawless, H. T. (1986). Putting out the fire:
New York: Academic Press. Effects of tastants on oral chemical irritation. Perception &
Sprafkin, C., Serbin, L. A., Denier, C., & Conner, J. M. Psychophysics, 39, 346-350.
(1983). Sex-differentiated play: Cognitive consequences and Stevens, J. C. (1979). Variation of cold sensitivity over the
early interventions. In M. B. Liss (Ed.), Social and cogni¬ body surface. Sensory Processes, 3, 317-326.
tive skills (pp. 167-192). New York: Academic Press. Stevens, J. C., & Cain, W. S. (1986). Smelling via the mouth:
Spreen, O. (1976). Neuropsychology of learning disorders: Post Effects of aging. Perception & Psychophysics, 40, 142—
conference review. In R. M. Knights & D. J. Bakker (Eds.), 146.
The neuropsychology of learning disorders (pp. 445-467). Stevens, J. C., & Marks, L. E. (1980). Cross-modality match¬
Baltimore: University Park Press. ing functions generated by magnitude estimation. Perception
Springer, S. P., & Deutsch, G. (1985). Left brain, right brain & Psychophysics, 27, 379-389.
(rev. ed.). San Francisco: Freeman. Stevens, J. C., & Stevens, S. S. (1960). Warmth and cold: Dy¬
Srulovicz, P., & Goldstein, J. L. (1983). A central spectrum namics of sensory intensity. Journal of Experimental Psy¬
model: A synthesis of auditory-nerve timing and place cues chology, 60, 183-192.
in monaural communication of frequency spectrum. Journal Stevens, K. N., & House, A. S. (1972). Speech perception. In
of the Acoustical Society of America, 73, 1266-1276. J. V. Tobias (Ed.), Foundations of modern auditory theory
Stark, L., & Bridgeman, B. (1983). Role of corollary discharge (Vol. 2, pp. 3—62). New York: Academic Press.
in space constancy. Perception & Psychophysics, 34, 371 — Stevens, S. S. (1935). The relation of pitch to intensity. Journal
380. of the Acoustical Society of America, 6, 150-154.
Stark, L., & Ellis, S. (1981). Scanpaths revisited: Cognitive Stevens, S. S. (1946). On the theory of scales of measurement.
models direct active looking. In D. Fisher, R. Monty, & I. Science, 103, 677-680.
Senders (Eds.), Eye movements: Cognition and visual per¬ Stevens, S. S. (1956). The direct estimation of sensory magni¬
ception (pp. 193-226). Hillsdale, NJ: Lawrence Erlbaum. tudes—loudness. American Journal of Psychology, 69, 1-
Stebbins, W. C. (1980). The evolution of hearing in the mam¬ 25.
mals. In A. N. Popper & R. R. Fay (Eds.), Comparative Stevens, S. S. (1959). Tactile vibration: Dynamics of sensory
studies of hearing in vertebrates (pp. 421-436). New York: intensity. Journal of Experimental Psychology, 57, 210—
Springer-Verlag. 218.
Steinberg, A. (1955). Changes in time perception induced by an Stevens, S. S. (1961). The psychophysics of sensory function.
anaesthetic drug. British Journal of Psychology, 46, 273- In W. A. Rosenblith (Ed.), Sensory communication (pp. 1-
279. 33). Cambridge, MA: MIT Press.
Steinfield, G. J. (1967). Concepts of set and availability and Stevens, S. S. (1975). Psychophysics: Introduction to its per¬
their relation to the reorganization of ambiguous pictorial ceptual, neural, and social prospects. New York: Wiley.
stimuli. Psychological Review, 74, 505-525. Stevens, S. S., & Galanter, E. (1957). Ratio scales and cate¬
Stelmack, R. M., Achom, E., & Michaud, A. (1977). Extravi¬ gory scales for a dozen perceptual continua. Journal of Ex¬
sion and individual differences in auditory evoked response. perimental Psychology, 54, 377-411.
Psychophysiology, 14, 368-374. Stevens, S. S., & Newman, E. B. (1934). The localization of
 References 575

pure tones. Proceedings of the National Academy of Sci¬ Swets, J. A. (1963). Central factors in auditory frequency selec¬
ences of the U.S.A., 20, 593-596. tivity. Psychological Bulletin, 60, 429-441.
Stevens, S. S., Volkman, J., & Newman, E. B. (1937). A scale Szentagothai, J. (1950). The elementary vestibulo-ocular reflex
for the measurement of the psychological magnitude of arc. Journal of Neurophysiology, 13, 395-407.
pitch. Journal of the Acoustical Society of America, 8, 185-
190. Tart, C. (1971). On being stoned. Palo Alto: Science & Behav¬
Stevens, S. S., & Warshovsky, F. (1965). Sound and hearing. ior Books.
New York: Time-Life Books. Taub, E., & Berman, A. J. (1968). Movement and learning in
Stoffregen, T. A. (1985). Flow structure versus retinal location the absence of sensory feedback. In S. J. Freedman (Ed.),
in the optical control of stance. Journal of Experimental The neuropsychology of spatially oriented behavior (pp.
Psychology: Human Perception and Performance, 11, 554- 173-192). Homewood, IL: Dorsey.
565. Tedford, W. H., Warren, D. E., & Flynn, W. E. (1977). Al¬
Stone, L. S. (1960). Polarization of the retina and development ternation of shock aversion thresholds during menstrual cy¬
of vision. Journal of Experimental Zoology, 145, 85-93. cle. Perception & Psychophysics, 21(2), 193-196.
Strange, W., & Jenkins, J. (1978). Role of linguistic experience Tees, R. C. (1974). Effect of visual deprivation on development
in the perception of speech. In R. Walk & H. Pick (Eds.), of depth perception in the rat. Journal of Comparative and
Perception and experience (pp. 125-169). New York: Physiological Psychology, 86, 300-308.
Plenum. Tees, R. C., & Midgley, G. (1978). Extent of recovery of func¬
Stratton, G. M. (1896). Some preliminary experiments on vi¬ tion after early sensory deprivation in the rat. Journal of
sion without inversion of the retinal image. Psychological Comparative and Physiological Psychology, 92, 768-777.
Review, 3, 611-617. Tees, R. C., & Werker, J. F. (1984). Perceptual flexibility:
Stratton, G. M. (1897a). Upright vision and the retinal image. Maintenance or recovery of the ability to discriminate non¬
Psychological Review, 4, 182-187. native speech sounds. Canadian Journal of Psychology, 38,
Stratton, G. M. (1897b). Vision without inversion of the retinal 579-590.
image. Psychological Review, 4, 341-360. Teeter, J. H., & Brand, J. G. (1987). Peripheral mechanisms of
Street, R. F. (1931). A gestalt completion test: A study of a gustation: Physiology and biochemistry. In T. E. Finger &
cross section of intellect. New York: Bureau of Publication, W. L. Silver (Eds.), Neurobiology of taste and smell (pp.
Columbia University. 299-330). New York: Wiley.
Streitfeld, B., & Wilson, M. (1986). The ABCs of categorical Teghtsoonian, M. (1987). The structure of an experiment. In M.
perception. Cognitive Psychology, 18, 432-451. Teghtsoonian & R. Teghtsoonian (Eds.), Fechner Day ’87
Stromeyer, C. F., III. (1978). Form-color aftereffects in human (pp. 49-52). Northampton, MA: International Society for
vision. In R. Held, H. Leibowitz, & H. L. Teuber (Eds.), Psychophysics.
Handbook of sensory physiology (Vol. 8, pp. 97-142). New Teghtsoonian, R. (1971). On the exponents in Stevens’ Law and
York: Springer-Verlag. the constant in Ekman’s Law. Psychological Review, 78,
Stroop, J. (1935). Studies of interference in serial verbal reac¬ 71-80.
tions. Journal of Experimental Psychology, 18, 624-643. Teghtsoonian, R. (1975). Review of psychophysics by S. S.
Stroud, J. M. (1955). The fine structure of psychological time. Stevens. American Journal of Psychology, 88, 677-684.
In H. Quastler (Ed.), Information theory in psychology: Teghtsoonian, R., Teghtsoonian, M., Berglund, B., & Ber-
Problems and methods (pp. 174-207). Glencoe, IL: Free glund, U. (1978). Invariance of odor strength with sniff
Press. vigor: An olfactory analogue to size constancy. Journal of
Stryer, L. (1987). The molecules of visual excitation. Scientific Experimental Psychology: Human Perception and Perfor¬
American, 257, 42-50. mance, 4, 144-152.
Suedfeld, P. (1980). Restricted environmental stimulation: Re¬ Teller, D. Y. (1981). Color vision in infants. In R. Aslin, J.
search and clinical applications. New York: Wiley. Alberts, & M. Petersen (Eds.), Development of Perception,
Supra, M., Cotzin, M. E., & Dallenbach, K. M. (1944). “Fa¬ Physiological Perspectives: Vol. 2 (pp. 298-312). New
cial vision:” The perception of obstacles by the blind. Amer¬ York: Academic Press.
ican Journal of Psychology, 57, 133-183. Teller, D. Y., & Bomstein, M. H. (1987). Infant color vision
Svaetichin, G. (1956). Spectral response curves of single cones. and color perception. In P. Salapatek & L. Cohen (Eds.),
Acta Physiologica Scandinavica, 1, 93-101. Handbook of infant perception: Vol. 1. From sensation to
Svaetichin, G., & MacNichol, E. F., Jr. (1958). Retinal mech¬ perception (pp. 185-237). Orlando, FL: Academic Press.
anisms for achromatic vision. Annals of the New York Acad¬ Teller, D. Y., & Movshon, J. A. (1986). Visual development.
emy of Sciences, 74, 385-404. Vision Research, 26, 1483-1506.
Swarbrick, L., & Whitfield, I. C. (1972). Auditory cortical Terenius, L., & Wahlstrom, A. (1975). Morphine-like ligand
units selectively responsive to stimulus “shape.” Journal of for opiate receptors in human CSF. Life Sciences, 16, 1759—
Physiology (London), 224, 68-69. 1764.
Swensson, R. G. (1980). A two-stage detection model applied Terheardt, E. (1974). Pitch, consonance and harmony. Journal
to skilled visual search by radiologists. Perception & Psy¬ of the Acoustical Society of America, 55, 1061-1069.
chophysics, 27, 11-16. Thomas, E. A. C., & Weaver, W. B. (1975). Cognitive
576 References

processing and time perception. Perception & Psychophys¬ Trehub, S. E., Schneider, B. A., & Endman, M. (1980). De¬
ics, 17, 363-367. velopmental changes in infants’ sensitivity to octave-band
Thomas, H. (1983). Parameter estimation in simple psycho¬ noises. Journal of Experimental Child Psychology, 29, 282-
physical models. Psychological Bulletin, 93, 396-403. 293.
Thomas, H., Jamison, W., & Hammel, D. D. (1973). Obser¬ Treisman, A. M. (1982). Perceptual groupings and attention in
vation is insufficient for discovering that the surface of still visual search for features and for objects. Journal of Exper¬
water is invariently horizontal. Science, 181, 173-174. imental Psychology: Human Perception and Performance,
Thombury, J. M., & Mistretta, C. M. (1981). Tactile sensitivity 8, 194-214.
as a function of age. Journal of Gerontology, 36, 34-39. Treisman, A. M. (1986a). Features and objects in visual pro¬
Timney, B. (1985). Visual experience and the development of cessing. Scientific American, 255, 114B-125.
depth perception. In D. J. Ingle, M. Jeannerod, & D. N. Treisman, A. M. (1986b). Properties, parts, and objects. In
Lee (Eds.), Brain mechanisms and spatial vision (pp. 147— K. R. Boff, L. Kaufman, & J. P. Thomas (Eds.), Handbook
174). Dordrecht: Martinus Nijhoff. of perception and human performance (pp. 35-1 to 35-70).
Timney, B., Mitchel, D. E., & Griffin, F. (1978). The devel¬ New York: Wiley.
opment of vision in cats after extended periods of dark rear¬ Treisman, A. M., & Davies, A. (1972). Divided attention to
ing. Experimental Brain Research, 31, 547-560. ear and eye. In S. Komblum (Ed.), Attention and perfor¬
Timney, B., & Muir, D. W. (1976). Orientation anisotropy: mance IV (pp. 101-118). New York: Academic Press.
Incidence and magnitude in Caucasian & Chinese subjects. Treisman, A. M., & Gelade, G. (1980). A feature-integration
Science, 193, 699-700. theory of attention. Cognitive Psychology, 12, 97-136.
Toch, H. H., & Schulte, R. (1961). Readiness to perceive vio¬ Treisman, A. M., & Gormican, S. (1988). Feature analysis in
lence as a result of police training. British Journal of Psy¬ early vision: Evidence from search asymmetries. Psycholog¬
chology, 52, 389-393. ical Review, 95, 15-48.
Todd, J. T. (1983). Perception of gait. Journal of Experimental Treisman, A. M. & Schmidt. H. (1982). Illusory conjunctions
Psychology: Human Perception and Performance, 9, 31 — in the perception of objects. Cognitive Psychology, 14, 107—
42. 141.
Todd, J. T., & Akerstrom, R. A. (1987). Perception of three- Treisman, M. (1963). Temporal discrimination and the indiffer¬
dimensional form from patterns of optical texture. Journal ence interval: Implications for the model of an internal
of Experimental Psychology: Human Perception and Perfor¬ clock. Psychological Monographs, 77, 1-31 (Whole No.
mance, 13, 242-255. 576).
Tomita, T. (1986). Retrospective review of retinal circuitry. Vi¬ Treisman, M. (1976). On the use and misuse of psychophysical
sion Research, 26, 1339-1350. terms. Psychological Review, 83, 246-256.
Torebjork, H. E., & Hallin, R. G. (1973). Perceptual changes Treisman, M., & Watts, T. R. (1966). Relation between signal
accompanying controlled, preferential blocking of A and C detectability theory and the traditional procedures for mea¬
fibre responses in intact human skin nerves. Experimental suring sensory thresholds: Estimating d' from results given
Brain Research, 16, 321-332. by the method of constant stimuli. Psychological Bulletin,
Torebjork, H. E., Ochoa, J. L., & Schady, W. J. L. (1983). 66, 438-454.
Role of single mechanoreceptor units in tactile sensation. In Treisman, M., & Williams, T. C. (1984). A theory of criterion
C. von Euler, O. Franzen, U. Lindblom, & D. Ottoson setting with an application to sequential dependencies. Psy¬
(Eds.), Somatosensory mechanisms (pp. 173-184). New chological Review, 91, 68-111.
York: Plenum. Tress, K. H., & Kugler, B. T. (1979). Interocular transfer of
Torgerson, W. S. (1961). Distances and ratios in psychophysi¬ movement aftereffects in schizophrenia. British Journal of
cal scaling. Acta Psychologica, 19, 201-205. Psychology, 70, 389-392.
Townshend, B., Cotter, N., Van Compemolle, D., & White, Troland, L. T. (1921). The enigma of color vision. American
R. L. (1987). Pitch perception of cochlear implant subjects. Journal of Physiological Optics, 2, 23-48.
Journal of the Acoustical Society of America, 82, 106— Tronick, E. (1972). Stimulus control and the growth of the in¬
115. fant's effective visual field. Perception & Psychophysics,
Townsend, J., & Ashby, F. G. (1982). Experimental test of 11, 373-376.
contemporary mathematical models of visual letter recogni¬ Tsai, Y. (1983). Movements of attention across the visual field.
tion. Journal of Experimental Psychology: Human Percep¬ Journal of Experimental Psychology: Human Perception and
tion and Performance, 8, 834-864. Performance, 9, 523-530.
Trehub, S. (1976). The discrimination of foreign speech con¬ Tsvetkova, L. S. (1972). Rehabilitative training in local brain
trasts by infants and adults. Child Development, 47, 466- lesions. Moscow: Pedagogika Publishing House.
472. Turnbull, C. (1961). Some observations regarding the experi¬
Trehub, S. E., & Schneider, B. A. (1987). Problems and prom¬ ences and behavior of the Bambuti pygmies. American Jour¬
ises of developmental psychophysics: Throw out the bath nal of Psychology, 74, 304-308.
water but keep the baby. In M. Teghtsoonian & R. Teght- Turner, P. (1968). Amphetamines and smell threshold in man.
soonian (Eds.), Fechner Day '87 (pp. 43-47). Northamp¬ In A. Herxheimer (Ed.), Drugs and sensory functions (pp.
ton, MA: International Society of Psychophysics. 91-100). Boston: Little, Brown.
 References 577

Uhlarik, J., & Johnson, R. (1978). Development of form per¬ Vogels, R., & Orban, G. A. (1986). Decision factors affecting
ception in repeated brief exposures to visual stimuli. In R. line orientation judgments in the method of single stimuli.
Walk & L. Pick, Jr., (Eds.), Perception and Experience. Perception & Psychophysics, 40, 74-84.
New York: Plenum. Volkmann, F. C., Riggs, L. A., Aimee, G., & Moore, R. K.
Ullman, S. (1986). Visual routines: Where bottom-up and top- (1982). Measurements of visual suppression during opening,
down processing meet. In E. C. Schwab & H. C. Nusbaum closing and blinking of the eyes. Vision Research, 22, 991 —
(Eds.), Pattern recognition by humans and machines: Vol. 996.
2. Visual perception (pp. 159-218). Orlando, FL: Academic Vurpillot, E. (1968). The development of scanning strategies
Press. and their relation to visual differentiation. Journal of Exper¬
Umezaki, H. & Morrell, F. (1970). Developmental study of imental Child Psychology, 6, 632-650.
photic evoked responses in premature infants. Electroen¬
cephalography and Clinical Neurophysiology, 28, 55-63.
Ungerleider, L. G., & Mishkin, M. (1982). Two cortical visual Waber, D. P. (1976). Sex differences in cognition: A function
systems. In D. J. Ingle, M. A. Goodale, & R. J. W. Mans¬ of maturation rate? Science, 192, 572-574.
field (Eds.), Analysis of visual behavior (pp. 549-586). Waber, D. P. (1977). Sex differences in mental abilities, hemi¬
Cambridge, MA: MIT Press. spheric lateralization and rate of physical growth at adoles¬
Uttal, W. (1981). A taxonomy of visual processes. Hillsdale, cence. Developmental Psychology, 13, 29-38.
NJ: Lawrence Erlbaum. Wade, N. J. (1984). Brewster & Wheatstone on vision. New
York: Academic Press.
Waespe, W., & Henn, V. (1977). Neuronal activity in the ves¬
Vallbo. A. (1981). Sensations evoked from the glabrous skin of tibular nuclei of the alert monkey during vestibular and op¬
the human hand by electrical stimulation of unitary me- tokinetic stimulation. Experimental Brain Research, 27,
chano-sensitive afferents. British Research, 215, 359-363. 523-538.
Vallbo, A. (1983). Tactile sensation related to activity in pri¬ Wahl, O. F., & Sieg, D. (1980). Time estimation among
mary afferents with special reference to detection problems. schizophrenics. Perceptual and Motor Skills, 50, 535-541.
In C. von Euler, O. Franzen, U. Lindblom, & D. Ottoson Wald, G. (1968). The molecular basis of visual excitation. Na¬
(Eds.), Somatosensory mechanisms (pp. 163-172). New ture (London), 219, 800-807.
York: Plenum. Walk, R. D., & Gibson, E. J. (1961). A comparative and ana¬
van der Heijden, A. H. C., Wolters, G., Groep, J. C., & Hag- lytic study of visual depth perception. Psychological Mono¬
enaar, R. (1987). Single-letter recognition accuracy benefits graphs, 75, 1-44.
from advance cuing of location. Perception & Psychophys¬ Wall, P. D. (1979). On the relation of injury to pain. Pain, 6,
ics, 42, 503-509. 253-264.
Varner, D., Cook, J. E., Schneck, M. E., McDonald, M., & Wallace, B., & Priebe, F. A. (1985). Hypnotic susceptibility,
Teller, D. (1985). Tritan discriminations by 1- and 2-month- interference and alternation frequency to the Necker cube
old human infants. Vision Research, 6, 821-831. illusion. Journal of General Psychology, 112, 271-277.
Vautin, R. G., & Berkley, M. A. (1977). Responses of single Wallace, P. (1977). Individual discrimination of humans by
cells in cat visual cortex to stimulus movement: Neural cor¬ odor. Physiology and Behavior, 19, 577-579.
relates of visual after-effects. Journal of Neurophysiology, Wallach, H. (1939). On sound localization. Journal of the
40, 1051-1065. Acoustical Society of America, 10, 270-274.
Verriest, G. (1974). Recent advances in the study of the ac¬ Wallach, H. (1948). Brightness constancy and the nature of a-
quired deficiencies of color vision. Fondazione “Gorgio chromatic colors. Journal of Experimental Psychology, 38,
Ranchi,” 24, 1-80. 310-324.
Verrillo, R. T. (1975). Cutaneous sensation. In B. Scharf (Ed.), Wallach, H. (1972). The perception of neutral colors. In R.
Experimental sensory psychology (pp. 150-184). Glenview, Held & W. Richards (Eds.), Perception: Mechanisms and
IL: Scott, Foresman. models: Readings from Scientific American (pp. 278-285).
Verrillo, R. T., & Bolanowski, S. J., Jr. (1986). The effects of San Francisco: Freeman. (Original work published in Sci¬
skin temperature on the psychophysical responses to vibra¬ entific American, 1963)
tion on glabrous and hairy skin. Journal of the Acoustical Wallach, H. (1987). Perceiving a stable environment when one
Society of America, 80, 528-532. moves. Annual Review of Psychology, 38, 1-27.
Verrillo, T. R., Fraioli, A. J., & Smith, R. L. (1969). Sensa¬ Wallach, H., & Becklen, R. (1983). An effect of speed on in¬
tion magnitude of vibrotactile stimuli. Perception & Psycho¬ duced motion. Perception & Psychophysics, 34, 237-242.
physics, 6, 366-372. Wallach, H., Becklen, R., & Nitzberg, D. (1985). Vector anal¬
Vierck, C. (1978). Somatosensory system. In R. B. Masterston ysis and process combination in motion perception. Journal
(Ed.), Handbook of sensory neurobiology: Vol. 1. Sensory of Experimental Psychology: Human Perception and Perfor¬
integration (pp. 249-310). New York: Plenum. mance, 11, 93-102.
Vimal, R. L. P., Pokomy, J., & Smith, V. C. (1987). Appear¬ Wallach, H., Newman, E. B., & Rosenzweig, M. R. (1949).
ance of steadily viewed lights. Vision Research, 27, 1309- The precedence effect in sound localization. American Jour¬
1318. nal of Psychology, 62, 315-336.
 578 References

Walley, A., Pisoni, D., & Aslin, R. (1981). The role of early Foundations of modern auditory theory: Vol. 1. (pp. 407-
experience in the development of speech perception. In R. 447). New York: Academic Press.
Aslin, J. Alberts, & M. Petersen (Eds.), Development of Ware, C. (1981). Subjective contours independent of subjective
perception: Psychobiological perspectives: Vol. 1. Audition, brighteners. Perception & Psychophysics, 29, 500-504.
somatic perceptions, and the chemical senses (pp. 219— Ware, C., & Cowan, W. B. (1987). Chromatic Mach bands:
256). New York: Academic Press. Behavioral evidence for lateral inhibition in human color vi¬
Walls, G. L. (1951). A theory of ocular dominance. AMA Ar¬ sion. Perception & Psychophysics, 41, 173-178.
chives of Ophthalmology, 45, 387-412. Warm, J. S., & McCray, R. E. (1969). Influence of word fre¬
Ward, L. M. (1971). Some psychophysical properties of cate¬ quency and length on the apparent duration of tachistoscopic
gory judgments and magnitude estimations. Unpublished presentations. Journal of Experimental Psychology, 79, 56-
doctoral dissertation, Duke University, N.C. 58.
Ward, L. M. (1972). Category judgments of loudness in the Warren, D. H. (1984). Blindness and early childhood develop¬
absence of an experimenter-induced identification function: ment. New York: American Foundation for the Blind.
Sequential effects and power function fit. Journal of Exper¬ Warren, R. M. (1970). Perceptual restoration of missing speech
imental Psychology, 94, 179-184. sounds. Science, 167, 392-393.
Ward, L. M. (1973). Repeated magnitude estimations with a Warren, R. M., Obusek, C. J., Farmer, R. M., & Warren,
variable standard: Sequential effects and other properties. R. P. (1969). Auditory sequence: Confusion of patterns
Perception & Psychophysics, 13, 193-200. other than speech or music. Science, 164, 586-587.
Ward, L. M. (1974). Power functions for category judgments Watanabe, T., & Katsuki, Y. (1974). Response patterns of sin¬
of duration and line length. Perceptual and Motor Skills, 38, gle auditory neurons of the cat to species-specific vocaliza¬
1182. tion. Japanese Journal of Physiology, 24, 135-155.
Ward, L. M. (1975). Sequential dependencies and response Watkins, L. R., & Mayer, D. J. (1982). Organization of endog¬
range in cross-modality matches of duration to loudness. enous opiate and nonopiate pain control systems. Science,
Perception & Psychophysics, 18, 217-223. 216, 1185-1192.
Ward, L. M. (1979). Stimulus information and sequential de¬ Weale, R. A. (1979). Discoverers of Mach-bands. Investigative
pendencies in magnitude estimation and cross-modality Ophthalmology and Visual Sciences, 18, 652-654.
matching. Journal of Experimental Psychology: Human Per¬ Weale, R. A. (1982). Focus on vision. Cambridge, MA: Har¬
ception and Performance, 5, 444-459. vard University Press.
Ward, L. M. (1982a). Mixed-modality psychophysical scaling: Weale, R. A (1986). Aging and vision. Vision Research, 26,
Sequential dependencies and other properties. Perception & 1507-1512.
Psychophysics, 31, 53—62. Weber, E. H. (1834). De pulen, resorptione, auditu et tactu:
Ward, L. M. (1982b). Determinants of attention to local and Annotationes anatomicae et physiologicae. Leipzig: Koeh¬
global features of visual forms. Journal of Experimental ler.
Psychology: Human Perception and Performance, 8, 562- Webster, W. R., & Atkin, L. M. (1975). Central auditory pro¬
581. cessing. In M. S. Gazzaniga & C. Blakemore (Eds.), Hand¬
Ward. L. M. (1983). On processing dominance: Comment on book of sensory psychobiology (pp. 325-364). New York:
Pomerantz. Journal of Experimental Psychology: General, Academic Press.
112, 541-546. Wegener, B. (Ed.). (1982). Social attitudes and psychophysical
Ward, L. M. (1985). Covert focussing of the attentional gaze. measurement. Hillsdale, NJ: Lawrence Erlbaum.
Canadian Journal of Psychology, 39, 546-563. Weil, A. T., Zinberg, E., & Nelson, J. N. (1968). Clinical and
Ward, L. M. (1986). Mixed-modality psychophysical scaling: psychological effects of marijuana in man. Science, 162,
Double cross-modality matching for “difficult” continua. 1234-1242.
Perception & Psychophysics, 39, 407-417. Weinstein, E. A., Cole, M., Mitchell, M. S., & Lyerly, O. G.
Ward, L. M. (1987). Remembrance of sounds past: Memory (1964). Anosognosia and aphasia. Archives of Neurology,
and psychophysical scaling. Journal of Experimental Psy¬ 10, 376-386.
chology: Human Perception and Performance, 13, 2lb- Weinstein, S. (1968). Intensive and extensive aspects of tactile
227. sensitivity as a function of body part, sex, and laterality. In
Ward, L. M., & Lockhead, G. R. (1970). Sequential effects D. R. Kenshalo (Ed.), The skin senses (pp. 195-218).
and memory in category judgments. Journal of Experimental Springfield, IL: Thomas.
Psychology, 854, 27-34. Weinstein, S., & Sersen, E. A. (1961). Tactual sensitivity as a
Ward, L. M., Porac, P., Coren, S., & Girgus, J, S. (1977). function of handedness and laterality. Journal of Compara¬
The case for misapplied constancy scaling: Depth associa¬ tive & Physiological Psychology, 54, 665-669.
tions elicited by illusion configurations. American Journal of Weisenberg, M. (1984). Cognitive aspects of pain. In P. D.
Psychology, 90, 609-620. Wall & R. Melzack (Eds.), Textbook of pain (pp. 162-172).
Ward, T. B. (1985). Individual differences in processing stim¬ Edinburgh: Churchill Livingstone.
ulus dimensions: Relation to selective processing ability. Weisstein, N. (1968). Rashevsky-Landahl neural net: Simula¬
Perception & Psychophysics, 37, 471-482. tion of metacontrast. Psychological Review, 75, 494-521.
Ward, W. D. (1970). Musical perception. In J. V. Tobias (Ed.), Weisstein, N. (1980). Tutorial: The joy of Fourier analysis. In
 References 579

C. S. Harris (Ed.), Visual coding and adaptability (pp. 365— 137). Princeton, NJ: Van Nostrand-Reinhold. (Original
380). Hillsdale, NJ: Lawrence Erlbaum. work published 1923, Psychologische Forschung, 41, 301—
Weisstein, N., Harris, C., Berbaum, K., Tangney. J., & Wil¬ 350)
liams, A. (1977). Contrast reduction by small localized Wertheimer, M. (1961). Psychomotor coordination of auditory
stimuli: Extensive spatial spread of above-threshold orienta¬ and visual space at birth. Science, 134, 1692.
tion-selective masking. Vision Research, 17, 341-350. Westheimer, G. (1979). Spatial sense of the eye. Investigative
Weisstein, N., Matthews, M., & Berbaum, K. (1974, Novem¬ Ophthalmology and Visual Science, 18, 893-912.
ber). Illusory contours can mask real contours. Paper pre¬ Wever, E. G. (1970). Theory of hearing. New York: Wiley.
sented at the meetings of the Psychonomic Society, Boston. Wever, R. A. (1979). The circadian system of man. New York:
Weisstein, N., Ozog, G., & Szoc, R. (1975). A comparison and Springer-Verlag.
elaboration of two models of metacontrast. Psychological Whalen, D. H., & Liberman, A. M. (1987). Speech perception
Review, 82, 325-343. takes precedence over nonspeech perception. Science, 237,
Weisstein, N., & Wong, E. (1986). Figure-ground organization 169-171.
and the spatial and temporal responses of the visual system. White, B. L. (1971). Human infants. Englewood Cliffs, NJ:
In E. C. Schwab & H. C. Nusbaum (Eds.), Pattern recog¬ Prentice-Hall.
nition by humans and machines: Vol. 2. Visual perception White, B. W., Saunders, F. A., Scadden, L., Bach-y-Rita, P.,
(pp. 31-64). Orlando, FL: Academic Press. & Collins, C. C. (1970). Seeing with the skin. Perception
Welch. R. B. (1969). Adaptation to prism-displaced vision: The & Psychophysics, 7, 23-27.
importance of target pointing. Perception & Psychophysics, White, C. (1963). Temporal numerosity and the psychological
5, 305-309. unit of duration. Psychological Monographs, 77, 1-37.
Welch, R. B. (1971). Prism adaptation: The "target pointing (Whole No. 575)
effect” as a function of exposure trials. Perception & Psy¬ White, C. W., Lockhead, G. R., & Evans, N. J. (1977). Mul¬
chophysics, 5, 102-104. tidimensional scaling of subjective color-blind observers.
Welch, R. B. (1978). Perceptual modification: Adapting to al¬ Perception & Psychophysics, 21, 522-526.
tered sensory environments. New York: Academic Press. White, C. W., & Montgomery, D. A. (1976). Memory colours
Welford, A. T. (1980). Reaction times. London: Academic in afterimages: A bicentennial demonstration. Perception &
Press. Psychophysics, 19, 371-374.
Well, A. D., Lorch, E. P., & Anderson, D. R. (1980). Devel¬ White, K. D., & Odom, J. V. (1985). Temporal integration in
opmental trends in distractability: Is absolute or proportional global stereopsis. Perception & Psychophysics, 37, 139—
decrement the appropriate measure of interference? Journal 144.
of Experimental Child Psychology, 30, 109-124. Whitfield, I. C. (1967). The auditory pathway. London: Arnold.
Wells, G. L., & Loftus, E. (1984). Eyewitness testimony: Psy¬ Whitfield, I. C. (1978). The neural code. In E. C. Carterette &
chological perspectives. Cambridge: Cambridge University M. P. Friedman (Eds.), Handbook of perception: Vol. 4.
Press. Hearing (pp. 163-183). New York: Academic Press.
Werker, J., Gilbert, J., Humphrey, K., & Tees, R. (1981). De¬ Whitfield, I. C. (1980). Auditory cortex and the pitch of com¬
velopmental aspects of cross-language speech perception. plex tones. Journal of the Acoustical Society of America, 67,
Child Development, 52, 349-355. 644-647.
Werker, J. F., & Logan, J. S. (1985). Cross-language evidence Whitfield, I. C., & Evans, E. F. (1965). Responses of auditory
for three factors in speech perception. Perception & Psycho¬ cortical neurons to stimuli of changing frequency. Journal
physics, 37, 35-44. of Neurophysiology, 28, 655-672.
Werker, J. F., & Tees, R. C. (1984). Cross-language speech Whitsel, B. L., Dreyer, D. A., Hollins, M., & Young, M. G.
perception: Evidence for perceptual reorganization during (1979). The coding of direction of tactile stimulus move¬
the first year of life. Infant Behavior and Development, 7, ment: Correlative psychophysical and electrophysiological
49-63. data. In D. R. Kenshalo (Ed.), Sensory functions of the skin
Werner, H. (1935). Studies on contour. American Journal of of humans (pp. 79-108). New York: Plenum.
Psychology, 47, 40-64. Whorf, B. L. (1956). Science and linguistics. In J. B. Carroll
Werner, J. S. (1979). Developmental change in scotopic sensi¬ (Ed.), Language, thought and reality: Selected writings of
tivity and the absorption spectrum of the human ocular me¬ Benjamin Lee Whorf (pp. 207-219). Cambridge, MA: MIT
dia. Unpublished doctoral dissertation, Brown University, Press.
RI. Whytt, R. (1751). An essay on the vital and other involuntary
Werner, .1. S., & Wooten, B. R. (1979). Human infant color motions of animals. Edinburgh: Balfour & Neill.
vision and color perception. Infant Behavior and Develop¬ Wickens, C. D. (1984). Processing resources in attention. In R.
ment, 2, 241-274. Parasuraman & D. R. Davies (Eds.), Varieties of attention.
Wertheimer, M. (1912). Experimentelle Studien uber das Sehen (pp. 63-101). Orlando, FL: Academic Press.
von Bewegung. Zeitschrift fur Psychologie, 61, 161-265. Wiener, N. (1961). Cybernetics (2nd ed.). Cambridge, MA:
Wertheimer, M. (1923). Principles of perceptual organization. MIT Press.
Abridged translation by M. Wertheimer. In D. S. Beardslee Wier, C. C., Jestead: W., & Green, D. M. (1977). Frequency
& M. Wertheimer (Eds.), Readings in perception (p. 115- discrimination as a function of frequency and sensation
580 References

level. Journal of the Acoustical Society of America, 61, servations of the neuro-historical basis of cutaneous pain.
178-184. Journal of Anatomy, 74, 413-440.
Wilier, J. C., Dehen, H., & Cambier, J. (1981). Stress-induced Worchel, P., & Dallenbach, K. M. (1947). “Facial vision:”
analgesia in humans: Endogeneuous opioids and naloxone- Perception of obstacles by the deaf-blind. American Journal
reversible depression of pain reflexes. Science, 212, 689- of Psychology, 60, 502-553.
690. Wright, L. L., & Elias, J. W. (1979). Age differences in the
Williams, D. R., MacLeod, D. I. A., Hayhoe, M. M. (1981). effects of perceptual noise. Journal of Gerontology, 34,
Foveal tritanopia. Vision Research, 21, 1341-1356. 704-708.
Williams, J. M. (1979). Distortions of vision and pain: Two Wright, M. J., & Johnston, A. (1985). Invarient tuning of mo¬
functional facets of D-Lysergic diethylamide. Perceptual & tion aftereffect. Vision Research, 25, 1947-1955.
Motor Skills, 49, 499-528. Wright, N. H. (1964). Temporal summation and backward
Williams, M. (1970). Brain damage and the mind. London: masking. Journal of the Acoustical Society of America, 36,
Penguin. 927-932.
Willis, W. D. (1983). Descending control of nociceptive trans¬ Wright, R. H. (1977). Odor and molecular vibration: Neural
mission by primate spinothalamic neurons. In C. von Euler, coding of olfactory information. Journal of Theoretical Bi¬
O. Franzen, U. Lindblom, & D. Ottoson (Eds.), Somatosen¬ ology, 64, 473-502.
sory mechanisms (pp. 296-308). New York: Plenum. Wright, R. H. (1978a). Specific anosmia: A clue to the olfac¬
Willis, W. D. (1985). The pain system. The neural basis of tory code or to something much more important? Chemical
nococeptive transmission in the mammalian nervous system. Senses and Flavor, 3, 235-239.
Basel: Karger. Wright, R. H. (1978b). The perception of odor intensity: Phys¬
Wilson, E. O. (1971). The insect societies. Cambridge, MA: ics or psychophysics? Chemical Senses and Flavor, 3, 73-
Harvard University Press. 79.
Wilson, H. R. (1986). Responses of spatial mechanisms can Wright, R. H. (1978c). The perception of odor intensity: Phys¬
explain hyperacuity. Vision Research, 26, 453-469. ics or psychophysics II. Chemical Senses and Flavor, 3,
Wilson, J. R., DeFries, J. C., McCleam, G. C., Vandenberg, 241-245.
S. G., Johnson, R. C., & Rashad, M. N. (1975). Cognitive Wright, R. H. (1982). The sense of smell. Boca Raton, FL:
abilities: Use of family data as a control to assess sex and CRC Press.
age differences in two ethnic groups. International Journal Wright, W. D. (1929). A re-determination of the trichromatic
of Aging and Human Development, 6, 261-275. mixture data. Medical Research Council (Great Britain),
Wilson, M. (1957). Effects of circumscribed cortical lesions Special Report Series, SRS-139, 1-38.
upon somesthetic and visual discrimination in the monkey. Wright, W. D. (1952). The characteristics of tritanopia. Journal
Journal of Comparative and Physiological Psychology, 50, of the Optical Society of America, 42, 509-521.
630-635. Wurtz, R. H., & Goldberg, M. E. (1971). Superior colliculus
Witkin, H. A., & Berry, J. W. (1975). Psychological differen¬ cell responses related to eye movements in awake monkeys.
tiation in cross-cultural perspective. Journal of Cross-Cul¬ Science, 171, 82-84.
tural Psychology, 6, 4-87. Wybum. G. M., Pickford, R. W., & Hurst. R. J. (1964). Hu¬
Wolff, H. G., & Goodell, B. S. (1943). The relation of attitude man senses and perception. Toronto: University of Toronto
and suggestion to the perception of and reaction to pain. Press.
Research Publications, Associaton for Research in Nervous Wyszecki, G., & Stiles. W. S. (1967). Color science: Concepts
and Mental Disease, 23, 434-448. and methods, quantitative data and formulas. New York:
Wong, C., & Weisstein, N. (1982). A new perceptual context- Wiley.
superiority effect: Line segments are more visible against a
figure than against a ground. Science, 218, 587-589. Yaksh, T. L. (1984). Multiple spinal opiate receptor systems in
Wong, E., & Weisstein, N. (1987). The effects of flicker on the analgesia. In L. Kruger & J. C. Liebeskind (Eds.), Neural
perception of figure and ground. Perception & Psychophys¬ mechanisms of pain (pp. 197-216). New York: Raven.
ics, 41, 440-448. Yamamoto, T., Yayama, N., & Kawamura, Y. (1981). Central
Woo, G., & Bader, D. (1978). Age and its effect on vision. processing of taste perception. In Y. Katsuki, R. Norgren,
Canadian Journal of Optometry, 40, 29-34. & M. Sato (Eds.), Brain mechanisms of sensation (pp. 197—
Wood, R. W. (1985). The “haunted swing” illusion. Psycho¬ 208). New York: Wiley.
logical Review, 2, 277-278. Yantis, S., & Jonides, J. (1984). Abrupt onsets and selective
Woodfield, R. L. (1984). Embedded figures test performance attention: Evidence from visual search. Journal of Experi¬
before and after childbirth. British Journal of Psychology, mental Psychology: Human Perception and Performance,
75, 81-88. 10, 601-621.
Woodrow, H. (1951). Time perception. In S. S. Stevens (Ed.), Yarbus, A. L. (1967). Eye movements and vision. New York:
Handbook of experimental psychology (pp. 1224-1236). Plenum.
New York: Wiley. Yarmey, A. (1979). The psychology of eyewitness testimony.
Woolard, H. H., Weddell, G., & Harpman, J. A. (1940). Ob¬ New York: Free Press.
 References 581

Yen, W. (1975). Sex-linked major gene influence on selected Zakay, D., Nitzan, D., & Glicksohn, J. (1983). The influence
types of spatial performance. Behavior Genetics, 5, 281 — of task difficulty and external tempo on subjective time es¬
298. timation. Perception <£ Psychophysics, 34, 451-456.
Yerkes, R. M., & Dodson, J. D. (1908). The relation of Zaporozhets, A. V. (1965). The development of perception in
strength of stimulus to rapidity of habit formation. Journal the preschool child. Monographs of the Society for Research
of Comparative Neurology and Psychology, 18, 459-482. in Child Development, 30, 82-101.
Yin, R. K. (1970). Face recognition by brain injured patients— Zeigler, H. P., & Leibowitz, H. (1957). Apparent visual size as
a dissociable ability. Neuropsychologia, 8, 395. a function of distance for children and adults. American
Yodogawa, E. (1982). Symmetropy, an entropy-like measure of Journal of Psychology, 70, 106-109.
visual symmetry. Perception Psychophysics, 32, 230- Zigler, M. J. (1932). Pressure adaptation time: A function of
240. intensity and extensity. American Journal of Psychology,
Yonas, A. (1981). Infants’ response to optical information for 44, 709-720.
collision. In R. N. Aslin, J. R. Alberts, & M. R. Petersen Zihl, J., von Cramon, D., & Mai, N. (1983). Selective distur¬
(Eds.), Development of perception (pp. 313-334). New bance of movement vision after bilateral brain damage.
York: Academic Press. Brain, 106, 313-340.
Yonas, A., Cleaves, W., & Pettersen, L. (1978). Development Zimmerman, M. (1983). Centrifugal control of somatosensory
of sensitivity to pictorial depth. Science, 200, 77-79. inflow into the spinal cord. In C. von Euler, O. Franzen, U.
Yonas, A., Goldsmith, L. T., & Hallstrom, J. (1978). Devel¬ Lindblom, & D. Ottoson (Eds.), Somatosensory mechanisms
opment of sensitivity to information provided by cast shad¬ (pp. 285-296). New York: Plenum.
ows in pictures. Perception, 7, 333-341. Zotterman, Y. (1959). Thermal sensations. In J. Fields, H. W.
Yonas, A., & Granrud, C. E. (1985a). Development of visual Magoun, & V. E. Hall (Eds.), Handbook of physiology:
space perception in young infants. In J. Mehler & R. Fox Section 1. Neurophysiology, 1, 431-458.
(Eds.), Neonate cognition: Beyond the blooming buzzing Zucker, I., Wade, G., & Ziegler, R. O. (1972). Sexual and
confusion (pp. 45-68). Hillsdale, NJ: Lawrence Erlbaum. hormonal influences on eating, taste preferences, and body
Yonas, A., & Granrud, C. E. (1985b). The development of sen¬ weight of hamsters. Physiology and Behavior, 8, 101-111.
sitivity to kinetic, binocular and pictorial depth information Zuidema, P., Gresnight, A. M., Bouman, M. A., & Koender-
in human infants. In D. Ingle, D. Lee, & M. Jeannerod ink, J. J. (1978). A quanta coincidence model for absolute
(Eds.), Brain mechanisms and spatial vision (pp. 113-145). threshold vision incorporating deviations from Ricco’s Law.
Dordrecht, Netherlands: Nijoff. Vision Research, 18, 1685-1689.
Yoneshige, Y., & Elliott, L. L. (1981). Pure-tone sensitivity Zurek, P. M. (1980). The precedence effect and its possible role
and ear canal pressure at threshold in children and adults. in the avoidance of interaural ambiguities. Journal of the
Journal of the Acoustical Society of America, 70, 1272— Acoustical Society of America, 67, 952-964.
1276. Zwicker, E. (1958). Uber psychologische und methodische
Young, L. R. (1971). Pursuit eye tracking movements. In P. Grundlagen der Lautheit. Acustica, 8, 237-258.
Bach-y-Rita, C. C. Collins, & J. E. Hyde. (Eds.), The con¬ Zwicker, E. (1986). A hardware cochlear nonlinear preprocess¬
trol of eye movements (pp. 429-443). New York: Academic ing model with active feedback. Journal of the Acoustical
Press. Society of America, 80, 146-153.
Young, R. A. (1977). Some observations on temporal coding of Zwislocki, J. J. (1978). Masking: Experimental and theoretical
color vision: Psychophysical results. Vision Research, 17, aspects of simultaneous, forward, backward, and central
957-965. masking. In E. C. Carterette & M. P. Friedman (Eds.),
Yuille, J. (1984). Research and teaching with police: A Cana¬ Handbook of perception: Vol. 4. Hearing (pp. 283-336).
dian example. International Review of Applied Psychology, New York: Academic Press.
33, 5-23. Zwislocki, J. J., Damianopoulos, E. N., Buining, E., & Glantz,
Yund, E. W., Morgan, H., & Efron, R. (1983). The micropat- J. (1967). Central masking: Some steady-state and transient
tem effect and visible persistence. Perception & Psycho¬ effects. Perception & Psychophysics, 2, 59-64.
physics, 34, 209-213.
 x

'
 Author Index
Page numbers in italics indicate figures; Arvidson, K., 215 Baylor, D. A., 69, 70
page numbers followed by t indicate Asa, C. S., 232 Beauchamp, G. K., 468
tabular material. Aschoff, J., 373, 374, 378 Bechtoldt, H. P., 301
Ashby, F. G., 331 Beck, J., 105, 319, 320, 422
Aantaa, E., 457 Aslin, R. N., 293, 301, 460, 461, 468, Beck, N. C., 270
Aaron, M., 520 480, 483, 505 Becker, W., 397
Abbs, J. H., 360 Aspinall, P. A., 146 Becklen, R., 383, 384
Abramov, I., 457 Atkin, L. M., 171, 172 Becklin, R., 436, 437
Abramson, A., 360 Atkinson, J., 301, 458, 461, 468 Bedford, W., 506
Achom, E., 527 Attili, G., 232 Beeler, W. J., 267
Adam, N., 378 Attneave, F., 315, 317, 320 Beersma, D. G. M., 378
Adams, A. J., 513, 515 Aubert, H., 383 Begleiter, S., 513
Adams, A. S., 515 Augenstine, L. G., 376 Beidler, L. M., 214, 215
Adams, J. A., 260 Avant, L. L., 307, 378, 379 Bekesy, G. von, 19, 157, 158, 163,
Adams, R. D., 378 Ayley, J. B., 506 194, 205, 241, 247, 249
Aimee, G., 104 Bell, B., 474
Akerstrom. R. A., 281 Bach-y-Rita, P., 250, 251 Bern, S. L., 524
Akil, H., 268, 269 Bacon, S. P., 183 Bender, D. B., 84
Albers, H. E., 374 Baddeley, A. D., 377 Benedetti, F., 245
Albrecht, D. G., 114 Bader, G. R., 167 Bennett, K. B., 449
Albrecht, R. E., 526 Bader, N., 475 Benson, A., 261
Albright, T. D., 84 Baggen, S., 331 Bentham, J. van, 372
Alexander, J. B., 526 Bailey, I. L., 289 Berbaum, K., 276, 299, 310, 312, 386
Alexander, K. R., 147 Bailey, P., 374 Berg, K. M., 465
Algom, D., 267 Baird, J. C., 16, 21, 42, 46, 49, 328 Bergeijk, W. A. van, 157
Allen, J. A., 412 Baker, C. L., 385 Bergen, J. R., 319, 321
Allison, A. C., 226 Baldwin, B. A., 84 Bergerone, C., 289
Allman, J., 381 Balint, R., 517 Berglund, B., 423
Allport, D. A., 438 Ball, K., 380 Berglund, U., 423
Allyn, M. R., 142, 493 Ball, W., 302 Bergman, T., 469
Amerson, T. L., Jr., 194 Balogh, R. D., 233 Berkley, M. A., 385, 391
Ames, A., Jr., 322, 495 Banks, M. S., 457, 458, 459, 460, Berlin, B., 147
Amoore, J. E., 226, 228, 230 461, 463, 469 Berman, A. J., 492
Amure, B. D., 513 Bannatyne, A., 473 Berman, N., 482
Anastasopoulos, D., 397 Baratz, S. S., 278 Bemhaus, I. J., 97
Andersen, G. J., 398, 399 Barbeito, D., 289 Bernstein, L. R., 188
Anderson, D. J., 168 Barclay, C. D., 387-88 Berry, J. W., 526
Anderson, D. R., 471 Baribeau-Braun, J., 527 Bertrand, D., 167
Anderson, J. E., 467 Barlow, H. B., 82, 85 Besser, G., 514
Anderson, N. H., 44, 51 Barr, A., 468 Besso, ML, 372
Anderson, N. S., 32 Barrera, M., 464 Best, H., 266
Andrews, B. W., 106 Bartleson, C. J., 146 Bever, T., 276
Anliker, M., 301 Bartlett, B., 506 Beverly, K. I., 398
Annis, R. C., 486 Barton, M., 517 Bharucha, J. J., 343
Anstis, S. M., 145, 385, 386 Bartoshuk, L. M., 213, 217, 219, 220 Bhatia, B., 381
Anstis, T., 492 Baruch, C., 198 Biddulph, R., 188, 189
Antes, J. R., 378, 379, 440, 469, 470 Basala, M., 361 Biederman, I., 324, 333, 334, 440,
Antonis, B., 438 Bashford, J. A., 365 442
Aoki, C., 449 Bass, E., 289 Bierman, K., 378
Applebaum, S., 232 Batteau, D. W., 191 Billings, B. L., 185
Arend, L. E., 103 Bauer, J. A., 399, 488, 489 Bindra, D., 227, 380
Arlin, M., 379 Baumgartner, G., 301 Birch, E. E., 301

583
584 Author Index

Birren, J., All Bregman, A. S., 345 Cain, D. P., 227

Bishop, P. O., 293 Brewster, J. M., 376 Cain, W. S., 219, 223, 228, 229, 233
Bissett, R. J., 41 Briand, K. A., 449 Calderone, L., 223
Blake, R., 178, 290, 484 Bridgeman, B., 287 Caldwell, R., 465
Blakemore, C., 458, 482 Bridges, C. D. B., 70 Calis, G., 320
Blakeslee, A. F., 218 Brill, S., 468, 486 Callaghan, T. C., 320
Blarney, P. J., 168 Brillat-Savarin, J. A., 222-23 Cambier, J., 268
Blasdel, G. G., 485 Broadbent, D. E., 447, 450 Cameron, S., 471
Blazynski, C., 69 Brodie, E., 261 Campbell, D. T., 503
Bliss, J. C., 251 Brodmann, K., 77 Campbell, F. W., 321, 381
Block, R. A., 378, 379 Brody, B. A., 517 Campbell, K. B., 526, 527
Blough, P. M., 521 Brooks, B., 255 Canevet, G., 196, 198
Boardman, W. K., 378 Brooks, R. A., 331 Cannon, M. W., Jr., 458
Bodis-Wollner, I., 301 Brooks, V., 315, 500 Carello, C., 274, 405, 410
Boer, L., 328 Brooksbank, B. W. L., 233 Carey, S., 464, 472, 473
Bolanowski, S. J., Jr., 247 Broota, K. D., 412 Carlson, C. R., 106
Boll, F., 69 Brou, P., 423 Carlson, N. R., 530
Bolton, T. L., 345 Brown, A. C., 267 Carlson, V. R., 418, 515
Bonnet,' C., 383 Brown, A. M., 168 Carmichael, L., 498
Bonsall, R. W., 232 Brown, B., 381, 513, 515 Caroll, D., 233
Booth, D. A., 233 Brown, E. L., 350 Caron, A., 465
Borbely, A. A., 378 Brown, J. L., 97 Caron, R., 465
Borg, G., 221 Brown, P. K., 133 Carpenter, D. L., 283
Boring, E. G., 408 Brown, P. E., 269 Carrell, T. D., 348
Bomstein, M. H., 135, 136, 147, 462, Brown, R. E., 231, 232 Carter, R. R., 44
463, 464, 475 Brown, S. W., 378, 379 Carterette, E. C., 345, 348
Borreson, C. R., 419 Brown, T. S., 225 Casey, K. L., 263, 264, 266, 270
Bose, H. E., 220 Brown, W., 33 Casseday, J. H., 193
Bossiere, E. La, 457 Brownell, W. E., 167 Cattell, J. M., 38
Bossom, J., 492 Browse, R. A., 320 Cavanagh, P., 143
Botte, M. C., 198 Bruce, C., 84 Cavanaugh, P., 321
Botwinick, J., All Brugge, J. F., 168, 194 Cavanaugh, P. H., 232
Bouffard, J. A., 520 Bruner, J. S., 146, 469 Cegalis, J. A., 527
Bouman, M. A., 98 Bruns, V., 193 Cei, A., 289
Bourne, L. E., 417 Brunswick, E., 299, 405, 495 Cerella, J., All
Bowen, R. W., 143 Brussell, E. M., 103 Ceridono, D., 289
Bowers, J. N., 194 Bryden, M. P., 358 Cemoch, J. M., 232, 233, 468
Bowker, D. O., 486 Buckhout, R., 498 Chambers, D., 383
Bowmaker, J. K., 133, 135 Bucy, P. C., 83 Chapman, C. R., 266, 269, 476
Bownds, M. D., 69 Buining, E., 184 Charman, W. N., 106
Boyd, I. A., 260 Bull, R., 508 Cheeseman, J., 450
Boynton, R. M., 135, 136, 138, 143, Bundesen, C., 386 Cheng, J. O., 269
147 Burg, A., 521 Cheng, P. W., 446
Brabyn, 521 Buri, M., 289 Cherry, E. C., 434
Braddick, O. J., 301, 385, 458, 461, Burnham, R. W., 146 Chevrier, J., 408
468 Bums, E. M., 341 Chocolle, R., 38, 181
Bradley, A., 114 Burt, G. S., 415 Cholewiak, R. W., 252
Bradley, D. R., 288, 390 Butler, D. L., 323 Chomsky, N., 355
Bradley, R. M., 467 Butler, R. A., 191, 194 Chou, C. L., 513
Braff, D. L., 515 Butter, C. M., 391 Choudhurt, B. P., 381
Braida, L. D., 31 Butters, N., 517 Chung, C. S., 276
Braine, L. G., 418 Butterworth, G., 429, 466 Clark, D. J., 120
Brand, J. G., 214 Byrd, R., 342 Clark, D. L., 378
Brandt, T., 108, 399 Clark, E. V., 349
Braun, C., 527 Cacace, A. T., 200 Clark, G. M., 168
Braunstein, M. L., 399 Caelli, T., 281, 321 Clark, H. H., 349
Brazelton, T., 461 Cagen, R. H., 225 Clark, J. C., 473
Breen, T. J., 446 Cahoon, D., 379 Clark, J. R., 146
 Author Index 585

Clark, S. E., 493 Craik, F., 477 Deptula, D., 527

Clark, T. B. Ill, 221 Cramon, D. von, 381 Deregowski, J., 501, 502, 504
Clark, W. C., 269 Crane, H. D., 165 Derrington, A. M., 114
Clarkson-Smith, L., 521 Craske, B., 492 Desaulniers, D. H., 195
Cleaves, W., 302 Cratty, B., 472 Desimone, R., 84
Clement, D. E., 317 Creed, R. S., 259 Deutsch, D., 341, 343, 344, 345, 346,
Clifford, B. R., 508 Creutzfeldt, O. D., 391 348, 450
Clifton, R. K., 466 Critchley, M., 516 Deutsch, G., 358, 472
Cogan, R., 268 Crockett, L., 524 Deutsch, J. A., 450
Cohen, A. J., 343 Crommelink, M., 381 DeValois, K. K., 76, 114, 138, 139,
Cohen, B., 400 Crook, C., 467, 468 293, 321
Cohen, K., 440, 470 Cross, I., 348 De Valois, R. L., 76, 79, 114, 138,
Cohen, S., 515 Crossey, A. D., 381 139, 293, 321
Cohen, W., 307 Crossman, E. R. F. W., 39 Devaney, K. O., 475
Cohen-Raz, L., 267 Crowley, W. P., Jr., 524 Dewsbury, 301
Cole, M., 517 Cruchfield, R. S., 526 Diamant, H., 215, 221
Cole, R, A., 355 Cuddy, L. L., 343 Diamond, R., 472, 473
Coles, M. G., 526 Cullen, J. K., Jr., 186 Diamond, R. M., 489
Collings, V. B., 217 Curcio, C. A., 68 Dichgans, J., 108, 399
Collins, C. C., 250, 251 Cutler, W. B., 234 Dickhaus, H., 264
Collins, J. F., 376 Cutting, J. E., 32, 358, 387-88, 398 Dickinson, R. G., 412
Comerford, J., 299 Cynader, M., 301, 482, 483 Diehl, R. L., 359, 360
Comfort, A., 230 Dijkstra, C., 373
Compemolle, D. Van, 168 Dineen, I. T., 464
Condon, W. S., 362 Daan, S., 373, 378 Ditchbum, R. W., 394
Conner, J. M., 524 Dacey, D. M., 73 DiZio, P. A., 400
Cook, J. E., 463 Dackis, C., 221 Djang, S., 497
Cook, V. T., 31 Dadson, R. S., 197 Dobelle, W. H., 133
Cooke, N. M., 446 Daher, M., 468 Dobson, V., 63, 457, 463
Cooper, B. Y., 263 Dallenbach, K. M., 193, 256, 257, 267 Dodson, J. D., 447
Cooper, F. S., 367 Dallos, P., 165, 167 Dodwell, P. C., 464, 496
Corbit, J. D., 360, 366 Dalton, K., 521 Doetsch, G. S., 215, 216
Coren, S., 11, 52, 63, 103, 104, 133, Dalziel, C. C., 62, 285 Doherty, M. E., 497
134, 144, 146, 269, 275, 288, 289, Damianopoulos, E, N., 184 Domey, R. G., 474
302, 311, 312, 315, 389, 390, 405, Daniels, J. D., 458 Donahy, T., 459
413, 414, 415, 422, 471, 474, 492, Dannemiller, J. L., 459, 460 Donaldson, I. M. L., 391
493, 494, 495, 503, 504, 507, 509, D’Aquila, J. M., 521 Doner, J., 283
512 Darian-Smith, I., 246 Doty, R. L., 232
Cormack, R. H., 299 Dark, V., 451 Douchin, E., 429
Comsweet, J. C., 307 Dartnall, H. M. A., 70, 133, 135 Dowd, J. M., 466
Comsweet, T. N., 101, 103, 109, 307, Darwin, C. J., 171 Dowell, R. C., 168
308, 420 DaSilva, J. A., 405, 412 Dowling, K., 461
Correia, M. J., 396 Davidoff, J. B., 516 Dowling, W. J., 348
Corso, J. F., 473, 474, 475, 521 Davies, A., 438 Drance, S. M., 146
Corwin, T. R., 486 Davies, P., 233 Dreyer, D. A., 246
Costanzo, R. H., 260 Dawson, J. L., 522 Droscher, V. B., 227
Costanzo, R. M., 225, 260 Day, M., 469 Drum, B., 94, 95
Costiloe, J., 520 Day, M. C., 471 Dubin, M. W., 457, 458
Cotter, N., 168 Day, R. S., 364, 412 Dubner, R., 266
Cotzin, M. E., 193 Deecke, L., 397 Duckman, R. H., 460
Covey, E., 222 Deffenbacher, K., 350 Duffy, S. A., 446
Cowan, W. B., 145 DeFries, J. C., 521 Dugan, M. P., 283
Cowart, B, J., 468 Degen, H., 268 Dumais, S. T., 293, 301, 444, 468
Cowen, G. N. J., 435 Delk, J. L., 146 Duncan, H. F., 502
Cowey, A., 85 Delman, G., 104 Duncan, P., 289
Cowley, J. J., 233 Delorme, A., 399, 408 Duncker, K., 383
Craig, J. C., 251, 252, 266, 269 Denier, C., 524 Durlach, N. I., 31
Craig, K. D., 266, 269, 467 Denny-Brown, D., 259 Dworkin, B., 267
586 Author Index

Easterbrook, J. A., 447 Farber, J., 500 Frey, M. von, 247, 248, 257
Easton, M., 288, 390 Farkas, M., 477 Freyd, J., 317
Easton, R. D., 361 Farmer, R. M., 348 Friberg, U., 215
Ebert, L., 212 Farrell, J. E., 384, 386 Friedman, A., 440, 469
Eccles, J. C., 259 Faurion, A., 213 Friedman, D. P., 242
Eddy, R. L., 132 Favreau, O. E., 143 Friedman, R. B., 331
Edmonds, E. M., 379 Fechner, G. T., 16, 17, 35, 41, 42 Friedman, S. L., 502
Edwards, A., 515 Fender, D. H., 390 Frisby, J. P., 82, 292
Edwards, R. S., 46 Fendrich, R., 383 Frishkopf, L. S., 207
Efron, R., 375, 376 pproe J 345 Froep, J. C., 433
Egan, D. J., 62, 285 Festinger, L., 103, 104, 142, 288, 390, Frost, B., 486
Egan, J. P., 21, 198-99 460, 493 Frykholm, G., 388
Egeth, H. E., 443 Field, J., 429, 466 Fuchs, A. F., 114
Ehrhardt, A. A., 524 Fields, R. W., 267 Fuld, K., 136
Eichengreen, J. M., 302 Filer, R., 379 Fuller, C. A., 374
Eijkman, E. G. J., 415 Fillenbaum, S., 146 Fuller, J. E., 391
Eilers, R., 515 Filsinger, E. E., 468 Fulton, J. F., 374
Eimas, P. D,, 360, 362, 366 Findlay, J., 440, 470 Funakoshi, M., 215, 217
Einstein, A., 372, 404 Finley, C., 400 Funkenstein, H. H., 173
Eland, J. M., 467 Fisher, D. L., 446
Elias, J. W., 477 Fisher, R., 378 Gaddes, W. H., 473
Elkind, D., 470 Fisher, R. P., 499 Gaes, G., 378
Ellingson, R., 458, 459 Fishman, M., 146 Galambos, R., 167, 171
Elliott, L. L., 468 Fisk, A. D., 447 Galanter, E., 11, 20, 35, 41, 44, 47
Ellis, H. C., 230, 498 Fisk, J. D., 288 Gale, A., 526
Ellis, S., 440, 470 Fitts, P. M., 32 Ganchrow, J. J., 215
Elman, J. L., 366 Fitzpatrick, V., 278 Ganchrow, J. R., 468
Emmerson, P. G., 144 Flandrin, J. M., 381 Ganz, L., 381
Emmerton, J., 63 Flannery, R., 191 Garbart, H., 443
Endman, M., 465 Flin, R. H., 472 Garcia, C. R., 234
Engel, R., 458 Flock, A., 165 Gardner, E. B., 260
Engen, T., 52, 222, 228, 229, 468, Flock, H. R., 104, 422 Gardner, E. P., 241, 260
520 Horn, M. C., 513, 515 Gardner, R. J., 212
Enns, J. T., 320, 471, 473 Florentine, M., 187 Gamer, W. R., 31, 32, 317, 318, 320,
Enroth-Cugell, C., 94 Flynn, W. E., 521 321, 329
Epley, J. M., 207 Fodor, J., 368 Garraghty, P. E., 242
Epstein, W., 251, 278, 386, 389, 405, Foley, J. E., 492 Gates, A., 193
412, 416, 419 Foley, K. M., 266 Geiselman, R. E., 499
Erickson, R. F., 221 Forman, E., 526 Geiwitz, J., 475
Erickson, R. P., 213, 215, 216, 221, Formby, C., 172 Gelade, G., 320
222, 227, 230 Forrester, L. J., 228 Gelb, A., 421
Eriksen, C. W., 31, 376, 432, 433 Foulke, E., 348 Geldard, F. A., 259, 266t, 395, 396
Ernst, M., 267 Fowler, T. J., 46 Gent, J. F., 219, 222
Erulkar, S. C., 194 Fox, R., 301, 312, 386, 468 George, E. J., 379
Essen, D. C. van, 77, 78, 85, 381 Fozard, J., 474 Gerbrandt, L. K., 517
Essock-Vitale, S. M., 232 Fraioli, A. J., 250 Gescheider, G. A., 250
Evans, C. R., 325 Fraisse, P., 372, 378 Gesteland, R. C., 225, 226, 227
Evans, E. F., 168, 173 Franchi, C., 233 Getchell, M. L., 225
Evans, N. J., 142 Frank, M., 216 Getchell, T. V., 225
Evans, P. M,, 252 Frankenhauser, M., 378 Giachetti, I., 227
Eysenck, H. J., 526 Frazier, N. I., 524 Gibbons, B., 223
Freedberg, E., 422 Gibson, E. J., 300, 301
Fabes, R. A., 468 Freeman, R., 483, 486 Gibson, J. J., 10, 274, 280, 281, 284,
Falk, J., 380 Freeman, R. B., 391 398, 405, 410, 500
Falmagne, J. C., 42 Freeman, S. K., 232 Gibson, M. M., 163
Fant, G., 366, 368 Fregnac, V., 458 Gibson, R. H., 47
Fantz, R. L., 459, 463, 464 French, J., 468 Gierke, S. M. von, 187
Farah, M., 451 Freud, S., 516 Gilbert, A. N., 223
 Author Index 587

Gilbert, J., 505 Grant, B. F., 16 Hall, J. W., Ill, 206-207

Gilchrist, A. L., 104, 421, 422 Gravetter, F., 31 Hall, M. J., 219
Gilinsky, A. S., 419 Graziadei, P. P. C., 225 Hall, W. G., 467
Gillam, B., 413 Green, B. G., 187, 188, 247, 255, Hallin, R. G., 263
Gilman, A., 516 257, 265 Hallstrom, J., 302
Gintzler, A. R., 268 Green, D. G., 94 Hallum, G., 506
Ginzberg, A., 475 Green, D. M., 21, 31, 36, 49, 185, Halper, F., 389
Girgus, J. S., 11, 52, 63, 146, 275, 186, 187, 188, 189, 205, 466 Halpem, B. P., 219
288, 315, 390, 413, 414, 415, Green, D. W., 440 Halpem, D. F., 312, 521, 524
471,474, 493, 495, 503, 504, 512 Green, M. A., 486 Halpem, D. L., 178
Glantz, J., 184 Green, M., 386 Halsam, D., 526
Glanville, E., 513 Green, P. A., 232 Hamalainen, H., 250
Glass, A. L., 440, 442, 445 Green, R. G., 526 Hamasaki, D. J., 458
Glicksohn, J., 379 Greenberg, M. J., 224 Hamemik, R. P., 167
Gliner, J. A., 513 Greene, H. A., 112 Hammel, D. D., 524
Glorig, A., 506 Greene, S. I., 418 Hammond, E. J., 440
Glucksberg, S., 435 Gregg, J. M., 285 Handel, S., 317, 318, 346
Gogel, W. C., 285, 384, 405, 412, Gregory, M., 447 Hanna, T. E., 187
422 Gregory, R. L., 382, 413, 415, 500, Hanson, S., 389
Goldberg, D., 312 503 Hardie, R. C., 63
Goldberg, M. E., 391 Gregory, W., 506 Hardy, J. D., 265, 266, 267
Golder, P., 312 Gresnight, A. M., 98 Harkins, S. W., 476, 526
Goldfoot, D. A., 232 Grice, G. R., 38 Harmon, L. D., 207, 327
Goldman, A., 248 Griffin, F., 484 Harper, R. S., 146
Goldsmith, L. T., 302 Griffith, B. C., 358 Harpman, J. A., 239
Goldstein, E. B., 205, 206 Grings, W., 506 Harris, C., 310
Goldstein, J. L., 206, 207 Groos, G., 373, 374 Harris, C. M., 70
Goldstein, R., 103 Gross, C. G., 84 Harris, C. S., 115, 139, 492
Goldstone, S., 378 Grossberg, J. M., 16 Harris, G., 266
Gombrich, E. H., 500 Grossberg, S., 103 Harris, J. P., 514
Goodell, B. S., 269 Grunau, M. von, 386 Harris, K. S., 358
Goodell, H., 265, 266 Grunau, R. V. E., 467 Harris, L. J., 521, 524
Gooding, K. M., 231 Grzegorczyk, P. B., 476 Harris, P., 460
Goodman, L., 516 Gudeman, H. E., 526 Harter, M., 458
Goodman, N., 500 Guedry, F. E., 396 Hartley, S., 483
Goodson, R., 507 Guilbaud, G., 264 Hartline, H. K., 72, 100-101
Goodwin, A., 246 Guitton, D., 381 Hartman, A., 515
Goodwin, M., 231 Gulick, W. L., 162, 172, 192, 198, Hartmann, W. M., 193
Goolkasian, P., 521 199, 200, 202 Harvey, L. O., Jr., 408, 419
Gopher, D., 451 Gumsey, R., 320 Harwood, D. L., 345, 348
Gordon, B., 485 Guttman, D., 436 Hastings, J. W., 374
Gordon, J., 136, 457 Guttman, N., 207 Hatfield, G., 386, 419
Gormican, S., 322 Guzman, A., 331 Hausfeld, S., 345
Gorta, A., 345 Gwiazda, J., 468 Hay, J. C., 492
Goto, H., 505 Hayhoe, M. M., 135
Gottlieb, G., 480 Haaf, R., 464 He, L., 269
Gottlieb, M. D., 97 Haake, R. J., 302 Head, H., 262
Gottschaldt, K., 239, 497 Haas, E. M., 217 Heaton, J. M., 517, 518
Gouras, P., 138, 139 Haber, R., 500 Hebb, D. O., 307
Gourlay, N., 502 Haber, R. N., 495 Hecht, S., 94, 98
Gower, E. C., 488 Haegerstrom-Portnoy, G., 515 Heckenmueller, E. G., 325
Goy, R. W., 232 Hagen, M., 501, 502 Heggelund, P., 79
Gracely, R. H., 266 Hagenaar, R., 433 Heijend, A. H. C. van der, 433
Graham, C. H., 132, 133, 285 Hahn, H., 219 Hein, A., 482, 487, 488, 489, 491
Graham, E., 526 Hainline, L., 457, 468, 469 Hein, S., 489
Graham, N., 115, 309 Haith, M. M., 464, 469 Heintz, C., 220
Granger, G. W., 513 Hake, H. W., 31 Held, R., 301, 399, 458, 460, 462, 468,
Granrud, C. E., 302, 460 Hakstian, A. R., 133, 134 483, 486, 487, 488, 489, 491, 492
 588 Author Index

Helekant, G., 513 Holmes, S. M., 142 Jansson, G., 386

Heilman, R., 196 Holst, E. von, 487 Jarvilehto, T., 250
Hellstrom, A., 35 Holway, A. F., 408 Jarvis, J. R., 142
Helmholtz, H. E. F. von, 11, 38, 127, Holyoak, K. J., 445 Javel, E., 167, 206
131, 204t, 205, 391, 405, 430 Holzman, P. S., 513, 527 Jeannerod, M., 381
Helson, H., 51 Honda, H., 288 Jeffress, L. A., 187
Henderson, A., 457 Horvath, S. M., 513 Jen, D. H., 167
Hendricson, A. E., 68 Hosick, E. C., 378 Jenkins, B., 486
Henmon, V. A. C., 38 Houck, M. R., 445 Jenkins, I., 505
Henn, V., 400 Houtsma, A. J. M., 206 Jennings, J. A. M., 106
Henning, H., 221, 230 Howard, I. P., 285, 391, 397, 492 Jensen, D., 458
Henry, J., 145 Howard, J. H., 449 Jesteadt, W., 19, 183, 186, 187, 189
Hensel, H., 239, 242, 252, 253, 257 Howard, R., 520 Johansson, G., 386
Hepler, N., 79 Howell, P., 348 Johnson, A. L., 233
Hering, E., 135, 136, 137, 405 Hoyer, W., 477 Johnson, C. H., 374
Herman, E., 288, 390 Hsia, Y., 132, 133 Johnson, D. H., 168
Hemandez-Peon, R., 432 Hubbell, W. L.,'69 Johnson, F., Jr., 228
Heron, W., 307 Hubei, D. H., 10, 76, 78, 79, 85, 139, Johnson, H. A., 475
Herscovitch, P., 172 381, 482 Johnson, M. A., 135, 136
Hershberger, W., 287 Hudson, W., 501, 502 Johnson, R., 480, 495
Hershenson, M., 415, 495 Hudspeth, A. J., 152, 165, 167, 179 Johnson, R. C., 521
Herskovits, M. J., 503 Huggins, G. R., 234 Johnston, A., 381
Hess, E. H., 300 Hughes, B., 251 Johnston, W., 451
Heuer, F., 383 Hughes, J., 167, 171 Jones, L. A., 260
Hey wood, J., 246 Hughes, K. C., 114, 328, 449 Jones, R., 501, 502
Hick, W. E., 39 Humphrey, D. E., 464 Jones, R. T., 513, 515
Hickey, T. L., 456, 457, 458, 482, Humphrey, G. K., 464 Jones, S. W., 476
483 Humphrey, K., 505 Jongsma, H. J., 415
Hicks, R. E., 378, 379 Humphreys, G. W., 418 Jonides, J., 430, 432, 433, 449
Hier, D. B., 524 Hunt, E., 444 Judd, D. B., 124
Higashiyama, A., 412 Hunzelmann, N., 381 Julesz, B., 294, 297, 301, 314, 319,
Hillenbrand, J., 178 Hurst, R. J., 213 320, 321, 327, 441
Hind, J. E., 163, 168 Hurvich, L. M., 76, 142, 144, 145, Jusczyk, P. W., 362
Hirsch, H. V., 482, 484, 485 295
Hirsch, J., 44 Hutman, L. P., 146, 475 Kaas, J. H., 242
Hirsh, I. J., 185 Hutz, C. S., 301 Kaess, D. W., 477
Hoagland, H., 376 Huxley, A., 515 Kahneman, D., 106, 309, 322, 430,
Hochberg, J., 285, 299, 315, 412, 422, Hyman, A., 223 438, 445, 450
469, 500 Kaiser, P. K., 135
Hockey, G. R., 447 Iacono, W. G., 527, 528 Kalat, J. W., 530
Hoenig, P., 288, 390, 493 Ikeda, H., 513 Kalina, R. E., 68
Hoffer, A., 515 Imbert, M., 457, 458 Kaneko, A., 73
Hoffman, D., 267 Ingram, R. M., 468 Kaplan, A., 513
Hoffman, H. A., 358 Intraub, H., 376 Kaplan, H., 268
Hoffman, J. E., 328, 432, 433, 445 Ippolitov, F. W., 520 Karita, K., 246
Hoffmann, K. P., 461, 482 Isbey, S. C., 193 Karmel, B. Z., 463
Hofsten, C. von, 386 Isensee, O., 264 Karvellas, 143
Hogan, H. P., 498 Ittelson, W. H., 277, 278, 494 Kasahara, Y., 217
Hogness, D. S., 132 Kasamatsu, T., 391
Hold, B., 232 Jacobs, G. H., 76, 138, 139 Katcher, M. H., 251
Holden, B. A., 59 Jacobsen, A., 104, 421 Katsuki, Y., 173
Holender, D., 450 Jahoda, G., 502 Kauer, J. S., 226, 227
Holland, H., 514 James, W., 372, 428 Kaufman, L., 172, 284, 285, 415
Holland, H. L., 499 Jameson, D., 145, 295 Kaufman, R., 302
Hollins, M., 246 Jameson, O., 142 Kausler, D. H., 477
Hollister, L., 515 Jamison, W., 524 Kawabata, N., 299
Holma, D., 378 Janowsky, D. S., 515 Kawamura, Y., 217
 Author Index 589

Kay, P., 147 Kolers, P. A., 376, 386 Leeuwenberg, E., 320
Kaye, H., 468 Komenda, J. K., 104 Lefton, L. A., 309
Kaye, ML, 301 Komoda, M. K., 422, 460 Lehman, J. R., 183
Kearny, J. K., 187 Konishi, M., 195 Lehmkuhle, S., 76, 82, 312
Keeley, S. M., 497 Konstadt, N., 526 Leibowitz, H. W., 108, 389, 408, 410,
Keith, B., 144 Kombrot, D. E., 44 412, 417, 419, 475, 502, 504
Keithley, E. M., 167 Koslow, M., 384 Leinonen, L., 246
Keller, H., 238 Kosterlitz, H. W., 268, 269 Lenel, J. C., 386
Kellman, P. J., 302, 460 Kozlowski, L. T., 387-88 Lennie, P., 73, 76
Kendrick, K. M., 84 Kratz, K. E., 76, 82 Leonard, D. G. B., 167
Kennedy, H., 381 Krauskopf, J., 98 LePage, E. L., 167
Kennedy, J., 500 Kravetz, M. A., 232 Lerman, S., 474
Kenshalo, D. R., 255, 256, 264 Kremenitzer, J. P., 461 Leshner, A. I., 232
Keren, G., 331 Krieger, A., 234 Lester, L. S., 328
Kessen, W., 136, 464, 469 Kries, J. von, 68 Lettvin, J. Y., 226, 227, 423
Kessler, E. J., 341, 342, 343 Kring, A. M., 323 LeVay, S., 482
Keuss, P., 328 Kropfl, W., 301 Leventhal, A., 482
Keveme, E. B., 232 Krueger, L. E., 250 Levi, D. M., 106
Khanna, S. N., 167 Kruger, J., 74, 139 Levin, S., 527
Kiang, N. Y. S., 162, 167 Krumhansl, C. L., 340, 341, 342, 343, Levine, J. M., 232
Kiefer, S. A., 194 346 Levine, M. W., 110
Kiesow, F., 247 Kryter, K. D., 506, 572 Levy, 438
Kietzman, M. I., 97 Kubovy, M., 106 Levy, D. L., 513
Kilbride, P. E., 146 Kuffler, S. W., 72 Levy, E. T., 194
Killbride, P. L., 502, 504 Kugler, B. T., 527 Lewis, C., 46, 49
Killeen, P. R., 359 Kuhl, P. K., 359, 362, 363, 465, 466, Lewis, I., 521
Kim, D., 513 468, 505 Lewis, J. W., 268
Kim, Y., 513 Kupchella, C., 147 Lewis, T. L., 458, 460, 468
Kimura, L., 215 Kurtz, D., 391 Lewis, V. G., 524
Kinchla, R. A., 328, 433 Kurtzberg, D., 461 Leyden, J., 232
King, L. E., 193, 194 Kutas, M., 487 Lhamon, W. T., 378
King, M. C., 329 Kuyk, T., 120 Liberman, A. M., 349, 358, 360, 364,
Kingham, J. R., 142 367, 368
Kinnear, P. R., 146 LaBerge, D., 433 Liberman, M. C., 162, 167, 357, 360
Kinsboume, M., 378 Lackner, J. R., 400 Lichte, W. H., 419
Kirk, B., 473 Ladefoged, P., 349, 352 Lichtenstein, M., 381
Kirk-Smith, R. A., 233 Lahey, M. A., 120 Liddell, E. G. T., 259
Kirschfeld, K., 63 Lakowski, R., 146, 475 Lie, I., 98, 106
Kitzes, L. M., 163, 194 Lamaze, F., 270 Liebelt, L., 440, 469
Klatt, D. H., 357, 366 Laming, D., 16 Liebeskind, J. C., 268, 467
Klatzky, R. L., 250 Lamour, T., 264 Lim, D. J., 162
Klein, G. S., 526 Land, E. H., 423 Linden, L., 423
Klein, R. M., 449 Landolt, E., 11 Lindsay, P. H., 68, 100, 159, 167,
Klein, S. A., 106 Lansman, M., 444 170, 186, 190, 192, 197, 202, 417,
Kligman, A., 232 Lappin, J. S., 283, 417, 419 438
Kline, P., 526 Larsen, A., 386 Linton, H., 526
Kloka, A. C., 267 Lasaga, M. L., 320 Lipsett, L. P., 467
Kluender, K. R., 359 Lathrop, G., 459 Lipsitt, L. P., 468
Kluver, H., 83 Lauter, J. L., 172 Lipton, R. B., 513, 527
Klymenko, V., 320, 321 Lawless, H. J., 221 Lisker, L., 360
Knudsen, E. I., 185 Lawless, H. T., 218 Livingstone, M., 76, 85
Kobler, J. B., 193 Lawley, H. J., 234 LNR Research Group, 330
Koenderink, J J., 98 Layton, W. M., 328 Lockhead, G. R., 31, 32, 142, 328,
Kohler, I., 489 Lederman, S. J., 250 329, 342
Kohler, W., 35, 313 Lee, J. R., 321 Loewenstein, W. R., 240
Kolb, B., 83, 358 Lee, M. H. M., 267 Loftus, E., 498, 499, 507
Kolb, H., 64, 73 Leehey, S. C., 486 Loftus, G., 469
590 Author Index

Logan, J. S., 368 Mariani, A. P., 64, 73 Meltzoff, A. N., 362, 363
Lohman, D. F., 521 Marks, L. E., 11, 47, 50, 181, 196, Melzack, R., 262, 264, 270, 467
Long, A. C., 391 250 Mentyer, T., 223
Long, G. R., 186 Marks, W. B., 133 Mergner, R., 397
Loomis, J. M., 251 Marley, A. A. J., 31 Merikle, P. M., 328, 450
Loop, M. S., 97 Marr, D., 10, 274, 292, 294, 331, Merkel, J., 39
Lorch, E. P., 471 334, 420 Mershon, D. H., 193, 194, 422
Lovitts, B. E., 419 Marre, M., 146 Merzenich, M. M., 168
Lowenstein, D., 396 Marshall, D. A., 227 Metzger, V. A., 250
Lu, R., 269 Marslen-Wilson, W. D., 366 Metzler, D. E., 70
Luce, 366 Martello, M., 464 Mewhort, D. J. K., 343
Luce, P. A., 357, 358, 365 Martin, C., 399 Meyer, G. E., 313
Luce, R. D., 31, 40, 49 Martin, D. K., 59 Meyer, W. J., 464
Lucia, H. C., 492 Martin, K. D., 513 Michael, C. R., 139, 141
Ludman, E. J., 137 Martin, M., 328 Michael, R. P., 232
Lummis, R. C., 207 Masica, D. N., 524 Michael, S., 384
Lumry, A. E., 527 Masland, R. H., 64, 73 Michaels, C. F., 274, 405, 410
Lumsden, E., 500 Massaro, D. W., 361, 366, 368 Michaud, A., 527
Luria, A. R., 516, 517, 519 Mather, G., 386 Michell, J., 40
Lydic, R., 374 Mather, J. A., 288 Michon, J., 372
Lyerly, O. G., 517 Matin, L., 287, 391, 394 Midgley, G., 301
Lyman P. J., 378, 379 Matthews, B. H. C., 258 Miezin, F., 381
Lynn, P. A., 167 Matthews, M., 312 Mihevic, P. M., 513
Mattingly, I. G., 349, 364, 367, 368 Mikaelian, H., 492
MacArthur, R. O., 513 Maunsell, J. H. R., 381 Miles, F. A., 391
MacDonald, D. W., 231, 232 Maurer, D., 457, 458, 460, 464, 465, Milewski, A. E., 468
MacDonald, J., 360 468 Mill, J. S., 306
MacDonald, J. A., 157 Maxwell, J. C., 58, 67, 127 Millan, M. J., 268
MacFarlane, A., 232, 460 Mayer, D. J., 269, 270 Millar, J. M., 358
Mach, E., 101, 103, 306 Mayhew, J. E. W., 292 Miller, D. L., 391
Machmer, J., 380 McBumey, D. H., 217, 220, 222, 232 Miller, G. A., 31, 36, 355
Mack, A., 288, 383, 390 McCall, R. B., 460 Miller, G. W., 378, 379
Mack, R., 449 McClain, L., 378, 379 Miller, J., 438
MacKain, K., 362 McCleam, G. C., 521 Miller, N. D., 106
MacKinnon, D. P., 499 McClelland, J. L., 366 Miller, R. J., 513
MacKinnon, G. E., 394 McClintock, M. K., 234 Millis-Wright, M., 323
Mackworth, N. H., 447, 469 McColgin, F. H., 381 Mills, A. W., 192
MacLeod, D. I. A., 94, 135 McCollough, C., 139 Milne, J., 179
MacLeod, P., 213, 227 McCray, R. E., 379 Milne, M., 179
MacLeod, R., 421 McCready, D., 415 Miranda, S. B., 463
MacNichol, E. F., 133 McDaniel, C., 386 Mishkin, M., 78, 242
MacNichol, E. F., Jr., 133, 137, 138 McDonald, M., 463 Mistretta, C. M., 476
Madden, D. J., 112 McElroy, A. E., 221 Mitchell, D. E., 301, 483, 484, 485,
Madden, T. M., 415 McFarland, R. A., 474 486
Maes, H., 381 McGee, M. G., 521 Mitchell, M. S„ 517
Maffei, L., 381 McGlone, J., 521 Mittelsteadt, H., 487
Mai, N., 381 McGlothlin, W., 515 Mohindra, I., 468
Mair, R. G., 520 McGuinness, 521 Monahan, J. S., 32, 329
Maisel, E. B., 463 McGuinness, D., 520 Monasterio, K. M. De, 138
Maland, J., 302 McGuinness, E., 381 Moncrieff, R. W., 228
Mallach, R., 483 McGurk, H., 360, 502 Money, J., 520, 524
Mandelbaum, M., 94 McHaftie, J. G., 78 Monroe, M. D., 135
Mandler, G., 324 McKnight, A. T„ 268, 269 Montellese, S., 97
Mandler, M. B., 486 McLean, J. P., 432 Montgomery, D. A., 146
Mangel, S. C., 76 Meals, D., 379 Montgomery, J. C., 157
Mangen, G. L., 526 Meer, H. C. van der, 295 Mooney, R. D., 458
Manning, K. A., 104 Meijer, J. H., 374 Moore, B., 466
Margolis, R. H., 200 Meiselman, H. L., 219, 220, 221 Moore, D., 410
 Author Index 591

Moore, J. D., 233 Nelson, R., 64, 73 Oster, P. J., 477

Moore, L. M., 476 Neuweiler, G., 193 Osterberg, G., 67
Moore, M. J., 469 Neville, H. J., 487 Ostfeld, A., 515
Moore, R. K., 104, 391 Newhall, S. M., 124, 146 Ostroy, S. E., 69
Moore, T., 505 Newland, J., 473 Ostry, D., 500
Moore-Ede, M. C., 374 Newman, E. B., 192, 193, 201 O’Toole, A. J., 449
Moran, J., 485 Newman, J. D., 173 Ottoson, D., 226
Moray, N., 434 Newport, K., 477 Over, R., 313
Morgan, H., 375 Nickerson, D., 124 Owens, D. A., 108
Morgon, B. A., 146 Nielson, C. R., 476 Owens, M. E., 467
Morish, R. B., 526 Nishimura, Y., 73 Owsley, C., 302
Morrell, F., 458 Nitzan, D., 379 Oyama, J., 49, 315
Morris, P. E., 499 Nitzberg, D., 383 Ozog, G., 309
Morris, V., 499 Noble, W., 193
Morrison, J. D., 285 Noda, H., 391 Packer, D., 68
Morrongiello, B. A., 466 Noell, W., 507 Padden, D, M., 359
Morrow, T. J., 263 Noma, E., 16, 21, 42 Palmer, A. R., 171
Morton, T., 520 Norgren, R., 216 Palmer, C., 346
Moskowitz, H., 515 Norman, D. A., 68, 100, 159, 167, Palmer, J., 383
Moskowitz-Cook, A., 463, 486 170, 186, 190, 192, 197, 202, 330, Palmer, S. E., 330
Mouloua, M., 447 330, 417, 435, 438 Pan, X., 269
Moulton, D. G., 227 Norman, J., 106 Pandya, P. N., 82
Mountcastle, V. B., 260, 263 Norman, J. L., 458 Paquet, L., 328
Movshon, J. A., 63, 458, 459, 462, Northrop, C. C., 162 Parasuraman, R., 447, 449
482, 484 Norton, T. T., 457, 458, 482 Parducci, A., 44
Mozel, M. M., 223, 227 Norwich, K. H., 31, 32, 36, 40, 220, Park, J. W., 416
Mroczek, K., 299 309 Parker, D. E., 389, 397
Muir, D., 429, 466 Nullmeyer, R., 38 Parker, S., 11, 47
Muir, D. W., 464, 485, 486 Nunez, A. A., 374 Parker, S. P. A., 191
Muise, G., 419 Nusbaum, 366 Parks, T. E., 323
Muller, D. G., 498 Nusbaum, H. C., 365 Parlee, M. B„ 520, 521
Mulligan, R. M., 379 Nusinowitz, S., 104 Pashler, H., 443, 450
Munsell, A. H., 124 Nykvist, W. F., 220 Pasnak, R., 278, 412
Murphy, C., 223 Pasternak, M., 476
Murphy, J. E., 432 Oakley, B., 220, 221 Pastore, R. E., 360
Murphy, T. D., 432 O’Brien, C. P., 268 Patterson, R. D., 205, 206
Mustillo, P., 293 Obusek, C. J., 348 Paulus, L., 217
Myers, A. K., 45 Ochoa, J. L., 239 Pauser, G., 264
Myles-Worsley, M., 451 O’Connell, R. J., 227 Peachly, N., 449
Odom, J. V., 295 Peduzzi, J. E., 456, 457, 458, 482
Nachmias, J., 187, 302 Ogasawara, K., 78 Peeples, D. R., 462
Nafe, J. P., 255 Ohlott, P., 50 Peichl, L., 106
Nagy, A. L., 144 Oldfield, S. R., 191 Peloquin, L. J., 527
Naka, K., 64 Olsho, L. W., 466 Pelosi, P., 228
Nakamura, C. Y., 526 Olson, R., 320 Penfield, W., 242, 244
Narens, L., 40 Olzak, L., 114 Penland, J., 440, 470
Nathans, J., 132 O’Mahony, M., 219, 220 Perfetto, G., 283
Navon, D., 326, 328, 451 Ommaya, A. K., 492 Perkell, J. S., 357
Neff, D. L., 185 Ono, H., 283, 289, 299, 412 Perl, E. R., 239, 262
Neff, W. D., 194 Ono, M. E., 283 Perrin, N. A., 331
Neisser, U., 32, 436, 437, 440, 441, Oppenheim, R. W., 467 Peters, R. W., 206-207
446 Orban, G. A., 76, 78, 381, 486 Peterson, A. C., 524
Nelson, B., 445 Omdorff, M. M., 232 Petrig, B., 301
Nelson, G., 459 Omstein, R. E., 374, 378 Petry, S., 313
Nelson, J. N., 378 Osaka, N., 95 Pettersen, L., 302
Nelson, L. M., 215 O’Shaughnessy, M., 514 Pettigrew, J. D., 293, 458, 483, 485,
Nelson, L. R., 268 Oshinsky, J. S., 346 486
Nelson, P. G., 173 Osmond, H., 515 Pfaff, D., 377
592 Author Index

Pfaffman, C., 216, 217, 220 Pribram, K. H., 517 Rhee, K., 513
Pfeiffer, R. R., 171 Price, J. L., 226 Rhein, L. D., 225
Phillips, C. G., 85 Priebe, F. A., 514 Rho, J. M., 162
Phillips, D. P., 194 Prinzmetal, W., 323, 329, 442 Rhodes, G., 432
Phillips, L. J., 460 Pritchard, R. M., 307 Ribaupierre, Y. de, 167
Phillipson, O. T., 514 Prkachin, K. M., 269 Rice, C. G., 506, 507
Piaget, J., 469 Procter, 301 Richards, W., 323, 420, 475
Piantanida, T. P., 132 Profitt, D. R., 388 Riesen, A., 482
Pick, 518 Puckett, J. deW., 390 Rieser, J., 468
Pick, A. D., 471 Puklin, J. E., 468 Riesz, R. R., 36, 185, 186, 187
Pick, H. L., Jr., 471, 492, 501, 502 Purcell, D. G., 521 Riggs, L. A., 104, 307
Picket, J. M., 365 Rijnsdorp, A., 373
Pickford, R. W., 213 Quigle, R., 506 Rivers, G., 103, 104
Pierce, J. R., 348 Quigley, S., 449 Rivest, J., 283
Pierce, L., 449 Quinn, P. C., 137 Robbins, M., 502
Piggins, D. J., 142 Roberts, J., 521
Pigion, R. G., 513 Raab, E., 301 Roberts, M., 362
Pinkers, A., 146 Rabbitt, P., 440, 477 Roberts, T. D. M., 260
Pionk, M., 521 Rabin, M. D., 223, 228 Robertson, P. W., 147
Pirenne, M. H., 98 Rahe, A., 507 Robey, J., 461
Pisanelli, A. M., 228 Raichle, M. E., 172 Robinson, D. W., 197
Pisoni, D. B., 348, 357, 358, 365, Rakerd, B., 193 Robinson, M. M., 73
366, 368, 505 Ram, C., 232 Robson, J. G., 106, 321
Pitts, W. H., 226, 227 Ramachandran, V. S., 276 Rocha-Miranda, E. C., 84
Pizorrolo, F. J., 518, 519 Rand, T. C., 360 Rock, I., 275, 284, 386, 389, 405,
Plastow, E., 418 Randsom-Hogg, A., 98 412, 415, 418, 436, 473
Plateau, M. H., 43 Raphaeli, N., 267 Rockland, K. S., 82
Plude, D., 477 Rashad, M. N„ 521 Rodieck, R. W., 69, 73
Podgomy, P., 433 Rasmussen, T., 242, 244 Rodrigues, J., 146
Podolsky, S., 474 Ratliff, F., 100-101, 103, 307 Roelofs, C. O., 287
Poggio, G. F., 260, 263 Rauth, J., 513 Rogel, M. J., 233
Poggio, T., 292 Rayleigh, Lord, 191 Rogers, B., 145
Pokomy, J., 143, 144, 146 Raymond, J. E., 389 Rogers, C. H., 251
Pollack, I., 31, 32, 341, 365 Read, J. S., 146 Rohrbaugh, J. W., 429
Pollack, R. H., 504 Reading, A. E., 266 Rojas, A., 226, 227
Pollatsek, A., 446 Reason, J., 445 Rollman, G. B., 266
Pollen, D. A., 106, 321 Redding, G. M., 492, 493 Romani, G. L., 172
Poltrock, S. E., 444 Reddy, D. R., 355, 364 Romano, J. A., 468
Pomerantz, J. R., 312, 322, 328, 329 Reddy, R., 364 Romano, P. E., 468
Pons, T. P., 242 Reed, C., 250 Romer, D., 46, 49
Poon, L., 477 Reed, M. A., 379 Root, W., 480
Poppel, E., 372, 374 Reeves, A., 98, 449 Rose, D. J., 389
Popper, R., 11 Regan, D. M., 398 Rose, J. E., 163, 167, 168, 171
Porac, C., 289, 312, 415, 471, 493, Regnier, F., 231 Rosenblum, L., 360
504, 509, 512 Rehn, T., 423 Rosenzweig, M. R., 193
Porjesz, B., 513 Reid, R. C., 103 Rosinski, R., 500
Porter, R. H., 232, 233, 468 Reilly, D. A., 221 Rosner, B. S., 378
Posner, M. I., 38, 448, 449 Reisberg, D., 514 Ross, H., 277
Post, B., 389 Reisch, A. M., 217 Ross, H. E., 144, 261
Post, R. B., 108, 417, 475 Remez, R. E., 348 Ross, N., 497
Postman, L., 146, 198-99 Remington, R., 449 Ross-Field, L., 524
Poulton, E. C., 46 Remington, R. W., 432 Rothblat, L., 482, 484
Powell, T. P. S., 260 Repp, B. H., 156 Roucoux, A., 381
Power, R. P., 345 Reschke, M. F., 261 Royster, J. D., 521
Powers, M. K., 94, 462 Restle, F., 51 Royster, L. H., 521
Poynter, W. D., 378 Reynolds, D. C., 146, 513 Rozin, P., 212, 223
Preble, L. D., 417, 419 Reynolds, P., 438 Rubin, E., 311
Preti, G., 234 Reynolds, R. I., 495 Rubin, P. E., 348
 Author Index 593

Ruble, D. N., 526 Schnizlein, J. M., 38 Shows, T. B., 132

Rudnicky, A. I., 355 Scholl, M., 461 Shulman, G. L., 432, 433
Ruggieri, V., 289 Schreiber, R. C., 167 Siddle, D. A., 526
Rumelhart, D. E., 330 Schubert, E. D., 161 Sidman, M., 473
Runeson, S., 388 Schull, J., 212, 268 Sieg, D., 527
Rusak, B., 374 Schuller, G., 193 Siegel, L. J., 270
Rushton, W. A. H., 134, 135 Schulman, P. H., 384 Silver, S. D., 504
Russell, M. J., 232 Schulte, R., 508 Silver, W. L., 215
Russoff, A. C., 457, 458 Schumer, R. A., 294 Silverton, L., 515
Ryugo, D. K., 162 Schvaneveldt, R. W., 446 Simmons, F. B., 207
Schwab, E. C., 365 Simon, E., 477
Saccuzzo, D. P., 515 Schwartz, C. B., 498 Sinclair, D. C., 263
Sachs, M. B., 167 Schwartz, M., 482, 484 Sinnot, J., 513
Sager, L. C., 328 Schwartz, S. H., 97 Siqueland, E. R., 362
St. James, J. D., 432, 433 Schwartz, T., 105 Sivian, L. S., 178, 181
Saito, S., 213 Sciascia, T. R., 423 Skottun, B. C., 114
Salapatek, P., 457, 458, 461, 463, Scott, H. A., Jr., 255, 256 Skowbo, D., 140
464, 465, 469 Scott, T. R., 215, 216, 217 Slavin, K., 521
Salmon, T. H., 218 Sedgwick, H., 500 Sloan, K. R., 68
Samuel, A. G., 360, 365 Segall, M. H., 503, 504 Sloane, 301
Sand, A., 396 Sekuler, R., 290, 380, 381, 475, 513 Sloboda, J. A., 348
Sander, L. W., 362 Selfridge, O. G., 331 Slowiaczek, 366
Sanders, B., 521 Seligman, P. M., 168 Sluyters, R. C. van, 458, 459, 482,
Sanders, G., 524 Semple, M. N., 194 484
Sanocki, T., 331 Senden, M. von, 485 Smallman, R. L., 214
Santos-Sacchi, J., 165 Serbin, L. A., 524 Smith, 143
Sapir, E., 497 Serpell, R., 473 Smith, A., 313
Saunders, F. A., 250, 251 Sersen, E. A., 520 Smith, D. V., 217
Sawusch, J. R., 360, 364 Settle, R. G., 232 Smith, J., 345
Sayers, B. M. A., 167 Shaffer, H. L., 438 Smith, M. C., 465
Scadden, L., 250, 251 Shaffer, L. H., 376 Smith, R. L., 250
Schady, W. J. L., 239 Shallice, T., 38, 39t Smith, V. C., 144, 146
Schaie, K. W., 475 Shankweiler, D. P., 367 Smith, W. S., 524
Scharf, B., 180, 183, 196, 198, 200, Shannon, C. E., 29 Snellen, H., 104
449 Shansky, M. S., 147 Snyder, S. H., 268
Schenkel, K. D., 181 Shapiro, K. L., 389 So, Y. T., 76
Scher, D., 521 Shapley, R., 76, 94, 103 Soares, M. P., 521
Schiano, D., 493 Sharma, S., 515 Sokolov, E. N., 429
Schiffman, H. R., 379 Sharpe, L. T., 97 Solis-Macias, V., 328, 433
Schiffman, S. S., 216, 221, 222, 227, Shavit, Y., 268 Southwick, E., 230
230, 231, 476 Shea, 301 Spelke, E. S., 460
Schilder, P., 497 Shea, S. L., 468 Sperling, H. G., 135
Schiller, P. H., 73, 76 Shebilske, W. L., 287, 423 Sperry, R. W., 299
Schindler, R. A., 168 Sheedy, J. E., 289 Spieker, S., 362
Schlaer, S., 98 Sheehy, J. B., 433 Spillmann, L., 98, 381
Schleidt, M., 232 Shefner, J. M., 110 Spinelli, D. N., 484, 517
Schmeckler, M. A., 360 Shepard, R. N., 340, 341, 342, 386, Spinnato, J. A., 268
Schmidt, A., 487 433 Spoendlin, H. H., 162
Schmidt, H., 322, 442 Shepard, R. P., 251 Sprafkin, C., 524
Schmiedt, R. A., 167 Sherman, S., 482 Spreen, O., 473
Schnapf, J. L., 69, 70 Sherman, S. M., 73, 74, 76, 82, 485 Springer, S. P., 358, 472
Schneck, M. E., 462, 463 Sherrick, M. F., 384 Srulovicz, P., 206, 207
Schneider, B. A., 41, 47, 456, 465 Sherrington, C. S., 259, 391 Stacey, E. W., Jr., 440, 442
Schneider, D. 226 Shiffrin, R. M., 443, 444, 445 Stark, L., 287, 440, 470
Schneider, G. E., 78 Shimai, K., 73 Stebbins, W. C., 157
Schneider, R., 520 Shimojo, S., 301, 323, 462 Steele, C. R., 167
Schneider, S. L., 251 Shockey, L., 364 Stein, B. E., 78
Schneider, W., 443, 444, 445, 447 Shower, E. G., 188, 189 Steinberg, A., 378
594 Author Index

Steiner, J. E., 468 Taub, E., 492 Vahle-Hinz, C., 239

Steinfield, G. J., 495 Tauchi, M., 73 Valentine, R. H., 527
Steinman, R. M., 390 Taylor, J. A., 321 Vallbo, A., 242, 247
Stelmack, R. M., 526, 527 Tedford, W. H., 521 Vandenberg, S. G., 521
Stephan, F. K., 374 Tees, R. C., 301, 363, 505 Varner, D., 463
Stephens, P. R., 394 Teeter, J. H., 214 Vaughan, H. G., 461
Stem, D., 362 Teghtsoonian, ML, 423, 456 Vautin, R. G., 385
Stem, I. B., 467 Teghtsoonian, R., 36, 50, 423 Veres, J. G., HI, 120
Stem, J. A., 477 Teller, D. Y., 63, 460, 462, 463 Verriest, G., 146
Stembach, R. A., 261, 266 Templeton, W. P., 492 Verrillo, R. T., 247, 250, 250
Sternberg, S., 376 Tepper, B., 374 Vickers, D., 38, 39t
Stevens, D. A., 218, 221 Terenius, L., 268 Vierck, C., 239, 259, 262
Stevens, J. C., 50, 219, 255, 257 Terheardt, E., 207 Vierck, C. J., Jr., 263
Stevens, S. S., 40, 41, 44, 45t, 47, 49, Terman, G. W., 268 Vigorito, J., 362
50, 92, 152, 192, 196, 201, 202, Tetewsky, E., 312 Vilardi, K., 383
203, 250, 257, 266 Tharp, D., 299 Vimal, R. L. P., 144
Stevenson, M., 502 Theodor, L. H., 509 Vincent, J., 415
Sticht, T., 348 Thomas, E. A. C., 378, 379 Virgi, R. A., 443
Stiles, W. S., 90 Thomas, H., 46, 524 Vogels, R., 486
Stoever, R. J., 328 Thomas, W. G., 521 Volkman, J., 201
Stoffregen, T. A., 399 Thorell, L. G., 114 Volkmann, F. C., 104
Stokes, B. A. R., 263 Thombury, J. M., 476 Vries, J. V. De, 278, 411
Stokinger, T. E., 185 Thornton, J. E., 232 Vries, H. De, 227, 228
Stolwijk, J. A. J., 267 Timney, B. N., 301, 483, 484, 486 Vurpillot, E., 302, 470
Stone, C. A., 471 Toch, H. H., 508
Stone, L. S., 299 Todd, J. J., 281, 388 Waber, D. P., 524
Strange, W., 505 Tomita, T., 64, 73 Wade, G., 520
Stratton, G. M., 489 Tomko, D. L., 47 Wade, N. J., 291
Street, R. F., 496 Torebjork, H. E., 239, 263 Waespe, W., 400
Streitfeld, B., 52, 359 Torgerson, W. S., 11 Wahl, O. F., 527
Strom, L., 215, 221 Townshend, B., 168, 207 Wahlstrom, A., 268
Stromeyer, C. F., Ill, 139 Townsend, J., 331 Wainwright, A., 285
Stroop, J., 446 Trehub, S. E., 456, 465, 468, 505 Wald, G., 69, 133
Stroud, J. M., 375 Treisman, A. M., 85, 320, 322, 323, Walk, R. D., 300, 301
Stryer, L., 69 329, 334, 432, 438, 441, 442, 444 Wall, P. D., 261, 262, 264, 267, 270
Stuart, G. W., 412 445, 450 Wallace, B„ 492, 493, 514
Studdert-Kennedy, M., 362, 367 Treisman, M., 37, 44, 374 Wallace, P., 232
Suedfeld, P., 378 Tress, K. H., 527 Wallach, H., 193, 383, 384, 391, 420
Sugitani, M., 246 Troland, L. T., 142 Walley, A., 505
Suitt, C., 458 Tronick, E., 460 Walls, G. L., 289
Sulzman, F. M., 374 Tsai, Y., 432, 449 Walter, A. A., 498
Summerfieid, A. Q., 362, 506 Tsvetkova, L. S., 519 Ward, D. C., 474
Supra, M., 193 Tuason, V. B., 527 Ward, L. M., 44, 46, 47, 49, 50, 266,
Supramaniam, S., 440 Tulunary, U., 52 328, 415, 433
Sussman, H. M., 360 Turnbull, C., 504 Ward, S., 524
Sutija, V. G., 458 Turner, P., 514 Ward, T. B., 526
Svaetichin, G., 137, 138 Tursky, B., 266 Ward, W. D., 201
Swarbrick, L., 173 Tversicy, B., 317 Ware, C., 145, 312
Swensson, R. G., 26 Ty, T. C., 262 Warm, J. S., 379
Swets, J. A., 21, 449 Tyer, Z. A., 278, 412 Warren, A. B., 474
Szczesiul, R., 50, 360 Tynan, P., 380 Warren. D. E., 521
Szentagothai, J., 396 Warren, D. H., 486
Szoc, R., 309 Warren, R. M., 348, 365
Uhlarik, J., 480, 495 Warren, R. P., 348
Tachibana, M., 73 Ullman, S., 335 Warshovsky, F., 152
Tanczos, 143 Umezaki, H., 458 Wassle, H., 106
Tangney, J., 310 Ungerleider, L. G., 78 Watanabe, T., 173
Tart, C., 515 Uttal, W., 11, 275, 276, 277, 405, 412 Watkins, L. R., 269, 270
 Author Index 595

Watson, S. J., 268, 269 Wickens, C. D., 451 Yaksh, T. L., 268
Watts, T. R., 37 Wiener, N., 29 Yamamoto, T., 217
Weale, R. A., 63, 101, 474, 475 Wier, C. C., 186, 187, 189 Yang, J. C., 269
Weaver, W., 29 Wiesel, T. N., 10, 78, 79, 139, 381 Yankell, S. L., 232
Weaver, W. B„ 378, 379 482 Yantis, S., 430, 432, 433
Webb, F., 524 Wikner, K., 468 Yarbus, A. L., 307, 440, 441
Weber, D. L., 31 Wilcox, S. B., 417 Yarmey, A., 507
Weber, E. H., 35, 247 Wilier, J. C., 264 , 268 Yayama, N., 217
Weber, E. U., 289 Willette, M., 379 Yeh, J., 463, 464
Webster, W. R., 171, 172 Williams, A., 310 Yeh, Y., 433
Weddell, G., 239 Williams, D., 477 Yeomans, D. C., 263
Wegener. B., 16 Williams, D. R., 135 Yerkes, R. M., 447
Weil, A. T., 378 Williams, J. M., 515 Yin, R. K., 473
Weinstein, E. A., 517 Williams, M., 477, 517 Yodogawa, E., 317
Weinstein, S., 246, 248, 520 Williams, T. C., 44 Yonas, A., 302, 460, 468
Weisenberg, M., 269 Williamson, S. J., 172 Yoneshige, Y., 468
Weiskopf, S., 136 Willis, W. D., 262, 263, 264, 265 Young, C., 446
Weiss, S., 465 Wilson, E. O., 230 Young, J. Z., 394
Weisstein, N., 110, 309, 310, 311, Wilson, H. R., 106, 114 Young, L., 400
312, 320, 321 Wilson, J., 433 Young, L. R., 390
Welch, R. B., 489, 491, 492 Wilson, J. R., 521 Young, M. G., 246
Welford, A. T., 38 Wilson, M., 52, 83, 359 Young, R. A., 142, 460
Well, A. D., 471 Wilson, W., 515 Yuille, J., 499, 507
Wells, A. M., 325 Winter, I. M., 171 Yund, E. W., 79, 375
Wells, G. L., 498 Winter, P. L., 173
Werker, J. F., 363, 368, 505 Witkin, H. A., 526 Zakay, D., 379
Werner, G., 260 Wolberg, Z., 173 Zaporozhets, A. V., 470
Werner, H., 309 Wolf, E., 474 Zare, S. L., 386
Werner, J. S., 463 Wolf, S., 520 Zaretsky, H. H., 267
Wertheimer, M., 313, 384, 466 Wolfe, J., 328 Zeigler, H. P., 412
West, R., 348 Wolff, H. G., 265, 266, 269 Zeki, S., 85
Westheimer, G., 106 Wolters, G., 433 Zhang, X., 269
Wever, E. G., 205, 206 Wong, E., 311, 312, 320 Zhuang, S., 269
Wever, R. A., 168, 373 Woo, G., 475 Ziegler, R. O., 520
Whalen, D. H., 360 Wood, R. W., 397 Zigler, M. J., 248
Whalen, J. J., 136 Woodfield, R. L., 524 Zihl, J., 381
Wheeler, D., 506 Woodrow, H., 372, 378 Zilbert, D., 482
Whishaw, I. O., 83, 358 Woods, B., 472, 473 Zimba, L. D., 449
Whitaker, H. A., 358 Woods, R., 477 Zimmerman, M., 264
White, B. L., 462, 489 Woodworth, D. G., 526 Zinberg, E., 378
White, B. W., 142, 250, 251 Woolard, H. H., 239 Zotterman, Y., 215, 221, 252, 253
White, C., 375 Wooten, B. R., 136, 137, 391, 463 Zrenner, E., 138
White, C. W., 142, 146, 493 Worchel, P., 193 Zucker, I., 374, 520
White, K. D., 295, 526 Wright, L. L., 477 Zue, V. W., 355
White, R. L., 168 Wright, M. J., 381 Zuidema, P., 98
White, S. D., 178, 181 Wright, N. H., 184 Zurek, E., 193
Whitehurst, G. S., 473 Wright, R. N., 224, 226, 228, 229 Zusho, H., 232
Whiteside, T. C. D., 285 Wright, W. D., 127, 132 Zwicker, E., 167, 181, 182 , 207
Whitfield, I. C., 166, 168, 173, 174 Wurtz, R. H., 391 Zwislocki, J. J., 167, 182, 184
Whitsel, B. L., 246 Wybum, G. M., 213, 214, 397
Whorf, B. L., 147, 497 Wysocki, C. J., 223
Whytt, R., 62 Wyszecki, G., 90
 -

>
 Subject Index
Page numbers in italics indicate figures; Aerial perspective, 302 Ascending stimulus series, 17
page numbers followed by t indicate as depth cue, 276-77 Assimilation, brightness, 103, 116
tabular material; page numbers in AFF (auditory flutter fusion), 514, 528 Astigmatism, 486, 509
boldface type indicate glossary Affect, 10 Attention, 428-51
definitions. Afference copy, 287, 302 aging and, 477
Aftereffects auditory, 432, 437-38
adaptation to prism distortion and, during childhood, 468-71, 469-71
A3 fibers, 242 491, 491, 509 covert, 430
pain perception and, 262, 264, 265 motion, 401 divided, 429, 436-39, 450-51, 452,
A 8 fibers, 242 Afterimage, 144-45, 148 452
pain perception and, 262, 264, 265 apparent movement and, 393, 393t auditory, 437-38
AB,H system, 213, 234 Age. See also Development visual, 436-37, 439
Abscissa, 17 color perception and, 145-46 expecting and, 447-50
Absolute distance, 274, 302 Agnosia, 516-18, 516-18, 528 costs and benefits of information
Absolute temperature, of skin, 252 visual, 83-84, 87 cues and, 448, 448-50
Absolute threshold, 17, 53 Alcohol, perceptual effects of, 513 filtering and, 433-38
brightness perception and, 92-93 Alpha androstenol, 230, 233 cocktail party phenomenon and,
thermal, 253, 255 Alveolar consonants, 350 434-35
for touch, 246, 246-47 Amacrine cells, 64, 86 divided attention and, 436-39
Acceleration, 395, 400 Ambiguity, speech perception and, video overlap phenomenon and, 436,
Accommodation, 62, 62-63, 64, 86, 355-56, 356 437, 438
302 Amplitude, loudness and, 195 focused, 429, 450-51, 452, 453
as depth cue, 284-85 Ampulla, 395, 395, 400 level of detail and, 433
in infants, 463 Analgesia, 267-68 orienting and, 429-33
Acoustic-phonetic invariance, 355-56, cognitive processes in, 269-70 attentional gaze and, 432-33
368 endogenous opiates and, 268-69, 270 covert, 429-32, 432
Across-fiber pattern, 216, 234 Anisotropy, 486 orienting reflex and, 429
Action potential, 532, 532-33, 534 Anomalous trichromatism, 132, 148 overt, 429
Active processing, 365, 368 Anosmia, 228, 234 searching and, 440-47
Acuity. See Visual acuity Anterior-insular cortex, 215, 234 automatic versus controlled, 442-46,
Acupuncture, 269, 269 Anticipation 445
Adaptation error of, 18, 53 feature versus conjunction, 440-42,
auditory, 198, 208 time perception and, 372 444, 444
brightness perception and, 92-94, 93 Aphasia, 519, 519t, 520, 528 vigilance and, 446-47, 448
chromatic, 144, 148 Apostilb, 90 visual, eye movements and, 440,
context and, 52, 52 Apparent distance, size constancy and, 441-43
dark, 92, 94, 116 409 selective, precedence effect and, 435
aging and, 474, 474 Apparent movement, 384-86, 400 size constancy and, 413
fatigue and, 114 afterimages and, 393 temporal versus nontemporal, 379-80
light, 92, 116 Apprehension, 406, 406t, 424 theories of, 450-51, 451, 452
of pain, 267, 267 of distance, 413 varieties of, 428, 428-29
to rearranged stimuli, 491-93 Aqueous humor, 59, 86 visual, 432-33, 436-37, 439
selective, 114, 116, 381, 401 Area, brightness perception and, 97-98 Attentional gaze, 432-33, 452
of spatial frequency channels, 115 Area 17, 77, 86 Attentional resources, 450-51, 452,
to smells, 228-29 Area 18, 77, 86 452
to tastes, 2/9, 219-20 Area 19, 77, 86 Aubert-Fleischl effect, 390, 400
to touch, 248-49 Area 22, 170 Audiometric units, 198t
Adaptation level, 50-52, 52, 53, 359 Area 41, 170 Auditory adaptation, 198, 208
Additive color mixture, 124-25, 125, Aristotle’s illusion, 245 Auditory attention, 432
126-27, 127, 148 Arousal, physiological, vigilance and, divided, 437-38
Adults, perceptual change in, 473-77 447, 448 Auditory canal, 158, 174
A endings, 258 Articulators, 350, 368 Auditory cortex, 172-74, 173

597
598 Subject Index

Auditory fatigue, 198-99, 199, 208 Binaural interactions, hearing and, 181 Brightness perception, 90-104
Auditory flutter fusion (AFF), 514, 528 Binaural presentation, 181, 208 brightness contrast and, 98-103, 99—
Auditory masking, 181-85, 182, 183 Binocular depth perception. See Depth 101
Auditory nerve, electrical activity of, perception, binocular cognitive factors in, 103-104, 104
164, 166, 166-69, 167, 169 Binocular disparity, 290, 291, 302 infants and, 462
Auditory system, 152-74. See also crossed, 290, 292-93, 302 maximum sensitivity and, 98
Hearing uncrossed, 290, 292-93, 303 retinal locus and, 94-95
auditory cortex and, 172-74, 173 Binocular-disparity-sensitive cells, in spatial frequency analysis and,
auditory pathways and, 169-72, 170 infants, 458 108-16
electrical activity of lower auditory Binocular perception, 285, 302 time and area and, 97-98
centers and, 171-72, 172 of depth, 289-95 visual acuity and, 104-108, 105—107
ear and, 157-66 cues for stereopsis and, 290, 291 — wavelength and, 95, 95-97
evolution and anatomy of, 157-58 93, 294 Bril, 92, 116
inner, 158, 159, 174 process of stereopsis and, 291-95, Bunsen-Roscoe law, 97, 116
mechanisms of transduction and, 295, 296, 297
165, 165-66 experience and, 482-83
middle, 158, 159, 174 Binocular rivalry, 437, 439, 452 Caffeine, taste blindness and, 219
outer, 158, 159 Biological clock, 373-78, 400 Carpentered world hypothesis, 503-
physiology of, 158-62, 159-63 biological pacemaker and, 376-78, 504, 509
vestibular sense and, 394-95, 395 377 Catch trials, 21, 53
wave motion on basilar membrane circadian rhythms and, 373-74, 375 Categorical perception, 358-60, 359,
and, 163, 163-64, 164 short-term timers and, 374-76 368
electrical activity of auditory nerve Biological motion, 386-87, 387, 400 Category scaling, 42, 43, 43-44,
and, 164, 166, 166-69, 167, 169 Biological reductionism, 10, 13 53
sound and, 152-56, 153. See also Bipartite target, 95 Cell(s), 316
Sound(s) Bipolar cells, 64, 86 Cell body, 530, 535
simple sound wave and, 152, 153, Bit, 29-30, 53 C endings, 259
154-57, 155, 155t, 157 Bitter taste, 213 Central masking, 184-85, 208
Auditory theories, of speech threshold for, 218 Central nervous system (CNS), 531,
perception, 366 Black-white opponent-process, 137 535
Autokinetic effect, 394, 400 Blank out, 307 Cents, 201
Automaticity, strong, 445 Blindness CFF (critical flicker fusion frequency),
Automatic processing, 438, 452 color, 131-33, 148 513-14, 528
Automatic searching, controlled day, 68-69, 86 C fibers, 242
searching versus, 442-46, 445 night, 68, 86 pain perception and, 262, 264, 265
Autotopagnosia, 518, 528 odor, 228 Change
Axon, 530, 535 psychic, 83, 87 object-relative, 383, 401
Azimuth, 189, 190, 208 snow, 307 subject-relative, 383, 401
Blind spot, 71, 86 time perception and, 378
Background, interaction with Bloch’s law, 97, 116 Change monitoring, 378, 400
luminance, 99, 99 Blondel, 90 Channel capacity, 30, 30-32, 3It, 32,
Background stimuli, 51, 53 Blur, as depth cue, 285 53
Backward masking, 184, 208, 270, 335 Bodycentric direction, 285-86, 287, Chlorolabe, 131, 148
contour emergence and, 309, 310 302 Choice reaction time, 38, 38-39, 39t.
touch and, 252 Body temperature, time perception and, 53
Bandwidth, 199-200, 200 376-78, 377 Chopper, 171, 174
Base stimulus, 360 Bony labyrinth, 394-95, 395 , 400 Chord, 343, 368
Basilar membrane, 161, 174 Braille, 250, 251 Chorda tympani nerve, 215, 220
skin as model for, 164 Brightness, 90-104, 116, 148 Chromatic adaptation, 144, 148
wave motion on, 163, 163-64, 164 adaptation and, 92-94, 93 CIE (Commission Internationale de
Beats, 185-86, 186, 208 of color, 123 l’Eclairage), 90, 116
Bel, 154, 174 photometric units and, 90-92, 9It, 92 CIE chromaticity space, 124, 127-30,
B endings, 259 relative, 303 128, 129, 148
Benefit, 452 aerial perspective and, 277 Cilia, olfactory, 225, 234
of information cues, 449 Brightness assimilation, 103, 116 Circadian rhythm, 373-74, 375, 400
Benham’s top, 143, 148 Brightness constancy, 104 Circumvallate papillae, 213, 214
Beta ((3), 25, 53 Brightness contrast, 98-103, 99-101 Clock theories, of time perception,
Bezold-Brucke effect, 144, 148 simultaneous, 98, 116-17 373-80
 Subject Index 599

Closure, law of, 314, 315, 336 culture and, 147 velar, 350
CNS (central nervous system), 531, memory color and, 146 voiced, 349
535 color stimulus and, 120-30, 121, 1211 Constancy, 404, 404-24
Coarticulation, 351 CIE color space and, 124, 127-30, brightness, 104
Cochlea, 158, 160, 174 128, 129 color (hue), 422-23, 424
Cochlear duct, 161, 174 color appearance systems and, 122- depth, stereoscopic, 299
Cochlear implant, 168, 174 24, 122-24 direction, 391, 400, 423, 424
Cochlear nucleus, 169, 174 color mixture and, 124-27, 125-27 illusion and, environmental and life
dorsal, 169, 170, 174 infants and, 462-63 history differences and, 502-504
ventral, 169-70, 175 intensity and duration and, 143-45 lightness (whiteness), 419-22, 420,
Cocktail party phenomenon, 434-35 physiology of color vision and, 130— 421, 424
Cognition, 9-10, 13 43 loudness, 423, 424
Cognitive clock. 373, 378-80. 400 opponent-process theory and, 135— odor, 423, 424
change and, 378 43, 136 position, 391, 401, 423, 424
processing effort and, 379 trichromatic color theory and, 130— shape, 416-19, 417, 418, 425
temporal versus nontemporal attention 35 size, 407, 407-16, 425
and, 379-80 spatial interactions and, 145 direct and constructive aspects of,
Cognitive factors Color solid, 124 410-12, 411, 413
in analgesia, 269-70 Color space, 124, 127-30, 128, 129 distance cues and, 407-10, 408
in brightness perception, 103-104, Color spindle, 124, 124, 148 illusion and, 408, 412-13, 414-16,
104 Color vision 415-16
in color perception, 146-48 defects in. 128, 131-33, 133, 134 Constancy scaling, 408, 408, 409, 424
color impressions and, 147-48 opponent-process theory of, models of Constant-ratio concept, lightness
culture and, 147 color-coding and, 141-43 142 constancy and. 419-21, 420, 421
memory color and, 146 screening inventory for, 134 Constructive perception, size constancy
Cognitive style, 526, 528 trichromatic color theory and, 130-35 and, 411-12, 413
perceptual effects of, 525-28 Color wheel, 122, 148 Constructive theories, 11, 13, 275,
Cohort theory, of speech perception, Commission Internationale de 302, 405, 424
366 l’Eclairage (CIE), 90, 116 Context, 405
Cold, paradoxical, 257, 257-58, 271 Common chemical sense, 212, 215 adaptation level and, 52, 52
Cold fibers, 252-53, 253, 254, 256- Comparison color, 95 identification and, 329, 329-31, 330
57, 270 Comparison stimuli, 33, 53 meaning and, 494-99, 495, 496, 498
paradoxical cold and warmth and, Complementary colors, 127, 148 eyewitness testimony and, 498-99,
257-58 Complex cells, 79-80, 81, 86 499
Color, 120, 120-46 Computational approach, 10-11, 13, shape constancy and, 417-18, 418
complementary, 127, 148 274-75, 302 speech perception and, 363-65
interaction with form, 140 to object perception and identification, visual, 383, 401
metameric, 124, 148 334-35, 335 effect on judged weight, 51
spectral, 149 Concentration test, 83, 86 Context stimuli, 406, 424
subjective, 142, 143, 149 Conceptually-driven processing, 325, Continuation, good, law of, 314, 314—
Color appearance systems, 122-24, 335 15, 336
122-24 Cones, 64, 68-71, 86 Contour!s), 307-10, 335, 368
Color atlas, 124, 148 foveal, 67 change and, 307-309
Colorblindness, 131-33, 148 in infants, 456-57 emergence of, 309-10, 310
Color circle, 122, 122, 123, 148 Configural feature, 329, 335 illusory, 312
Color-coding, models of, 141-43 142 Conjunction search, 441, 452 of piece of music, 343, 344
Color column, 139 feature search versus, 441-42, 444, subjective, 312, 312-13
Color constancy, 422-23, 424 444 textural, 318
Color impressions, 147-48 Consonance, 195, 208 Contrast
Color mixture, 124-27, 125-27 Consonants, 349-50, 368 brightness, 98-103, 99-101
additive, 124-25, 125, 126-27, 127, alveolar, 350 simultaneous, 98
148 fricatives and, 349, 350 color, simultaneous, 145, 149
subtractive, 125, 126, 126-27, 149 labial, 350 Contrast matching. 111, 111
Color perception, 143-48 nasal, 349-50 Controlled processing, 444, 452
age and physical condition and, 145— palatal, 350 Controlled searching, automatic
46 in speech spectrogram, 353-54 searching versus, 442-46, 445
cognitive factors in, 146-48 stop, 349 Convergence, 302
color impressions and, 147-48 unvoiced, 349 as depth cue, 285, 286
600 Subject Index

Convergence angle, 285, 286 d' (sensitivity), 25, 27, 27, 28, 53 experience and, 480, 481, 482-87
Copulins, 231 computation of, 28 restricted and selective rearing and,
Comea, 59, 61, 86 Dark adaptation, 92, 94, 116 482-87
aging and, 474 aging and, 474, 474 in infancy, 456-68
astigmatism and, 486 Data-driven processing, 324-25, 335 brightness and color and, 462-63
Correct negative, 22, 53 Day blindness, 68-69, 86 eye movements and spatial vision
Corresponding retinal points, 290, 302 Decibel (dB), 154, 174 and, 460-61, 461
Cortex Dendrites, 530, 535 hearing and, 465-67
auditory, 172-74, 173 Density, of sound, 195, 208 pattern discrimination and, 463-65,
extrastriate, 77, 86 Depolarization, 531, 535 464-66
inferotemporal, 83, 86 Depressants, perceptual effects of, 514 psychophysical methods and, 459,
somatosensory, 242, 246, 271 Depth constancy, stereoscopic, 299 459-60
kinesthesis and, 260 Depth perception. See also Space taste and smell and, 467-68
pain perception and, 264 perception touch and pain and, 467
striate, 77, 87 binocular, 289-95 visual acuity and, 462, 462, 463
temporal, superior, 84, 87 cues for stereopsis and, 290, 291— visual system and, 456-59, 457t
visual. See Visual cortex 93, 294 perceptual change in adults and, 473—
Cortical gray, 19 process of stereopsis and, 291-95, 77
Cost, 452 295 , 296, 297 global, 476-77
of information cues, 449 interaction of cues in, 295, 298, 299 hearing and, 475-76, 476
Covert attention, 430 monocular, 290 taste and smell and, 476, 476t
Covert orienting, 429-32, 432, 452 perspective as cue for, 276-77 visual function and aging and, 473-
Cranial nerves, 242 physiological cues for, 284-85 75, 474, 475
eighth, 396 accommodation as, 284-85 Developmental approach. 456, 477
Crista, 395, 395, 396, 400 convergence and divergence as, 285, Dichotic listening, 358, 434, 452
Criterion, 24-27, 53 286 Dichotic presentation, 436, 452
Critical flicker fusion frequency (CFF), pictorial cues for, 275-84, 303 Dichromats, 131-32, 148
513-14, 528 aerial perspective as, 276-77 Difference threshold, 33-34, 34, 53
Critical period, 483, 509 height in the plane as, 281, 282 Diplopia, 290, 292-93, 303
Cross-adaptation interposition or overlay as, 275, 276 Direction
of smell, 228, 229, 234 linear perspective as, 278, 279, 280 bodycentric, 285-86, 287, 302
to tastes, 219-20 motion and motion parallax as, 283, headcentric, 286, 287, 303
Crossed disparity, 290, 292-93, 302 283-84 Directional acuity, 105, 106
Cross-modality matching, 47, 49, 49- retinal and familiar size as, 277, Direction constancy, 391, 400, 423,
50, 53 277-78, 278, 279 424
Cross-model integration, 359, 360-62 shading as, 275-76, 276 Direct perception, 10, 13, 274, 303,
Crystalline lens, 62 , 62-63, 64, 86 texture gradients as, 280-81, 280-82 405, 424
accommodation of, as depth cue, size constancy as cue for, 410 size constancy and, 410-11, 411
284-85 subjective contours as cue for, 312 Direct scaling, 41, 42-50, 53
aging and, 474 types of, 274-75 category judgment and, 42, 43, 43-44
C-type horizontal cells, 137 Dermis, 239, 270 cross-modality matching and, 47, 49,
Cues, 275, 303 Descending stimulus series, 17 49-50
depth. See Depth perception Detail perspective, 280-81, 280-82 magnitude estimation and, 44-47,
information, 432, 447-48, 453 Detail set, of attention, 433 45t, 46, 47, 48
costs and benefits of, 448, 448-50 Detection, 16-27, 17, 53 Discrimination, 16, 33-40, 34, 53
effect on attention, 449-50 method of constant stimuli and, 17, pattern
monocular, 275, 303 19-20, 20 during childhood, 472-73
physiological, 303 method of limits and, 17-19, 18t in infants, 463-65, 464-66
stimulus, 432, 453 signal detection theory and, 2It, 21- reaction time and, 37-40, 38, 39t
structural, 303 27, 22t, 23, 24t, 25-27, 28 signal detection theory in, 37
Cued trials, 448 Deuteranomaly, 132, 148 of sounds, 185-89, 186-88
Culture, 503 Deuteranopia, 132, 148 Weber’s law and, 35, 35-36, 36t, 37
color perception and, 147 Development, 456-77 of weights, 260
Cupola, 395 in childhood, 468-73 Disembedding, 521, 522, 528
Cyanolabe, 131, 148 attention and search and, 468-71, Dissonance, 195, 208
Cycle, 174 469-71 Distal stimulus, 404-405, 424
of sound waves, 154 encoding and memory and, 472-73, Distance. See also Space perception
Cycles per second, 154, 174 474 absolute, 274, 302
Cyclopean eye, 286, 288, 303 filtering and, 471, 472 apparent, size constancy and, 409
 Subject Index 601

apprehension of, 413 Entrainment, 374, 400 motion perception and, 388, 388-92,
relative, 274, 303 Environmental differences, 499-509 392, 393t, 393, 394
Distance cues illusion and constancy and, 502-504 spatial vision and, in infants, 460-61,
registered, 413 occupational, 506, 506-507, 507 461
size constancy and, 407-10, 408 perceptual set and, 507-509, 508 vestibular sense and, 396, 397
Distance perception, types of, 274-75 picture perception and, 500-502, 502, visual searching and, 440, 441-43
Divergence, 303 503, 503 voluntary pursuit, 388, 389
as depth cue, 285 speech and, 504-506 Eyewitness testimony, 498-99, 499
Divided attention, 429, 436-39, 450- Epidermis, 239, 271
51, 452, 452 Epithelium, olfactory, 224, 225, 234 Facilitation, 480, 481, 509
auditory, 437-38 Equal-interval scaling, 42, 43, 43-44, False alarm, 22, 53
visual, 436-37, 439 53 Familiar size, 303, 501, 509
Dolorimeter, 265-66, 271 Equal loudness contour, 197, 197-98, as distance cue, 278, 278, 279
Dol scale, 266, 271 208 size constancy and, 412
Dominant wavelength, 124, 148 Equal pitch contour, 202, 203, 208 Farsightedness, 63, 64
Dorsal cochlear nucleus, 169, 170, 174 Equal temperament scale, 201, 208 Fasciculus, 531
Dorsal roots, 242 Error Fatigue
Double pain, 262-63, 271 anticipation, 18, 53 adaptation and, 114
Drugs negative time, 35, 54 auditory, 198-99, 199, 208
perceptual effects of, 513-16, 514 perseveration, 18, 53 time perception and, 378
time perception and, 378 refractive, 63, 87 Feature(s), 321-22, 335
Duplex perception, 360, 368 Erythrolabe, 131, 148 configural, 329, 335
Duplex retina theory, 68, 86 Eustachian tube, 160, 174 emergent, 322, 335
Duration. See also Time Event processing, 378, 400 Feature detectors, 365
color perception and, 144-45 Evoked potential, visual, 458 Feature integration theory, 322-23,
estimation of, 372 Evoked response, 526-27, 528 335, 442
of musical notes, 345 Exafference, 487, 509 Feature search, 441, 453
perceived, 195, 208 Excitation, contour emergence and, 309 conjunction search versus, 441-42,
Dynamic range, 178, 179-80, 208 Excitatory responses, auditory, 172-73, 444, 444
Dyschromatopsias, 146, 148 173 Fechner’s law, 41-42, 42, 53
Dyslexia, 473, 477 Excitatory transmitter, 531-32, 535 Feedback
Expansion, focus of, 398, 400 from accommodation, as depth cue,
Expectations, 494-99 285
Ear. See Auditory system; Hearing; effect on attention, 449 from eye movements
Sound(s) payoff matrix and, 24 direction and, 287-88
Eardrum, 158, 174 Expecting, 429, 447-50, 448, 453 motion perception and, 391-92, 392
Early selection, 450, 451, 452 Experience Fidelity, 29
Eccentric cell, 100 depth perception and, 297, 301-302 Field dependency, 526, 528
Echolocation system, 193, 208 development and, 480, 481, 482-87 Field independency, 526, 528
Efference copy, 287, 303 restricted and selective rearing and, Figural goodness, 315
Efferent fibers, 396, 400 482-87 information and symmetry and, 315-
Egocenter, 286, 288, 303 Extent, 453 17, 316-18
Egocentric localization, 274, 303 of attention, 433 Figural integration, 323
Electrical activity, of lower auditory External illuminance, 419, 424 Figural organization, 313, 313-15,
centers, 171-72, 172 Extrastriate cortex, 77, 86 314, 316
Electromagnetic radiation, 58, 59 Extrovert, 526, 527, 528 spatial frequencies and, 320-21, 321
Electrophysiology, of smell, 226-27, Eye. See Visual system Figure, 310-11, 335
227 Eye chart, 104-105 Figure-ground organization, 311-12
Embedded figures test, 521, 522, 528 Eye dominance, 81, 86 Filiform papillae, 213
Emergent feature, 322, 335 Eye drift, 394 Filled duration illusion, 378, 379, 400
Emmert’s law, 409, 424 Eye-head system, 382, 388, 389, 389- Filtering, 428, 433-38, 453, 477
Emmetropic eye, 63, 64, 86 92, 392 , 393t, 400 during childhood, 471, 472
Encoding, 472, 477 Eye movement(s) cocktail party phenomenon and, 434-
memory and, during childhood, 472- during childhood, 469-70 35
73, 474 passive, motion perception and, 391 divided attention and, 436-39
Endogenous opiates, 268-70, 271 reflex pursuit, 389, 389 video overlap phenomenon and, 436,
Endorphins, 268, 271 saccadic, 460, 477 437, 438
Enhancement, 480, 481, 509 schizophrenia and, 527-28 visual, 436-37, 439
Enkephalins, 268, 271 smooth-pursuit, 460-61, 477 First formant, 353
602 Subject Index

Flow (time), of time, 372, 373, 378, pitch and, 195 detection of sounds and, 178, 178-95
400 Frequency principle, 204, 205, 208 auditory masking and, 181-85, 182,
Focal attention, 322, 335 Frequency sweep detector, 173, 174 183
Focal stimulus, 51, 53, 406, 424 Fricatives, 349, 350 sound discrimination and, 185-89,
object and situation properties of, 406 Fundamental frequency, 203, 208 186-88
Focused attention, 429, 450-51, 452, missing, 204, 208 sound localization and, 189-95
453 Fungiform papillae, 213 temporal, frequency, and binaural
Focus of expansion, 398, 400 Fusion, 290, 303 interactions and, 180-81, 181
Foliate papillae, 213 effects of drugs on, 513
Force, perception of, 260-61 effects of occupational noise on, 506,
Forced-choice preferential looking, Ganglion cells, 64, 86 506
460, 477 Ganzfeld, 307, 308, 335 gender differences and, 521
Formants, 352-54, 353-55, 369 Gate-control theory, 264, 265, 271 high-frequency limits of, 180
first, 353 cognitive processes and, 269-70 in infants, 465-67
second, 353 Gaze, attentional, 432-33, 452 subjective dimensions of sound and,
Formant transitions, 352, 354, 354-55, Gender differences, 520-25 195-208
369 in visual-spatial abilities, 521-25, 522 loudness and, 196, 196-200, 197,
Form perception, 306-35 physiological factors and, 521-22, 199, 200
contour and, 307-10 524 pitch and, 197, 198t, 200-203, 202,
change and, 307-309 psychosocial factors and, 524-25 203
emergence of, 309-310, 310 Geniculostriate system, 76-77, 75, 77, theories of pitch perception and, 197,
interaction with color and, 140 86 203-208, 204t, 207
object recognition and identification Geons, 333, 333-34, 335 Heat grill, 254
and, 323-31, 324 Gestalt, 313-15, 314, 316, 335-36 Height, relative, 281, 282
data-driven versus conceptually- Gestures, phonetic, 367-68 Height in the plane, 281, 282, 303
driven processing and, 324-31, 326 Glabrous skin, 238-39, 271 Helicotrema, 160, 174
recognition versus identification and, nerve fibers in, 242 Hemifield, 449, 453
324 Glaucoma, 516 Hemiretinas, 76
perceptual object and, 310-23 Glial cells, 530, 535 Hertz (Hz), 154, 174
feature integration theory and, 322- Global arrangement, 336 Hick’s law, 40, 53
23 Global precedence, 328 Hidden figures test, 521, 522, 528
features and, 321-22 Global processing, 325-26, 326, 328 High even harmonics, 203
figural organization and, 313, 313- Global stereopsis, 294, 296, 303 Hit, 22, 53
15, 314, 316 Glossopharyngeal nerve, 215 Homophenes, 364, 369
figure and ground and, 310-13, 311, Golgi tendon organs, 259, 271 Homophones, 364, 369
312 Good continuation, law of, 314, 314— Horizontal cells, 64, 86
information, symmetry, and good 15, 336 Horopter, 290, 294, 303
figures and, 315-17, 316-18 music and, 345 Hot taste
spatial frequencies and figural Graded potential, 532, 535 quenching, 221
organization and, 320-21, 321 Grating acuity, 105, 106 threshold for, 218
texture segregation and, 317-20, 319 Gravity, weight perception and, 260- Hue, 122, 148. See also Color
theories of object identification and, 61 perception
331-35 Gray matter, 531 Hue constancy, 422-23
visual field and, 306, 306-307 Ground, 310-11, 336 Hydrophobicity, smell and, 224
Forward masking, 182-84, 208, 271 Gustatory sense. See Taste Hyperacuity, 106, 116
touch and, 252 Hypercolumn, 81, 83, 86
Fourier analysis, 110 Hypercomplex cells, 80, 81, 86
spatial, 108-10, 109, 110 Ffabituation, 136, 148, 460, 477 Hypermetropia, 63, 64, 86
Fourier components, 156, 157, 174 of orienting reflex, 429 Hyperpolarization, 69, 86, 531, 535
Fourier’s theorem, 108-10, 109, 110, Hair bundle, 162, 174 Hypothalamus, suprachiasmatic nucleus
116 Hairy skin, 238, 239, 111 of, 374, 375, 401
Fourier synthesis, 110 Hallucinogens, perceptual effects of, Hz. See Hertz
Fovea, 65-68, 66-68 515
cones of, 67 Harmonics, 203-204, 208 Identification. See Object identification
in infants, 457 high even, 203 Identification-by-components, 331, 333,
Fovea centralis, 67, 86 Headcentric direction, 286, 287, 303 333-34
Free nerve endings, 239, 271 Hearing, 178-208. See also Auditory Illuminance, 90, 91, 116
Frequency (f), 154, 174 system; Sound(s) external, 419, 424
fundamental, 208 aging and, 475-76, 476 retinal, 90, 116, 419, 424
 Subject Index 603

Illumination, visual acuity and, 107- Inhibitory responses, auditory, 173, Lateral geniculate nucleus, 76, 86
108 173 color perception and, 138-39, 139
Illusion(s), 6-8, 8, 13 Inhibitory-surround organization, of in infants, 458
Aristotle’s, 245 skin, 240-41 Lateral inhibition, 100, 101, 116
constancy and, environmental and life Inhibitory transmitter, 532, 535 Lateral line, 157
history differences and, 502-504 Inner ear, 158, 159, 174 Lateral olfactory tract, 226, 234
filled duration, 378, 379, 400 Inner hair cells, 161, 162, 174 Lateral posterior nucleus, 78, 86
moon, 415-16, 416, 424 In phase, 154, 174 Late selection, 450, 451, 453
Mueller-Lyer, 415, 415, 424, 471, Insects, pheromones and, 230-31 Law of closure, 314, 315, 336
472, 477, 493, 494, 504 Instruction Law of good continuation, 314, 314—
drugs and, 515 objective, 418, 424 15, 336
one-speaker stereo, 193 projective, 418, 424 music and, 345
Poggendorff, 512, 512, 528 Integral stimuli, 328-29, 336 Law of Pragnanz, 315, 316, 336
Ponzo, 413, 414, 415, 424, 471, 472, Integration, 469, 477 Law of proximity, 313-14, 314, 336,
477 Intensity. See Stimulus intensity 344
size constancy and, 408, 412-13, Intensity difference, 189, 208 Law of similarity, 314, 314, 336, 345
414-16, 415-16 Intemeuron, 530, 535 Learning, 10
visual-geometric, 471, 477 Interposition, 303, 501, 509 sensory-motor, 487-94, 488, 489
Illusion decrement, 493-94, 494, 509 as depth cue, 275, 276 illusion decrement and, 493-94, 494
Illusory conjunction, 322, 336 Interpretation, in information perceptual rearrangement and, 489-
Illusory contours, 312, 336 processing, 10 93, 491
Image, retinal, 306-307, 336 Intersections, 319 Lemniscus, 531, 535
stabilized, 308-10, 336, 393, 393t Interstimulus interval, 182-83, 208, Lens
Image-retina system, 382-88, 387, 384, 400 crystalline, 62 , 62-63, 64, 86
388, 393t, 400 Interval of uncertainty, 34, 54 accommodation of, as depth cue,
Incus, 158, 174 Interval scale, 40-41, 54 284-85
Indirect scaling, 41-42, 42, 54 Introvert, 526, 527, 528 aging and, 474
Individual differences, 512, 512-28 Invalid trials, 448 zoom, 432
gender differences and, 520-25 Invariance, speech perception and, Lesion, in visual cortex, 79
visual-spatial abilities and, 521-25, 355-56, 356 Levels-of-processing analysis, 10, 13
522 Inversion, optical, 490 Lexical Access From Spectra (LAFS)
personality and cognitive style and, Iodopsin, 70, 86 model, 366
525-28 Iris, 59, 86 Life history differences. See
physiological, 512-20 Isosensitivity curve, 23, 23 Environmental differences
drugs and, 513-16, 514 Light, 58. See also Brightness;
physical pathology and, 516-20, Jet lag, 374 Brightness perception; Vision;
516-18, 519t, 520 Just noticeable difference (jnd), 34, 34, Visual system
Induced motion, 383-84, 400 54 neural responses to, 71-74
Induction, 480, 481, 509 Fechner’s law and, 42 retinal responses and receptive fields
Infants. See Development Weber’s law and, 35, 36 and, 72 , 72-73 , 73
Inference, unconscious, 405, 425 W, X, and Y cells and, 73, 73-74,
Inferior colliculi, 170, 174 74
Inferotemporal cortex, 83, 86 Kinesthesis, 258-61, 271 Light adaptation, 92, 116
Inflow theory, 391-92, 392 , 400 neural responses in, 259-60, 260 Lightness constancy, 419-22, 420,
perception of weight and force and, 421, 424
Information, bit of, 29
defined, 29 260-61 Light reflex, pupillary, 62
stimuli and receptors for, 258-59, 259
Information channel, 428, 453 Limbic system, 242, 271
Kinetic depth effect, 284, 303
Information cue(s), 432, 447-48, 453 Limits, method of, 17, 18, 54
costs and benefits of, 448, 448-50 Linearity, 369
effect on attention, 449-50 Labeled-line theory speech perception and, 355
Information masking, 185 of smell, 227, 234 Linear perspective, 303
Information processing, 10, 13 of taste, 216-17 as depth cue, 278, 279, 280
Information theory, 29-32, 30t, 54 Labial consonants, 350 Local arrangement, 336
Information transmission, 30, 30-32, Labyrinth, 158, 174 Localization(s)
311, 32, 54 LAFS (Lexical Access From Spectra) egocentric, 274, 303
Inhibition model, 366, 369 eye movement and, 287-88
contour emergence and, 309 Lamaze technique, 270 object-relative, 274, 303
lateral, 100, 101, 116 Lambert, 90 of sounds, 189—95
two-tone, 167, 175 Language, effect on percepts, 497-98 complex sounds and, 192, 192-94
604 Subject Index

Localization(s) (continued) eyewitness testimony and, 498-99, thresholds for, 383

physiological mechanisms of, 194— 499 vestibular sense and, 392, 394-97,
95 Medial geniculate, 170, 174 395
simple tones and, 189-92, 190, 192 Medial lemniscus, 215, 234, 242, 271 neural responses in, 396-97
of touch sensation, 247-48, 248, 249 Meissner corpuscles, 239 stimuli and receptors for, 395, 395-
Local processing, 325-28 Mel, 201, 208, 340 96, 396
Location, perceived, 195, 208 Melody, 343 Motivation, time perception and, 379-
Lock-and-key theory, of smell, 226, Memoric phase, in information 80
234 processing, 10 Motives
Locus, 453 Memory musical, 344
of attention, 433 encoding and, during childhood, 472- payoff matrix and, 24
Long-range process, 385-86, 401 73, 474 Motor neuron, 530, 535
Loudness, 195, 196, 196-200, 197, visual shadowing and, 438 Motor theory, of speech perception,
198t, 199, 200, 208 Memory color, 146 367-68
magnitude estimation of, 48 Mental rotation task, 521, 522, 523, Movement, 380-400. See also Eye
Loudness constancy, 194, 423, 424 528 movement(s); Kinesthesis; Motion
LSD, perceptual effects of, 515 Merkle disks, 239 perception
L-type horizontal cells, 137 Metacontrast, 309, 310, 336 apparent, 384-86, 400
Lumen, 90, 116 Metameric colors, 124, 148 afterimages and, 393
Luminance, 90, 116 Metathetic continuum, 41, 54 biological, 386-87, 387, 400
interaction with background, 99, 99 Method of constant stimuli, 17, 19-20, induced, 383—84, 400
Luminosity curve, 95 , 95, 116 20, 54 motion parallax and, 283, 283-84
Method of limits, 17-19, 18, 18t, 54 perceptual, 401
Mach bands, 102, 103, 116 Micropattems, 375 real, 384, 401
Macula, 66-67, 86, 395-96, 396, 397, Microsaccades, 307 reflex pursuit, 401
401 Microspectrophotometer, 133, 148 smooth-pursuit, 401
Macula lutea, 66-67, 86 Microvillae, 214 voluntary pursuit, 401
Macular spot, 67 Middle ear, 158, 159, 174 Mueller-Lyer illusion, 415, 415, 424,
Magnitude estimation, 44-47, 45t, 46, Millimicron (mp,), 58, 86 471, 472, 477, 493, 494, 504
47, 48, 54 Mind’s eye, 432 drugs and, 515
Magnitude matching, 50, 54 Minimum audible angle, 192, 208 Multidimensional interaction. 405, 424
Maintenance, 480, 481, 509 Minimum audible field, 178, 208 Multidimensional scaling, taste qualities
Malleus, 158, 174 Minimum audible pressure, 178, 208 and, 221
Mammals, pheromones and, 231-34 Miss, 22, 54 Muscle spindles, 258-59
Marijuana, perceptual effects of, 515 Missing fundamental, 204, 208 Music, 340-48
Masker, 181, 208 Mixed-modality scaling, 50, 54 musical forms and, 343-48, 344,
Masking, 181, 208 Modulation transfer function, 109, 346, 347
auditory, 181-85, 182, 183
backward, 184, 208, 270, 335
,
110-12, 111 112 musical tones and, 340-43, 341
Modulus, 44, 54 Musical forms, 343-48, 344, 346, 347
contour emergence and, 309, 310 Molar concentration, 217, 234 Musical interval, 343, 369
touch and, 252 Moments, perceptual, 375-76 Musical notes, duration of, 345
central, 184-85, 208 Monaural presentation, 181, 208 Musical space, 342
contour emergence and, 309-10, 310 Monochromat(s), 131 Musk, smell of, 224
forward, 182-84, 208, 271 Monochromatic stimuli, 122-23, 148 Myelin sheath, 530-31, 535
touch and, 252 Monocular cues, 275, 303 Myopia, 63 , 64 , 86
information, 185 Monocular depth perception, 290 night, 108
simultaneous, 181, 209 Moon illusion, 415-16, 416, 424
touch and, 252 Motion. See Movement
Matching Motion aftereffect, 401 Naloxone, 268-69
contrast. 111, 111 Motion parallax, 283, 283-84, 303 Nanometer (nm), 58, 86
cross-modality, 47, 49, 49-50, 53 Motion perception Nasal consonants, 349-50
magnitude, 50, 54 apparent motion and, 384-86 Nativist versus empiricist question,
Matrix, 316 biological motion and, 386-88, 387 depth perception and, 299
Maturation, 480, 481, 509. See also eye movements and, 388, 388-92, Near point distance, 63, 86
Development 389, 392 , 393t, 393, 394 Nearsightedness, 63, 64
Maxwell’s spot, 67 physiological motion detectors and, Necker cube, 514, 514-15, 528
McGurk effect, 359, 360-62, 369 380-82 Negative time error, 35, 54
Meaning, context and, 494-99, 495, self-motion and, 397-400, 398 Neonates. See Development
496, 498 stimulus factors in, 382-84 Neospinothalamic pathway, 242, 271
 Subject Index 605

Nerve(s), 531, 535. See also specific Object identification, 16, 27-32, 53, Optic nerve, 71, 87
nerves 324, 336, 323-31, 324 Optic radiations, 77, 87
cranial, 242, 396 context and, 329, 329-31, 330 Optic tract, 76, 87
olfactory, 234 data-driven versus conceptually-driven Optokinetic nystagmus, 461, 477
Nerve endings processing and, 324-31, 326 Order, 372
free, 271 gender differences and, 524, 525 Ordinal scale, 40, 54
in skin, 239 information theory and, 29-32, 30t Ordinate, 17
in muscle, 258-59, 259 channel capacity and, 30, 30-32, Organ of Corti, 161-62, 162, 175
in skin, 239, 239-41, 240 3It, 32 Orientation specificity, 79, 87
Nerve fibers, 242 recognition versus identification and, Orienting, 429-33, 453
pain perception and, 262, 264, 265 324 attentional gaze and, 432-33
Nervous system, 530, 530-32 theories of, 331-35 covert, 429-32, 432, 452
Neural activity, 530, 531-33, 532 Objective instruction, 418, 424 orienting reflex and, 429
Neural coding, of temperature, 252-53, Object properties, 406, 424 orienting response and, 428, 453
253, 254 Object-relative change, 383, 401 overt, 429
Neural function, techniques to measure, Object-relative localizations, 274, 303 Orienting reflex, 429, 453
533, 533-34, 534 Oblique effect, 486-87, 509 Orienting response, 428, 453
Neural pathways, skin senses and, Occipital lobe, 77, 86 Oscilloscope, 533, 533, 535
240-42, 243, 244, 245-46 Occupation, perceptual effects of, 506, Ossicles, 158-59
Neural responses 506-507, 507 Otocysts, 394, 401
in kinesthesis, 259-60, 260 Octave, 340 Otoliths, 394, 401
to pain stimuli, 263-64, 265 Ocular dominance, experience and, 482 Otosclerosis, 516
to smell, 225, 226-27, 227 Oculomotor system, size perception Outcome matrix, 22, 22t, 54
to taste, 215-17, 216 and, 410 Outer ear, 158, 159, 175
in vestibular sense, 396-97 Odor blindness, 228 Outer hair cells, 162, 175
Neural satiation, 116 Odor constancy, 423, 424 Outer spiral fibers, 162, 175
selective adaptation and, 114 Off responses, 72, 86, 171, 175 Outflow theory, 391-92, 392, 401
Neurons, 530, 530-31 auditory, 172, 173 Out of phase, 155, 175
postural, 260 visual, 72, 72 Oval window, 160, 175
Neurophysiology, 530-34 Ogives, 20 Overlay, 303
nature of neural activity and, 530, Ohm’s acoustical law, 156, 157, 175 as depth cue, 275, 276
531-33, 532 Olfactory bulb, 225, 226, 234 Overtones, 203-204
neurons and nervous system and, 530, Olfactory cilia, 225, 234
530-32 Olfactory epithelium, 224, 225, 234 Pacemaker, 376-78, 377, 401
techniques to measure neural function Olfactory nerve, 225, 226-27, 227, Pacinian corpuscle, 239, 239-41, 240,
and, 533, 533-34, 534 234 259, 271
Neutral trials, 448 Olfactory rod, 224-25, 234 kinesthesis and, 258
Neutral zone, 253, 271 Olfactory sense. See Smell Pain, 261-70
Night blindness, 68, 86 Olfactory tract, lateral, 226, 234 analgesia and endogenous opiates and,
Night myopia, 108 One-speaker stereo illusion, 193 267-70, 269
Nociceptors, polymodal, 262 On neurons, 171, 175 double, 262-63, 271
Nodes of Ranvier, 531, 535 On-off responses, 72, 87 evolutionary significance of, 261-62
Noise, 19 auditory, 172, 173 gender differences and, 521
Nominal scale, 40, 54 visual, 72, 72 in infants, 467
Normative orientation, shape constancy On responses, 72, 87 neural responses and, 263-64, 265
and, 418 auditory, 172, 173 scale of intensity for, 266
Now, 372, 401 visual, 72, 72 stimuli and receptors for, 261-63
Nuclei, 531, 535 Opiates, 268 summation of, 267
Nystagmus, 397, 401 endogenous, 268-70, 271 thresholds, intensity, and adaptation
optokinetic, 461, 477 Opponent process, 135-43, 136, 148 and, 264-67, 266t, 267
models of color-coding and, 141-43 Pain points, 264-66, 266t
142 Palatal consonants, 350
Object(s). See also Perceptual object black-white, 137 Palate, 214
recognition of, 323-31, 324 Opsin, 69, 87 Paleospinothalamic pathway, 242, 271
data-driven versus conceptually- Optacon, 251, 271 Pandemonium, 331, 332, 336
driven processing and, 324-31, 326 Optical inversion, 490 Panum’s area, 290. 294, 303
recognition versus identification and, Optic axis, 66, 68, 87 Papillae, 213, 234
324 Optic chiasm, 74, 75, 87 Paradoxical cold, 257, 257-58, 271
Object file, 322 Optic disk, 71, 87 Paradoxical warmth, 257-58
606 Subject Index

Parallel search, 443, 453 Phase-locking, 168, 175 Poggendorff illusion, 512, 512, 528
Parietal region, 242, 244, 271 Phenylthiocarbamide (PTC), 218, 234 Point of subjective equality, 34, 54
Passive processing, 365, 369 threshold for, 218, 218 Poly modal nociceptors, 262
Pattern discrimination Pheromones, 223, 230-34, 234 Ponzo illusion, 413, 414, 415, 424,
during childhood, 472-73 sensitivity to, 230-34 471, 472, 477
in infants, 463-65, 464-66 Phi movement, 384 Pore, taste, 214
Pattern perception, tactile, 250-52, 251 Phone, 350-51, 369 Position constancy, 391, 401, 423, 424
Pattern recognition, local features in, Phoneme, Phonemes, 350-51, 35It, Position information, motion perception
327 369 and, 391-92, 392
Pauser neurons, 171, 175 environmental and life history Positron emission tomography, 172
Payoff matrix, 24, 24t, 54 differences and, 504-505 Postsynaptic membrane, 532, 535
Peduncle, 531 Phonemic boundary, 358, 369 Postural neurons, 260
Penumbra, 102 Phonemic restoration effect, 365, 369 Potentiation
Perceived duration, 195, 208 Phonemic shadowing, 434, 453 of smell, 234
Perceived location, 195, 208 Phonetic gestures, 367-68 of tastes, 219-20
Perception, 9, 13 Phonetic refinement theory, of speech Power law, 44-47, 45t, 46, 47, 48,
categorical, 358-60, 359, 368 perception, 366 54
constructive, size constancy and, 411 — Phonetic space, 366 Practice, divided attention and, 438
12, 413 Photometric units, 90-92, 9It, 92 Pragnanz, law of, 315, 316, 336
direct, 10, 13, 274, 303, 405, 424 Photometry, 90, 116 Preattentive process, 322, 336
size constancy and, 410-11, 411 Photon, 58, 87 Precedence effect, 192-93, 209
duplex, 360, 368 Photopic vision, 68, 87, 93, 116 selective attention and, 435
intelligent, 11, 13, 275, 303 luminosity curve and, 95-96 Preferential looking, 459, 459-60, 477
tactile, of patterns, 250-52, 251 Photopsin, 70, 87 forced-choice, 460, 477
task of, 404, 404-405 Photoreceptors, 64, 87 Presbyopia, 63, 87
theories of, 10-11 Phrases, musical, 344 Present, subjective, 372
Perceptual distortions. See Illusion(s) Phrase shadowing, 434, 453 Pressure amplitude, 154, 175
Perceptual learning approach, 456, 477 Physical condition, color perception Pressure sense, 258
Perceptual moments, 375-76 and, 146 Presynaptic membrane, 531, 535
Perceptual movement, 401 Physiological arousal, vigilance and, Primaries, 127-28, 128, 148
Perceptual object, 310-23, 336 447, 448 Primary auditory projection area, 170,
feature integration theory and, 322-23 Physiological cues, 303 175
features and, 321-22 Physiological differences, 512-20 Primarylike neuron, 171, 175
figural organization and, 313, 313-15, drugs and, 513-16, 514 Primary sketch, 334
314, 316 physical pathology and, 516-20, Primary visual cortex, 77, 80-82, 82,
figure and ground and, 310-13, 311, 516-18, 519t, 520 87
312 Physiological zero, 253, 271 in infants, 458
information, symmetry, and good dependence of thermal sensation on, Primers, 230, 234
figures and, 315-17, 316-18 256 Principle axis, 418
spatial frequencies and figural Pictorial depth cues, 275-84, 303. See Probability distribution, 24, 25, 54
organization and, 320-21, 321 also Depth perception Processing dominance, 328, 336
texture segregation and, 317-20, 319 Picture perception, environmental and Processing effort model, 379, 401
Perceptual rearrangement, 489-93, life history differences and, 500- Profile analysis, 187-88, 209
491, 509 502, 502, 503, 503 Projective instruction, 418, 424
Perceptual set, 507-509, 508, 509 Pigment, 69 Proprioceptive information, motion
Perseveration, error of, 18, 53 Pigment epithelium, 63-64, 87 perception and, 391-92, 392
Personality, perceptual effects of, 525- Pinna(ae), 158, 175 Prosopagnosia, 518-19, 528
28 sound localization and, 191 Protanomaly, 132, 148
Perspective Piper’s law, 98, 116 Protanopia, 132, 148
aerial, 302 Pitch, 195, 197, 198t, 200-203, 202, Prothetic continuum, 41, 54
as depth cue, 276-77 203, 208, 340-41 Proximal stimulus, 404, 405, 424
detail, 280-81, 280-82 perfect, 342-43 Proximity, law of, 313-14, 314, 336,
linear, 303 relations between, 340-43 344
streaming, 401 theories of perception of, 197, 203- Psychic blindness, 83, 87
Phase, 175 208, 204t, 207 Psychoactive drugs, perceptual effects
of sound waves, 154-55 Pixels, 316 of, 515
Phase angle, 155, 175 Place principle, 204, 209 Psychophysics, 16-53
Phase difference, 190-91, 208 Planning, time perception and, 372 detection and, 16-27, 17
 Subject Index 607

method of constant stimuli and, 17, Reduction conditions, 278, 303 Rod(s)
19-20, 20, 54 Reference level, 50 olfactory, 224-25, 234
method of limits and, 17-19, 18, Reflectance, 90, 116, 419, 424 retinal, 64, 68, 68-71, 70, 87
18t, 54 Reflecting tapetum, 64, 87 in infants, 456-57
signal detection theory and, 21t, 21- Reflex(es) Rooting response, 467, 477
27, 22t, 23, 24t, 25-27, 28, 54 orienting, 429 Rounded vowels, 350
direct, 42-50 pupillary, to light, 62 Round window, 160-61, 175
adaptation level theory and, 50-53, Reflex pursuit movement, 389, 389, Ruffini endings, 239
52 401
discrimination and, 33-40, 34 Refraction, 121
reaction time and, 37-40, 38, 39t Refractive error, 63, 87 Saccades, 440, 453
signal detection theory in, 37 Refractory period, 532, 535 Saccadic eye movements, 460, 477
Weber’s law and, 35, 35-36, 36t, 37 Registration, 406, 406t, 424 Saccharin, taste blindness and, 219
identification and, 27-32 in information processing, 10 Saccule, 395, 395, 401
information theory and, 29-32, 30t Reissner’s membrane, 161, 175 Salty taste, 213
scaling and, 40-53 Relative brightness, 303 threshold for, 218
indirect, 41-42, 42 aerial perspective and, 277 Satiation, neural, 114, 116
visual, in infants, 459, 459-60 Relative distance, 274, 303 Saturated neuron, 168, 175
PTC (phenylthiocarbamide), 218, 234 Relative height, 281, 282 Saturation, 123, 123, 148
threshold for, 218, 218 Releasers, 230, 234 Scala media. See Cochlear duct
Pulvinar nucleus, 78, 87 Residual stimuli, 51, 54 Scales, 40-41
Pupil, 59, 62, 87 Resolution acuity, 105, 106, 116 interval, 40-41, 54
Pupillary light reflex, 62 Response pattern, payoff matrix and, 24 nominal, 40, 54
Purity, 123, 148 Response probability, 20,20 ordinal, 40, 54
Purkinje shift, 96-97, 116 Resting potential, 531,535 ratio, 41, 54
Restricted environmental stimulation Scaling, 16, 40-53, 54
Quantum, 58, 87 technique, 378,401 adaptation level theory and, 50-53,
Restricted rearing, 482-87, 509 52
Radial fibers, 162, 175 human studies of, 485-87 constancy, 408, 408, 409, 424
Radiance, 90, 91, 116 neurophysiological effects of, 482-84, direct, 41, 42-50, 53
Random-dot stereopsis, 294-95, 296 483 category judgment and, 42, 43, 43-
Ratio scale, 41, 54 perceptual effects of, 484-85, 485 44
Reaction time, 54 Retina, 63-65, 66, 87 cross-modality matching and, 47, 49,
choice, 38, 38-39, 39t, 53 color sensitive zones on, 136 49-50
discrimination and, 37-40, 38, 39t, duplex retina theory and, 68, 86 magnitude estimation and, 44-47,
40 in infants, 456-57 45t, 46, 47, 48
simple, 38, 54 locus on equal-interval, 53
stimulus discriminability and, 39 brightness perception and, 94-95 indirect, 41-42, 42, 54
Reafference, 487—89, 488, 489, 509 visual acuity and, 106-107, 107 mixed-modality, 50, 54
Real movement, 384, 401 mapping blood vessels of, 65 Schemata, 469, 477
Rearrangement, 489-93, 491, 509 Retinal, 69, 87 Schizophrenia, 527-28, 528
Receiver operating characteristic (ROC) Retinal illuminance, 90, 116, 419, 424 Schwann cells, 530, 535
curve, 23, 23, 54 Retinal image, 306-307, 336 Sclera, 59, 87
Receptive fields, 87 stabilized, 308-10, 336 SCN (suprachiasmatic nucleus), 374,
of skin, 241 afterimages and, 393, 393t 375, 401
visual, 72, 79-80, 80, 81 Retinal image size, as distance cue, Scotoma, 79, 87
Receptors __ 277, 277-78, 278, 303 Scotopic vision, 68, 87, 93, 116
kinesthetic, 258-59, 259 luminosity curve and, 96
for pain, 262-63 Retinal points, corresponding, 290, 302 Scotopsin, 70, 87
for skin senses, 238-41, 239, 240 Reverberation, 194, 209 Search, 469, 477
smell, 224-26, 225 Reversible figures, 422 during childhood, 468-71, 469-71
generalist, 226 drug effects and, 514, 514—15 conjunction, 452
stretch, 258-59 Rhodopsin, 69, 87 feature, 453
taste, 213-15, 214, 215 Rhythm, 345, 347,369 parallel, 453
vestibular, 395-96, 396 circadian, 373-74,375,400 serial, 453
Recognition, 324, 336 Ricco’slaw,98,116 Searching, 428-29, 440-47, 453
of patterns, local features in, 327 ROC (receiver operating characteristic) automatic versus controlled, 442-46,
Recognition acuity, 116 curve, 23,23,54 445
608 Subject Index

Searching (continued) discrimination and, 37 Smell primaries, 230

feature versus conjunction, 440-42, vigilance and, 447 Smell prism, 230, 230
444 Signal present presentation, 21 Smell qualities, 229-30, 230, 231
vigilance and, 446-47, 448 Similarity, law of, 314, 314, 336, 345 Smoking, perceptual effects of, 513
visual, eye movements and, 440, Simple cell, 79, 80, 87 Smooth-pursuit eye movement, 460*
441-43 Simple reaction time, 38, 54 61, 477
Secondary visual cortex, 77, 87 Simultagnosia, 517, 517, 518, 528 motion perception and, 388, 389,
Second formant, 353 Simultaneity, 372 389-92, 392, 393t, 401
Selection, early, 450, 451, 452 Simultaneous brightness contrast, 116- Snellen eye chart, 104-105
late, 450, 451, 453 17 Snow blindness, 307
Selective adaptation, 114, 116, 381, Simultaneous color contrast, 145, 149 Social behavior, pheromones and, 232-
401 Simultaneous masking, 181, 209 33
of spatial frequency channels, 115 Sine wave grating, 109, 109, 117 Sodium-potassium pump, 531, 535
Selective attention, precedence effect Sinusoidal sound wave, 152 Soft palate, 214
and, 435 SI System (Systeme International Solitary tract, 215, 234
Selective rearing, 482-87, 509 d’Unites), 90, 117 Somatosensory cortex, 242, 246, 271
human studies of, 485-87 Situation properties, 406, 425 kinesthesis and, 260
neurophysiological effects of, 482-84, Size, familiar, 303, 501, 509 pain perception and, 264
483 as distance cue, 278, 278, 279 Somatosensory map, 242, 244, 245,
perceptual effects of, 484-85, 485 size constancy and, 412 246
Self-adaptation, to smells, 228, 229, Size constancy, 407, 407-16, 425 Sone, 196, 209
234 direct and constructive aspects of, Sound(s), 152-56, 153. See also
Self-motion, 397-400, 398, 401 410-12, 411, 413 Hearing
induced, 399 distance cues and, 407-10, 408 detection of, 178, 178-95
Semicircular canals, 395, 395, 401 illusion and, 408, 412-13, 414-16, auditory masking and, 181-85, 182,
Semitones, 201 415-16 183
Sensation, 9, 13 Skill, 446 sound discrimination and, 185-89,
Sensitive period, 301, 303 Skin 186-88
Sensitivity (d')» 25, 27, 27, 28, 53 absolute temperature of, 252 sound localization and, 189-95
computation of, 28 glabrous, 238-39, 271 temporal, frequency, and binaural
maximum, brightness perception and, nerve fibers in, 242 interactions and, 180-81, 181
98 hairy, 238, 239, 271 discrimination of, 185-89, 186-88
Sensory-motor learning, 487-94, 488, inhibitory interactions on, 240-41 frequency of, 180-81, 181
489 as model for basilar membrane, 164 hearing limits and, 180
illusion decrement and, 493-94, 494 temperature of pitch and, 195
perceptual rearrangement and, 489- below physiological zero, 253 threshold and, 179
93, 491 touch threshold and, 247 localization of, 189-95
Sensory convergence, 400, 401 Skin senses, 238-46. See also complex sounds and, 192, 192-94
Sensory neuron, 530, 535 Kinesthesis; Pain; Temperature; physiological mechanisms of, 194-
Separable stimulus, 328-29, 336 Touch 95
Sequence, 372 neural pathways and responses and, simple tones and, 189-92, 190, 192
Serial search, 442, 453 240-42, 243, 244, 245-46 pitch of, relations between, 340-43
Set, perceptual, 509 stimuli and receptors and, 238-41, simple sound wave and, 152, 153,
Sexual attraction, pheromones and, 239, 240 154-57, 155, 155t, 157
230, 231 Smell, 222-34, 223, 468 speech. See Speech perception
Shading, as depth cue, 275-76, 276 adaptation to, 228-29 subjective dimensions of, 195-208
Shadowing, 430-31, 434-35, 453 aging and, 476, 476t loudness and, 48, 195, 196, 196—
phonemic, 434, 453 gender differences and, 520 200, 197, 198t, 199, 200
phrase, 434, 453 intensity and qualities and, 229-30, pitch and, 197, 198t, 200-203, 202,
visual, memory and, 438 230, 231 203
Shape, 307, 336 interaction with taste, 222-23, 223, theories of pitch perception and, 197,
Shape constancy, 416-19, 417, 418, 224 203-208, 204t, 207
425 neural responses in, 225, 226-27, 227 Sound pressure level (SPL), 154, 155t,
Short-range process, 385, 401 odor constancy and, 423 175
Sighting-dominant eye, 289, 303 sensitivity to pheromones and, 230-34 Sound shadow, 189, 209
Signal absent presentation, 21 stimuli and receptors for, 224-26, Sound waves
Signal detection theory, 2It, 21-27, 225 motion on basilar membrane, 163,
22t, 23, 24t, 25-27, 54 thresholds for, 227-28, 228 163-64, 164
 Subject Index 609

simple, 152, 153, 154-57, 155, 155t, Spatial vision, eye movements and, in Stereotaxic instrument, 533-34, 534,
157 infants, 460-61, 461 535
Sour taste, 213 Species differences, in development of Stevens’s law, 44-47, 45t, 46, 47, 48,
Space perception, 274-302 space perception, 299-301,300 55
binocular depth perception and, 289- Specificity, orientation, 79, 87 Stimulants, perceptual effects of, 514
95 Spectral colors, 123, 149 Stimulus(i)
cues for stereopsis and, 290, 291 — Spectrum, of sound, 194 adapting, 114
93, 294 Speech perception, 348-68 ascending series of, 17
process of stereopsis and, 291-95, categorical, 358-60, 359 background, 51, 53
295, 296, 297 context and, 363-65 base, 360
development of, 299-302 cross-model integration and, 359, color, 120-30, 727, 121t
experience and, 297, 301-302 360-62 CIE color space and, 124, 127-30,
species differences in, 299-301,300 development of, 362-63, 363 128, 129
interaction of cues and, 295, 298, 299 duplex, 360 color appearance systems and, 122—
physiological cues for depth and, environmental and life history 24, 122-24
284- 85 differences and, 504-506 color mixture and, 124-27, 125-27
accommodation and, 284-85 issues in, 355-65 comparison, 33, 53
convergence and divergence and, ambiguity and invariance and, 354, concentration of, taste and, 217
285, 286 355-56, 356, 357 constant, method of, 77, 19-20, 20,
physiological cues for direction and. context and, 363-65 54
285- 89, 287 development and, 362-63, 363 context, 406, 424
eye dominance and perceived segmentation of signal and, 357 descending series of, 17
direction and, 289 stimulus and, 349-55 discriminability of, reaction time and,
eye movements and, 287-88 acoustic properties of speech and, 39
pictorial depth cues and, 275-84 351, 352-54, 353-55 distal, 404-405, 424
aerial perspective and, 276-77 consonants and vowels as, 349-50 filtering, 433-38
height in the plane and, 281, 282 phonemes as, 350-51, 35It focal, 51, 53, 406, 424
interposition or overlay and, 275, theories of, 365-68, 367 object and situation properties of,
276 Speech spectrogram, 351, 352, 353- 406
linear perspective and, 278, 279, 280 55, 369 integral, 328, 329, 336
motion and motion parallax and, Spike potential, 532, 166, 535 kinesthetic, 258
283, 283-84 Spindle organs, 258-59, 271 micropattems and, 375
retinal and familiar size and, 277, Spinothalamic pathway, 242, 271 monochromatic, 122-23, 148
277-78, 278, 279 Spiral ganglion, 162, 175 motion perception and, 382-84
shading and, 275-76, 276 SPL (sound pressure level), 154, 155t, for pain, 261-62
texture gradients and, 280-81, 280- 175 proximal, 404, 405, 424
82 S potentials, 137, 148 residual, 51, 54
types of space and depth perception Spotlight, 432 separable, 328-29, 336
and, 274-75 Square wave grating, 109, 109, 117 for skin senses, 238
Spatial agnosia, 517, 518, 528 Stabilized retinal image, 308-310, smell, 224
Spatial frequencies 336 speech perception and, 349-55
figural organization and, 320-21, 321 afterimages and, 393, 393t acoustic properties of speech and,
low and high, 110 Staircase method, 18-19, 55 351, 352-54, 353-55
Spatial frequency analysis, 108-16 Standard, 33, 55 consonants and vowels as, 349-50
Fourier’s theorem and, 108-10, 109, Standard candle, 90 phonemes as, 350-51, 35It
110 Standard color, 95 taste, 212-13
modulation transfer function and, 109, Standard units, 90, 117 vestibular, 395, 395, 396
110-12, 111, 112 Stapes, 158, 175 vibratory
neural spatial frequency channels and, Statocysts, 394, 401 adaptation for, 249
709,112-14,775 Statolith, 394, 401 intensity of, 250, 250
selective adaptation and, 772, 114-16 Steady-state responses, 253 touch threshold for, 247
Spatial frequency channels Stereopsis, 290, 303 Stimulus cue, 432, 453
neural, 112-14, 113 cues for, 290, 291-93, 294 Stimulus dimensions, 32
selective adaptation of, 115 global, 294, 296, 303 Stimulus intensity
Spatial interactions, color perception process of, 291-95, 295, 296, 297 color perception and, 143-44
and, 145 random-dot, 294-95, 296 for pain, 266
Spatial modulation transfer function, Stereoscope, 291-92, 295, 303 smell and, 229
111, 117 Stereoscopic depth constancy, 299 taste and, 217, 220-21
610 Subject Index

Stimulus intensity (continued) receptors and, 213-15, 214 thermal, 253, 255-57, 255-57
thermal, 253, 257, 257-58 salty, 213, 218 for touch, 244, 246, 246-48, 249
for touch, 250, 250 sour, 213 absolute, 246, 246-47
visual acuity and, 107-108 stimuli and, 212-13 two-point, 247-48, 248, 249, 271
Stop consonants, 349 sweet, 213, 218 variability of, 19
Storage size model, 220-21, 401 thresholds for, 217, 217-19, 218 Threshold distance, for hearing, 180
Streaming perspective, 398, 398, 401 Taste blindness, 218-19 Threshold response curve, 166, 166—
Strength, of behaviors, 20 Taste buds, 213-15, 214, 235 67, 175
Stretch receptors, 258-59, 271 Taste pore, 214, 235 Timbre, 155, 155, 156, 175, 195, 203,
Striate cortex, 77, 87 Taste qualities, 221-22, 222 209, 342
Strong automaticity, 445 Taste sensitivity, 217-18 tonal grouping and, 345
Stroop effect, 445, 446, 453 T cells (transmission cells), 264, 265, Time, 372-80. See also Duration
Structural cues, 303 271 biological clocks and, 373-78
Structural theories, 453 Tectopulvinar system, 74, 75 , 77-78, biological pacemaker and, 376-78,
of attention, 450, 451 78, 87 377
Subjective colors, 142, 143, 149 motion perception and, 381 circadian rhythms and, 373-74, 375
Subjective contours, 312, 312-13, 336 Tectorial membrane, 161, 175 short-term timers and, 374-76
Subjective present, 372 Tectum, 77, 87 brightness perception and, 97-98
Subject-relative change, 383, 401 Temperature, 252-58 cognitive clocks and, 378-80
Substantia gelatinosa, 264, 265, 271 body, time perception and, 376-78, change and, 378
Subtractive color mixture, 125, 126, 377 processing effort and, 379
126-27, 149 heat grill and, 254 temporal versus nontemporal
Successiveness, 372 neural coding of, 252-53, 253, 254 attention and, 379-80
Superior colliculus, 78, 87 of skin contour emergence and, 309-10, 310
in infants, 458 below physiological zero, 253 hearing and, 180
motion perception and, 391 touch threshold and, 247 perceived duration and, 195, 208
Superior olives, 170, 175 thermal intensity and qualities and, Time difference, 189, 190, 191, 209
Superior temporal cortex, 84, 87 253, 257, 257-58 Tinnitus, 193
Suprachiasmatic nucleus (SCN), 374, thermal thresholds and adaptation and, Tone(s), musical, 340-43, 341
375, 401 253, 255-57, 255-57 Tone chroma, 341, 369
Sweet taste, 213 Template matching, 365 Tone height, 340, 369
threshold for, 218 Tempo, 345-46, 369 Tonotopic response. 171, 175
Symmetry, figural goodness and, 316- Temporal attention, 379-80 response, 171
17, 317 Temporal lobes, 77, 82-84, 84 Touch, 246-52
Synapse, 531-32, 532, 535 motion perception and, 381 adaptation of, 248-49
Synaptic cleft, 531, 535 Temporal processing model, 379, 401 gender differences and, 520
Synaptic knob, 531, 535 Terminators, 319 in infants, 467
Synaptic vesicle, 531, 535 Textons, 319, 319-20, 336 stimulus intensity and, 250, 250
Systeme International d’Unites (SI Textural contour, 318, 336 tactile pattern perception and, 250-52,
System), 90, 117 Texture, visual, 317-18 251
Texture gradient, 303 thresholds for, 244, 246, 246-48, 249
Target, 181, 209 as distance cue, 280-81, 280-82 TRACE, 366-67, 369
bipartite, 95 Texture segregation, 317-20, 319 Tracking lag, 390
localization of, eye movement and, Thalamus, 76, 87, 215, 271, 235, 242 Tract, 531, 535
287-88 3-D model, 334 Transactional viewpoint, 494-99, 495,
Taste, 212-22, 467-68 Three-dimensional response, 501-502, 496, 498, 509
adaptation and, 219, 219-20 509 Transduction, 165, 165-66
aging and, 476, 476t Threshold(s) of light, 64
bitter, 213, 218 absolute, 17, 53 olfactory, 225-26
effects of drugs on, 513 brightness perception and, 92-93 Transmission (T) cells, 264, 265, 271
gender differences and, 520 thermal, 253, 255 Transmitter substance, 531-32, 535
hot, 218, 221 for touch, 246, 246-47 Trichromatic theory, 130-35, 149
intensity and qualities and, 220-22, difference, 53 color vision defects and, 128, 131 —
222 for movement, 383 33, 133, 134
interaction with smell, 222-23, 223, for pain, 264-66, 266t physiological basis of, 133-35, 135,
224 smell, 227-28, 228 136
introversion-extroversion and, 527 sound frequency and, 179 Trichromatism, anomalous, 132, 148
neural responses in, 215-17, 216 taste, 217, 217-19, 218 Trigeminal nerve, 215
 Subject Index 611

Tristimulus values, 130, 149 photopic, 68, 87, 93, 116 Visual-spatial abilities, gender
Tritanopia, 132, 149 luminosity curve and, 95-96 differences and, 521-25, 522
Tuned neuron, 166, 175 scotopic, 68, 87, 93, 116 physiological factors and, 521-22,
Tuning curve, 167, 167, 175 luminosity curve and, 96 524
Tunnel of Corti, 161-162, 162, 175 Vision substitution system, 250-51, psychosocial factors and, 524-25
2‘/2-D sketch, 334 271 Visual straight ahead, 286, 289
Two-dimensional response, 501-502, Visual acuity, 60, 86, 104-108, 105- Visual system, 58-85. See also Visual
509 107, 117 cortex
Two-point threshold, 247-48, 248, aging and, 475 development in infants, 456-59, 457t
249, 271 directional, 105, 106 effects of lights on, 507
Two-tone inhibition, 167, 175 as function of retinal location, 106— eye and, 58-71, 60
Tympanic canal, 160 107, 107 aging and, 473-75, 474, 475
Tympanum, 158, 175 grating, 105, 106 crystalline lens of, 62, 62-63, 64
hyperacuity and, 106 Cyclopean, 286, 288, 303
Unconscious inference, 405, 425 illumination and, 107-108 fovea of, 65-68, 66-68
Uncrossed disparity, 290, 292-93, 303 in infants, 462, 462, 463 mind’s, 432
Unrounded vowels, 350 intensity and, 107-108 movement of, direction and, 287-88
Utricle, 395 , 395 , 401 recognition, 104, 116 retina of, 63-65, 66
resolutional, 105, 106, 116 rods and cones of, 66, 68-71, 70
VI, 77, 87 retinal position and, 106-107, 107 sighting-dominant, 289, 303
V2, 77, 87 Vernier, 105, 106, 117 light and, 58, 59
V3, 77, 87 Visual agnosia, 83-84, 87 neural responses to light and, 71-74
Vagus nerve, 215 Visual angle, 105, 105, 117 retinal responses and receptive fields
Valid trials, 448 Visual attention, 432-33 and, 72 , 72-73 , 73
Vanishing point, 280, 303 divided, 436-37, 439 W, X, and Y cells and, 73 , 73-74,
Vection, 399, 401 Visual capture, 430, 431, 432, 453 74
Velar consonants, 350 Visual cliff, 300, 300-301, 303 tectopulvinar system and, 75 , 77-78,
Ventral cochlear nucleus, 169-70, 175 Visual context, 383, 401 78
Ventral posterior nuclei, thalamic, 215 effect on judged weight, 51 visual pathways and, 74-77, 75
VEP (visually-evoked potential), 458, Visual cortex experience and, 482
477 aging and, 475 in infants, 457, 457t
Vergence movements, 285, 303 alternate visual maps and, 85, 85 Visual texture, 317-18, 336
Vernier acuity, 105, 106, 117 color perception and, 139 Vitreous humor, 63, 87
Vestibular canal, 160, 175 motion perception and, 381, 391 Vocorder, 353
Vestibular nuclei, 396-97, 401 organization of, 80-82, 82, 83 Voiced fricatives, 350
Vestibular sense, 392, 394-97, 395 primary, 77, 80-82, 82, 87 Voiceless fricatives, 350
neural responses in, 396-97 in infants, 458 Voice onset time, 358, 369
stimuli and receptors for, 395, 395- receptive fields in, 79-80, 80, 81 Volley principle, 206, 209
96, 396 secondary, 77, 87 Volume, of sound, 195, 209
Vestibular system, 394, 401 temporal lobes and, 77, 82-84, 84 Voluntary pursuit movements, 388,
motion perception and, 389 Visual field, 61, 306, 306-307, 336, 389, 401
self-motion and, 399-400 484-85, 485, 509 Vowels, 349, 369
Vibration theory, of smell, 226, 235 Visual filtering, 436-37, 439 rounded, 350
Video overlap phenomenon, 436, 437, Visual-geometric illusions, 471, 477 in speech spectrograms, 353
438 Visual information, self-motion and, unrounded, 350
Vigilance, 446-47, 448, 453 399-400
Vision Visually-evoked potential (VEP), 458, Warm fibers, 252-53, 253, 254, 256-
aging and, 473-75, 474, 475 477 57, 271
color, 130-43 Visual maps, 85, 85 paradoxical cold and warmth and,
defects in, 128, 131-33, 133, 134 Visual object agnosia, 516, 516-17, 257-58
opponent-process theory and, 135— 528 Warmth, paradoxical, 257-58
43, 136 Visual pathways, 74-77, 75 Water
screening inventory for, 134 experience and, 482 smell and, 224
trichromatic color theory and, BO¬ geniculostriate, 76-77, 77 taste adaptation and, 220
SS in infants, 457, 457t Wave, 152, 153, 175
duplex retina theory of, 68 Visual response, in infants, 457-458 Wavelength, 175
effects of drugs on, 513-14 Visual searching, eye movements and, brightness perception and, 95, 95-97
gender differences and, 521 440, 441-43 dominant, 124, 148
612 Subject Index

Wavelength (continued) Weber’s law, 35, 35-36, 36t, 37, 55 X cells, 73, 87
of sound waves, 154 with weights, 260 in infants, 458
primary, 127 Weight
W cells, 73, 87 judged, effect of visual context on, 51 Y cells, 73, 87
Weber fraction, 35-36, 36t, 37, 55 perception of, 260-61 in infants, 458
pain stimuli and, 266 White matter, 531 Yerkes-Dodson law, 447, 448, 453
smell and, 229 Whiteness constancy, 419-22, 420,
sound discrimination and, 188, 188 421 Zeitgeber, 374, 401
taste intensity and, 220 Whytt’s reflex, 62 Zoom lens, 432
with weights, 260, 261
 , \
V

> »
 .
*