Learning and Memory

Learning and Memory provides a balanced review of the core methods and the latest
research on animal learning and human memory. Topical coverage ranges from the basic
and central processes of learning, including classical and instrumental conditioning and
encoding and storage in long-term memory, to topics not traditionally covered, such as
spatial learning, motor skills, and implicit memory. The general rules of learning are
reviewed along with the exceptions, limitations, and best applications of these rules.
Alternative approaches to learning and memory, including cognitive, neuroscientific, func-
tional, and behavioral, are also discussed. Individual differences in age, gender, learning
abilities, and social and cultural background are explored throughout the text and pre-
sented in a dedicated chapter.

The relevance of basic principles is highlighted throughout the text with everyday exam-
ples that ignite reader interest in addition to more traditional examples from human and
animal laboratory studies. Research examples are drawn from education, neuropsychol-
ogy, psychiatry, nursing, and ecological (or everyday) memory. Each chapter begins with
an outline and concludes with a detailed summary. Applications and extensions are show-
cased in text boxes as well as in distinct applications sections in every chapter; and review
and recapitulation sections are interspersed throughout the chapters.

W. Scott Terry is Professor Emeritus at the University of North Carolina at Charlotte

where he taught psychology of learning to undergraduates and graduate students in
industrial-organizational psychology, clinical psychology, and education. He received his
Ph.D. from Yale University, where he studied learning, cognitive psychology, and psycho-
biology. He has published research in the Journal of Experimental Psychology, Learning
and Motivation, Journal of Educational Psychology, and Teaching of Psychology. Over the
years, his research interests have included classical conditioning, comparative cognition,
personality and memory, and everyday lapses of cognition.
Learning and Memory
Basic Principles, Processes, and Procedures

Sixth Edition
W. Scott Terry
Designed cover image: © Getty Images
Sixth edition published 2023
by Routledge
605 Third Avenue, New York, NY 10158
and by Routledge
4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2023 W. Scott Terry
The right of W. Scott Terry to be identified as authors of this work has been
asserted in accordance with sections 77 and 78 of the Copyright, Designs and
Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in writing from the
publishers.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks and are used only for identification and explanation without intent to
infringe.
First edition published by Pearson Education Inc. 2000
Fifth edition published by Routledge 2017
ISBN: 978-1-032-12973-0 (hbk)
ISBN: 978-1-003-22709-0 (ebk)
DOI: 10.4324/9781003227090
Typeset in Univers
by SPi Technologies India Pvt Ltd (Straive)
Access the Support Material: www.routledge.com/9781032129730
To:

Lorraine Piersanti Terry

Christopher Scott Terry

Jason Scott Terry

BRIEF CONTENTS

Preface xvii
Acknowledgments xviii

1 Introduction 1

2 Habituation and Other Forms of Stimulus Learning 24

3 Classical Conditioning 49

4 Instrumental Conditioning: Reward 81

5 Instrumental Conditioning: Nonreward, Punishment, and Avoidance 115

6 Verbal Learning 145

7 Human Memory Conceptual Approaches 174

8 Short-Term Memory 203

9 Encoding 233

10 Storage and Retrieval 266

11 Spatial, Motor-Skills, and Implicit Learning 299

12 Individual Differences in Learning and Memory 329

Glossary 358
References 371
Author Index 409
Subject Index 418
CONTENTS

Preface xvii
Acknowledgments xviii

1 Introduction 1
The Origins of the Study of Learning 2
Epistemology 2
Evolution 3
Contemporary Influences 4
The Definition of Learning 4
Some Caveats 7
The Learning/Performance Distinction 8
Learning: Recapitulation 9
The Relationship Between the Terms Learning and Memory 9
The Science of Learning and Memory 12
Basic Research 13
Applied Research 13
Common Sense and Common Knowledge 15
Why Animals? 16
Conceptual Approaches to the Study of Learning 17
The Functional Approach 17
The Behavioral Approach 17
The Cognitive Approach 18
The Neuroscience Approach 18
Applications 20
Studying: Recommendations from Experts 20
Summary 21

2 Habituation and Other Forms of Stimulus Learning 24

What Is a Stimulus? 25
The Orienting Response 25
Habituation 26
Methods of Studying Habituation 26
Parametric Features of Habituation 29
Explanations of Habituation 32
Nonlearning Explanations 32
Neuroscience Theory 33
Evolution 35
Cognitive Theories 35
Sensitization 37
x Contents

Perceptual Learning 38
Factors Affecting Perceptual Learning 39
Other Effects of Stimulus Exposure 40
Preference for Familiar Stimuli 40
Recapitulation: The Effects of Repeated Stimulus Presentation 43
Applications 43
Exposure Therapy for Fear 44
Needle Fear 45
Summary 46

3 Classical Conditioning 49
The Definition of Classical Conditioning 50
Historical Note 52
Methods of Studying Classical Conditioning 52
Representative Procedures 53
What Stimuli Can Serve as CSs? 54
Basic Phenomena of Conditioning 55
Acquisition 55
Extinction 57
Generalization 58
Discrimination 59
Second-Order Conditioning 59
The Role of Contiguity 60
Summary of the Basic Phenomena 61
Other Factors Affecting Conditioning 61
Compound CSs 61
CS–US Relevance 65
Conditioned Inhibition 66
What Is Learned in Classical Conditioning? 67
So, What Is Learned in Pavlovian Conditioning? 68
The Role of Awareness in Conditioning 69
Extensions of Conditioning 70
Evaluative Conditioning 70
Conditioning with Drug USs and the Development of Tolerance 71
Applications of Conditioning 74
The Conditioning Theory of Phobias 74
Overview of the Conditioning Theory of Phobias 76
Extinction as Therapy 76
Summary 77

4 Instrumental Conditioning: Reward 81

Definition and History 82
Thorndike and Trial-and-Error Learning 82
Skinner and Operant Learning 83
Elements of Instrumental Conditioning 84
 Contents xi

Methods of Study 85
Positive Reinforcement 86
Reinforcement Variables Affecting Acquisition 87
Amount of Reinforcement 87
Drive87
Schedules of Reinforcement 89
Delay of Reinforcement 90
Secondary Reinforcement 91
Social Reinforcement 91
Theories of Reinforcement 92
Drive Reduction 92
Incentive Motivation 93
The Neural Basis of Reinforcement 93
Reinforcers as Behaviors 94
Reinforcers as Strengtheners 95
Reinforcers as Information 95
So, What Is Reinforcement? 96
Is Reinforcement Necessary for Learning? 96
Awareness in Human Instrumental Learning 98
Criticisms of the Use of Reinforcement 98
Response Learning 99
Shaping 99
Limitations of Response Learning 100
Discriminative Stimulus Control 102
Generalization and Discrimination 102
Summary of Response Learning and Stimulus Learning 104
What Is Learned in Instrumental Conditioning? 104
Response–Reinforcer Learning 105
Stimulus–Response Learning 105
Stimulus–Reinforcer Learning 106
What Is Learned? Stimulus–Response–Reinforcer 106
Applications 107
Habits 107
Behavior Modification 109
Summary 111

5 Instrumental Conditioning: Nonreward, Punishment, and Avoidance 115

Defining the Contingencies: Nonreward, Punishment, and Avoidance 116
Extinction 117
The Partial Reinforcement Extinction Effect 118
Punishment 120
When Does Punishment Work? 121
Side Effects of Punishment 124
Punishment or Nonreward? 125
Persistence 127
xii Contents

Avoidance Learning 128

Theories of Avoidance Learning 129
Approach–Avoidance Conflict 130
Approach or Avoidance as a Coping Response 131
Prediction, Control, and Helplessness 133
Extensions of the Learned-Helplessness Concept 135
Summary of Learned Helplessness 137
Neuroscience and Aversive Learning 137
The Amygdala and Aversive Learning 137
Avoidance Conditioning 138
Applications of Aversive-Learning Contingencies 140
Pet Containment Systems 140
Treatment of Obsessive-Compulsive Disorder 140
Summary 141

6 Verbal Learning 145

The Ebbinghaus Legacy 146
Serial Learning 148
Serial Position 148
The Ubiquitous Serial-Position Curve 148
Remote Associations 150
Serial Learning: A Summary 151
Paired-Associate Learning 151
Analysis of Paired-Associate Learning 152
Direction of Associations 154
Paired-Associate Learning: A Summary 154
Free Recall 156
Serial-Position Effects 156
Rehearsal 157
Organization 158
Free Recall: A Summary 159
Recognition Memory 160
Recognition: Remembering Versus Knowing 160
Accessible Memories 161
Recall Versus Recognition Versus Relearning 162
Implicit Learning 163
Relationships Among the Verbal-Learning Tasks 164
Statistical Learning 164
Propositional Learning 165
Statistical Language Learning 165
Applications 167
Assessment of Cognitive Impairment 167
Mnemonics 167
Summary 170
Recognition Memory 171
 Contents xiii

7 Human Memory Conceptual Approaches 174

Partitioning Memory 175
Components of Memory Approach 177
Dual-Store Theory: Short-Term and Long-Term Memories 177
Divisions of Long-Term Memory: Episodic and Semantic 178
Divisions of Long-Term Memory: Procedural Learning and Priming 180
The Organization of Long-Term Memory 184
Stages of Memory Approach 184
Dissociating Stages 185
Processing Approaches to Memory 187
Depth of Processing 187
Artificial Neural Networks 188
Modeling Configural Learning 189
Connectionism and the Other Approaches 191
Applications 191
The Study of Abnormal Memory: Amnesia 191
Types of Amnesias 193
Everyday Forgetting and the Models of Memory 197
Summary 199

8 Short-Term Memory 203

Some History 204
Short-Term Memory Tasks 205
The Brown–Peterson Distractor Task 205
Memory Span 207
Characteristics of Verbal Short-Term Memory 208
Acoustic Encoding 208
Limited Capacity 209
Limited Duration 210
Forgetting: Short-Term Memory Is Sensitive to Disruption 210
Transfer to Long-Term Memory 210
Control Processes 211
Summary of the Features of Short-Term Memory 211
Other Modalities of Short-Term Memory 211
Visual Short-Term Memory 212
Spatial Short-Term Memory 212
Short-Term Memory for Actions 212
Short-Term Memory for Odors 213
Short-Term Memory in the Hearing Impaired 213
Working Memory 215
Theoretical Overview of Working Memory 215
Measuring Working Memory 218
Executive Functions 219
Working Memory and Consciousness? 220
Working Memory and Culture 220
xiv Contents

Individual Differences in Working Memory 221

Aging and Working Memory 221
Dementia and Working Memory 221
Anxiety and Working Memory 222
Multitasking 222
Is There Really a Separate Short-Term Memory? 225
A Single-Memory Approach? 225
Neuropsychological Dissociations of Two-Memory Systems 226
One Memory or Several? 227
Applications 227
Comprehending and Using Language 228
Problem Solving 228
Summary 229

9 Encoding 233
Separating Encoding from Retrieval 234
Some Basic Variables in Encoding 234
Elaborative Rehearsal 235
What Exactly Is Elaborative Processing? 236
Imagery and Memory for Pictures 239
Meaningfulness 240
Presentation Variables 241
Testing Effects 241
Isolation Effects 243
Spacing Effects 245
Learner Variables 247
Incidental Versus Intentional Learning 247
Incentives 248
Arousal 249
Emotions and Encoding 254
Schemas 257
Metamemory 258
Encoding: Summing Up 261
Applications 261
Academic Learning and Encoding 261
Summary 262

10 Storage and Retrieval 266

Storage 267
Long-Term Memory for Naturalistically Learned Material 268
The Nature of Storage 270
Retrieval 276
Retrieval from Episodic Memory 278
What Makes a Good Retrieval Cue? 279
 Contents xv

Emotional Arousal and Retrieval 283

Prospective Memory 284
Metamemory and Partial Retrieval 285
Feeling of Knowing 285
False Retrieval 286
False Memory 286
Source Memory 287
The Effect of Postevent Information 289
Debunking Misinformation 290
Recovered Memory 291
Retrieval Versus Reconstruction 292
Applications 293
Strategies for Searching Memory 293
Context-Specific Learning 295
Reconsolidation 295
Summary 295

11 Spatial, Motor-Skills, and Implicit Learning 299

Spatial Learning 300
Rats, Mazes, and Psychology 300
Routes Versus Cognitive Maps 300
Maze Learning and the Brain 305
Schemas in Spatial Memory 307
The Development of Spatial Memory in Children 309
Motor-Skills Learning 309
Practice 311
Knowledge of Results 314
Long-Term Retention of Skills 316
Implicit Learning 316
Some Implicit-Learning Tasks 317
Dissociating Categories of Implicit Learning 320
Expertise 321
From Declarative to Procedural 322
Applications 323
Implicit Learning 323
Spatial Memory 324
Developing Memory Skill 326
Summary 327

12 Individual Differences in Learning and Memory 329

The Nature of Nurture: The Genetics of Learning Ability 330
Animal Studies 330
Human Studies 332
Age Differences in Learning and Memory 333
Conditioning 333
xvi Contents

Memory Development in Children 335

Aging and Memory 340
Exceptional Memory: The Mnemonists 342
Highly Superior Autobiographical Memory 343
Gender and Cognitive Abilities 345
Learning Styles 347
The Visualizer–Verbalizer Dimension 347
Kolb348
Sternberg 348
Learning Styles? 349
Self-Control 349
Social and Cultural Differences 350
Epic Memories 351
Experimental Studies 351
Contemporary Cultural Psychology 352
Summary 355

Glossary 358
References 371
Author Index 409
Subject Index 418
PREFACE

Learning and remembering what we have learned are fundamental psychological processes.
The scientific study of learning and memory has produced a sizable body of principles, laws,
and heuristic rules. Since the very beginning of psychology as a discipline, researchers who
study learning have been asked to provide instruction in these principles. What do we know
about learning, and how can this knowledge be usefully applied? My approach to writing a
book on learning and memory is to present the basic methods and results of our research
and to emphasize the relevance of our research to other disciplines.
My purpose in writing this book has remained the same since the first edition.
I wanted to write what I teach in my own courses. I have taught Psychology of Learning
to college undergrads, Master’s level I/O psychology, and students in Education Master’s
program. I had a unique background that coincided with the resurgence in behavioral
Learning Theory, the development of Cognitive Psychology, and new discoveries in the
Biopsychology of Learning.
I had several goals in mind in writing this book. The topical coverage is restricted to
basic and central processes, such as classical and instrumental conditioning, and encod-
ing and storage in long-term memory. Yet today, the basic processes should include
implicit memory, spatial learning, and remembering in the world outside the laboratory. In
addition to presenting some general rules for learning, it is just as important to specify the
exceptions and limitations to those rules. When is spacing practice or giving immediate
reinforcement better, and when might massed practice or delayed reward be better?
I have tried to write a book that will appeal to a broad audience by including a range
of research examples, from education, neuropsychology, psychiatry, nursing, marketing,
and ecological (or everyday) memory. I value basic research on learning and memory. This
includes laboratory studies, often of animals, as fundamental means for determining the
principles by which learning occurs and memory persists. Yet in my own teaching, I have
found that the relevance of these basic principles needs to be made clear consistently.
Lecturers often use everyday examples to maintain student interest. I believe that a more
important purpose is to illustrate how basic principles can be translated into applications.
To be skilled and knowledgeable in the basics is not enough. For instance, experts in
animal conditioning are not necessarily good dog trainers. (Just ask my dogs). We also
need to consider the extent to which our laboratory results will generalize to other subject
populations that differ in age, learning abilities, cultural background, and even species (!).
A book on learning should also include aids to facilitate learning. Each chapter begins
with an outline and concludes with a detailed summary. Review and recapitulation para-
graphs are interspersed in the chapters. Some topics and terms are repeated across chap-
ters. Essential terms are in boldface, and these also appear in the glossary at the end of
the book. Other important terms are in italics. An instructor’s website will be available, to
include lecture outlines, additional material for use in class, suggested readings, student
activities, and links to online videos. A separate set of test questions is also available.
ACKNOWLEDGMENTS

A number of people have influenced me, either directly or indirectly, in writing this book.
I had the good fortune to have had mentors both in college and graduate school. At
Fairfield University, the psychology faculty were enthusiastic, devoted, and student ori-
ented. They determined my choice of a career as a teacher and researcher in the field of
learning. At Yale University, the freedom to pursue individual interests, and close faculty
interactions, offered an exciting environment in which to learn. Allan Wagner, in his class-
room and in his laboratory, became an enduring role model and mentor. (A note to stu-
dents: My guess is that many of your professors and instructors had similar relationships
that led to their own decisions to become teachers. Ask them about it).
I would also like to thank my extended Terry family for encouragement, patience in
living with an absent-minded professor, and leaving me alone for long periods of time. The
next generation of Terry children have renewed my wonder in the potential for learning. A
wise professor once said that if you want to learn about Learning, have children. I’m still
learning.
I would like to acknowledge the editorial and production staff at Taylor & Francis /
Routledge for their dedicated and professional work on this book. Special thanks to
Georgette Enriquez, psychology editor at Taylor & Francis, who made this new edition
possible. I appreciate her encouragement and support. Thanks also to senior editorial
assistant Lakshay Gaba. The production staff corrected my errors and updated the text’s
design. Thank you senior production editor Kris Šiošytė, and copyeditor Lorraine Savage,
proofreader Sandra Stafford, and indexer Merideth Murray.
Several university and college instructors commented on chapters and successive
drafts of the several editions of this book, providing extensive commentary and their own
insights into this field. For their efforts, I am both grateful and respectful. Any errors and
mistakes that remain are mine, and the reviewers can safely reply, “I told you so.”
Finally, I would like to thank my students, both undergraduate and graduate, from the
last too-many years, for letting me try out ideas on them. I have found that if you present
things in a good-natured manner, students will respond similarly. They have been a valua-
ble source of feedback.
 CHAPTER

1
Introduction

CONTENTS

The Origins of the Study of Learning 2 Applied Research 13

Epistemology 2 Common Sense and Common
Evolution 3 Knowledge 15
Contemporary Influences 4 Why Animals? 16
The Definition of Learning 4 Conceptual Approaches to the
Some Caveats 7 Study of Learning 17
The Learning/Performance The Functional Approach 17
Distinction 8 The Behavioral Approach 17
Learning: Recapitulation 9 The Cognitive Approach 18
The Relationship Between the Terms The Neuroscience Approach 18
Learning and Memory 9 Applications 20
The Science of Learning and Studying: Recommendations from
Memory 12 Experts 20
Basic Research 13 Summary 21

Students who take a course in the psychology of learning are usually knowledgeable
about learning by the time they have reached this point. Through instruction and on-the-
job training, they have already picked up numerous everyday, common sense principles
about how to learn. Students can readily tell their instructors that it is better to spread
studying over several days rather than to cram it all into one day (what psychologists call
the spaced versus massed practice effect), and that temporary forgetting for otherwise
well-known information occurs, especially on exam days (what we otherwise call retrieval
failure). Students are aware of what psychologists call context-dependent learning, that it
is better to study in the place in which you will take the test. These practical principles are
accurate as broad generalities, but they are also only partially true. They are half-truths.
This is a book full of half-truths.
Let me quickly explain what I mean. There are numerous facts, laws, and principles
of learning that have been uncovered over the past 120 years that psychology has for-
mally been in existence. However, these principles are more complex than the simple
statements we popularly use to describe them (e.g., spaced practice is better than

DOI: 10.4324/9781003227090-1
2 Introduction

massed practice). Statements of these principles almost always require qualifiers; they
are true under certain conditions. In this book, I will attempt to tell both halves, and thus
in the end something closer to the truth as we know it now.
Take some well-known popular generalizations. Spaced practice produces better
learning than does massed practice. Well, yes, usually. But much depends on what we
are attempting to learn and how long we will have to retain it, just two of several possible
qualifiers we might add here. Actually, one line of research on remembering people’s
names suggests that it is better to mass repetitions of a given name at first, and then
gradually lengthen the interval between successive presentations (Landauer & Bjork,
1978). (Note to readers: The standard format for noting sources in psychology is to list the
authors’ last name and the year of publication. Complete source information is provided
in the References at the end of the book). Repeat the new name immediately; repeat it
again after a little while; and keep increasing the interval to the next repetition. As a sec-
ond example, common sense seems to say that feedback is more effective when it is
given immediately and consistently after each performance of a behavior. Yet, this is not
always so. Skilled movements are sometimes learned better with delayed or only occa-
sional feedback (see Chapter 11).
Other forms of learning pose questions that have alternative correct answers. Do
subliminal audio messages, such as suggestions to induce self-control or weight loss,
work? Both yes and no answers can be defended. Some data indicate they are effective,
but more probably due to a placebo effect. Does this mean that there is no such thing as
subliminal learning? No, learning can occur at many levels of awareness or conscious-
ness. Does sleep learning occur? Instead of buying the hard-copy version of this text,
should you get the audio version to listen to throughout the night? The answer depends
on what you mean by learning. Research conducted in sleep labs indicates that factual
information is probably not being learned, but possibly some other forms of learning (such
as conditioning of the Pavlov variety) might occur.
The point of these examples is to give a sample of what real principles of learning
look like. The goal of this book is to present a scientifically accurate and sophisticated
view of the principles of learning. This includes the qualifying statements: when a given
principle holds and when exceptions occur. We can simplify a description to aid compre-
hension, but it can be simplified only so much before it becomes inaccurate.

THE ORIGINS OF THE STUDY OF LEARNING

The field of research broadly described as learning has its origins in philosophy, particu-
larly the philosophical movement of empiricism in the eighteenth century; and in science,
particularly the development of evolution theory within biology in the nineteenth century.
These movements are still active influences in contemporary psychology.

Epistemology
The nature–nurture question asks how we are affected by biology on the one hand (i.e.,
nature) and by environment on the other (i.e., nurture). If a child were raised in isolation
from others, what would that child know? Would the child instinctively develop into a kind
 The Origins of the Study of Learning 3

and just person, or one who is cruel and selfish? The the French philosopher and social
commentator Jean Rosseau (1712–1788) was wary of the degrading influences of civiliza-
tion, and believed that “noble” peoples would be discovered living beyond the reach of
so-called civilization. However, the discovery of feral children, living apart from other
humans, were intellectually and emotionally disabled (Candland, 1993).
The branch of philosophy known as epistemology studies how we come to have
knowledge. This is also the central question for the field of learning. One view is that
some knowledge is innate or instinctive. Humans were controlled by instinctive motives
such as sociability, aggression, competitiveness, or protectiveness of family. Further
knowledge was developed by reasoning as illustrated by the derivation of geometric axi-
oms and algebraic logic. In each case, knowledge is present independent of particular
experiences with the world.
By contrast, the British philosopher John Locke (1632–1704) said that the origin of
knowledge is in experience, as provided to the mind through the senses. This is the phi-
losophy of empiricism. For instance, our notion of cause and effect is learned by our fre-
quent experiences in which one event in the world is regularly followed by another event.
Such regularities lead to the conclusion that the first event is the cause of the second.
Empiricism led psychologists to investigate how we acquire knowledge through
experience. What is the origin of our association from STOP to GO or TABLE to CHAIR?
Locke said tables and chairs are contiguous: They occur together in time or space. Beyond
that, tables and chairs are frequently contiguous. Therefore, our mental representations
of these separate ideas become associated in our minds.
Early 20th century psychologists were receptive to the study of association learning.
For example, by adopting Ivan Pavlov’s method of conditioning, experimenters paired a tone
with food several times, and recorded the dog’s reactions to the tone. In Germany, Hermann
Ebbinghaus studied associations that developed from one item to another in the learning of
verbal lists. Here again we see the associative principles of contiguity and frequency.

Evolution
The field of learning was also influenced by advances in the sciences. One of the most
significant influences was Charles Darwin’s theory of evolution. In his On the Origin of
Species, published in 1859, Darwin explained how organisms could change over genera-
tions to better adapt to the environment in which they lived. Darwin first noted that there
were differences among members of a species; not all individuals were identical. Some
of these differences increased the likelihood of survival and reproduction. If these differ-
ences were inherited, then the evolution of adaptive specializations could accumulate
across generations.
Today we acknowledge that learning and memory evolved in living organisms as
adaptive specializations. Whereas evolution theory at first stressed anatomical changes
as a means of adapting to the environment, psychologists emphasized learning as a
means of adapting within the organism’s lifetime. Better learners were more likely to
survive and reproduce, and pass on their capacity to learn. In addition, the belief in a
common evolutionary history suggested that animals other than humans could be stud-
ied, with generalizations proceeding in either direction along the phylogenetic scale.
4 Introduction

Contemporary Influences
This discussion of philosophy and biology may seem to be of historical interest at best,
but each has had a continuous influence on the field of learning. In one contemporary
example, instinct, empiricism, and evolution are represented in a theory of biological
preparedness for learning. For example, language is thought to be a biologically pre-
pared form of learning in humans, something we learn quickly and readily due to our
evolutionary history. This is shown by several aspects of human language: it is universal;
it has a common developmental progression in children across cultures; it is readily
acquired even in language-poor environments; and certain areas of the brain seem ded-
icated to language (Pinker, 1994). Environment is also essential to language develop-
ment, determining the particular language we learn and the specific rules of our native
language. But the fact that we even learn a language, as complex as this is and as
intellectually immature as we are as infants, suggests the existence of a biological
predisposition.
Another example of nature–nurture interaction is the theory that evolution has pro-
duced several memory systems by which organisms learn. There are some specialized
systems, such as one for song learning in birds or face memory by primates. Other sys-
tems might serve incremental learning of habits versus the memory for individual
moments that so characterize human memory (Sherry & Schacter, 1987). The biology of
the brain determines the kinds of learning that can occur, but experience provides the
actual knowledge and memories.

THE DEFINITION OF LEARNING

Learning is the acquisition of knowledge. Knowledge is broadly defined to include verbal
knowledge, and also skills, attitudes, feelings, or behavior outside conscious awareness
(see Table 1.1). Beyond this general description of knowledge, the scientific study of learn-
ing requires a precise, operational definition of what can be observed as indicators that
learning has occurred. Thus, the study of learning is guided formally by an objective defi-
nition, as well as informally by the actual practices and interests of the researchers.
Learning may be defined as a relatively permanent change in behavior or behavioral
repertoire that occurs as a result of experience. This formal definition specifies what is
included under the rubric of learning, and just as important, what is to be excluded. This
definition has several components.
First, learning involves an observed change in behavior. The point here is that the
detection of learning requires some objective evidence. Learning and memory them-
selves are not directly observed; they occur in brain or mind. As much as we may be
interested in the inner workings of the brain or mind, we often need to observe the
organism’s behavior to see what is going on inside.
Neuroscientists are coming closer to identifying the actual neural changes that
accompany learning or remembering. And these neural changes can be correlated with
behavioral changes. Learning and memory are biological processes and psychological
processes that intervene between the environment (which we can manipulate) and
behavior (which we can measure).
 The Definition of Learning 5

Table 1.1 The Breadth of Learning

The everyday use of the term learning does not describe all of the diverse phenomena that are
studied in the Learning sciences. Here are some of what is included by the term Learning.

1. Learning is a widespread phenomenon. It includes both animals and humans. It is applicable to

the behavior of intact organisms, to isolated sets of neurons, and to synaptic subsystems of
the brain; and to models of the brain such as computer simulations, neural networks, and
algorithms.
2. Learning involves events as diverse and simple as conditioning of an isolated muscle twitch, to
more complex phenomena such as acquiring a prejudice, a symbolic concept, or a neurotic
symptom.
3. Learning includes the external responses of the organism; internal bodily and physiological
responses; and mental reactions such as thoughts and feelings.
4. Learning is concerned with the original acquisition of a response or knowledge, with its later
loss (forgetting or extinction), its retention over time (memory), and its possible value in the
acquisition of new responses (transfer and generalization).
5. Learning intersects with other fields of psychology such as motivation, perception, develop-
ment, personality, and social and cultural factors.
6. Learning more often deals with the behavior of the average of a group of individuals, but
sometimes with differences among individuals or among groups of individuals.
7. The study of learning is associated with a long philosophical, academic, and scholarly tradition;
and has long served as a source of practical application and technology.
8. The learning process is continuous with the more general linguistic, cognitive, information-
processing, and decision-making activities of the organism.

What kinds of behaviors can we use to measure learning? Learning outcomes are
multidimensional. Consider an experience you may have had as a child: a sibling jumping
out of a darkened room or closet to scare you. The fear learned from such an episode
could be expressed verbally in your recollections of the event years later; physiologically
by increased heart rate in fearful anticipation of a repeat of the episode; and behaviorally
by the avoidance of entering dark hallways or rooms in the house.
The measurement of learning is nicely illustrated by an example from personnel psy-
chology. For instance, say a training workshop has been conducted in an employment
setting. How do we know what the workshop participants learned? Kraiger, Ford, and
Salas (1993) suggested three types of assessments might be appropriate. One outcome
of training is the factual knowledge that the participants can recall. Another outcome is
skill learning, represented by some behavior that the participants can now do more quickly
or accurately. A final outcome would be affective (or attitude) changes. Do the employees
now feel more competent, confident, or committed after training?
Next in the definition, learning involves changes in behavioral repertoire, or the stock
of behaviors that might be performed. Not all learning is immediately evidenced by overt
behaviors. What you have just learned from this text is probably not affecting your behavior
now. Thus, the definition of learning includes the potential for seeing a change in behavior
to be demonstrated when testing conditions prompt the display of this new knowledge.
The distinction between potential and actual changes in behavior is demonstrated in
a classic study of socially learned aggression. Albert Bandura showed that children imitate
6 Introduction

aggressive behaviors they see adult models perform (e.g., Bandura, 1965). Children
watched a videotape in which the models (grad students) punched an inflated clown bal-
loon, or BoBo doll, by kicking it or throwing it. The children were later allowed to play with
the BoBo doll. In one condition of the experiment, the model in the tape had been praised
for playing aggressively. The children who saw this video later imitated many of the spe-
cific aggressive behaviors. In another condition, the model was scolded for misbehaving,
and the children who saw this version performed fewer aggressive responses (see
Figure 1.1). So far, we have a difference in the observed behavior between the two exper-
imental conditions: Children imitated the praised model more and the scolded model
less. Then the experimenter offered a reward for each aggressive response the child could
reproduce. For the children who had seen the scolded model, the incentives increased
imitation of the aggressive behaviors. For these children, the aggressive behaviors were
part of the behavioral repertoire, even though they were not immediately displayed.
The Bandura study also showed that a gender difference in aggressive behavior dis-
appeared when incentives were offered to demonstrate what the model had done. The
girls remembered the aggressive behaviors they had observed, but they inhibited imitat-
ing these responses until it was acceptable to do so.
Learning occurs as a result of experience. This book describes what some of these
learning-producing experiences are. They may be as varied as a conditioning experiment in
the laboratory, a lecture heard, a skill practiced, or an attitude developed due to some now
unrecalled event. The definition of learning excludes changes in behavior that are not due
to experience. As we will see in what follows the line between learning and other sources
of change is often blurred. The point of the experience phrase in the definition is to ask us
to consider what is the source of a behavior change: What counts as experience?

Girls Boys
4

3
No. of Responses

0
Model Model Incentive
Praised Scolded Offered

Figure 1.1
Imitation of Aggressive Behavior. Mean number of different aggressive responses imitated by children during
the first phase of testing as a function of the consequences to the model they had observed; and the number
of responses imitated when an incentive was offered to perform.
Source: Bandura (1965).
 The Definition of Learning 7

Finally, learning is said to be relatively permanent. This may seem contrary to every-
day experiences in which we all too frequently forget facts, names, appointments, and so
on. But we may remember more than we realize. Forty years after high school graduation,
alumni still could identify pictures and names of classmates, although at first, only 20
percent of their graduating class could be named (Bahrick, Bahrick, & Wittlinger, 1975).
Thus, much more was learned than was apparent on tests of the ability to recall names.
The “relatively permanent” phrase is intended to exclude transient changes in behav-
ior, changes that are not due to learning. Responding could temporarily fluctuate due to
increases or decreases in, as examples, arousal, fatigue, or motivation. Rats run faster in
a maze if they are hungrier and slower if they are satiated. This does not mean they sud-
denly know more when hungry and know less when full. We have to separate the passing
effects of variables such as arousal and motivation from their permanent effects on learn-
ing (Kimble, 1967).

Some Caveats
The study of learning includes phenomena that do not fit precisely within the formal defi-
nition. There are gray areas. One is the distinction between biology and environment.
The attribution of behavioral changes to either environment or biology is a false
dichotomy. There is an inseparable interplay between the two, and the line between
learning and other experientially based changes is not always clear. For instance, we say
learning is based on experience. Yet experience affects the development of the brain, too.
Exposing immature rat pups to an enriched environment, one with toys and other rats,
enhances development of areas of the brain important for learning. Both the number of
nerve cells and the number of connections among cells are increased (Rampon et al.,
2000). The enriched environment rats will be better at learning certain tasks than rats
reared in more standard environments (Kemperman, Kuhn, & Gage, 1997). Here rearing
experience causes permanent changes that increase learning capacity, but not what is
actually learned.
One non-learning source of behavioral change is maturation. Some behavioral changes
occur because of the physical, neural, or cognitive maturation that takes place over time.
For example, when sparrows reach a certain age and at a certain time of year, they begin
to sing. In some species, singing does not depend on having heard other birds sing.
Singing, and even the particular song, is innate. When we casually talk about children
learning to walk, we are wrong in thinking that these skills are dependent only on learn-
ing. In human infants walking is dependent on maturation. Physical maturation in the
muscles and the bones and cognitive maturation of coordination allow walking to occur.
At one extreme, there are some human behaviors that are substantially influenced by
maturation. Infant development of sitting upright, standing, and eventually walking are
primary examples. Gesell and Thompson (1929) conducted a classic experiment in which
one infant twin of a pair received several weeks of practice at stair climbing. The other
twin, denied this explicit practice, later took only a week to equal the proficiency the
practiced sibling had achieved in four weeks. Similarly, Lenneberg (1967) describes a
pre-linguistic child who was blocked from practicing language sounds for several months
by a tracheal tube. When the tube was removed the child immediately showed
8 Introduction

age-appropriate language development. In Gesell’s case, early practice gave little benefit,
and in Lenneberg’s case, the absence of practice produced little decrement.
Other behaviors clearly illustrate the interaction of experience and maturation.
Marler’s (1970) study of white-crowned sparrows is especially instructive here. The male’s
song during the breeding season shows variability across geographic regions. Marler
raised some birds in isolation from others of their type. When singing began several
months after hatching, the birds sang a song that was, in outline, the appropriate song for
the white-crowned sparrows. However in detail, the song was significantly different or
abnormal. Exposing the birds to a song of their own type during the period from 10 to 50
days after hatching leads to normal song development. Thus, song is determined by the
interaction of maturation (an innate predisposition) and learning (experience with specific
songs).

THE LEARNING/PERFORMANCE DISTINCTION

Earlier, we noted that learning itself is not directly observed. This process occurs in the
mind or the brain, and we infer that learning has occurred based on the behavior of the
organism. Performance refers to the measures of behavior used to indicate whether learn-
ing occurred, knowledge was acquired, and memory is present. However, these behavioral
measures are sometimes imperfect and indirect. There is not always a one-to-one corre-
spondence between what the organism knows and what the organism does or says.
Performance may be adversely affected by a number of factors, including the sub-
jects’ attention, effort, interest, motivation, or uncertainty about the nature of the task.
Sometimes no behavioral change is observed even though (we realize later) learning
has occurred. The classic example of this is Tolman and Honzik’s (1930) study of latent
learning. Rats which were fed in the goal box learned to run directly to the goal box. (No
big deal. Kids’ stuff for rats). Other rats were placed in a maze but were not given food
when they reached the goal box. Not surprisingly, these rats persisted in wandering
throughout the maze, entering dead-ends day after day. Then one day food was suddenly
present in the goal box, and there was an immediate improvement in performance. The
animals now made few wrong turns on their way to the goal box. In those previous trials
without food reward the rats had indeed learned the layout of the maze, but this knowl-
edge remained hidden until the subjects were motivated to complete the maze quickly.
Latent learning is knowledge that is not displayed in performance. Similarly, your knowl-
edge of this chapter may remain latent until an exam is given. In our learning studies, the
absence of performance has been aptly referred to as the “problem of behavioral silence”
(Dickinson, 1980). If there is no change in behavior, we really do not know whether learn-
ing has not occurred or learning has occurred and is latent.
There are parallel distinctions for knowledge versus performance, and memory ver-
sus performance. Our performance assessments may not indicate the presence of
knowledge or memory that is actually there.
Test anxiety is one too-familiar illustration of the learning–performance distinction.
Students who truly know the material can perform poorly on the exam because of exces-
sive anxiety. Their test performance does not accurately assess their underlying learning.
(To cite one extreme case, Capretta and Berkun [1962] noted that soldiers crossing an
 The Relationship between the Terms Learning and Memory 9

unstable rope bridge over a deep ravine performed worse on a memory task than when
tested under nonstress conditions). The phenomenon of stereotype threat similarly
shows that performance does not always match underlying ability. There are negative
stereotypes about the abilities of certain groups, for example that the elderly are forget-
ful. Reminding someone who is a member of the stereotyped group can negatively affect
their performance. Thus, instructions to a senior citizen that state “we are going to test
your memory” and “we are interested in how good your memory is” can prime aging-for-
getfulness worries. If neutral instructions are given, age differences are smaller or absent
altogether (Rahhal, Hasher, & Colcombe, 2001).

Learning: Recapitulation
Let’s review the key ideas of the previous sections. Research on learning is guided by a
formal definition that makes our study more objective: Learning is a relatively permanent
change in behavior, or behavioral repertoire, that is due to experience. This definition
excludes changes in behavior that are transient and are thus likely to reflect fluctuations
in attention, motivation, or arousal level. The study of learning intersects with studies of
innate or maturationally determined behaviors. Although our formal definition emphasizes
changed behavior as an indicant of learning, we also acknowledge that behavioral perfor-
mance can be a misleading indicator of what has been learned.

THE RELATIONSHIP BETWEEN THE TERMS LEARNING AND

MEMORY
The words learning and memory in everyday language have related but distinct uses. The
same holds for the technical meanings within psychology. There are both formal and infor-
mal distinctions that psychologists make. In the past, learning referred to conditioning and
reinforcement, to (nonhuman) animal subjects, or to skills requiring repeated practice.
Memory was used in reference to verbal recall in studies of human subjects.
A more formal distinction is to say that learning refers to acquiring knowledge or
behavior, whereas memory refers to retaining that knowledge or behavior. As a researcher
or student, one could primarily be interested in the acquisition, or encoding, of new infor-
mation: learning associations among stimuli, learning skills, or learning facts. After these
things have been learned, one could be interested in the storage and retention of the
associations, skills, or facts. Essentially, we make a distinction between two phases:
getting information into memory and then maintaining the information.
For instance, in studying learning, we would consider those factors that affect acqui-
sition, such as the amount of reinforcement, the number of study or practice trials, or the
spacing of the trials. The development or progression of learning could be illustrated by a
learning curve (see Figure 1.2). The learning curve is a graphic plot of a measure of
behavior on the Y, or vertical, axis (e.g., number or size of the correct responses) as a
function of the number of practice trials shown on the X, or horizontal, axis (see Box 1.1).
On the other hand, if the primary interest is in memory, we would consider factors that
affect the retention of already learned material, such as the length of the retention interval
or the presence of distracting activities during that interval. We could graph the course of
10 Introduction

BOX 1.1 THE LEARNING CURVE

The phrase learning curve has entered everyday language, often used as a metaphor in com-
paring individuals. One person is said to be further along the learning curve than another, for
instance. An advertisement for computer software claims it will put you farther ahead of your
competitors on the learning curve. What exactly is the learning curve?
The phrase refers to a particular shape of the curve that develops over training trials, par-
ticularly as described by Clark Hull, a prominent Yale learning theorist of the 1940s (Hull,
1943). He said the basic learning curve is a negatively accelerated curve. This means that
learning (or rather performance, which is what is actually measured) starts off with a period
of very rapid growth, in which each trial produces large increments in performance. These
increments get smaller and smaller on later trials, which is what negative acceleration means.
There is a point of diminishing returns, such that continued practice has smaller benefits.
Figure 1.2 shows the hypothetical increments across successive trials as Hull depicted them.
The vertical axis represented hypothetic units of Habit Strength, a measure of learning. The
horizontal axis represents successive practice trials. As can be see, the increase in learning
from trial 1 to trial 2 goes from 10 to 20, an increase of ten units. The increase between trials
6 and 7 is five units; the increase between trials 14 and 15 is only about three units.
We could make an analogy to learning to play tennis. At first, the improvements with each
lesson may be fairly large. With yet more practice, improvements seem smaller. Performance

100

80
Units of Habit Strength (SHR)

60

40

20

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Succesive Reinforcements (N)

Figure 1.2
Hull’s Theoretical Learning Curve. Notice that the increases in height in the curve are smaller and
smaller across trials. This is a negatively accelerated learning curve.
Source: From Principles of Behavior (pp. 108, 116, 117), by C. L. Hull, 1943, New York: Appleton-Century-Crofts. Adapted
with permission.
 The Relationship between the Terms Learning and Memory 11

may eventually reach an asymptote, or plateau, after which little or no further improvement is
seen.
Learning curves are not always negatively accelerated. Sometimes performance improves
very slowly at first, and then the phase of accelerated increments kicks in. This produces an
S-shaped curve: small increases at the start of training, large increases in the middle trials,
and a return to slow growth at the end. To say that you are farther along the learning curve
means that you are in the phase of rapid acceleration or have passed through it, whereas
someone else is still stuck in that early phase of slow growth.
This prototypical learning curve has been documented in many situations and is incorpo-
rated in contemporary theories of learning (e.g., the Rescorla-Wagner theory; see Chapter 3).
Although the curve may be an accurate description of what we observe, its interpretation can
be challenged. By averaging over subjects whose individual performances vary, we may pro-
duce a curve that does not represent any one subject (Gallistel, Fairhurst, & Balsam, 2004).
Other theorists suggest that the rate of growth is better described by a power curve (e.g.,
Newell & Rosenbloom, 1981). This means that the trials units along the horizontal axis are
expressed in log numbers. That is, the first point is trial 1, the next point is trial 10, then trial
100, etc. This compresses the larger numbers and produces a straight line rather than a
curve.
The opposite of the learning curve is the forgetting curve. Here we would plot the amount
remembered at different intervals of time after learning has been completed. Over time there
is a decline in what can be recalled, the opposite of the learning curve. Forgetting curves can
also take different forms, depending on whether memory loss is rapid or gradual. (A forget-
ting curve is shown in Figure 6.1, Chapter 6).

memory by a forgetting curve. This would be a plot of the measure of behavior (again,
the number or size of correct responses) as a function of time since learning was
completed.
The learning–memory distinction can be illustrated by considering the role of taking a
test on learning and memory. We often measure the progress of learning during study or
shortly thereafter. Certain variables lead to faster learning. But do the same conditions
that enhance learning also produce better retention days or weeks later? The answer is
not always.
For example, say that you have just studied in preparation for a test. Would it be better
to study the material one more time, or take a practice test? A first guess might be that
additional study would be a better strategy. However, research on the “testing effect”
(see Chapter 9) shows that a practice test can be more beneficial than additional study
time. Roediger and Karpicke (2006) had college students study new information in the
form of textbook-like paragraphs. One group of subjects (or participants) studied the par-
agraphs twice. A second group read the paragraphs once and then took a practice test on
what they had just read. Everyone took a test, the real test, five minutes later. This is how
learning is often measured, shortly after the study trial. The results of this test are shown
in the left set of bars in Figure 1.3. The height of the bars indicates the number of facts
recalled. The students who studied the material twice remembered slightly more facts
(81 percent) than did students who had read the material once and took a practice test
(75 percent). Learning seems to be better with additional study.
12 Introduction

0.8 Study-Study

Study-Test

Proportion Recalled
0.6

0.4

0.2

0
5 Minutes 2 Days 1 Week
Retention Interval

Figure 1.3
Effects of Additional Study or a Practice Test on Learning. Subjects studied twice or studied once and took
a practice test. Learning was tested 5 minutes after, two days later, or one week later.
Source: From “Test-enhanced Learning,” by H. L. Roediger III and J. D. Karpicke, 2006, Psychological Science, 17, p. 250.
Copyright American Psychological Association.

Which condition produces better memory? Other students were brought back two
days or one week later to take the final test. The middle and right sets of bars in Figure 1.3
show that, at these times, the study-and-practice test groups remembered more facts
than did the study-twice groups. For instance, the students who took the practice test
recalled more a week later (56 percent correct) than did those who had studied twice (42
percent). That is, although studying twice may have produced better recall in the short
term, replacing additional study with a practice test produced more correct answers in
the long-term.
In this example, we might say that learning is being studied during the initial session,
whereas memory is studied in the second session. You might argue (validly) that the
learning phase also tested memory by requiring recall even after 5 minutes. Indeed, some
psychologists assert that any test of learning also involves a test of memory (e.g., Spear
& Riccio, 1994). Schmidt and Bjork (1992) said that “Rather than viewing learning and
post training retention as separable… we argue that the effectiveness of learning is
revealed by… the level of retention shown” (p. 209). Their preference is to emphasize the
“relatively permanent” clause in the definition of learning and focus on longer-term reten-
tion as the best measures of learning.

THE SCIENCE OF LEARNING AND MEMORY

Empiricism says that knowledge comes from experience; from observations of the real
world. However, simple observation is subject to bias, misperception, and the influence
of unknown extraneous variables. Science offers a method of discovering valid findings
 The Science of Learning and Memory 13

and principles. A scientific approach to knowledge states that observations need to be

made in a systematic and controlled fashion. This means that sometimes research is
conducted in controlled situations, oftentimes a laboratory, to minimize the influence of
contaminating factors. Specific types of research subjects might be selected, such as
children, older adults, rats, or mice. The measures of learning or memory are chosen to
be reliable, valid, and documented. These are all part of the meaning of systematic and
controlled.
Science is characterized as quantitative. We can quantify the magnitude of the varia-
bles we manipulate, such as the loudness of a tone or the amount of food used as a
reward. We can also quantify behavior, learning, or memory. How many words were cor-
rectly recalled? How many drops of salivation did Pavlov’s dog make during hearing the
tone?
Science strives for objectivity. Science tries to be public, replicable, objective, and
transparent. We try not to bias the outcome of a study. We set the experiment in motion
and see where the results fall. We publicly report the details on how the study was con-
ducted, and the results of the study. Anyone else has enough information to reproduce a
given study to verify, qualify, or extend the results.
One might expect a textbook on learning to include detailed information on how to
learn the material in the book. Instead, learning is illustrated by preparations such as
eyeblink conditioning or word-list memorization. This discrepancy reflects the different
purposes of research. Research on learning is sometimes categorized as being either
basic or applied.

Basic Research
Basic research is an interest in understanding the fundamental processes of learning and
memory. It seeks to demonstrate cause-and-effect relationships between key variables.
To demonstrate the causality, we must control or eliminate other contaminating variables.
This is often accomplished using a simple task, a simpler organism, and a highly control-
lable setting. For instance, if you want to discover whether synaptic changes occur in
learning, you might begin your study on organisms that have few and large neurons, such
as worms or snails.
The questions asked in basic research do not always have obvious and immediate
applications outside of the lab. Basic science researchers believe in the potential useful-
ness of their research, even if the applications are not known until the research is com-
pleted. As one physicist put it, basic research sometimes provides a solution in search of
a problem (Lemonick, 1995). For example, lab research on the learning of aversions to
new tastes by rats has been applied to controlling sheep poaching by coyotes (Gustavson
& Garcia, 1974), and to blocking the development of food aversions in people undergoing
chemotherapy (Bernstein, 1991). Neither application was anticipated by those doing the
initial laboratory research.

Applied Research
Applied research is relevant to, or will apply to, solving a specific practical problem. The
distinction between basic and applied research might be better thought of as a continuum
rather than a dichotomy. (As with many of the terms encountered so far in this chapter,
14 Introduction

we first set out a dichotomy and then suggest the truth lies somewhere in between). Any
given piece of research falls somewhere along the basic—applied continuum depending
on the relevance of the study to a specific target population, task, and/or setting to which
we wish to apply our results. Some examples may illustrate.
What is the effect of caffeine on memory? Caffeine is a known stimulant, often used
by students to boost alertness, and logically should facilitate learning. Research on rats
has shown varied results: Caffeine sometimes facilitates maze learning, but sometimes
inhibits it (Lashley, 1917; Terry & Anthony, 1980). This is basic research. In other experi-
ments, college students are asked to remember a list of words. These subjects and this
task have greater relevance if we are trying to generalize to humans. Yet this study still
retains aspects of basic research, in that the caffeine is given under controlled conditions
in the lab, using blind-run and placebo conditions, and so on. The result of one such exper-
iment was that caffeine impaired the immediate retention of the lists (Erikson et al., 1985).
One final study to consider is possibly the most relevant to student learning: What is
the relationship of caffeine intake to grade point average (GPA)? Now we are getting to
the important question. Gilliland and Andress (1981) surveyed University of Oklahoma
students to determine the amount of caffeine consumed and correlated consumption
with the students’ GPAs. The results showed a negative correlation: Higher caffeine con-
sumption was associated with lower overall grades, and, conversely, lower caffeine went
along with higher grades. This example would seem to present the most applicable and
the most ecologically valid of the caffeine findings presented so far. Yet one can imagine
reasons other than caffeine for these results. Do procrastinating students drink lots of
coffee while cramming for exams and papers? Does self-reported caffeine consumption
accurately reflect actual consumption? By leaving the lab for the actual world, we lose
control over certain variables in our attempt to simulate naturalistic conditions. There can
be a trade-off between experimental rigor and ecological validity in research, as occurs in
the study of everyday remembering (see Box 1.2).

BOX 1.2 STUDYING EVERYDAY MEMORY

In 1978, Ulric Neisser, recipient of an American Psychological Association award for distin-
guished scientific contributions, criticized experimental psychology for its failure to study
memory and cognition as they are used in our lives. “If X is an interesting or socially signifi-
cant aspect of memory, then psychologists have hardly ever studied X” (Neisser, 1978, p. 4).
Neisser was criticizing psychologists for conducting too much laboratory research and not
enough research on how memory works outside the lab. Neisser’s remarks were soon fol-
lowed by a proliferation of research on memory in the everyday world.
Ten years later, Banaji and Crowder (1989) published their reply titled “The Bankruptcy of
Everyday Memory,” words equally provocative to Neisser’s remarks. Banaji and Crowder
argued that studying memory in natural settings does not automatically ensure that general
principles of memory will be found. These authors pointed to an analogy in chemistry. No one
criticizes chemists for doing controlled lab studies instead of studying everyday compounds
found in the kitchen or bathroom. Banaji and Crowder note that because there are so many
uncontrolled variables in the everyday world, researchers must resort to the laboratory to
produce valid findings about cause and effect.
 The Science of Learning and Memory 15

For example, consider the scenario in which eyewitnesses to a traffic accident are ques-
tioned. Who can say which witness is more accurate, or why, when so many variables are
uncontrolled? Witnesses observed from different perspectives and saw different things;
their first descriptions could contaminate those repeated later; and the delay until witnesses
are questioned varies. How can valid results be obtained under such poor experimental
conditions?
Somewhere between the two extremes of Neisser and of Banaji and Crowder probably
lies the truth. Several commentators pointed out that the research setting, laboratory versus
field, does not determine the scientific validity of the results; that real-world research can
produce generalizable principles; and the study of memory in everyday life can test theories
derived in the laboratory.
The study of everyday memory has since flourished. There has been an increase in scien-
tific rigor. In addition to methods of naturalistic observation and self-reports, controlled exper-
iments are used to study ecological memory. There are numerous new journals that publish
this research, such as Applied Memory Research, Journal of Applied Cognitive Psychology,
Journal of Applied Research in Memory and Cognition, and the Journal of Experimental
Psychology: Applied.
Importantly, the study of everyday memory suggests practical remedies for real problems,
such as determining the veracity of eyewitness testimony, the validity of recovered memo-
ries, or the remediation of memory loss produced by injury, illness, or aging.

Common Sense and Common Knowledge

As noted at the start of this chapter, many readers already possess sophisticated knowl-
edge about how to learn. After all, students are professional learners. Is much of what
psychologists teach about learning already common knowledge?
We present people with a scenario, in which a child has been feeding pigeons at the
windowsill for several weeks. What would happen if the feedings stopped? Most people
thought that the pigeons would stop coming to the window (what we learning psycholo-
gists refer to as experimental extinction). When tested on 20 similar learning scenarios,
people (both students and random people in a park) answered about 75 percent correctly
(Houston, 1983). Does this show that we are teaching the obvious? What everyday knowl-
edge does not account for are the qualifications on these everyday findings, which might
predict reversals or paradoxical results. For instance, there are situations in which with-
holding rewards can lead to more persistent responding, rather than a decrease.
People in general also have incorrect beliefs about memory, sometimes referred to as
memory myths (Klatzky, 1984). These include distorted beliefs about amnesia, hypnosis,
aging, or photographic memories. For example, a common conception of amnesia is that it
involves such extensive forgetting even to the extent of losing your personal identity. Nearly
48 percent of surveyed Americans agreed that amnesiacs forget who they are. Yet none of
the memory experts surveyed agreed with this statement (Simons & Chabris, 2011).
Can memory be improved by training? If physical exercise makes the body stronger,
then mental exercise should make memory stronger. A Scandinavian survey found that
over 90 percent of the respondents believed that memory exercises would improve over-
all memory ability (Magnussen et al., 2006). However, research shows only that people
16 Introduction

can improve their ability to remember specific types of information through practice. Thus,
you can improve your memory for names, or random sequences of the digits 0 and 1, if
this is what you practice remembering. Someone who can memorize pi to one hundred
decimal places does not remember names and faces any better than the rest of us.
Even students have incorrect perceptions about how memory works, the best strat-
egies for learning, and how well they individually remember. Generations of students
have used the study strategies of highlighting, underlining, and rereading (Miyatsu,
Nguyen, & McDaniel, 2018). However, these are fairly passive activities and the repetition
adds little to long-term learning. Students also overestimate how well they have learned
before a test; and overestimate how well they did after the test.

Why Animals?
Humans are of course animals. A distinction is sometimes made between research on
humans and that with non-human animals. In this book, the word humans is used as a
shorthand term. The word animals usually refers to all the nonhuman animals.
The reasons for using nonhuman animals in experiments on learning can be simply
stated. First, the experiences of animal subjects can be highly controlled, obviously within
the experiment itself but also prior to the experiment in terms of the genetic and life
history of the organism. Second, given our shared evolutionary history, animals and
humans have similar nervous systems and therefore an assumed generality in the basic
principles of learning. Granted, there are exceptions to those generalities.
A third, and controversial, reason for using animals is that procedures are used on
animals that cannot ethically be applied to humans. This justification is controversial
because some would question why animals are not given similar protection from painful
or dangerous procedures. Any proposed study using animals must go through a review
process similar to that for human research. The rationale for exposing animals to pain or
distress needs to be justified, and measures taken to minimize. The procedures should
impose as little as possible on the well-being of the subjects. Still, there is not the free-
dom that humans have to refuse to participate.
How prevalent is dissatisfaction with animal use in psychological research? A repre-
sentative survey of 1,200 psychology majors found fairly strong support for the continued
use of animals (Plous, 1996). About 70 percent supported the use of animals in psycho-
logical research and believed such research was necessary. A greater number (85 per-
cent) believed that before a proposed study is approved, the investigators should be
required to assess the degree of pain the animals will experience. In fact, such research
does require prior review by an ethics committee. Incidentally, the survey found that
fewer faculty in psychology departments are using animals than in previous years, and
fewer psychology students take lab courses using animals.
Neal Miller (1985), in accepting the Distinguished Professional Contribution Award
from the American Psychological Association, listed some contributions of behavioral
research on animals. These include the development of behavioral therapies for psycho-
logical disorders; applications to behavioral medicine, such as in the control of cardiovas-
cular and asthmatic responses; research on the effects of early experience on neural
development; the psychoactive effects of drugs; and benefits to animals themselves,
both for those under our care and for wildlife.
 Conceptual Approaches to the Study of Learning 17

The contribution of animal research in psychology is not always acknowledged. Some

introductory psychology textbooks reference certain findings to later studies of human
participants, when in fact the phenomena first emerged from animal laboratories (Domjan
& Purdy, 1995).

CONCEPTUAL APPROACHES TO THE STUDY OF LEARNING

When rats (who, along with college students, are psychologists’ favorite research sub-
jects) learn a maze, what exactly do they learn? Does the rat’s natural history determine
that spatial learning will lead to the greatest likelihood of survival? Is it a list of specific
turns, like a memorized set of directions? Do they acquire a sort of cognitive map of the
layout of the maze? Or should we describe the neural changes that underlie the learning?
These questions illustrate four broad approaches to studying learning and memory. The
functional approach emphasizes the necessity of learning for adaptation to changing envi-
ronments. The behavioral approach focuses on learning specific behaviors. The cognitive
approach emphasizes the acquisition of knowledge and expectancies. The neuroscience
approach studies the changes that learning produces in the brain.
These approaches, along with others, play a major role in our understanding of learn-
ing and memory. The several approaches are not mutually exclusive, and contemporary
research commonly derives from a combination of perspectives.

The Functional Approach

Animals (people included) are adapted to their environments, e.g., the freezing tundra
versus the steaming desert. In addition to the obvious physical adaptations, the capacities
to learn and remember are other adaptations. Learning evolved as a way for organisms to
adapt more quickly to the changes and inconstancies in their environment. The func-
tional approach studies how learning and remembering aid survival.
One focus of the functional approach is the evolution of learning across species.
Animals with a common evolutionary history would likely share certain kinds of learning
or memory abilities. For instance, all animal species have the capacity to acquire associ-
ations: to link one stimulus to another or link behavior and its consequences. A second
focus is on the unique adaptations that differentiate species. Thus, face recognition,
language learning, and autobiographical memory may be specialized adaptations for
humans.

The Behavioral Approach

The behavioral approach emphasizes the relationship among, first, observable behav-
iors; second, the antecedent stimuli that precede behavior; and, third, the consequences
that follow behavior. What are the environmental stimuli and conditions that come to
evoke behavior? What are the consequences or outcomes that affect the likelihood of
repeating the behavior? And what behaviors themselves are learned? The goal of behav-
ioral psychology is to predict and control behavior on the basis of knowledge of the ante-
cedents, the behavior, and its consequences.
One version of this approach, known historically as radical behaviorism, shuns theo-
rizing about unobservable (and therefore speculative) processes within the organism’s
18 Introduction

mind. Instead, behaviorism attempts to describe the lawful relationships among stimuli,
responses, and consequences. An example of a behavioral law is “The likelihood that a
behavior will recur increases if the behavior has been followed by a reinforcing stimulus
in the past.” So, if aggressive behavior is rewarded, the likelihood of aggression in the
future will increase. If these empirical relationships correctly and accurately describe
behavior, there is no need to hypothesize unobserved thought processes to explain the
behavior (i.e., the child has a bullying personality, or aggression is cathartic).

The Cognitive Approach

The cognitive approach derives from information-processing approaches to the mind.
Information, or knowledge, is encoded, transformed, stored, and retrieved. The influence
from computer science is obvious: These are analogous to processes within a computer.
The basic tenet of the cognitive approach is the postulation of an internal representation.
That is, the organism is said to form an internal representation that is used as the basis
for guiding behavior (Pearce, 1997). As examples, the representation could be the mem-
ory of a stimulus, an association, our dictionary of words and their connections, or the
spatial layout of a particular environment. (Cognitive researchers will often talk about this
representation being mental or in the mind, but this does not necessarily refer to a mental
mind apart from the physical representation in the brain). This internal representation is
inferred on the basis of behavior.
Cognitive psychology is not exclusive to humans. So, in (the other) animals we can
study their short-term and long-term memory, cognitive maps, the organization of mem-
ory, etc.

The Neuroscience Approach

The neuroscience approach seeks to determine the underlying biological basis for learn-
ing and memory. What changes occur in the nervous system during learning? Where are
memories located? What is the chemistry and physiology that allows the brain to encode,
store, and retrieve memory?
Neuroscience often combines with other approaches. As examples, the psychologist
Karl Lashley attempted to determine which areas of the rat’s brain were necessary for
learning and remembering mazes (e.g., Lashley, 1929). He did this by systematically remov-
ing various regions of the brain before or after learning. The neurosurgeon Wilder Penfield
studied memory localization by stimulating the brain of his human patients with weak elec-
tric current (Penfield & Rasmussen, 1950). The patients, who were conscious during this
portion of the operation, reported sights and sounds that felt like memories. These classic
research programs illustrate the strategy of combining approaches: behavioral (maze learn-
ing), cognitive (memory recall), and biological (brain lesions and brain stimulation).
Contemporary neuroscience has additional methods such as brain scanning and case
studies of brain-injured individuals. For example, positron emission tomography (PET) and
functional magnetic resonance imagery (fMRI) scans measure the relative levels of activ-
ity in the brain. You have probably seen photographs of scans in which the brain is color-
coded to show which areas are most active. An experimental subject performs various
memory activities while the brain is scanned. For instance, someone could be asked to
remember a list of words, such as DOG, TABLE, GLASS, and so on. This is a memory
 Conceptual Approaches to the Study of Learning 19

Figure 1.4
Peak Activation Areas in the Left and Right Hemispheres During Two Memory Tasks. The left prefrontal
cortex shows more activation during memory encoding, whereas the right prefrontal cortex shows more
activation during memory retrieval.
Source: From “PET Studies of Encoding and Retrieval: The HERA Model,” by L. Nyberg, R. Cabeza, and E. Tulving, 1996,
Psychonomic Bulletin & Review, 3, p. 143. Copyright © 1996 by the Psychonomic Society. Reprinted with permission.

encoding task: It involves putting a list of words into memory. Later the subjects are
asked to recall that list. This test is a memory retrieval task. Comparisons of the scans
show that some areas are more active during encoding of the list, and other areas are
more active during retrieval. Figure 1.4 shows drawings derived from scans of the left and
right halves of the brain, with markers showing points that were particularly active during
encoding versus during retrieval (Nyberg, Cabeza, & Tulving, 1996). As can be seen, when
learning the list, many more points in the left-front part of the brain are active; when
recalling the list, there are more points active in the right-front part.
Newly discovered methods allow researchers to tag, or mark, groups of neurons
when they are active. A rat or mouse is given a learning experience, such as pairings of a
tone stimulus followed by an electric shock. Cells that are active during learning are chem-
ically tagged. Later, when the animals are re-exposed to the tone, cells that become active
are also chemically marked. The two populations of cells, those that were active during
learning and those active at testing, overlap. This indicates that some of the same cells are
involved in encoding and retrieval. Using still another chemical, the activation of cells can
be blocked. Now the tone does not elicit fear. Finally, chemical activation of the learned
cells (not using the tone) triggered the fear response. The learned response was seen in
the absence of the tone stimulus. These studies suggest that same system of cells is
involved in the encoding and retention of a learned fear (Josselyn & Tonegawa, 2020).
Historically, there has been tension among the approaches, usually over which one
was better. In the example that began this section, we posed the question of whether
maze learning should be described as a series of turns (the behavioral approach) or acquir-
ing a mental map (the cognitive approach). Neuroscience research offers a reconciliation
between the two by showing that there are both habit learning and cognitive learning
systems in the brain (Petri & Mishkin, 1994).
20 Introduction

APPLICATIONS
Educators are keenly interested in applications of neuroscience to instruction. Research
on the Mozart effect illustrates one such extrapolation. In a well-publicized study, college
students who listened to 15 minutes of music by Wolfgang Amadeus Mozart showed an
increase of several points in their spatial intelligence test scores (Rauscher, Shaw, & Ky,
1993). (Note: Finish reading this paragraph. Do not stop to download Mozart in prepara-
tion for tomorrow’s math exam). The publicity surrounding this finding led some to advo-
cate Mozart’s music as an IQ booster.
All of us would like to think that something like listening to Mozart (or playing chess or
learning a new language) somehow reprograms our brains and make us more intelligent.
Unfortunately, other studies did not replicate the Mozart effect or instead found less inter-
esting explanations (e.g., Steele, Bass, & Crook, 1999). Was the Mozart selection simply
more exciting and stimulating, or was the control music too relaxing and calming?
Educational implications are often drawn from neuroscience research. Ideas such as
educating both halves of the brain, or identifying a critical period in child development
during which the brain is ready to learn, have also been popular. However, we need to be
cautious in generalizing findings from one discipline to another. Brain research may be a
bridge to memory research, and in-turn memory research may be a bridge to educational
innovations. But the link from brain science to educational application may be “a bridge
too far” (Bruer, 1997, p. 4).

Studying: Recommendations from Experts

Most psychologists who study learning and memory also teach learning and memory.
They are interested in applying what has been learned in their research to academic
learning. Various study groups have reported on the advice these teacher-scientists offer
(e.g., Dunlosky et al., 2013). Below are a few of their recommendations. I start with the
one that may be most fundamental.
1 Elaboration. Effective learning requires an active mind and not passive exposure to
the material. Learning has less to do with the intention to learn (“I really need to learn
this for the test”) and more about how you think about the information. Elaboration
includes strategies such as: Can you restate the ideas in your words? Can you think
of an example? Can you connect it to something else you know? Does this idea make
sense? Elaboration also involves choosing to use strategies, such as those listed as
numbers 2 and 3 below.
2 Spaced repetition. Spreading out or spacing exposures to material leads to better
learning than does massing exposures. Sometimes a second presentation that
occurs too close to the first (what we call massed repetition) adds nothing at all to
memorability. The repetitions can be in the form of a review of the material, a sum-
mary, or a practice test, but in general, spaced is better.
3 Interleaving. Intermix the study of different materials. In a sense, this promotes some
spacing. If you have a few topics to learn, intermix studying of them rather than
focusing just on the first, then the second, etc.
 Summary 21

4 Practice testing. A practice test or self-test of your knowledge can convey more ben-
efit than simply studying the material a second or even third time.
5 Beware of self-judgments about learning. Your intuition about whether you know
something can be wrong. When we review just-studied material, a feeling of familiar-
ity falsely suggests that we know it. Students are often wrong in their judgments
about whether, and how well, they have learned.
For example, if the material is easily learned, we assume it will also be easily remem-
bered. Given a list of word pairs to memorize, the pair TABLE-TREE will be learned quicker
than the pair TABLE-AARDVARK. Surely the easily learned pair will be remembered better.
Unfortunately, if something is too easily learned, it may be too readily forgotten. Remember
that password you made up on the spur of the moment, the one that seemed so obvious?
The one you can no longer recall? The difficult word pairs engage more mental resources
(rehearsal, imagery, meaningfulness, elaboration) to help us learn the connection.

SUMMARY
The goal of this book is to present a scientifically accurate and sophisticated view of the
principles of learning. This includes adding qualifying statements to general principles:
Specifying when a given principle holds and when exceptions occur.

The Origins of the Study of Learning

The branch of philosophy known as epistemology studies how we come to have knowl-
edge. This is also central to the field of learning. The nature–nurture question asks how we
are affected by biology on the one hand (i.e., nature) and by environment on the other
(i.e., nurture). The philosopher John Locke said the origin of all knowledge comes from
experience. This is empiricism.
Darwin’s theory of evolution suggests that learning evolved as an adaptive speciali-
zation. Learning is a means of adapting to the environment within the organism’s lifetime.
That different species are related through a common evolutionary history suggests that
learning by animals can offer insights into how humans learn.

The Definition of Learning

Learning is defined as a relatively permanent change in behavior or behavioral repertoire
that occurs as a result of experience. Each phrase in the definition is significant. Because
learning itself, in the mind or in the brain, is not directly observable, behavioral change is
necessary to provide evidence that learning has occurred. Measures of learning can be
physiological, behavioral, or verbal.
The phrase behavioral repertoire acknowledges that not all learning is immediately
evidenced in behavior. Learning includes the potential for a change in behavior, to be
demonstrated when conditions prompt the display of this new knowledge. Learning pro-
duces relatively permanent changes in behavior, to exclude transient changes to fluctua-
tions in arousal, fatigue, or motivation.
22 Introduction

Biologically determined maturation illustrates gray areas around the edge of our defi-
nition of learning. Learning occurs in the context of biological physical, neural, cognitive,
and social development.

The Learning/Performance Distinction

There is not always a one-to-one correspondence between what an organism knows (i.e.,
learning) and what an organism does (performance). Tolman and Honzik’s study (1930)
of latent learning showed that rats learned the layout of the maze without reward, but this
knowledge remained hidden until the subjects were motivated with food to complete the
maze quickly. In an exam situation, students who have truly learned the material can still
perform poorly on the exam (due to excessive anxiety, for instance).

The Relationship between the Terms Learning and Memory

Learning refers to the acquisition, or encoding, of knowledge or behavior. We could illus-
trate its development with a learning curve, a graphic plot of a measure of behavior on the
vertical axis (e.g., number or size of the correct responses) as a function of the number
of trials given, shown on the horizontal axis. Memory refers to the retention of knowl-
edge or behavior after it has been learned. We could illustrate the course of memory by a
forgetting curve, again plotting a measure of behavior on the vertical axis (e.g., number or
size of the correct responses), as a function of the passage of time since learning was
completed, on the horizontal axis. For example, students given extra study trials seemed
to learn more; but when memory was tested later, the students given practice tests
recalled more.

The Science of Learning and Memory

The scientific investigation of Learning is characterized as empirical (based on observa-
tions), quantitative (measurable), and objective (public, replicable, unbiased).
Basic research is an interest in understanding the fundamental processes, by demon-
strating cause-and-effect relationships between key variables. We must often use artificial
situations or tasks in order to control or eliminate contaminating variables that could affect
behavior. Applied research, the other end of the continuum, is designed to be relevant
to, or to apply to, a specific practical problem. Each type of research is appropriate for
answering certain kinds of questions.
Aren’t the principles of learning simply common knowledge? In fact, there are dis-
crepancies between what learning-professionals and what laypeople believe about mem-
ory. These memory myths include distorted beliefs about amnesia, hypnosis, and
forgetting in the aged.
Nonhuman animals are used in experiments on learning because their experiences
can be highly controlled and because there is a presumed similarity in learning processes
between animals and humans. Research using animals has made numerous contribu-
tions to the welfare of both animals and people.
 Summary 23

Conceptual Approaches to the Study of Learning

There are several broad approaches to the study of learning. The functional approach
proposes that learning and remembering evolved as means for organisms to adapt to the
changes and inconstancies in their environment and thus to aid survival. A behavioral
approach focuses on the acquisition of specific responses or behaviors. It emphasizes the
relationship between these observable behaviors to the stimuli that precede behavior,
and to the consequences that follow behavior. A cognitive approach emphasizes internal
(mental or neural) cognitions and expectancies. Information is encoded, transformed,
stored, and retrieved. A neuroscience approach studies the changes that learning pro-
duces in the brain. It seeks to determine the underlying biological basis of learning and
memory within the nervous system.

Applications
Those psychologists who study learning and memory can offer advice to educators. We
see this in particular in attempts to generalize from neuroscience research to student
applications. However, research on topics such as the Mozart effect, critical periods in
neural development, and environmental enrichment, may not yet be applicable to educa-
tion. Advice from experts on how to study, such as self-testing and spaced repetition, and
being cautious of our own estimates of what we know do have empirical support.
 CHAPTER

2
Habituation and Other Forms of
Stimulus Learning

CONTENTS

What Is a Stimulus? 25 Perceptual Learning 38

The Orienting Response 25 Factors Affecting Perceptual Learning 39
Habituation 26 Other Effects of Stimulus
Methods of Studying Habituation 26 Exposure 40
Parametric Features of Habituation 29 Preference for Familiar Stimuli 40
Explanations of Habituation 32 Recapitulation: The Effects of
Nonlearning Explanations 32 Repeated Stimulus Presentation 43
Neuroscience Theory 33 Applications 43
Evolution 35 Exposure Therapy for Fear 44
Cognitive Theories 35 Needle Fear 45
Sensitization 37 Summary 46

Researchers often start with a simple model, system, or paradigm with which to derive
some first principles, and progress to more complex phenomena and laws. Learning
about single stimuli, uncomplicated by associations or rules relating them to other events,
would seem to offer such a simple learning situation.
As a starting point, suppose an innocuous stimulus is presented some number of
times. Your behavior at first indicates that you notice the stimulus, but since it has no
apparent significance, your reactions to the stimulus decrease. A noise in the house at
night awakens you, but you realize it is just the house settling or the furnace humming,
and so you come to ignore these house sounds. However, you can probably think of
times when you became more reactive to a stimulus after a first exposure. A noise in the
house awakens you; and as you worry about what it might be, the behavioral and physio-
logical reactions increase each time the noise recurs. Rather than habituating, you are
more responsive to the noise.
Exposure to a stimulus can affect our behavior in other ways, even if we are not
aware of the previous exposure. For instance, a stimulus could be flashed on a screen so
briefly (only a fraction of a second) that it cannot be consciously recognized. Yet this una-
wares presentation can affect reactions to the stimulus later, possibly inducing a feeling

DOI: 10.4324/9781003227090-2
 What Is a Stimulus? 25

of familiarity or even liking of the stimulus. Repetition also increases our preference for
particular songs, for works of art, and for particular foods.
The starting point for this chapter is that stimulus repetitions often create responsive-
ness to a stimulus. However, stimulus exposures can produce a number of other reac-
tions: greater responsiveness, preference and liking, and speeded reactions. “The main
lesson to be learned from the study of habituation—and this makes it an even more
appropriate subject to start with—is that habituation is almost never as simple as it first
seems” (Walker, 1987, p. 34).
Since we will be talking about stimulus learning in this, and next few, chapters, it is
useful to consider the question of what do we mean by stimulus?

WHAT IS A STIMULUS?
What is a stimulus? As used in biology, the term refers to a change in the physical or
chemical environment that elicits a reaction. A tap on the check would be a stimulus for
an eyeblink response. An odor elicits wrinkling of the nose, orienting, and attention. In
psychology, the stimulus might be a tone (specified by its loudness, pitch, and duration).
In some situations, the stimulus might be seen as the “cause” of the response or reac-
tion, as in the touch–blink reflex, or the salivation by Pavlov’s dog to a tone. Stimuli can be
external, such as tones and lights; or internal, such as hormone secretions, muscle ten-
sion, or difficulty breathing. A stimulus does not have to be physical; it could originate in
the mind (or in the brain). A thought or memory might be the stimulus for a response.
“I remember what happened that day…” followed by anxiety; or “I just remembered it’s
her birthday” followed by a phone call.

The Orienting Response

The occurrence of a novel or an unexpected stimulus elicits an orienting response (OR).
Pavlov aptly described the OR as an investigatory reflex. The organism reacts to identify
the nature and source of the stimulation, which may be important to the survival of the
organism. The OR is actually not a single response but is a composite of several physiolog-
ical and behavioral reactions. First, a stimulus may evoke a startle response. This may
involve whole body startle to extremely loud noises, or simply an eye blink to milder stim-
uli. In addition, the novel stimulus produces sense receptor orienting. We turn to look in the
direction of a sight or sound; dogs perk up their ears to hear better. There is also a readi-
ness for a fight or flight. Orienting reactions are a preparation for danger, and so defensive
behaviors are primed. Finally, there is increased arousal. This arousal could be described
physiologically, as measured by heart rate or breathing changes, and also psychologically,
as measured by degree of attention and alertness (Siddle, Kuiack, & Kroese, 1983).
Not each of these components occurs to every new stimulus. A starter’s pistol that
goes off unexpectedly can elicit a head-jerking startle, whereas a tone stimulus presented
in a lab experiment may simply elicit an eye blink. Thus, the OR depends on stimulus
intensity, the situation, the potential for danger, and other factors.
The novelty of a stimulus is just one factor determining whether an OR is made, but
it is not a necessary factor. A familiar but meaningful stimulus can also elicit an OR.
26 Habituation and Other Forms of Stimulus Learning

Hearing one’s name mentioned unexpectedly is a potent elicitor of an OR, although this
is hardly a novel stimulus. A predatory animal will orient to the sound of its usual prey, a
familiar stimulus.

HABITUATION
Habituation is the decrease in orienting reactions to a stimulus that is repeatedly pre-
sented. An often-used phrase describes habituation as “the waning of responsiveness”
to repetitive stimulation. An initially new stimulus becomes familiar with repeated pres-
entations, and thus becomes less likely to elicit an OR. The word habituation is used in
the research literature both to refer to the procedure (repetitive presentation of a stimu-
lus) and to the effect or outcome (a decrease in responding).
Habituation is a simple form of learning in which the organism learns something
about a single stimulus. Unlike most other indications of learning that we will consider in
later chapters, habituation is indicated by less responding over trials. This is just the oppo-
site of the learning curve described earlier, which increases over trials.
If the stimulus turns out to have some significance to the organism, because it is
followed by other events or consequences, orienting may be replaced by learned adaptive
responses. For example, in Pavlov’s experiment, the tone stimulus is followed by food.
Orienting to the first several tones is gradually replaced by salivating to the tone, as the
dog learns that the tone is followed by food. In most habituation experiments, only the
to-be-habituated stimulus is presented so that the addition of other acquired responses
does not complicate interpretation.
Habituation also occurs to intense and/or painful stimuli. The responses to noxious (or
aversive) stimuli are called defensive responses. These defensive reactions can also habit-
uate: Reflex responses to potent stimuli may decrease across repeated presentations of
the stimulus (Wetherington, 1982). Although some researchers differentiate orienting
responses from defensive responses, other researchers treat both sorts of reactions as
being on a continuum, with the strength or kind of response being more a function of
stimulus intensity.
Habituation is ubiquitous: It is seemingly found everywhere. Habituation occurs
across the phylogenetic scale, from snails to humans. It occurs in various segments of
nervous systems, including isolated spinal neurons and at synapses. The study of habitu-
ation brings together diverse groups of researchers, from those interested in the physiol-
ogy of habituation to those studying attention and memory in children. These investigators
share a common terminology, similar principles of learning, and sometimes the same
theoretical explanations. Such cross-fertilization is desirable for the scientific advance-
ment of learning.

Methods of Studying Habituation

Much research on habituation studies some form of startle reaction to sudden or noxious
stimuli. In small animals, whole body startle to a loud noise burst can be measured by the
movements in the small chamber the animal is in. In humans, startle can be recorded via
an eyeblink or muscle potentials recorded on the skin around the eye. Heart rate may
jump at the sound of a startle-eliciting noise.
 Habituation 27

In humans, other physiological responses are often used to monitor reactions to new
stimuli. The skin conductance response (or SCR) measures subtle changes in electrical
conductivity in the skin that are associated with arousal or emotionality. The SCR is probably
familiar to you as a component of the lie detector test, in which electrodes are placed on
the hand or arm. Illustrative data from a habituation experiment measuring SCR are shown
in Figure 2.1. College student participants were tested individually in a laboratory setting.
They were presented with 15 brief tones, one every minute or so. Some students heard a
high-pitched tone, and the others heard a low-pitched tone. As can be seen in the graph,
there was a large reaction to the first few tones, and then the skin conductance response
decreased across later stimuli for each group. After 15 stimulus presentations, the tones
were switched, so the low-pitched tone replaced the high-pitched tone, and vice versa.
Orienting was reinstated when the subjects heard a different tone (Siddle et al., 1983).
In the novelty recognition task, mice or rats are allowed to explore some objects,
such as blocks, in a test environment. New objects are then added before the next test is
conducted. The animals explore the objects that are not recognized, touching, sniffing,
and even climbing on them. This active exploration indicates orienting. Objects that are
remembered as familiar are explored less. The reduced attention given to the old objects
indicates habituation. The novelty recognition task is widely used among neuroscience
researchers to assess how brain lesions, genetic alterations, new drugs, and environmen-
tal manipulations affect memory (e.g., Tang et al., 1999).
One widely used measure of orienting is that of eye fixations to novel visual stimuli.
We visually explore new stimuli, and gradually shift our gaze away from these stimuli

12 Group HL
Group LH
Means SCR Magnitude (mhos/cm2)

10

2 4 6 8 10 12 14 16
Trials

Figure 2.1
Habituation of the Skin Conductance Response to a Tone Stimulus. One group received 15 presentations
of a high-pitch (HL) tone, the other group received a low-pitch (LH) tone. On trial 16, the opposite tone was
presented to each group.
Source: From “The Orienting Reflex,” by D. A. T. Siddle, M. Kuiack, and B. S. Kroese, in Physiological Correlates of Human
Behavior (p. 161), edited by A. Gale and J. A. Edwards, 1983, London: Academic Press. Copyright 1983 by Academic Press.
Reprinted with permission.
28 Habituation and Other Forms of Stimulus Learning

once they become familiar. Measuring the change in duration or number of eye fixations
to a stimulus is a good indicator of habituation. In a sense, the eyes become the windows
of the mind. Measuring eye movements is especially useful for studying learning by
infants. A novel object or image is shown repeatedly, and habituation is shown by a
decrease in fixations across trials. We could interpret this as an indication that the child
comes to recognize the stimulus as familiar. Following habituation to one stimulus, a
second is presented for comparison, one old and one new stimulus. The child shows
more inspection of the new stimulus than of the old one (see Box 2.1).

BOX 2.1 HABITUATION AS A MEASURE OF INFANT PERCEPTION AND COGNITION

Habituation tasks have been used to study perception, learning, and reasoning in infants. The
cognitive capabilities of preverbal children are difficult to measure directly. Measuring the
duration of eye fixations to familiar and novel stimuli provides a valuable tool for investigating
learning in infants. In this method, one stimulus is habituated by presenting it frequently,
leading to a decrease in the duration of eye fixations to the stimulus. Then an altered stimulus
is presented to see whether fixations increase, indicating that it is noticed as different from
the familiar stimulus.
Adults typically divide the light spectrum into several distinct primary colors, e.g., red,
green, blue). Intermediate hues can be reliably assigned to one or another of these catego-
ries. For example, between those colors labeled as blue or green are colors we label as
greenish blue and bluish green, which adults consistently categorize (if forced to) as either
blue or green. How would preverbal infants, who have not learned these linguistic divisions
of the color spectrum, classify intermediate hues? Bornstein, Kessen, and Weiskopf (1976)
first presented 4-month-old infants with repetitions of a 15-second light. Different groups of
infants were shown lights of different colors, for example, greenish BLUE for one group and
bluish GREEN for another group (the color in capital letters is the adult category label).
Fixation times declined from 8 to 9 seconds on the first presentations to 3 seconds over the
last few (see the left side of Figure 2.2). Thus, habituation occurred. The infants were then
given test presentations of the habituated color, and a color on either side from the color
spectrum (see right side of Figure 2.2). In each case, less looking occurred to the hue within
the same adult color category, even though each tested hue was equidistant from the habit-
uated color. For example, after habituating to bluish GREEN, the infants spent even less time
looking at GREEN, and spent more time fixating to greenish BLUE.
Similar results were shown with the other color boundaries. Variations of reds, yellows,
greens, and blues are categorized as equivalent, whereas there are sharper boundaries
between these color categories.
This same technique is used to study infants’ perception of number (Wynn, 1992). The
infant was shown a fixed number of puppets (either one puppet or two) across several trials,
and looking time decreased. When a puppet was added or subtracted on test trials there was
an increase in looking. In another experiment, the infants were shown a puppet that jumped
two times or three times on the habituation trials. A change in the number of jumps on the
test trial increased looking time (Wynn, 1995).
The starting point for research like this is, indirectly, the philosophical question of nature
versus nurture: Do we innately perceive colors? Do we learn the concept of number?
Curiously, a technique used to study learning (habituation) can be used to show evidence for
native perceptual abilities.
 Habituation 29

I Exposure II Test
10
Group 1
480 nm
Mean Looking Time (sec) 8
Group 2
510 nm

0
3 6 9 12 15 450 480 510 540
Trials B gB bG G

Figure 2.2
Design and Results of One of the Color Conditions. Infant mean eye fixation times to habituated
colors (Phase I) and test colors (Phase II).
Source: Bornstein et al. (1976).

Parametric Features of Habituation

Thompson and Spencer (1966) described a list of parametric features of habituation.
Parametric refers to taking one dimension of an independent variable and systematically
varying it to see the effect on the response. For example, one can vary the spacing of
stimulus presentations across a range of values and measure the amount of response at
each spacing. The Thompson and Spencer features have become a benchmark for evalu-
ating whether response changes that occur in other situations are really habituation. (The
features were revisited and reasserted in Rankin et al., 2009). A consideration of several
of these features illustrates the process of habituation.

Frequency of Repetition
Habituation increases as a function of the number of repetitions of a stimulus. (Remember:
As habituation increases, responding decreases.) The exact number of repetitions neces-
sary to produce a substantial response decrement varies considerably. As shown in
Figure 2.1 earlier, human habituation occurs after a few presentations of a mild tone
stimulus. Rat startle response to loud tones continues to decline across hundreds of
presentations.

Spontaneous Recovery
If the stimulus is withheld for a period of time, the response tends to recover. The next
time the stimulus is presented, the response will be larger than before the delay inter-
val. The phrase spontaneous recovery refers to the return of the response after an
interval of time without stimulations. Indeed, if the delay interval is sufficiently long
30 Habituation and Other Forms of Stimulus Learning

(e.g., overnight in an animal experiment), the reaction to the previously habituated stim-
ulus may completely recover.
Frings et al. (2006) studied habituation of the acoustic startle response to a loud tone
burst. Startle was measured from electrodes placed around the eye to record eyeblinks.
The human subjects were tested across four sessions, with 42 loud tone presentations
daily. The size of the OR is expressed as a percentage of the amplitude on the first trials.
Spontaneous recovery in this experiment is illustrated in Figure 2.3, which shows habitu-
ation and recovery of the eyeblink. Between the final trials of Day 1 and the first trials of
Day 2, there is a recovery of startle blinks. Actually, for many years, this feature was used
to argue that habituation was not an example of learning. If habituation produced only a
temporary change in responding, then habituation did not fit the relatively permanent
clause in the definition of learning. Why did we come to believe that habituation is learn-
ing? The answer is in the next feature.

Effects of Repeated Habituations

If habituation is conducted over a series of sessions, each separated by some interval of
time, habituation occurs more quickly during successive sessions. The spontaneous
recovery (or loss of habituation) is less, and habituation requires fewer stimulus
presentations as compared to the initial and previous sessions. Habituation accumulates
across trials and sessions (see Figure 2.3).

Figure 2.3
Habituation of Human Eyeblink Startle Response. Percent of initial blink amplitude across blocks of seven
trials on four successive days.
Source: From “Involvement of the Human Cerebellum in Short-term and Long-term Habituation of the Acoustic Startle
Response: A Serial PET Study,” by M. Frings et al., 2006, Clinical Neurophysiology, 117, p. 1290–1300. Data derived from
Figure 2.2, p. 1294.
 Habituation 31

Short-Term Habituation and Long-Term Habituation

A distinction can be made between short-term habituation and long-term habituation.
Short-term habituation is that which occurs within a closely-spaced block of trials, or from
one stimulus presentation to the next. In Figure 2.3, short-term habituation is the reduc-
tion in OR size from the first trials on Day 1 to the last trials on Day 1; and ditto for each
day thereafter. Long-term habituation is the overall decrease in response across succes-
sive sessions. Short-term habituation is the decrease in the responding within each ses-
sion. Long-term habituation is the reduction in responding from day to day; overall, the
response is smaller on Day 2 than on Day 1, and so one across the next days.
Spontaneous recovery is also demonstrated in Figure 2.3 On the first few trials of
Days 2, 3, and 4, responding has recovered from the lower levels at the end of the
preceding sessions. But the response does not completely return back to the initial level
at the start of Day 1. Some habituation is retained over the 24 hours from one session to
another. This is long-term habituation.
Differentiating between short-term and long-term habituation is important because
they likely have different biological mechanisms.

Spacing of Stimulations
Responding decreases more quickly across shorter intervals between stimulus presenta-
tions. Basically, massed stimulus repetitions produce more habituation than do spaced
repetitions.
This massed-spaced effect is more complicated than the above statements suggest.
Massing repetitions may produce less responding than spacing repetitions during the
habituation phase. But the difference may disappear or even reverse when tested again
afterwards. This can be shown in a study by Gatchel (1975) in which college student par-
ticipants heard a series of tones, one tone presented every 20 seconds for one group or
every 100 seconds for the second group. Across 15 tone presentations, there were
smaller responses to stimuli spaced 20 seconds apart than to stimuli 100 seconds apart,
which is a massed-presentation effect. The subjects were given a 15-minute rest interval
and then the tones started again. The 20 second massed group showed more recovery to
the tone presentations following the delay than did the 100 second spaced group. The
lesson here may be that massed presentations produce more short-term habituation,
whereas spaced presentations can be more effective in producing long-term habituation.

Dishabituation
So far we have been talking about only one stimulus in a habituation experiment. The
occurrence of other stimuli can disrupt habituation to the target stimulus. For instance,
after the response to a first stimulus has been habituated, the presentation of a second
novel stimulus will reinstate orienting to the first stimulus. This is called dishabituation.
Dishabituation is the reinstatement of orienting to a habituated stimulus (stimulus #1)
by presentation of a different stimulus (stimulus #2).
For example, in one study, college student participants heard 15 presentations of the
same 4-second tone. The size of the OR decreased, which is simply habituation. Then a
new stimulus, say, a patch of red light was projected on a computer screen. The light itself
elicits an OR (but we are not interested in the reaction to the light). When the tone is next
32 Habituation and Other Forms of Stimulus Learning

presented, its sixteenth presentation, there is a larger response to it than to the previous
(fifteenth) occurrence. The light stimulus acted as a dishabituator; the reinstated response
to the sixteenth tone shows dishabituation. Dishabituation is a transient aftereffect of the
dishabituating stimulus. Presenting the tone again (the seventeenth presentation) should
lead to the smaller, habituated response (Siddle, 1985).
Dishabituation is a sensible reaction to changed conditions of stimulation. The disha-
bituator reinstates arousal, investigatory reflexes, sense receptor orienting, and so on, in
preparation for potential changed environmental conditions. (Dishabituation is shown in
Figure 2.5, described in a later section).

Stimulus Specificity of Habituation

Habituation occurs to the particular stimulus that is presented. This is the principle of
stimulus specificity. Habituation, or the absence of orienting, does not occur to any or all
novel stimuli that have themselves not been presented repeatedly. The recurring sound
your refrigerator makes, that you now ignore, does not block orienting to a new noise
from the fridge.
It is important to remember that it is the stimulus that habituates, not the response.
The OR is readily available for other unexpected stimuli.
The habituation of one stimulus may spread to other similar stimuli. That is, after
habituating one stimulus there can be generalization of habituation to other stimuli that
are similar. The degree of generalization depends on the similarity of the test stimuli to
the habituated stimulus. For example, in the Siddle et al. (1983) study noted earlier in this
chapter, habituation to a high- or low-pitched tone did not generalize to the opposite tone
(Figure 2.1). The different tone elicited a large OR. However, if a test tone of medium pitch
had been presented, likely a smaller or weaker OR would have occurred.

Summary of the Parametric Features of Habituation

The response to a novel stimulus decreases as a function of the number of repetitions of
the stimulus. The habituated response spontaneously recovers over time if the stimulus
is withheld for a while, and dishabituation occurs after the occurrence of another novel
stimulus. Habituation cumulates across repeated series of stimulus presentations.
Habituation is stimulus specific, although there may be generalization from the habituated
stimulus to similar stimuli. Closely spaced presentations may suppress the orienting
response quickly, but there is evidence that more widely spaced presentations produce
more long-term habituation.

EXPLANATIONS OF HABITUATION
Nonlearning Explanations
Why does responding decrease after a certain number of stimulus presentations? One
possibility is that habituation is not learning at all but instead is due to changes in either
the sensitivity of the sense receptors or the muscle effectors. One nonlearning explana-
tion suggests that habituation is due to sensory adaptation. The sense receptors become
less sensitive after repeated stimulations. For example, on entering a room you detect a
distinct odor or smell but later it is no longer noticed. Here, the olfactory receptors lose
 Explanations of Habituation 33

their sensitivity to detect a constant odor. Alternatively, maybe the response system is
fatigued. After so many startle responses, the system is too depleted to continue. Either
sensory adaptation or motor fatigue will build up over trials, and both will recover with
rest, thus producing what looks like habituation and spontaneous recovery.
Control procedures are necessary to determine the contribution of either factor to
any response decrement we observe. One such control procedure is changing the stim-
ulus, as in the Siddle et al. study shown earlier in Figure 2.1. If responding occurs to the
changed stimulus then, obviously, the sense receptors detected the stimulus and the
effectors are capable of responding.
Learning theories of habituation fall into two broad categories of explanation: neuro-
science and cognitive.

Neuroscience Theory
Aplysia: A Model System
A physiological approach to habituation seeks to discover the underlying synaptic events
involved in the learning. Eric Kandel, initially trained to be a psychiatrist but decided to go
into basic research, studying habituation in a giant marine snail, Aplysia californica. This
particular snail is used because it has a simple nervous system with few neurons. Many
of these nerve cells are relatively large, and their functions are known. But what behav-
iors can Aplysia perform that we can study? Aplysia uses a siphon to take in seawater
from which food and oxygen is filtered. Stimulation of the siphon, either by touch or by a
squirt from a water jet, elicits retraction back into the body cavity in times of danger (see
Figure 2.4). Siphon withdrawal shows habituation and many of the features listed by
Thompson and Spencer (1966).
Habituation, dishabituation, and spontaneous recovery of siphon withdrawal to a tac-
tile stimulus are illustrated in Figure 2.5. After six touch stimulations, withdrawal has
nearly ceased. This indicates habituation. A light is then flashed as a dishabituator (indi-
cated by L on the abscissa), which reinstates responding to the next presentation of the
tactile stimulus. The light stimulus is repeated again later, but it has a smaller dishabituat-
ing influence when given after the ninth tactile stimulus presentation. After a 60-minute
rest interval, there is spontaneous recovery to the touch stimulus: Presentation of the
tactile stimulus produces a large response, which then rehabituates. Presenting the light
stimulus again produces dishabituation to the tactile stimulus. (Remember, dishabituation
is an increase in response to the touch stimulus. It is not a response to the light).
By measuring nerve potentials on the stimulus and the response sides of the siphon
circuit, Kandel and his colleagues have isolated the change that occurs in habituation.
Basically, habituation occurs at the synapse (or junction) between two cells. The sensory
neuron, which detects the touch, retains its sensitivity; the motor neuron controlling with-
drawal retains its responsivity. The learned change is in the flow of synaptic chemicals
between the two nerve cells, so that the sensory neuron does not activate the motor
neuron.
Biological psychologists have long sought the memory engram, a word used to refer
to the change that occurs in the nervous system to encode new learning. The work by
Kandel and others illustrates how contemporary researchers are closing in on describing
34 Habituation and Other Forms of Stimulus Learning

Figure 2.4
Aplysia, an Ocean Dwelling Slug. Looking down at the Aplysia from above. The mantle is spread to show the
gill and siphon within. Touch by the tactile stimulus causes withdrawal of the gill and siphon. Repeated tactile
stimulation produces habituation of withdrawal. A shock stimulus produces sensitization, or increased
responding.
Source: From “Relationships between Dishabituation, Sensitization, and Inhibition of the Gill- and Siphon-Withdrawal Reflex in
Aplysia Californica: Effects of Response Measure, Test Time, and Training Stimulus,” by R. D. Hawkins, T. E. Cohen, W. Greene,
and E. R. Kandel, 1998, Behavioral Neuroscience, 112, p. 25. Copyright © 1998 by the American Psychological Association.

Figure 2.5
Habituation of Siphon Withdrawal of Aplysia to a Tactile Stimulus. The response measured is the tension
of the siphon withdrawal to tactile stimulation. A light stimulus (L) was presented between some of the tactile
stimuli. Dishabituation of the response occurred to the tactile stimuli following presentation of a light stimulus.
Source: Lukowiak and Jacklet (1972).
 Explanations of Habituation 35

the engram. This is obviously basic research, not applied, and it has contributed substan-
tially to our understanding of the neural basis of learning. Eric Kandel was awarded the
Nobel Prize in Medicine or Physiology in 2000. His life and work are described in his
autobiography In Search of Memory (Kandel, 2006).

Evolution
Research on Aplysia shows one possible mechanism for habituation in the nervous sys-
tem. Habituation varies with responses and different species. The field of comparative
evolution compares species that have had different evolutionary histories. (Overlapping
disciplines include comparative anatomy, comparative neuroscience, comparative physi-
ology, etc. My favorite is EvoDevo: evolutionary developmental biology). Comparing
habituation in different species suggests there are multiple mechanisms and sites
(Ramaswami, 2014). At one end of the phylogenetic spectrum, habituation may be a
transient decrease in sensitivity of the sense organs. Bacteria or protozoans (one-celled
organisms), which do not have a nervous system, might temporarily adapt to aversive
tastes of substances.
For animals with a primitive nervous system, such as the Aplysia, habituation may
occur at the synapse between sensory and motor nerve cells. At a still higher level are
organisms with complex nervous systems that have both excitatory and inhibitory cir-
cuits that can intervene at multiple sites along a response circuit. These latter systems
can also draw on memory to enhance activation to novel stimuli, or inhibit activation trig-
gered by familiar stimuli.

Cognitive Theories
The cognitive approach to habituation suggests that orienting responses are elicited by
stimuli that are not recognized, that is, by stimuli that are not already in memory. Repeated
presentations of a new stimulus lead to the formation of a memory for the stimulus.
Habituation indicates the acquisition of that memory.
The Russian psychologist and physiologist Sokolov introduced many Western
researchers to the study of orienting reflexes (e.g., Sokolov, 1963). Sokolov hypothesized
that there is a comparator mechanism, which compares the current sensory input to
images of stimuli stored in memory. An OR is made if the comparator does not find a
match in memory, whereas the OR is inhibited if the comparator finds a match. This idea
of comparing a stimulus in the environment to one in memory is a characteristic of the
cognitive theories.
The theories of Wagner (1981; Whitlow & Wagner, 1984) and Olson (1976), developed
from experiments on animals and human infants, respectively, elaborate on the memory
system. They state that a stimulus can be remembered in short-term memory, long-term
memory, or both. The more permanent, durable habituation seen across sessions or over
long intervals reflects the long-term memory for the stimulus. This is the learning we have
been talking of when we say that the stimulus becomes known or familiar. However, a
stimulus may also be recognized if it has occurred very recently and is still remembered
in short-term memory. Thus, a current stimulus can be recognized if it matches a rep-
resentation in either short-term memory or long-term memory (see Figure 2.6).
36 Habituation and Other Forms of Stimulus Learning

Stimulus Stimulus ORIENT

Comparator Response
Input Analyzer or NOT

STM

LTM

Figure 2.6
A Generalized Cognitive Model of Habituation. A hypothesized comparator mechanism compares stimulus
input to representations in short-term memory and in long-term memory. A match between input and what is
stored in memory leads to inhibition of responding. A mismatch leads to an orienting response.

This dual-memory theory describes why massed stimulus presentations can produce
more habituation than spaced presentations. When the stimuli are repeated closely
together, a given stimulus can be recognized as familiar either from short-term or long-
term memory. When stimuli are spaced apart, the short-term representation will have
been forgotten, and recognition can only occur if the stimulus is represented in long-term
memory (Whitlow & Wagner, 1984).
Memory also records that the stimulus occurred in a particular setting or at a particu-
lar time. Cognitive theories incorporate the notion of expectancy, or an active prediction,
for what stimulus will occur, where it will occur, and when. You have probably habituated
to the bell that rings in the school halls to signal class changes. If the bell now rings off
schedule, or if the bell rang in your home, it would no doubt elicit a startle reaction. This
notion is incorporated in Wagner’s (1976) idea that the surprising occurrence of a stimulus
elicits orienting. An expected occurrence of that same stimulus does not elicit orienting,
even though the stimulus is objectively the same in both cases.
An example of the role of expectancy is the missing stimulus effect (e.g., Siddle,
1985; Siddle & Lipp, 1997). After exposure to a fixed sequence of a pair of stimuli, such
as Tone-Light (T-L), on test trials the second stimulus is omitted, so just T occurs. The
absence of the expected light in this case evokes an OR, just after the time when the light
should have occurred. What is interesting here is that an OR is triggered not by the occur-
rence of something, but by its absence. (An orienting reaction occurs to—nothing!)
Cognitive theory says that, based on our experiences, we learn expectancies for particu-
lar patterns of stimulation. When events violate an expectancy, orienting occurs.
The missing stimulus effect occurs in daily life when the sudden realization that what
was expected did not happen: The alarm clock didn’t go off! The missing stimulus effect
is vividly illustrated by my own experiences at opposite ends of the child-rearing spec-
trum: suddenly awakening at night because the baby did not cry for a late feeding; and 17
years later, awakening because I did not hear him come home yet.

Explanations of Habituation: Summary

There are two nonlearning explanations that are alternatives to learning-based habitua-
tion. A response may decrease because the stimulus is not detected (sensory adaptation)
 Explanations of Habituation 37

or because the response apparatus is temporarily inhibited (effector fatigue). Neuroscience

theories of habituation, such as Kandel’s from the study of Aplysia, seek the physiological
and biochemical basis for the changes that underlie habituation. Cognitive theories, those
of Sokolov or Wagner, attribute habituation to the acquisition of a memory for the repeated
stimulus. This memory eventually includes detailed information about the stimulus, such
as its temporal patterning, and the relationship to other stimuli.

Sensitization
The repetition of a given stimulus does not always, and only, produce a smaller response.
There can in some cases be an increase across repetitions. Sensitization is an increase in
orienting and defensive responses that occurs with repeated stimulation. Stimuli that
sensitize are usually intense, painful, or emotional. That tingle you noticed on your arm
habituates when you realize it is just your sleeve rubbing. If the tingle was a spider walk-
ing up your arm, sensitization occurs, and you overreact to the next tingle.
Richard F. Thompson introduced the dual-process theory of habituation, which states
that the overall behavioral response to stimulus repetition depends on the balance between
two factors: habituation and sensitization. (e.g., Groves & Thompson, 1970). The stimuli
used in habituation experiments can elicit two separate reactions. Habituation is one reac-
tion and is due to the familiarity of the repeated stimulus. Sensitization is usually produced
by stimuli that are intense (such as a loud noise burst), painful (such as electric shock), and
emotional (such as fearful). These stimuli increase the level of arousal in the nervous sys-
tem. This arousal counteracts habituation, and the balance between the two may be a
decrease, no change, or an increase in the response amplitude. When mild and innocuous
stimuli are presented, habituation alone occurs. When strong or intense stimuli are pre-
sented, sensitization can increase responsiveness and overshadow any habituation.
Groves and Thompson found two types of neurons in their studies of spinal cord
reflexes. Type H (for habituation) neurons are in the circuit directly involved in the reflex
response: sensory neurons that detect the stimulus, and motor neurons that control the
response. The type H produce habituation. Type S neurons (for sensitization) reflect
the level of arousal of the nervous system. This sensitization-produced arousal increases
the excitability of the motor neurons involved in the response.
Habituation and sensitization can be differentiated by experimental manipulations.
One is the effect of background noise level on the startle response (Davis, 1974). Two
groups of rats received the same series of startle-eliciting tones. The difference was that
one group heard these against a background noise level of 60 decibels; the second group
had a background level of 80 decibels. The rats with the quieter background showed
habituation, i.e., a decrease in startle over trials; the rats with the louder background
showed sensitization, i.e., an increase in startle across trials. The loud background noise
increased the arousal to a degree that counteracted habituation.
A magnified startle reaction can occur in the presence of other sources of arousal.
Anxiety can combine with the arousal elicited by a loud noise burst to produce a larger
startle reaction. The experimental procedure of fear potentiated startle is useful for
detecting the presence of fear: the more fear present, the greater the enhancement of
the startle. For instance, startle is enhanced in individuals with some types of anxiety,
38 Habituation and Other Forms of Stimulus Learning

such as panic disorder or post-traumatic stress disorder (PTSD). In the context of the
laboratory setting, anxious individuals startle more to the noise burst than do less-anxious
subjects. The startle response is also potentiated in the presence of a stimulus that sig-
nals electric shock (Grillon, 2002).

PERCEPTUAL LEARNING
If we are exposed to variations of a stimulus—such as different font styles, brands of
coffee, or varieties of wines—we usually learn the differences among them. Perceptual
learning refers to learning to perceive differences and similarities among stimuli based
on exposure to the stimuli. Perceptual Learning, both the subject and the title of an
award-winning book by Eleanor Gibson (1969), is a robust phenomenon in animals and
humans.
In a classic study of perceptual learning, human subjects, both children and adults,
were given a deck of “scribble” cards (Gibson & Gibson, 1955). The scribbles differed in
dimensions, such as the tightness of the coils and left–right orientation (see Figure 2.7).

Figure 2.7
Scribbles Used to Assess Perceptual Learning.
Source: From “Perceptual Learning: Differentiation or Enrichment?” by J. J. Gibson and E. J. Gibson, 1955, Psychological
Review, 62, p. 36. Copyright © 1955 by the American Psychological Association.
 Perceptual Learning 39

One coil was labeled the standard. After viewing it for 5 seconds, the participants then
decided whether the other scribbles in the deck matched the standard. The subjects were
not informed whether their choices were correct. Over successive sortings of the deck of
cards, subjects became more accurate in identifying matching versus different scribbles.
On the basis of such experience, the participants learned to perceptually differentiate
among the scribbles.
Perceptual learning of stimuli can be useful for later learning about those stimuli. For
example, it should be easier to learn someone’s name if the name is a familiar word rather
than an unfamiliar name. Once we have learned to perceive, recognize, or identify a stim-
ulus, it is easier to learn other things about it, such as its name or its use. In a classic
perceptual learning experiment by Gibson and Walk (1956), young rats were continuously
exposed to cutouts of a triangle and a circle for 90 days in their home cages. Subsequently,
these stimuli were used as cues in a discrimination experiment. One shape was placed
to mark the correct turn in a maze that led to the food-rewarded goal box. The other shape
marked the choice that led to the nonrewarded goal. Gibson and Walk found that subjects
given preexposures learned the maze faster. These subjects had already learned to distin-
guish between the two shapes. The control subjects without prior exposure to the shapes
first had to master this discrimination between circle and triangle.
Researchers have studied perceptual learning using numerous classes of stimuli,
from colors and faces to reading X-ray scans. Exposure alone to different stimuli does not
ensure learning to differentiate them. When asked to select the picture of a penny from
among eight versions shown, few people could identify the real one (Nickerson & Adams,
1979). (Does Lincoln face right or left?) Perceptual differentiation among stimuli seems to
require attentional weighting: increasing our attention to features that distinguish among
stimuli (Goldstone, 1998). In the penny example, we assign a higher weight to the color
“copper” to differentiate a penny from a dime, than to a feature such as which direction
Lincoln is facing. (He is facing to our right).
Perceptual learning occurs in learning to discriminate language sounds, such as “ba”
and “da,” or the r/l difference that is present in English but not in Japanese. The language
Hindi has a pair of t sounds, which adult speakers of Hindi differentiate but which English-
speaking adults do not. When do we learn to differentiate such sounds in our language?
Werker (1989) found that infants in English-speaking environments can discriminate the
two Hindi ts at 6 months of age but lose this ability between ages 8 and 12 months.
Experience may both promote new categories of knowledge and may also prevent the
loss of existing categories.

Factors Affecting Perceptual Learning

Presenting Contrasting Stimuli
Stimuli differ along a number of dimensions, and perceptual learning requires identifica-
tion of which dimensions are relevant for discrimination. Thus, presentation of both posi-
tive and negative instances seems to be necessary. Simply repeating a single standard
stimulus does not lead to efficient learning to distinguish that stimulus from others.
In the Gibson and Gibson (1955) scribbles experiment mentioned earlier, the different
scribble cards were presented in an unsystematic order. Does the sequence of
40 Habituation and Other Forms of Stimulus Learning

presentation of correct and incorrect exemplars affect ease of learning? This question
was addressed in a study in which subjects learned to differentiate similar looking faces
(Dwyer, Mundy, & Honey, 2011). The face pictures were sequenced to alternate pairs of
similar faces (say face A and then face A′), or a block of face A pictures was shown fol-
lowed by a block of A′ faces. Because the two pictures were not presented simultane-
ously, the comparison had to be made between the memory of the previous face with the
presently viewed face. Stimulus learning was faster when similar items were contrasted
more often in the alternation sequence.

Transfer from Easy to Difficult Stimuli

Initial experience with an easy discrimination can facilitate learning of a more difficult
discrimination. For instance, one perceptual-learning task requires the detection of a line
that has a different vertical orientation than the remaining lines—for example, the last
upright line in the following: // /// /// \ . An easy differentiation occurs when the test line is
rotated 45 degrees from the orientation of the remaining lines, such as //// \ . In a page full
of lines, a more difficult detection is finding a test line that is rotated only 30 degrees from
the other lines, such as //// l . Practice with the more obvious differentiation facilitates later
learning of the more difficult differentiation.

Attention and Feedback

Does the subject need to be actively attempting to distinguish among the stimuli, as in
sorting them into same–different categories? Perceptual learning seems to require some
intention to focus on differences among the stimuli. Gibson (1969) said that perceptual
learning is an active process of exploring stimuli in an attempt to obtain information about
them. Perceptual learning does occur in the absence of experimenter feedback about
performance. Yet the demands of the tasks we impose on our subjects suggest there are
dimensions that separate specific stimuli. The experimenter’s instructions “Are these
scribbles the same or different?” suggests we should look for differences and thus insti-
gate active processing.

OTHER EFFECTS OF STIMULUS EXPOSURE

Simple exposure to a stimulus produces learning that can be revealed by a variety of other
behavioral outcomes. Habituation is just one outcome. Learning from exposure to a stim-
ulus can be manifested in several other phenomena. Research on other types of learning,
such as the mere exposure effect shows that learning about stimuli occurs in the absence
of intention to learn.

Preference for Familiar Stimuli

Exposure to a stimulus sometimes leads to an affective, or emotional, increase in the
preference for the stimulus. This is known as the mere exposure effect (Zajonc, 1968),
so-called because the stimulus has merely been presented in the absence of rewards,
problems, or other tasks that might actively evoke stimulus processing. Both animals and
humans show an increased liking of familiar stimuli. Abraham Maslow (1937), later known
 Other Effects of Stimulus Exposure 41

for his contributions to humanistic psychology, conducted a study in which his experimen-
tal participants performed a number of tasks across several evenings in a laboratory set-
ting. Some tasks were repeated, for example, copying sentences, completing personality
inventories, or reading lists of foreign names. Other tasks were done once. At the end of
the study, the participants were asked to rate their preference and liking for the tasks. The
subjects liked the repeated tasks, but more than that: They liked the familiar lab, the
familiar experimenter, and even the familiar pictures on the wall!
Animals also develop a preference for the familiar, especially places or foods (Hill,
1978). Rats are at first wary of a new food, taste, or odor. This neophobia, or literally “fear
of the new,” declines with continued (safe) exposure to the food. Rats exposed to a sac-
charine solution will at first drink very little, but eventually come to prefer it over less
familiar tastes. Apparently, familiarity breeds liking and not contempt. See Box 2.2 for a
further discussion of food neophobia.

BOX 2.2 AVOIDANCE AND PREFERENCE FOR NEW FOOD TASTES

Many animals are reluctant at first to consume new foods. Taste neophobia makes good
evolutionary sense, to be cautious about new foods whose effects on the body are unknown.
Some phenomena we have considered in this chapter—habituation and mere exposure—
suggest that taste neophobia could be overcome by simple exposure to the new food.
Indeed, animal researchers often must incorporate this as a preliminary stage in a research
study. We realize the animal subjects will need to become familiar with the various food
reinforcers to be employed.
For example, Domjan (1976) gave rats daily periods of access to a solution of saccharin in
water, a new taste for the rats. Even though saccharin is sweet tasting even to rats, con-
sumption was weak at first. It increased across the 20 days of the study. Neophobia was
reduced by exposure.
Is the reduction of taste neophobia purely a habituation effect? Taste learning might involve
something more, specifically learning that the food is safe to eat. This is referred to as the
learned safety hypothesis. Learning that a food is not poisonous is different from learning to
ignore a stimulus that is without consequence, as a habituation explanation might suggest.
Rats given a single exposure to a distinctive flavor later had difficulty in learning an aversion
to this taste when it did cause illness (Siegel, 1974).
Getting children to eat new foods is often a challenge. Can exposure to a new food
increase liking? Most of us have heard the line “I tried it once; I didn’t like it.” The problem is,
once is not enough. Studies of adults and children have shown that many exposures, 10 or
more, may be necessary to induce acceptance of new cheeses or juices (Pliner, 1982; Birch
& Marlin, 1982).
By contrast, a single exposure to a taste in 4- to 7-month-olds sufficed to make a new food
acceptable, with some generalization to another food within the same category (i.e., between
baby food peaches to pears or from carrots to corn; Birch, Gunder, Grimm-Thomas, & Laing,
1998). Infants have had less experience with tastes than toddlers have had, so one exposure
may have a greater impact than does a single taste for a 4-year-old. So, new foods may be
accepted more readily at an earlier age.
Habituation can play a role in regulating meal duration and meal termination. Repeated or
extended exposure to a taste over short time periods decreases preference. Thus, habitua-
tion can occur to the taste of a single food sampled several times within a tasting session,
42 Habituation and Other Forms of Stimulus Learning

leading to a decrease in preference or consumption (Booth, 1990; Rolls, 1990). Sometimes

that first bite of your favorite food is the best tasting. For example, after consuming one meal
of crackers and cheese, subjects will eat less of a second meal of the same food but will eat
more if a different food is offered. This taste habituation is transient, and the overall liking will
return after a period of abstinence from the taste.
The implication of this finding is that meals with few different tastes will produce more
taste habituation, leading to less overall consumption and less likelihood of meal resumption.
Your cafeteria’s strategy of making everything taste the same might actually promote health-
ier eating.

In a demonstration of the mere exposure effect, college-aged and elderly subjects

were shown a series of Japanese ideograms (Wiggs, 1993). These are complex symbols
representing words and were unfamiliar to the American subjects. Later, these ideograms
and some new ones were shown and the subjects were asked to rate their “liking” of
each stimulus on a 7-point scale. The results, shown in Figure 2.8, show average liking of
stimuli by the two age groups. Stimuli that had been presented three or nine times previ-
ously were more liked than stimuli seen once or never before. These liking effects were
found with both age groups.
According to the cognitive theories of stimulus learning, exposure leads to the forma-
tion of a memory for the stimulus. We could have tested memory by simply asking sub-
jects if they remembered the stimulus (e.g., have them choose the familiar one from an
array of distractors). But what if the subjects did not remember the stimulus that had been
presented? Would they still prefer it? The intriguing possibility is that emotional affect,

5.0 Young
4.9
4.8
4.7
Elderly
Mean Liking Rating

4.6
4.5
4.4
4.3
4.2
4.1
4.0

0 1 3 9
Frequency

Figure 2.8
Liking for Japanese Ideogram Stimuli. Graph plots mean liking rating by elderly and young adult subjects who
had previously seen each stimulus 1, 3, or 9 times, or had never seen it.
Source: From “Aging and Memory for Frequency of Occurrence of Novel, Visual Stimuli: Direct and Indirect Measures,” by C. L.
Wiggs, 1993, Psychology and Aging, 8, p. 406. Copyright © 1993 by the American Psychological Association. Reprinted with
permission.
 Applications 43

expressed as a liking or preference, is a separable judgment from conscious memory for

the stimulus. Kunst-Wilson and Zajonc (1980) tested this hypothesis by flashing pictures
of irregularly shaped nonsense objects at a speed faster than their subjects could con-
sciously perceive them. Later, the subjects were shown pairs of shapes and were asked
one of two questions: Which shapes had they seen before and which shapes did they like
better? Performance on the memory test for the previously seen shapes was at chance
level: The subjects could not correctly distinguish between the old shapes and new ones
presented during testing. This finding seems perfectly reasonable, because the images
had initially been presented for such a short duration that they could not be consciously
perceived, much less remembered. However, when asked which objects they liked, there
was a consistent preference for the previously presented shapes. The subjects seemed to
prefer these objects even though they did not recognize the shapes as familiar.
The study by Kunst-Wilson and Zajonc (1980) is an important demonstration of the
fact that experience can have a variety of effects, not all of which are available to con-
scious recall.
Why does mere exposure make a stimulus more pleasing? The perceptual fluency
hypothesis states that we affectively prefer stimuli that are easily perceived. That is,
images, sounds, tastes, and so on that are more readily perceived because of previous
experience are more preferred. For example, several exposures to a difficult-to-read word
increases our fluency in reading that word, and also increases preference (Reber,
Winkielmanm, & Schwartz, 1998).

Recapitulation: The Effects of Repeated Stimulus Presentation

Exposure to a given stimulus produces varied, and sometimes conflicting, results. A
series of repetitions of a meaningless stimulus can lead to habituation, or a decrease in
orienting to the stimulus. Exposure can also increase the liking or preference for the
now-familiar stimulus, what we called the mere exposure effect. Familiarization with a
stimulus can facilitate new learning, what Gibson called perceptual learning. Presentation
of a fear-associated stimulus can produce sensitization, or an increase in the orienting
response to the stimulus; or potentiated startle, an enhanced response to another stim-
ulus. This pattern of effects from ostensibly the same manipulation, the simple presenta-
tion of a stimulus, is disconcerting. However, there are rules making the outcome of a
particular manipulation more predictable.

APPLICATIONS
Habituation certainly occurs outside the laboratory, often in some fairly mundane ways.
The reminders you posted on the refrigerator, too far in advance, are unnoticed by the
time you need to act on them. You decide to study in a new place thinking there will be
less distraction, but in fact there is more distraction as you orient to this new environ-
ment. In a laboratory simulation, students attempted to study while office noises and
speech sounds were played in the background (Banbury & Berry, 1997). The noise initially
disrupted learning, but after 20 minutes of exposure, the noise was less disruptive. As
with other forms of habituation, a brief interval without stimulation restored the disruptive
effect of the noise, that is, there was spontaneous recovery.
44 Habituation and Other Forms of Stimulus Learning

Other instances of habituation have considerably more significance. Warning labels

have been on cigarette packages for decades. Heavy smokers may be exposed to as
many as 7,000 warnings per year, but they have probably habituated to the messages.
Several countries introduced new warnings that are larger and are more graphic (e.g.,
pictures of diseased lungs). A cross-national study found that in countries with graphic
warnings, such as Canada and the United Kingdom, more people noticed the warnings,
even several years after introduction, than noticed the blander warnings in the United
States (Hammond et al., 2007).
Another source of concern are all those warnings we receive on our computers and
phones daily. IT people (information technologists) worry we have become so accus-
tomed to them that we do not read them; we just click OK to remove the warning from
our screen. In doing so we miss changes in the warnings or new urgency to the mes-
sages. Vance et al. (2018) conducted a real-world simulation by presenting warnings
across multiple sessions. In one study, there was a decline in eye movements directed
towards warnings, both within each daily session, and overall across the five days of the
study. Clearly the participants were paying less attention to the messages. In a second
study, the researchers were tracked in their granting permissions for downloading, shar-
ing, and other risky behaviors across 15 days. In this case, adherence to the warning
declined. However, an important finding was that attention could be maintained by vary-
ing the warnings with different versions, formats, or graphics. Attention still declined, but
there was less habituation than to static warnings.

Exposure Therapy for Fear

An application of clinical importance is the use of stimulus exposure in treating anxiety
and phobias. Exposure therapy is the use of repeated exposures to fear eliciting stimuli,
situations, or events to reduce the fear. The method uses controlled exposure to feared
stimuli or situations. If you have a fear of flying, your therapist might at some point pre-
scribe a long plane flight. If you fear snakes, attendance at the Indiana Jones Film Festival
may be suggested.
Exposure therapy often starts off with presentations of a milder version of the phobic
object and gradually increases its intensity or proximity. An example from animal behav-
ioral therapy is a report of a large dog named Goliath who was fearful during thunder-
storms. The dog ran around the room, jumping on furniture, knocking things over, and
generally causing mayhem. Exposure consisted of playing a thunderstorm recording,
starting at a low volume that did not elicit fear. The volume was gradually increased as the
dog came to tolerate the noise without adverse reactions, until the recording could be
played at high volume. The researchers said the dog did not react to real thunder sounds
afterwards (Tuber, Hothersall, & Voith, 1974).
One proponent of exposure therapy is David Barlow (e.g., Barlow, 2000). Rather than
trying to gradually acclimate to the fear, Barlow advocates maximal arousal and contact
with the feared stimulus or situation. (This form of exposure therapy is sometimes called
flooding or implosion therapy). For example, for a young man who feared closed places,
the advice was to shut himself in a small space for as long as he could take it. The man
repeatedly locked himself in the trunk of his car until the last such experience simply left
him bored (Slater, 2003). Patients are sometimes encouraged to “juice up” on coffee to
 Applications 45

increase their feelings of anxiety during exposure treatment—for example, just before
giving a public speech. (This could be the basis for a new reality TV show).
A novel technique for exposure is virtual reality therapy. For example, to treat the fear
of flying, an individual is presented with flying-related stimuli while wearing a virtual reality
helmet. Computer-generated visual images are displayed on the inside of the helmet’s
visor, and airplane sounds are presented over headphones or from large bass speakers
(which also produce vibrations that mimic an actual plane). Takeoffs and landings can be
simulated, as well as different flying conditions. In one study, four sessions of virtual real-
ity therapy were as effective as a control therapy conducted in actual (nonflying) airplanes.
The treated subjects in both conditions reported less fear and anxiety about flying, and 76
percent actually took a post-treatment air flight to prove the point (Rothbaum et al., 2006).
Do exposure therapies work? They have proved difficult to evaluate because of the
many variations across clinical research studies: different versions of exposure therapy;
exposure to actual phobic stimuli or exposure to imagined stimuli; students who volun-
teer as subjects versus clinic patients who seek therapy; and the criteria for evaluating
success of the therapy. However, the consensus is that exposure treatments for specific
phobias (heights, animals, flying, etc.) are effective by several measures: self-reported
distress; decreased arousal in skin-conductance or heart rate; and greater willingness to
approach feared stimuli (e.g., a snake or flying in a plane; Choy, Fyer, & Lipsitz, 2007).
Exposure is also effective in treating PTSD. In these days of evidence-based medicine,
exposure therapies have strong scientific support for their efficacy. However, there is
reluctance by some people to expose themselves to the thing they most fear, and by
therapists to employ exposure (Foa, Gillihan, & Bryant, 2013).

Needle Fear
The coronavirus pandemic sparked concern over needle and injection fears. One pre-
Covid survey found that avoidance of flu vaccination because of needle fear occurred in
27 percent of hospital employees, 18 percent of long-term care workers, and 8 percent of
hospital health case workers (McLenon & Rogers, 2019). Needle phobics try to avoid all
injections, including those medical procedures in which an injection or blood draw may be
required. Needle fears encompass a group of different fears. Some people fear the pain
of injection; some have a blood phobia; and some fear they will faint (and may have
fainted in the past).
Fear of needles and injections is often treated with exposure therapy. Individuals are
exposed to a hierarchy of needle related images or videos, working up to exposure in
real-life settings: handling a needle or watching someone receive an injection. Exposure
is a component of cognitive behavioral therapy, which also includes instruction, partici-
pant modeling, and discussing the clients’ beliefs about the fear or the negative conse-
quences of injection.
Research on the efficacy of needle exposure therapy has not been encouraging (see
McMurtry et al., 2016). One problem is the quality of the scientific research, similar to the
problems mentioned earlier in evaluating other forms of exposure therapy (e.g., inade-
quate control conditions, varied participant characteristics). The measures of effectiveness
are often patient self-reports. Most people will say their fear has decreased. Since needle
phobia is a collection of different fears, each might require different implementation.
46 Habituation and Other Forms of Stimulus Learning

Positive results are sometimes evident immediately after therapy when the client actually
permits an injection. Unfortunately, treatment effects often dissipate in delayed testing.
If exposure to injection-related images was sufficient to habituate fear, the intensive
news coverage in the years following introduction of the Covid vaccines should have
eliminated needle fear of everyone in America. “Each night it’s the same. Story after story
on the TV news is about the Covid vaccination effort, and they are all illustrated with
footage of needles sinking into exposed upper arms” (Appleby, 2021). One would think
that so much exposure would substitute for therapy.

SUMMARY
What Is a Stimulus?
As used in biology, the term refers to a change in the physical or chemical environment
that elicits a reaction. In psychology, the stimulus might be external, such as tones and
lights; or internal, such as hormone secretions, muscle tension, or difficulty breathing; or
mental, such as a thought or memory.

The Orienting Response

A novel stimulus elicits an orienting response, a composite of physiological and behavio-
ral reactions that includes startle, sense-receptor focusing, increased arousal, and a read-
iness for fight or flight. A familiar stimulus that is meaningful or unexpected can also elicit
orienting.

Habituation
Habituation is the decrease in size or frequency of the orienting reaction to a stimulus that
is repeatedly presented. Habituation is a simple form of learning. The orienting response
to neutral or innocuous stimuli can habituate. Defensive reactions to painful and noxious
stimuli also show habituation.
Habituation is studied via responses such as startle, eyeblink, or the skin conduct-
ance response. A method used to study habituation in human infants is to measure the
duration of eye fixations to visual stimuli. Novel stimuli elicit more fixations, which
decrease with repetition of the stimulus.

Parametric Features of Habituation

Thompson and Spencer described several characteristic features of habituation. These
features represent a standard set of criteria for evaluating habituation across different
species, tasks, or responses. More frequent repetitions of a stimulus produce a greater
decrement in the size or frequency of the response.
If the stimulus is withheld for a period of time, the response tends to spontaneously
recover.
If habituation is repeated over a series of sessions, habituation occurs more quickly.
Short-term habituation is the decrease in response from one stimulus presentation to
the next. Long-term habituation is the enduring overall decrease in response across suc-
cessive sessions.
 Summary 47

Habituation is affected by the spacing between successive stimulus presentations.

Massed stimulations produce more habituation in the short term; spaced repetitions
sometimes produce more long-term habituation as assessed hours or days later.
The presentation of a different stimulus will temporarily restore orienting to the first
stimulus, a phenomenon called dishabituation.
Habituation is stimulus-specific. Responding diminishes specifically to the repeated
stimulus, not to all novel stimuli. Habituation may generalize to similar stimuli.

Explanations of Habituation
A response decrement could occur for reasons other than learning. Sensory adaptation
or effector fatigue are two such nonlearning possibilities.
A biological approach is to study habituation in Aplysia. Siphon withdrawal, controlled
by a few identifiable neurons, shows many of the parametric features of habituation. By
measuring nerve potentials on the stimulus and the response sides of the siphon circuit,
Kandel found that habituation is a decrease in the sensitivity of the motor neuron to stim-
ulation by the sensory neuron.
Cognitive theories hypothesize that a memory of the repeated stimulus is formed.
Subsequent sensory input is compared to this memory. Orienting occurs when a new
stimulus differs from what is in memory, and habituation occurs when there is a match to
what is in memory. Cognitive theory also says that a stimulus can be remembered in
either short-term or long-term memory. Orienting occurs in the absence of an expected
stimulus (the missing stimulus effect).

Sensitization
Dual-process theory states that the response to repetitive stimulation reflects the combi-
nation of habituation and sensitization. Some neurons show habituation, whereas other
neurons react to intense levels of stimulation and increase responsiveness. For example,
the startle response actually increases across stimulations if tested against a sensitizing
loud background noise level.
Fear potentiated startle is a magnified startle reaction due to the sensitization or the
presence of other sources of arousal, such as anxiety or other related stimuli.

Perceptual Learning
Exposure to variations in a stimulus leads to learning about differentiating features of that
stimulus. After experience with a number of similar stimuli, such as the scribbles in Gibson’s
research, we can learn to perceptually differentiate one scribble from a different one.

Factors Affecting Perceptual Learning

Perceptual learning is enhanced by exposure to contrasting stimuli. Simple repetition of a
single standard stimulus does not lead to discriminative perceptual learning.
Initial experience with an easy discrimination can facilitate learning of a more difficult
discrimination.
Learning to differentiate among similar stimuli requires some focus on the features
that distinguish among stimuli.
48 Habituation and Other Forms of Stimulus Learning

Other Effects of Stimulus Exposure

Stimulus repetition does not always lead to a decrement in responding as in habituation.
Mere exposure to a stimulus sometimes leads to a preference or liking of the stimu-
lus. Animals develop a preference for familiar places or foods, in contrast to neophobia, or
fear of the new.

Applications
Habituation occurs to warning labels on foods and cigarette packages. Warnings pre-
sented on phones and computers also habituate through repeated exposures.

Exposure Therapy for Fear

Stimulus exposure is used in treating anxiety and phobias. Exposure therapy uses con-
trolled exposure to feared stimuli or situations. Exposure begins with a milder version of
the phobic object and gradually increases in intensity or proximity across presentations.
Exposure may be to the actual phobic object itself, or the subjects can be asked to imagine
the phobic objects or situation.
The effectiveness of exposure therapy is difficult to evaluate because of many varia-
tions in subject populations, procedures, and assessment criteria used across clinical
research studies. Extensive research has shown that exposure therapy is effective in
reducing specific phobias and in the treatment of PTSD.
Fear of needles and injections is often treated with exposure therapy, although the
efficacy of needle exposure therapy has not been encouraging.
 CHAPTER

3
Classical Conditioning

CONTENTS

The Definition of Classical CS–US Relevance 65

Conditioning 50 Conditioned Inhibition 66
Historical Note 52 What Is Learned in Classical
Methods of Studying Classical Conditioning? 67
Conditioning 52 So, What Is Learned in Pavlovian
Representative Procedures 53 Conditioning? 68
What Stimuli Can Serve as CSs? 54 The Role of Awareness in
Basic Phenomena of Conditioning 55 Conditioning 69
Acquisition 55 Extensions of Conditioning 70
Extinction 57 Evaluative Conditioning 70
Generalization 58 Conditioning with Drug USs and
Discrimination 59 the Development of Tolerance 71
Second-Order Conditioning 59 Applications of Conditioning 74
The Role of Contiguity 60 The Conditioning Theory of Phobias 74
Summary of the Basic Phenomena 61 Overview of the Conditioning
Other Factors Affecting Theory of Phobias 76
Conditioning 61 Extinction as Therapy 76
Compound CSs 61 Summary 77

Most students have some general familiarity with Pavlov’s experiments. Ivan Pavlov (1849–
1946) was a Russian physiologist who received the Nobel Prize for his work in digestion. In
a typical salivary conditioning experiment on dogs, the sound of metronome ticking was
followed a few seconds later by the presentation of some food. The dog salivated when fed.
Several of the metronome–food trials were administered and salivation was monitored.
After several pairings, the dog began to salivate during the metronome (Pavlov, 1927/1960,
which is the standard reference to G. V. Anrep’s translation). Pavlovian conditioning, or clas-
sical conditioning, suffers from something of a multiple-personality problem. On the one
hand, it is considered to be a simple, almost reflexive form of learning. Conditioning simply
transfers a response from one stimulus to another. These beliefs underlie popular culture
references to someone who responds like “Pavlov’s dog,” a caricature promoted by writers
such as Aldous Huxley (in his Brave New World) to the Rolling Stones.

DOI: 10.4324/9781003227090-3
50 Classical Conditioning

On the other hand, researchers have long known that conditioning is more complex
than any popular stereotype suggests. Osgood, in the standard textbook of experimental
psychology, wrote, “Naivete with regard to conditioning is due…more to the fond hope
that the process would prove as simple as conventional diagrams imply. The phenomena
of conditioning are actually very complicated” (1953, p. 316). Contemporary theories have
a distinctly cognitive flavor in describing conditioning.

“Pavlovian conditioning is not a stupid process by which the organism willy-nilly

forms associations between any two stimuli that happen to co-occur. Rather, the
organism is better seen as an information seeker using logical…relations among
events…to form a sophisticated representation of the world”.
(Rescorla, 1988, p. 154)

Using the knowledge learned in conditioning, organisms can flexibly respond in an adap-
tive fashion.
Interest in classical conditioning goes through cycles. At first eagerly adopted as a
tool to study learning, Pavlovian conditioning was replaced by reinforcement learning with
its emphasis on the modification of voluntary behavior. However, classical conditioning is
once again popular. Some researchers view conditioning from an ecological or evolution-
ary perspective and consider its role in ensuring survival (Hollis, 1997). The application to
areas such as drug dependence and tolerance connect conditioning research to the field
of health psychology. And classical conditioning offers model systems for neuroscientists
to study the biology of learning, on biological levels from synapses to genetics.
Possibly foremost among our reasons for studying classical conditioning is to study
associative learning. Association refers to the connections that are formed between the
internal representations of events, such as stimuli and responses. A simple illustration is
word association. The word TABLE often evokes CHAIR as a response. But how are asso-
ciations formed? Classical conditioning is one tool to study the conditions under which
associative learning occurs.

THE DEFINITION OF CLASSICAL CONDITIONING

Simply put, a neutral stimulus, such as a tone, is paired with a significant stimulus, such
as food or shock. The development of a new response to the neutral stimulus (such as
salivation during the tone) indicates learning. This is more-or-less the definition you learned
in Psychology 101. In actual practice, classical conditioning includes a number of variations
and violations of the definition. The learning that occurs in classical conditioning can be
described on several levels: behavioral, as the learning of a new response; cognitive, as
the acquisition of knowledge about the relationship between stimuli; or neural, as the
pattern of synaptic changes that underlie conditioning.
Say we are to perform an experiment in which mild but aversive electric shocks are
to be presented randomly in time. Because of our ethical discomfort with shocking ani-
mals or college students, let us suppose the participants are all faculty in the Economics
Department. The shocks, delivered to the participants’ forearm, are unavoidable, but our
participants would desperately like to know when each is about to occur. We sound a tone
for a few seconds before each shock. What will the participants learn? After several
 The Definition of Classical Conditioning 51

pairings of tone followed by shock, the tone will probably come to elicit a behavioral reac-
tion of hand flexion; physiological reactions such as muscular tensing or bracing; and
verbalizable knowledge of the tone–shock relationship.
Classical conditioning is a procedure that incorporates several features illustrated in
the preceding example. First, two stimuli are presented. One stimulus, labeled the
unconditioned stimulus (or US), is motivationally significant to the subject at the start
of the experiment; for example, food or shock. The unconditioned stimulus elicits an
unconditioned response (or UR), often reflexively without prior training. In Pavlov’s
experiments, food as the US produced salivation as the UR. The second stimulus, the
conditioned stimulus (or CS) is a neutral stimulus such as a tone, light, or ticking metro-
nome. This stimulus is actually the to-be-conditioned stimulus at the onset of the experi-
ment. The CS is neutral in the sense that it does not at first elicit the same response as
does the US. For example, the tone does not evoke salivation. Certainly, if the tone is
novel, it will elicit orienting reactions (as described in Chapter 2).
The two stimuli are presented in a specified sequence, normally the CS first followed
by the US. In Pavlov’s experiments, a metronome ticked for a few seconds and then food
powder was blown via a tube into the dog’s mouth (see Figure 3.1). After a sufficient
number of pairings, the CS elicited some of the responses originally elicited by the US.
Thus, salivation occurred during the tone. This response during the tone is called the
conditioned response (or CR). The development of a conditioned response to the CS is
an indication that conditioning has occurred.

Figure 3.1
Pavlov’s Salivary Conditioning Apparatus.
Source: From “The Method of Pawlow in Animal Psychology,” by R. M. Yerkes and S. Morgulis, 1909, Psychological Bulletin, 6,
p. 257–273; their Figure 2, p. 264. Copyright © 1909 by the American Psychological Association. Reprinted by permission.
52 Classical Conditioning

Because the conditioned and unconditioned responses are both salivation, how do
we determine which stimulus is eliciting the salivation? Conditioned salivation usually
occurs during the CS, before the onset of the US. That is, the dog salivates during the
several seconds the metronome is on before food is presented. The CR was sometimes
referred to as an anticipatory response; salivation occurs in anticipation of food. We can
also intersperse CS-alone test trials, omitting the US, among the conditioning trials. In this
manner we can see response to the CS in the absence of any response elicited by the US.

Historical Note
In the early 1900s psychologists used a variety of methods to study learning, including
trial and error, problem solving, Type S and Type R conditioning, and the eponymous
Pavlovian conditioning. Hilgard and Marquis, in their influential text of 1940, introduced
the now accepted term Classical Conditioning for Pavlov’s method. Hilgard and Marquis
did not say why, but the context suggests they were distinguishing the original or initial
form of conditioning as the classic method.
More appropriate translations of Pavlov’s terms would be unconditional and condi-
tional. A response is unconditionally elicited by the US at the start of the experiment, but
a response to the CS is conditional on the pairing operation. However, psychologists’
habit of using the -ed ending is hard to break.

METHODS OF STUDYING CLASSICAL CONDITIONING

Pavlov’s procedure of salivary conditioning is rarely used today. The tendency is to use
smaller animals or to use humans, both of which are more economical; or to use prepa-
rations that are relevant to a specific application. A few exemplar methods referred to
often in this chapter are outlined in Table 3.1.

Table 3.1 Examples of the Stimuli Used and Responses Measured in Some Conditioning Tasks
Described in This Chapter

TASK Salivary Eyeblink Taste-Aversion Magazine

Conditioning Conditioning Learning Approach

A
Conditioning
Trials
CS ➔ US tone ➔ food tone ➔ airpuff saccharine ➔ poison tone ➔ food
\ \ \ \ \
UR salivation blink illness approach
B
Test Trials
CS only tone tone saccharine tone
\ \ \ \
salivation eyeblink avoid saccharine approach food
CR salivation blink aversion approach
 Methods of Studying Classical Conditioning 53

Representative Procedures
Eyeblink Conditioning
In eyeblink conditioning, the US is a puff of air directed toward the eye, which elicits an
eye-blink as the UR (and other responses as well, such as heart rate changes, but typically
only one response is measured at a time). A CS, such as a tone, can be presented just
before the airpuff US. The CR is an eyeblink that occurs to the CS. The eyeblink can be
monitored by recording electrical activity in the eye muscles from surface electrodes on
the skin around the eye. Alternatively, movement of the lid can be detected when the
blink breaks a light beam reflected off the eye. In either case, a signal is fed into a com-
puter. From this record, the size of the blink can be measured.
Eyeblink conditioning with human subjects is convenient in that participants are often
available, the apparatus is portable, and an experiment can be completed within a single
session. The eyeblink response has also been used with dogs, cats, and rabbits
(Gormezano, 1966; Osgood, 1953). Rabbits are especially accommodating (well, maybe
passive) and have been the source of much of our recent data. The human and rabbit
preparations have been useful in mapping out the neural circuitry involved in eyeblink
conditioning.

Skin Conductance Response

Another measure of conditioning is the change in the amplitude of the skin conductance
response, or SCR. The electrical conductivity of the skin varies with changes in the level
of emotionality of the subject. The SCR is measured via electrodes usually placed on the
arm or palm. A US such as a loud noise or a mild shock will cause an increase in the SCR
as the unconditioned response. On conditioning trials, a CS such as a tone or a light is
followed with the US. After several pairings, a conditioned elevation of skin conductance
occurs during the CS.
Eyeblink and SCR conditioning fit the traditional definition of classical conditioning,
using well-defined stimuli and responses. In modern research, however, the label of
“classical conditioning” is often applied to situations that deviate from the traditional in
any of several ways. In some procedures the CR is not the same as the UR, or condition-
ing is indirectly assessed by the effects that the conditioned stimulus has on other behav-
iors. These characteristics are illustrated in the following preparations.

Taste Aversion Learning

At one time or another, most of us have developed a food aversion after becoming ill, even
if the food itself did not cause the illness. Experimental studies of taste aversion typically
employ rats as participants. In taste aversion learning, a novel taste, saccharin-flavored
water for instance, is presented as the CS and is followed by the administration of an
illness-inducing drug as the US. The UR is politely referred to as “gastrointestinal distress,”
although typically no direct measurements are made of the magnitude of the rat’s illness.
After the animal has recovered, the taste CS is re-presented to assess the degree of
conditioning. Learning is indicated by a decrease in the amount of saccharin-flavored
water consumed, as compared to consumption prior to conditioning, or by avoidance of
the saccharin flavor altogether. Note that gastrointestinal distress to the taste CS is not
54 Classical Conditioning

directly measured as a CR. Instead, conditioning is indirectly assessed by the participant’s

tendency to avoid the taste, hence the label “taste aversion learning.” A variety of other
behaviors have been observed as indications of an aversion to the food. Cats shake it off
their paws, coyotes bury it, and rats will tip over the food cup (Gustavson, 1977).
Unlike more traditional responses, such as salivation or the eyeblink, taste aversions
are often acquired after a single taste–illness pairing, and can be learned with wide tem-
poral separation between the taste and the illness. For these reasons taste aversions can
be learned inadvertently. Medical patients undergoing radiation or chemotherapy some-
times develop an aversion to foods consumed prior to treatment (Bernstein, 1991).

Magazine Approach Conditioning

In contrast to aversive conditioning methods such as taste aversion, there are methods
used to study appetitive conditioning, that is, learning about the occurrence of desirable
stimuli such as food or water. In approach conditioning, a neutral stimulus, such as a tone
of a flash of light, is followed by the presentation of food in a specific location. After a few
pairings of the tone with food, the animal approaches the food location when the tone
sounds.
Approach conditioning usually occurs in a conditioning chamber, a Plexiglas box. (The
conditioning chamber is a version of the Skinner box described in the next chapter). A food
dispenser, called a magazine, is located in a recessed niche in one wall of the chamber.
Food pellets can be delivered into a tray. Delivery of foods pellet (the US) is preceded by
a tone (the CS). The subject, usually a rat or a mouse, learns to poke its head into the niche
during the tone, and this head-poke is recorded as the conditioned response (the CR).

Summary of the Methods of Conditioning

The traditional definition of classical conditioning is exemplified in salivary or eyeblink
conditioning: Reflex-eliciting stimuli are used as USs, the food or the airpuff, and the
conditioned response is similar in form to the unconditioned response, salivation or an
eyeblink. Broadening the domain of conditioning has demonstrated that the laws and
regularities of conditioning apply to a wide range of behaviors. In addition to the taste
aversions described here, conditioning has also been shown to play a role in aggressive,
territorial, and mating behaviors in animals (Hollis, 1997). Some methods represent mix-
tures of several forms of learning. For instance, the initial acquisition of a taste aversion
is classical conditioning. The subsequent avoidance of that taste is an instance of rein-
forcement learning (see avoidance learning in Chapter 5).

What Stimuli Can Serve as CSs?

Laboratory stimuli selected as potential CSs have several characteristics: they can be
precisely controlled (e.g., duration, intensity); they can be exactly described (pitch of a
tone, the color wavelength of a light); and they are reproducible (in your lab, and by some-
one else in their lab). The usual textbook examples of to-be-conditioned stimulus is the
onset of a discrete signal such as light or tone. Stimuli including touch, smells, and tastes
have been successfully conditioned. The sudden offset of an already present stimulus can
also serve as an effective CS (e.g., Logan & Wagner, 1962). In a film, the forest suddenly
becoming quiet is a sure signal that something dramatic is about to happen.
 Basic Phenomena of Conditioning 55

An important advance in contemporary research is the recognition that diffuse stimuli

of time or place can act as conditioned stimuli. Contextual stimuli, the place or environ-
ment in which training occurs, are readily conditioned (e.g., Dweck & Wagner, 1970;
Siegel, Hinson, Krank, & McCully, 1982). Thus, we may fear the dentist’s office, just as our
pets fear the veterinarian’s.

What Stimuli Can Serve as USs?

Unconditioned stimuli are typically events that have some biological significance to the
organism such as food and shock (Gunther, Miller, & Matute, 1997). The US can also have
acquired significance, such as photos, music, or words in evaluative conditioning.
Even though we usually measure a single unconditioned response (salivation to food,
or an SCR to shock), USs are complex events. They have specific sensory components,
for example, taste, touch, sound, as well as general emotional components, for example,
fear. Thus, we should expect conditioned responses to be multidimensional also.
We can ask the participant to imagine receiving the US rather than actually presenting
one. That is, “imagine you have been shocked” may be the aversive US (Dadds, Bovbjerg,
Redd, & Cutmore, 1997). The use of mental imagery and imagination form the basis of
some behavioral therapies such as systematic desensitization or aversion therapy.
Alternatively, the subject in an experiment might only observe another participant receiv-
ing the US. In a taste-aversion procedure, rats learned an aversion when they saw another
rat become ill after eating a certain food (the “poisoned partner” effect, Revusky, Coombes,
& Pohl, 1982).

BASIC PHENOMENA OF CONDITIONING

Acquisition
Acquisition refers to the development of a conditioned response as a result of CS–US
trials. Some procedures, such as eyeblink conditioning, require many pairings to produce
CRs and are ideal for plotting learning curves. Other procedures produce such rapid con-
ditioning that a single pairing is sufficient. Not surprisingly, taste-aversion learning is one
of these.

BOX 3.1 ANXIETY AND CONDITIONING

Different individuals show different amounts of conditioning. This holds for humans and ani-
mals. Pavlov described temperamental differences among his dogs that he believed affected
ease of conditioning. Surely people differ from one another also.
The theory of British psychologist Hans Eysenck (e.g., 1981) said that anxiety produces
higher levels of arousal in certain brain areas that should facilitate conditioning. Measures of
autonomic nervous system activity are elevated by anxiety, e.g., heart rate and skin conduct-
ance. Anxious individuals experience higher levels of emotional responsiveness, which would
add to the anticipation of an aversive US such as the electric shock used in lab experiments.
Individuals diagnosed with posttraumatic stress disorder (PTSD) are characterized by
heightened anxiety. Do they classically condition more rapidly? Orr et al. (2000) compared
conditioning in PTSD and non-PTSD individuals. The members of the non-PTSD group had also
56 Classical Conditioning

experienced traumas, such as combat experiences or motor vehicle accidents. The CS+, a
colored circle projected on a computer monitor, was paired with an electric shock to the fin-
gers as the US (described by the investigators as “highly annoying”). Conditioning was meas-
ured by the size of the skin-conductance response during the CS. During a habituation phase,
only the color CSs were presented. Even there, PTSD group made larger SCRs than the non-
PTSD controls, indicating greater reactivity to stimuli in an emotional situation (see Figure 3.2).
During the conditioning phase, the PTSD group showed larger reactions during the color
CS than did the control subjects. The responses of the non-PTSD participants actually became
smaller across conditioning trials, probably reflecting habituation to the shock. The PTSD sub-
jects showed increasing SCRs over conditioning trials. During a final, extinction phase, the
PTSD subjects persisted in responding to the CS, even though the shock was no longer given.
A cognitive approach to conditioning suggests that anxious individuals are more likely to
rehearse aversive experiences, such as the unpleasant US. To test this idea, Davey and
Matchett (1994) gave subjects conditioning with a loud noise US. Following this, an explicit
rehearsal stage was added in which the participants were encouraged to imagine, as vividly
as possible, the noise US. Testing showed that the anxious individuals had become more
fearful of the CS. The conditioned response increased without further conditioning experi-
ence, but through rumination about the stimuli.
The fact that anxiety can enhance conditioning may sound paradoxical at first. We usually
assume that anxiety inhibits learning. However, anxiety may retard performance in more

1.4 PTSD
Control
1.2

1.0
Mean SCR

0.8

0.6

0.4

0.2

5 10 15
Habituation Condition Extinction

Figure 3.2
Conditioning in PTSD and Control Participants. Mean size of the SCR (skin conductance response)
during the CS-only habituation trials, CS-shock conditioning trials, and CS-only extinction trials.
Source: From “De Novo Conditioning in Trauma-Exposed Individuals with and without Posttraumatic Stress Disorder,”
by S. P. Orr, L. I. Metzger, N. B. Lasko, M. L. Macklin, T. Peri, and R. K. Pitman, 2000, Journal of Abnormal Psychology,
109, p. 294. Copyright 2000 by the American Psychological Association. Adapted with permission.
 Basic Phenomena of Conditioning 57

complex tasks. For example, in discrimination training a CS+ is paired with the US, but a
CS− is not. High anxiety participants respond more to CS+ (as they did in the Orr et al. study
described above), but the anxious also responded more to CS−. They showed less discrimi-
nation in response to CS+ versus CS– (Orr et al., 2000).
Eysenck (1981) said that while high- and low-anxious individuals are exposed to aversive
life experiences, the high anxious unfortunately learn more fear from those experiences. The
anxious also discriminate less between stimuli that do or do not predict dangers. The result
is more generalized anxiety, provoked by more stimuli in the world.

Control Procedures
Although a conditioned response to the CS may indicate that conditioning has occurred,
the response could occur for other reasons. A rat that has been given poison may shy
away from all new foods and not only the conditioned taste. Eyeblinks and SCRs occur as
orienting and startle responses to the novel stimuli used as the CSs. Therefore, control
conditions are needed to demonstrate that the response we observe is due to the condi-
tioning procedure. Responses to the CS in a control condition can serve as a baseline
with which to compare responding by the conditioning group.
In one control procedure, both the CS and the US are presented but the two stimuli
are not presented together. Experimental participants receive CS–US pairs, and control
participants receive unpaired CSs and USs. In the random control procedure (Rescorla,
1967), the CS and the US are each separately programmed to occur at random times
during the experimental sessions. The idea is that the occurrences of the CS and US are
not correlated; they have a zero correlation. This equates the overall exposure to the
stimuli between the conditioning and control groups.

Extinction
Extinction is a procedure that would follow acquisition. Extinction is the presentation of
an already conditioned CS alone, but without the US. The result is a decrease and maybe
the eventual disappearance of the CR, which we call extinction of the response.
One might think (wrongly!) that extinction is the opposite of acquisition: If pairing the
CS and US leads to conditioning, then it seems logical that the process could be reversed
to remove conditioning. However, extinction does not eliminate the CS–US association,
but only suppresses it. We see this when the extinguished response reappears after the
CS has been withheld for a while. Then, re-presentation of the CS leads to a recurrence
of the conditioned response. This spontaneous recovery shows that responding to the
CS was only inhibited by extinction.
Extinction and spontaneous recovery are illustrated in Figure 3.3 using data from
salivary conditioning in dogs (Wagner, Siegel, Thomas, & Ellison, 1964). Within each ses-
sion, the response to the CS alone extinguishes. From the end of one session to the start
of the next, spontaneous recovery occurs. Extinction proceeds more rapidly day by day,
and less spontaneous recovery occurs, so eventually the CR nearly disappears.
The pattern just illustrated has a practical implication for what to expect if you try to
extinguish an unwanted conditioned response. Repeated extinctions likely will be
58 Classical Conditioning

Figure 3.3
Spontaneous Recovery of Conditioned Salivary Response in Dogs. Following CS-US conditioning, CS-alone
extinction trials were given across six days. Spontaneous recovery of the response, from the last trial of each
session to the start of the next session, is shown each day.
Source: Adapted from Wagner et al. (1964, p. 356).

necessary. For example, after a single session in which snake or spider fear were extin-
guished by exposure to the phobic stimuli, there was significant relapse when partici-
pants were retested after four weeks (Rachman & Lopatka, 1988). This is spontaneous
recovery. Additional sessions of exposure are needed to more permanently reduce the
fear.
Extinguished responses return in other ways. If the US is presented alone sometime
after extinction, the CR will return. This is called reinstatement; the lost response is rein-
stated. Also, extinction is context specific. If extinction occurs in a context different from
the conditioning context, the extinction will not carry back to the conditioning context.
This is called renewal.
One way to think of extinction is to consider that two associations are learned: first a
CS-to-US connection during acquisition, and then a CS-to-absence of US association dur-
ing extinction. Afterwards, the CS could potentially retrieve either of these two associa-
tions. An analogy is the need to learn a new password. At first there are two memories:
the old and the new passwords. You will slip and use the old password by mistake for a
while, until the new one becomes dominant.

Generalization
Stimuli that are similar to the CS will often also trigger a conditioned response, even
though the stimuli have never been paired with the US. New tones similar in pitch to the
 Basic Phenomena of Conditioning 59

tone CS will elicit a CR, although maybe not as large as that to the original tone CS. This
is called generalization: the conditioned response is elicited by stimuli that are similar to
the trained CS. To test for generalization, a novel CS, which has not been paired with the
US, is presented to see whether a CR occurs to it. Care needs to be taken in conducting
a generalization test and in interpreting the results. Extinction or learning that the test
stimuli are not paired with the US could counteract generalization that occurs.
Generalization may occur along many different dimensions of similarity, for example,
size, shape, color, and meaning. A story in the local newspaper described an alligator that
had become accustomed to eating marshmallows offered by residents in a golf course
community. The alligator soon took to eating golf balls, too.

Discrimination
Not all of the stimuli in the environment are paired with unconditioned stimuli. So, we
need to discriminate among stimuli that are, or are not, paired with USs. In discrimina-
tion training, one CS (sometimes labeled the CS+) is followed by the US and another CS
(labeled the CS−) is not. For example, a tone is paired with food, but a light occurs alone.
CS+ US trials and CS− trials are usually intermixed within conditioning sessions. We
would conclude a discrimination has been learned when the subject makes a conditioned
response to the CS+ but not to the CS−.
During the first trials of discrimination training, some responding to the CS− will
occur due to generalization from the CS+. Continued training counteracts the tendency
to generalize.

Second-Order Conditioning
In second-order conditioning, a CS that has already been conditioned is now used as a
US to condition another CS. For example, in the first phase a tone is paired with food. In
the second phase, a light is then paired with the tone. Test trials with the light alone are
used to detect the transfer of conditioning to it (see Table 3.2). Second-order conditioning
is shown by a conditioned response (such as salivation) to the light CS, even though the
light was never directly paired with the food US. Second-order conditioning is interesting
as an example of a behavioral syllogism: If light equals tone, and tone equals food, then
does light equal food?

Table 3.2 Phases in a Second-Order Conditioning Experiment

First-Order Conditioning Second-Order

Conditioning

Phase I Phase II Phase III

Stimuli presented CSI ➔ US CS2 ➔ CS1 Test CS2

CRs made CR to CS1 CR to CS1 CR to CS2?
Example tone ➔ food light ➔ tone light
\ \ \
salivation salivation salivation?
60 Classical Conditioning

It may be possible to use the light to condition a third CS, which would be third-order
conditioning. The general label “higher-order conditioning” is used to encompass all
orders of distance from the original US.

THE ROLE OF CONTIGUITY

We have repeatedly referred to CS–US pairings in classical conditioning, but so far without
precisely defining what a pairing means. The general idea is that the two stimuli need to be
contiguous, or close together in time, to become associated. The role of contiguity raises
two important questions: Does the sequence of stimuli matter? How close is close enough?
Pairing the CS and US usually means that the CS begins a short time before the US.
This is the default meaning. In forward conditioning the onset of the CS precedes the
onset of the US, for example, tone and then food, or tone and then airpuff (see Figure 3.4).
The duration of the CS-to-US interval that is effective depends on the response being
conditioned. As examples, eyeblink conditioning occurs rapidly when tone precedes the
airpuff by half a second. Salivary and SCR conditioning are obtained with intervals of 5 to
10 seconds between the CS and the US. Taste-aversion learning can occur effectively
with intervals of hours separating taste from the illness. Although there are specific
CS-to-US intervals that produce the fastest learning, conditioning occurs across a range
of forward intervals.
Does contiguity demand this forward sequence? In simultaneous conditioning, the
CS and US have onset at the same time: They are simultaneous. The typical finding is
that there is a poor conditioning to this CS. (Test trials in which the CS is presented alone
are needed to assess conditioned responding that would otherwise be masked by the

(a) Forward
CS
US

(b) Simultaneous
CS
US

(c) Backward
CS
US

(d) Unpaired Control

CS
US

Time

Figure 3.4
Sequential Arrangements of CS and US in Classical Conditioning. The temporal onsets of the CS and US in
four contiguity variations: Forward, Simultaneous, Backward, and Explicitly Unpaired (Control).
 Other Factors Affecting Conditioning 61

unconditioned response). In backward conditioning, the US onset occurs first and then
the CS is presented. Backward US-CS pairings produce conditioned responses under
certain circumstances (e.g., Wagner & Terry, 1975). However, conditioned responding in
the backward procedure is not as strong or enduring as is forward conditioning.
Why is there such a difference between forward versus simultaneous and backward
pairings? One explanation is that the purpose of conditioning is to produce an adaptive
response. A CS that precedes the US allows opportunity for a response to be made in
anticipation of the US. Second, maybe simultaneous and backward pairings do produce
associative learning, but these arrangements are not conducive to producing conditioned
responses. This is the learning–performance distinction. To phrase it casually, the organ-
ism may know the tone follows the shock, but it does no good to prepare for a shock that
has already come and gone.

Summary of the Basic Phenomena

Already we have seen how classical conditioning is sensitive to several procedural varia-
bles. Many events can serve as CSs or USs, both discrete stimuli (such as tone or food)
and diffuse stimuli (such as context). Conditioning occurs rapidly in taste-aversion learning
procedures, and more slowly in the eyeblink preparation. The ease of conditioning is
affected by the sequence of the CS and US and the interval of separation between the
two stimuli. Conditioning generalizes from the trained CS to similar stimuli, but a discrim-
ination can be learned between stimuli that are or are not paired with the US. Through
second-order conditioning, one CS that has already been conditioned can be used to
transfer learning to new CS.

OTHER FACTORS AFFECTING CONDITIONING

In the real world, a significant stimulus occurs in the presence of many stimuli but not all
available stimuli become conditioned. There is a process of stimulus selection by which
only certain of the stimuli become associated with the US. For instance, if you become ill
after eating, you develop an aversion to one of the foods, but not to the plate and fork you
used, the music that played in the background, or the other foods you ate. Each of these
stimuli is contiguous with the US, but only certain stimuli become conditioned. Several
factors determine which stimuli acquire conditioning: the presence of competing CSs,
how predictive each CS is of the US, and the relevance of the CS to the US.

Compound CSs
When two or more conditioned stimuli occur together before the US, each may become
conditioned but to varying degrees. In a compound CS procedure, two CSs, a tone and
a light as examples, are presented together and followed by the US. After a sufficient
number of pairings, the tone + light compound elicits a conditioned response. Each stim-
ulus is then assessed by separately presenting the tone and the light. There is often a
weaker conditioned response to each stimulus alone than to the compound (see
Figure 3.5). There is less salivation, for example, to the tone by itself or the light alone
than to the tone + light compound.
62 Classical Conditioning

Compound
Conditioning Overshadowing Blocking

CS1 + strong CS1 + CS1 US

CS2 US weak CS2 US CS1 + CS2 US
100
CS1 + CS2

80 CS1
CS2
Percent Conditioning

60

40

20

0
Compound Overshadow Blocking

Figure 3.5
Conditioning in Three Compound - CS Conditioning Procedures. Illustrative outcomes of amount of
responding to the compound (CS1 + CS2 together) and to each element (CS1 or CS2) alone.

The balance of conditioning that occurs to each CS can be altered in several ways,
most of which make intuitive sense. One CS that is more salient than another, because it
is louder, brighter, or otherwise more attention-getting, can overshadow the less-salient
stimulus. After conditioning to a compound made up of CSs of unequal intensities, the
more salient stimulus will produce a larger response than the less salient one.
Even a simple CS is actually a compound stimulus. A discrete CS, such as a tone or
light, occurs in the presence of contextual stimuli, the conditioning chamber. The discrete
CS usually overshadows contextual stimuli because it more accurately predicts the occur-
rences of the US.

The Blocking Effect

Leon Kamin observed that under certain conditions, one CS in a compound failed to
acquire conditioning. A new CS was added to another stimulus that already predicted the
US. For instance, first a tone was paired with the US, then a light was added to the tone.
Blocking occurs when the presence of an already conditioned stimulus blocks learning to
a new CS (e.g., Kamin, 1969).
During the first phase of a blocking study, one CS is conditioned. For example, a tone
is paired with the food (see Figure 3.5). In the second phase, the tone and a new stimulus,
a light, are presented simultaneously and followed by food. Our interest is in the amount
of conditioning to the added cue, the light. When the light is tested alone, there is little if
any response to it. The tone apparently blocked conditioning to the light. Even though
frequent light–food pairings were given, conditioning to the light was severely inhibited.
Why was conditioning to the added stimulus blocked? Kamin suggested that what
normally occurs in conditioning is that the sudden occurrence of the US is surprising,
 Other Factors Affecting Conditioning 63

which causes the participant to “retrospectively review” in memory what recent events
might have caused this US. (If you got a static electricity shock, you might think back:
“What did I just do to cause that?”) In the blocking procedure, the occurrence of the US
in the second phase was no longer surprising: It was signaled by the trained tone.

Prediction Error
The brain (mind) constantly makes predictions about the world around us and our interac-
tions in it. When the predictions are confirmed, life proceeds smoothly. In reaching for an
object on the table, the brain is (unconsciously) predicting the convergence of your hand
with the object. Unpredicted or surprising events set the occasion for new learning to
occur. The discrepancy between what is expected and what actually occurs leads to
adjustments in knowledge to bring future predictions in-line with the actual.
This adjustment of predictions has evolved into the concept of prediction error, an
important concept in neuroscience, neural networks, and mathematical models of condi-
tioning. When an individual first encounters a stimulus followed by an unexpected US,
there is a large discrepancy between what is predicted (probably nothing) and what actu-
ally occurs (maybe food or shock). This is a large prediction error. After learning that a
particular cue predicts the food there is no longer a discrepancy between expectations
and actuality. When this prediction error equals zero, new learning ceases, even if pairings
of tone and food continue.

Rescorla-Wagner Model
The idea that conditioning depends on the surprisingness of the US has been captured,
both conceptually and mathematically, in a formula by Robert Rescorla and Allan Wagner
(Rescorla & Wagner, 1972; Wagner & Rescorla, 1972). The Rescorla-Wagner model (or
the Wagner-Rescorla model, as it is known to Wagner’s students) provides a trial-by-trial
description of the learning that accrues to the CS. The theory particularly focuses on the
case in which multiple CSs are presented, as in the compound CS and blocking
experiments.
According to the Rescorla-Wagner model, “Organisms only learn when events violate
their expectations” (Rescorla & Wagner, 1972, p. 75). Conditioning is based on how sur-
prising or unexpected the US is. If the CS does not well predict the US, as would be the
case early in training, large gains in learning occur on those trials. As the predictive
strength of the CS increases across trials, smaller gains would accrue. This follows the
form of the traditional learning curve.
In the model, the maximum amount of conditioning (or prediction) of the US is repre-
sented by lambda (λ). (A larger or more intense US will have a higher value of λ). The
predictive strength of the CSs present is represented by V. The prediction error is the
difference between λ and V, or λ − V. If the difference is large, because the CSs have little
predictive power to begin with, each pairing will produce large increments in
conditioning.
The mathematical formula is written ΔV = f (λ − V ), which reads in English as delta V
(or the change in V ) is a function of lambda minus V.
An illustration of the course of conditioning to a tone + light compound is shown in
the left panel of Figure 3.6. On each trial, the learning rate arbitrarily is set at 50% for this
 64 Classical Conditioning

Compound-CS Control Blocking Condition

Tone Tone
Light Light
Lambda Lambda
100 100

V V
80 80

Percent Conditioning
Percent Conditioning

60 60

40 40

20 20

0 0
T1 T2 T3 T4 T5 T6 T1 T2 T3 T4 T5 T6
Trials Trials

Figure 3.6
Hypothetical Growth of Conditioning to Each Element of a Compound CS and Their Summed Strengths
(V), According to the Rescorla-Wagner Model. The left panel shows conditioning when V starts at zero. The
right panel shows the blocking effect when V has a positive value because of preconditioning to the tone.

example. (The rate determines the speed of learning, and would be affected by the inten-
sity of the US, the salience of the CS, etc.). Each trial produces new conditioning equal to
half the difference between the maximal amount of conditioning (λ) and the strength of
existing conditioning to the CSs (V ). Thus, on the first trial when conditioning (or V ) is
assumed to be zero, the increment is 50 percent, or halfway to 100 percent. On the sec-
ond trial, conditioning increases half the distance again, to 75 percent, half the distance
between 50 and 100. What is important to notice is that conditioning is divided between
the tone and the light equally in this example.
How does the Rescorla-Wagner model explain the blocking effect? As a result of the
first phase, the tone has acquired some associative strength. The right panel of Figure 3.6
shows the hypothetical course of conditioning during the second phase of a blocking
study, when both tone and light are presented. Trial 1 starts with a high predictive value
because of preconditioning to the tone. The difference between the amount of condition-
ing possible (λ) and the amount present (V, summing tone and light, even though the
latter has no strength to start with) is smaller. The result is that the trial-by-trial increments
in conditioning are small, and what little new learning occurs is divided between the tone
and light. The net result is little conditioning to the light.
The Rescorla-Wagner model has been successful in describing a variety of condi-
tioning phenomena. The model has also influenced recent developments in neurosci-
ence and neural network models of learning. (Rescorla and Wagner have each been
awarded the Distinguished Scientific Contribution Award by the American Psychological
Association).
 Other Factors Affecting Conditioning 65

CS–US Relevance
Another factor that affects conditioning is the relevance of the CS to the US. Referring
back to our taste-illness example, how likely is it that the illness was caused by the food
versus by the plates? In learning about cause-and-effect relationships in the world, organ-
isms may have a bias toward perceiving that something eaten is more likely to cause
illness than something seen or heard. A symmetrical property of relevance asserts that
sights or sounds in the environment are relevant signals for external stimuli such as
shocks or airpuffs.
Although the relevance of tastes to illness makes common sense, it may be that
tastes readily condition because they are more salient or intense CSs. We need a research
design that shows that interoceptive and exteroceptive cues are equally adequate, but
they differ in their relevance depending on the type of US used. Garcia and Koelling (1966)
demonstrated this in their now classic studies. (John Garcia was awarded the Distinguished
Scientific Contribution Award by APA in 1979). Rats were presented with a compound CS
consisting of external stimuli of tone and light, and an interoceptive stimulus of taste, all
simultaneously. This was arranged by allowing the rats to lick saccharin-flavored water from
a drinking tube. Each lick completed a circuit that beeped the tone and flashed the light.
Garcia and Koelling referred to this as bright-noisy-tasty water. This CS was paired with
either of two USs: for one group of animals, electric shock was delivered through the floor
of the conditioning chamber; for the second group of animals, an illness-inducing drug was
injected. (The design of this study is outlined in Table 3.3). Following this conditioning treat-
ment, the two groups were tested for fear of the exteroceptive stimuli (the tone and light)
or aversion to the interoceptive stimulus (the saccharin taste). The results showed that rats
that had been made ill after exposure to the compound CS refused to drink saccharin but
did not react to the tone/light. Conversely, rats shocked after the compound were fearful in
the presence of the tone/light but did not avoid drinking the saccharin flavored water.
Each type of CS, exteroceptive and interoceptive, was conditionable. But condition-
ing was selective: taste selectively associated to illness, whereas tone and light selec-
tively associated with shock. The Garcia and Koelling study nicely illustrates the advantages
of basic research in a laboratory setting. In this situation, the conditioned stimuli could be
highly controlled, ensuring equal exposure and contiguity of each with the US.

Table 3.3 Design and Results of the Garcia and Koelling Study on Stimulus Selectivity in Taste
Aversion Learning

Conditioning Phase Test Phase Response

Group 1 light/tone licking –> light/tone aversion to light/tone

+ foot shock or
licking –> sacc no aversion to taste
saccharine
Group 2 light/tone licking –> light/tone no aversion to light/tone
+ toxin or
licking –> flavor aversion to taste
saccharine
66 Classical Conditioning

What determines CS-to-US relevance? The theory of prepared learning asserts that
the evolutionary history of different species has prepared them to form certain associa-
tions (Rozin & Kalat, 1971; Seligman, 1970). Taste-aversion learning may be one of several
instances of prepared learning, which also include the human capacity to learn language
and biases toward acquiring certain types of phobias (e.g., fear of snakes, heights, and the
dark).
An alternative explanation of relevance is that the bias toward associating certain
classes of stimuli is learned (Davey, 1995). Selective associating occurs because an indi-
vidual’s prior experience with taste stimuli has shown them to be related to illness, and
other types of cues were not.
Conditioning can occur between nonrelevant CSs and USs, although maybe not as
readily as with prepared stimuli. Animals learn to avoid color, feeding containers, places,
and other exteroceptive cues (see Mackintosh, 1983).

Conditioned Inhibition
The anticipation of food or danger is obviously important for an organism’s survival.
Detecting the absence of food or danger is also important. Our references to conditioning
so far have actually been to excitatory conditioning. Simply put, in excitatory conditioning
a CS is associated with the occurrence of the US. By contrast, in inhibitory conditioning a
CS is associated with the absence of the US. Imagine a situation relevant to student life:
predicting the dreaded pop quiz. Unfortunately, cues are rarely available to signal on what
days the quiz (the US) will occur, hence the name “pop quiz.” Inhibitory conditioning
would occur if a cue reliably signals when the quiz will not occur. If the teacher arrives
empty-handed, then a quiz is not imminent. Instead of remaining tense and anxious, the
students can relax!
How is an inhibitor produced? Inhibitors must be actively trained by arranging what is
essentially a negative correlation between the CS and a specific US. This means this CS
occurs without the US; and the US occurs without this CS. The standard procedure is to
present the potential inhibitor, call it CSx (where x is the mystery stimulus), simultane-
ously with a conditioned excitor, but leave out the US. That is, the training sequence would
alternate CS+US trials and CS+/CSx–no US trials. To describe this procedure casually, at
those times when the US fails to appear after CS+, its absence is attributed to the CSx.
To make a stimulus an inhibitor, it must be presented when there is an expectation
that the US will occur, but the US does not occur. That is why CSx is presented simulta-
neously with CS+. The expectation of food or shock is disconfirmed.
Conditioned inhibitors often do not produce a measurable response: no eyeblinks, no
salivation. So how can the presence of inhibition be detected? One method is based on
the idea that inhibitory stimuli inhibit responding to excitatory stimuli. The suspected
inhibitor (CSx) is presented at the same time as an excitatory CS, to see if it will reduce
responding otherwise seen to the excitor (e.g., Neumann, Lipp, & Siddle, 1997). The
rationale is like adding a negative number to a positive number: The net result will be less
than the positive number was to begin with, for example, the sum of −3 and +5 is +2. So,
adding an inhibitor to an excitor reduces the size of the response to the excitor.
 What Is Learned in Classical Conditioning? 67

WHAT IS LEARNED IN CLASSICAL CONDITIONING?

Classical conditioning is an example of associative learning. Given the presence of four
elements in a traditional Pavlovian trial—the CS, the US, the UR, and eventually the CR—
we can ask what is associated with what? There are several theories of what is learned in
classical conditioning. (The two theories discussed are summarized in Table 3.4).
The stimulus–response (S–R) theory says the CS becomes associated with the UR.
What is important about the classical conditioning procedure is the CS (e.g., tone) is
paired with a response (e.g., salivation). Through conditioning the CS comes to elicit the
unconditioned response, or some portion of the UR. The presence of a US ensures that
an UR will occur contiguous with the CS. This response-learning theory seemingly fits a
simple description of what happens during Pavlovian conditioning: a tone now elicits an
eyeblink, salivation, or leg flexion, and so on. It even applies to subtle aspects of the
response. In one preparation, pigeons were conditioned to make a pecking response as
the CR by using either food or water as the US. Interestingly, the pecks took the form of
biting movements when the US was food (and eating is the UR), but sucking movements
when the US was water (and drinking is the UR). The pigeons were making the same
response that was elicited by the US (Jenkins & Moore, 1973). The appeal of S–R theory
is that it provides a simple and direct account of what is conditioned in classical
conditioning.
Although the S–R theory adequately describes some forms of the conditioned
response, there are several problems with it. One is that the CR is not always the same
as the UR. In heart rate conditioning, a shock will often elicit acceleration as the UR, but
the CR will sometimes be deceleration. If the drug morphine is used as a US, heart rate,
general activity, and sensitivity to pain all decrease. The CR, however, is the opposite in
each case: Heart rate, pain sensitivity, and activity all increase (Siegel, 1991; these studies
are described later in this chapter).
An alternative theory states that conditioning produces an association between the
CS and the US. This is the stimulus–stimulus (S–S) theory. The modern version of the S–S
theory is decidedly cognitive in orientation. The first step in learning is the formation of
memory representations of the CS and the US. Conditioning further involves acquiring an
association between the two representations, just as you have acquired an association
between the representations of TABLE and CHAIR.

Table 3.4 Comparison among Two Theories of What Is Learned in Classical Conditioning

Theory S–R S–S

Associated elements CS–UR CS–US

Example Tone–salivation Tone–expectancy for food
Evidence for CR often same as UR US devaluation
Criticisms CR often different from UR Does not specify what response will be
made
68 Classical Conditioning

An advantage, as well as a disadvantage, of the S–S theory is that it does not specify
what the conditioned response will be or what form it will take. Instead, knowledge is
acquired, with some flexibility as to how that knowledge is applied. The CR is not a rigid,
reflexive reaction that becomes attached to the CS.
Support for the S–S theory comes from studies of US devaluation. If the US is altered
after conditioning has occurred, will responding to the CS change to match the new value
of the US? For example, say a tone is first paired with an electric shock US. (This is psy-
chology; you knew there would be a shock eventually). Then, in a second phase, milder
shocks are given. This manipulation should devalue the US and its representation in mem-
ory. When we go back and retest the tone CS, will there be as much fear as occurred
during the first phase? Or will the fear be weaker, reflecting the less-intense US experi-
enced in the second phase? The results suggest that there is often a weaker CR to the
tone after as compared to before devaluation (Rescorla, 1987). According to the cognitive
version of the S–S theory, a CR of a specific size and shape is not acquired. Rather, a
response is available that can be flexibly altered depending on the current status of the
organism’s knowledge.
There are other theories of how conditioned responses are acquired. For instance,
one theory is that classical conditioning is a form of reinforcement learning. Conditioned
responses are reinforced, or more colloquially “rewarded,” because they make the US
more palatable or less aversive (Perkins, 1968). An eyeblink CR is reinforced because it
blocks the airpuff US from striking your eyeball. A salivary CR makes the dry food powder
US more palatable.

So, What Is Learned in Pavlovian Conditioning?

Among several theories of what is learned in Pavlovian conditioning, which explanation is
correct? Although each may be fine in some cases, none of the theories appears to be
sufficient for all situations. Part of any resolution to this question goes back to the learn-
ing-versus-performance distinction: What has the participant learned versus what does
the participant do? Stimulus–stimulus theory describes the knowledge about stimulus
relationships that is learned, but the stimulus–response theory describes the behavior
that is shown.
Other research suggests that conditioning produces multiple conditioned responses.
For instance, Wagner and Brandon (1989) point out that conditioning leads to both a gen-
eral, emotional response and a specific, localized response. When using a shock US, there
is a diffuse emotional conditioned reaction to the CS, such as fear; and a specific motoric
reaction, such as an eyeblink. As another example, we can point to the varied roles of
conditioning in advertising. One claim is that conditioning associates an affective feeling
with a product: By pairing the right images and sounds with the product name in a com-
mercial, the viewer will associate the product with pleasure, action, or fun. However, the
conditioning procedure may also associate the product with factual information, for exam-
ple, that the product is fast, effective, or long-lasting (Kim, Allen, & Kardes, 1996).

Evolutionary/Adaptive Approach to Conditioning

Evolution theory provides a rationale for comparing conditioning across species. It has
two significant, but opposite, implications. First, organisms with a shared evolutionary
 What Is Learned in Classical Conditioning? 69

history should have commonalities, in form and in psychological processes. Second,

adaptation implies that specializations in learning will evolve between species. This fea-
ture suggests studying the unique ways that conditioning has been modified to fit particu-
lar environments. Both of these features have been hinted at in this chapter. The general
principles of conditioning are in fact, general. One could broadly identify the principles of
classical conditioning without naming a particular species. Species specializations are
also evident: taste aversion learning in animals, human phobias, or differencent neural
mechanisms across species.

Learning in the Brain

A different approach to answering the question “what is learned” is to attempt to map out
the changes in the brain that correspond to conditioning.
The neural circuit involved in eyeblink conditioning has been mapped by Richard F.
Thompson and his colleagues and students (e.g., Thompson, 2005). Although a tone CS
and an airpuff US produce activity in several places in the brain, it appears that a final
common destination for the association is in the cerebellum. The cerebellum, literally the
“little cerebrum,” is in the lower back portion of the brain. An experimentally produced
lesion in the cerebellum of an animal prevents tone-to-eyeblink conditioning. The tone CS
is heard and registered in the brain. The airpuff US produces an eyeblink UR. However, the
tone–airpuff pairings do not give the tone the capacity to produce an eyeblink CR. Similar
results are found in case studies of people who have sustained damage to their cerebel-
lums. Normal eyeblinks occur; reflexive eyeblinks occur; but there is impaired eyeblink
conditioning (Daum et al., 1993).
Do Thompson’s results show that classical conditioning is localized in the cerebellum?
Yes and no. Eyeblink conditioning may operate through the cerebellum, but conditioning
of other reactions involves different areas. For instance, conditioned fear produced by
pairing a tone CS with a painful US involves the amygdala. If we measure skin conduct-
ance responses, individuals with cerebellum damage will make conditioned responses to
the tone, reactions that are mediated through other areas of the brain (Daum & Schugens,
1996). Even learning as simple as classical conditioning is represented in multiple areas
of the brain.

The Role of Awareness in Conditioning

If classical conditioning is considered to be a simple, possibly even primitive form of
learning, then cognitive factors and conscious awareness are not necessary for condition-
ing to occur. What is the role of conscious awareness in conditioning? Evidence from
several different sources can be considered.
It is certainly the case that some college student participants can verbally report the
CS–US contingency that they have experienced in eyeblink and SCR conditioning (e.g.,
Shimp, Stuart, & Engle, 1991). But is awareness of the CS–US contingency necessary
for conditioning? The data from unaware participants are inconsistent. In some cases,
experimental participants who did not show conditioning also reported that they were
not aware of the CS–US arrangement (see review by Dawson & Schell, 1987). But in
another study, unaware and aware participants responded comparably (Papka, Ivy, &
70 Classical Conditioning

Woodruff-Pak, 1997). Interestingly, participants who reported that the tone was followed
by an airpuff were not aware that they had made eyeblinks during the tone.
Conscious awareness of the conditioning procedure can be minimized by using dis-
tracting secondary tasks. For example, eyeblink conditioning occurs even when the par-
ticipants are simultaneously reacting to words presented visually or watching a video
(Papka et al., 1997). We can also cite the case of a man with spinal cord injury who
showed conditioned control of his bladder, even though neither the stimuli nor the
responses involved could be felt by the patient (Ince, Brucker, & Alba, 1978). Together,
these sorts of observations suggest that Pavlovian conditioning in humans does not
require conscious awareness.
In some instances, awareness of the CS–US relationship does determine responding
to the conditioned stimulus. In a conditioning preparation such as the eyeblink or SCR,
simply telling our human participants that extinction is about to begin, and reinforcing this
belief by disconnecting some of the wires, leads to an immediate cessation of the CR.
There is no reaction to the very next CS presentation. However, awareness that an aver-
sive US will no longer occur does not always override all conditioned responses.
The question about the role of awareness is not an either/or choice. Classical condi-
tioning occurs throughout the nervous system, at levels which obviously involve con-
scious awareness and levels which are obviously beyond awareness. For instance,
classical conditioning can occur:

• at the level of individual neurons, as in isolated spinal cells;

• in sub-cortical regions, as shown in animals without cortex and by people with dam-
aged cortex;
• in the autonomic nervous system as hormonal and immunological responses;
• in the cortical areas, as occurs in some instances of human conditioning, and in many
complex phenomena in animals such as discrimination and second order conditioning.

EXTENSIONS OF CONDITIONING
Evaluative Conditioning
In evaluative conditioning an affectively neutral stimulus, one that is neither particularly
liked nor disliked, is presented along with another stimulus that already evokes a strong
affective evaluation. We then see if the emotional tone of the neutral stimulus changes as
a function of this conditioning experience. The neutral and affective stimuli are often
words, names, or pictures. In evaluative conditioning the CR and UR are measured by the
participants’ ratings (or evaluations) of the stimuli (Martin & Levey, 1987).
In one example (Hammerl, Bloch, & Silverthorne, 1997), the USs were scenic photo-
graphs of city, park, and public locations that were rated very high or very low on a scale of
“liking.” Other pictures that had neutral ratings were used as the CSs. Each neutral picture
was presented for 2 seconds, followed by a liked or disliked picture for 2 seconds. Five
pairings were given. What the participants actually see is a long series of slides arranged in
a particular sequence in which a neutral slide is immediately followed by one of the liked
 Extensions of Conditioning 71

(or disliked) slides. A cover story is used to try to mask the true purpose of the experiment.
The participants are later asked to rate the individual pictures for liking. The neutral pictures
that had been paired with liked stimuli received higher positive ratings than they had
before; the neutral slides paired with disliked pictures now received lower ratings.
Media advertisements for pharmaceuticals often pair a product with “feel good”
images. The goal is to associate positive dispositions with the product. In a simulation
study, participants saw a flu remedy that was paired with background images that elicited
positive feelings. The participants rated the drug as more effective and safer. Note that the
advertisements said nothing about safety or efficacy. The intent was simply to change
attitudes (Biegler & Vargas, 2016).
Evaluative conditioning is not without its skeptics. Some theorists question whether,
in the absence of traditional CRs and URs, evaluative conditioning is really classical con-
ditioning. Others question whether the preferences are based on conscious knowledge
(De Houwer, Baeyens, & Field, 2005). As we will see, some of these same concerns are
raised about other classical conditioning procedures.

Conditioning with Drug USs and the Development of Tolerance

Pavlov noted that after repeated injections of morphine, the dog became nauseous sim-
ply at the sight of the hypodermic needle. Contemporary work on conditioned drug reac-
tions has broadened our conception of classical conditioning in several ways: Drugs can
act as unconditioned stimuli; contextual stimuli can function as CSs; and the conditioned
response is sometimes opposite to the UR.
These factors are illustrated in the development of drug tolerance. Repeated admin-
istrations of a drug can lead to a reduction in the drug’s effectiveness. Rats given a series
of injections of opiate drugs such as morphine or heroin develop a tolerance, as measured
by respiration, body temperature, and pain threshold. What role might classical condition-
ing play here? Is tolerance a CR?
One way to test for classical conditioning after drug exposure is to replace the drug in
the injection with an inert saline solution. This is a placebo injection. The idea is to monitor
the body’s conditioned response to the injection procedure (the CS) in the absence of the
drug (the US). For some drugs, the physiological reaction is opposite of that initially elic-
ited by the drug. For example, morphine raises body temperature, but the saline injection
lowers temperature. Morphine has analgesic effects, in reducing pain; the placebo
increases pain sensitivity (see Siegel, 1991, for a review).
Why is the reaction to a saline injection opposite to that of the morphine? One theory
proposes that conditioning can lead to the development of conditioned responses that
are the opposite of the unconditioned responses. The UR to a stimulus stays the same
over presentations. But a second response, the CR, gradually develops over trials, and is
opposite in direction from the UR. Siegel (1991) refers to this conditioned response as a
compensatory response, one that counteracts the effects of the drug itself to maintain
bodily homeostasis.
Another feature of the drug conditioning experiments is the recognition that the con-
text in which the drug occurs is a CS. This means that the original reaction to the drug can
be reinstated by giving the drug in a different context where the compensating response
72 Classical Conditioning

Table 3.5 Tolerance to the Effects of Morphine on Pain

Test Situation Paw-withdrawal Pain Threshold Hypothesized

Latency (seconds) Elements Present

Saline given each day 11 Normal pain threshold No CR, no UR

Morphine Day 1 28 Pain suppression UR to morphine
Morphine Day 4 11 Tolerance (pain threshold UR to morphine +
returned to baseline) compensatory CR
Day 5, switched to 4 Ouch! Pain threshold is Compensatory CR
saline lower than baseline only

(or CR) is not evoked. An illustrative pattern of results is shown in Table 3.5. After animals
had been given frequent morphine injections, they had the same threshold for pain as
animals given saline injections. This demonstrates tolerance to morphine. Giving the mor-
phine in a different room reinstated morphine’s usual pain-suppressing property. The
effectiveness of the drug returned simply by testing the rat in the “nondrug” room.
Normal pain threshold is measured before morphine is given. Morphine at first
reduces pain sensitivity, producing a long latency before reacting to pain, across four days
of morphine administrations pain returns (what we call tolerance to morphine). The sub-
stitution of saline in morphine tolerant animals makes them even more sensitive to pain,
indicating a compensating CR. (Based on Siegel, 1975).
In one particularly dramatic demonstration of room-specific tolerance, rats were given
heroin or saline in different rooms. Rats that had a tolerance to an otherwise lethal dose
of heroin in the heroin room showed a 50 percent increase in the death rate when injected
in the nondrug room (Siegel et al., 1982). This may explain why some drug users overdose
when taking amounts they previously tolerated. If the drug is taken in a different environ-
ment from usual, the compensating responses may not be evoked, and so the drug has
its full effect. In a retrospective interview study of overdose victims who lived to tell about
their experience, 7 out of 10 reported changed environmental conditions associated with
the overdose reaction (Siegel, 1984).
By extension, the compensatory-response model also addresses withdrawal symp-
toms. Say that an addicted individual is confronted with cues associated with drug taking:
the place where opiates have frequently been used, at the time of day in which the drug was
usually taken, or after the usual number of hours since the last administration. These charac-
teristics define context, circadian, and temporal CSs, respectively. If the drug is not taken,
then the compensating CRs are seen in unadulterated form. The responses may be a drop
in body temperature, speeded respiration, cramping and nausea, and hyperactivity. These
symptoms of withdrawal, evoked in the absence of the drug, are opposite to the reactions
elicited by opiate drugs. The conditioning theory also suggests that detoxification that takes
place in a very different environment from the drug-taking environment does not extinguish
the context–drug association. Relapse occurs because on returning to the original living
environment associated with drugs, withdrawal and craving are elicited (Siegel, 1982).
The theory of compensating responses has its share of criticisms. In particular, only
some drugs show CRs that oppose the UR. Eikelboom and Stewart (1982) suggested that
 Extensions of Conditioning 73

some drugs have different effects on the central nervous system than on other parts of
the body. Only the central nervous system effects would be conditioned.
Research on drug conditioning illustrates three important points: first, that drugs can
act as USs; second, that contextual stimuli can act as CSs; and third, that the conditioned
response does not have to mimic the unconditioned response.
One other extension of classical conditioning is modeling causality learning, or the
learning of cause and effect relationships. This application is described in Box 3.2.

BOX 3.2 DETECTING CAUSALITY

Pavlovian conditioning has sometimes been described as an example of causal learning: a

stimulus event that regularly precedes the US could logically be thought of as being a cause
(Hall, 1994; Young, 1995)
The explanation of cause-and-effect relationships has a long history in philosophy and
psychology. Some philosophers said that we instinctively perceive these sorts of patterns,
just as with other innate categories of knowledge and perception. David Hume, a British
empiricist philosopher, argued instead that our belief that one event causes another derives
completely from our past experience with those stimuli. “When we look about us toward
external objects, and consider the operation of causes, we are never able…to discover any
power of necessary connection. …We find only that the one does actually follow the other”
(Hume quoted in Jones, 1952, p. 778). Hume argues that our expectation that the “cause”
will be followed by an “effect” is simply an inference and not necessarily a fact.
The scientists’ conditions for demonstrating cause and effect match the contingencies for
producing classical conditioning: The CS and US occur together, the CS closely precedes the
US, and the US occurs in the presence of the CS but not in its absence (Hall, 1994).
There are parallels between classical conditioning and learning about other kinds of con-
tingencies between stimuli. For example, in medical diagnoses, one may look for symptom–
disease correlations. What kinds of information do you need to conclude that a particular
symptom is diagnostic of the disease? First, look for the conjunction of the two events. For
example, must a fever be present in order to diagnose the flu? How often does a fever occur
without the flu, of the flu without a fever? Logically, we would need to tally all the possible
co-occurrences of the two events. This sort of contingency analysis can be arranged in a
2 × 2 table: presence or absence of the symptom and presence or absence of the illness (see
Table 3.6).
In evaluating the information from these sorts of contingencies, we humans do a rather
poor job. We have a bias to attribute cause based mainly on the conjunction of two events
and neglect the information obtained from the other cells of the table. Wasserman, Dorner,
and Kao (1990) asked subjects about the relative importance of these four cells in making
inferences about hypothetical problems, such as the relationship of a symptom to an illness.
The cell in which the symptom and illness are both present was judged to be the most impor-
tant fact in making a judgment (chosen by nearly 100 percent of the subjects). The cells in
which the symptom or the illness occurs alone were judged to contribute less information to
the decision, and the cell in which neither occurs was rated least important (by 50 percent of
the subjects).
There are ways contingencies can be misperceived. If the two events each have high
frequencies of occurrence, a pseudo-contingency might be perceived. For instance, if tones
74 Classical Conditioning

Table 3.6 2 × 2 Tables to Assess Causality

Effect Present Effect Absent

Cause Present Cause, effect Cause, no effect

Cause Absent No cause, effect No cause, no effect

and shocks each occur often, the tone might be seen as the cause (or predictor) of shock. The
infrequent occurrences of tones or shocks alone is not sufficient to judge the true relation-
ship. Also a few coincidental conjunctions of the putative cause and effect might be per-
ceived as the rule. This is a sort-of “first impression” phenomenon. A tone-followed-by-shock
early in training has more impact than coincidental pairings later.
Studies of human contingency learning have found parallels to animal conditioning. For
instance, a CS will be ignored if it is not a reliable predictor. Similarly, public security alerts or
storm warnings that turn out to be false alarms reduce future attention to such warnings. The
result of all this research has been a fertile cross-pollination of ideas between animal and
human learning theories.

APPLICATIONS OF CONDITIONING
The Conditioning Theory of Phobias
A phobia is an excessive and intense fear, usually of a specific object or situation, such
as a fear of snakes and spiders, of heights, or of speaking in public. Where do phobias
come from? Psychologists can offer a variety of explanations. Maybe phobias are instinc-
tive or innate reactions to certain stimuli, just like our reactions to sudden loud noises.
Maybe fear reflects an unremembered trauma. There is one other explanation that psy-
chologists point to whenever they want to demonstrate the significance of Pavlovian
conditioning: Phobias originate through classical conditioning. An initially neutral stimulus
becomes phobic because it has been paired with something traumatic, painful, or
frightening.
The initial statement of the conditioning theory of fear learning was pretty simple.
John Watson and Rosalie Rayner (1920) set out to condition an 11-month-old child named
Albert to fear a laboratory rat. The first exposures to the rat showed that fear was not
innate; Albert readily attempted to touch and grasp the rat. Presentations of the rat (used
here as the CS) were followed by the hammering of a steel bar (the US). Over the course
of a few pairings, Albert became more tentative in his reaching for the rat; his lip began to
quiver; he cried; and finally, he turned and crawled away. This fear of the rat was a condi-
tioned fear, the CR.
Watson and Rayner’s demonstration of fear conditioning was both powerful and influ-
ential. Their study served as the basis for speculations in literature on the frightening
applications to future societies. In Brave New World Aldous Huxley describes children
who are conditioned to fear books by electrically shocking them when books are touched.
The idea was to produce a working class that would not be distracted by ideas and
education.
 Applications of Conditioning 75

Although widely cited, the Little Albert case study leaves much to be desired as a
valid scientific experiment. The story has been distorted over the years, like a rumor that
is exaggerated with each retelling (Harris, 1979). Watson and Rayner’s conditioning theory
has been rightly criticized for being far too simple.
Contemporary knowledge about classical conditioning has been applied to explaining
phobias and other anxiety disorders (e.g., Mineka & Zinbarg, 2006). A significant revision
to the conditioning model is the addition of the notion of prepared learning (Seligman,
1972). Preparedness suggests that evolution has predisposed us to acquire certain fears
that have high survival value. Earlier, we saw that taste-aversion learning might be an
instance of prepared learning. Similarly, phobias do not develop to any arbitrary object
that happens to occur in proximity to trauma. Instead, people become fearful of the dark,
heights, and enclosed spaces; of snakes and spiders; or of other people. These are stimuli
that have represented dangers during human and primate evolution. There are other
classes of stimuli with which we surely have more unpleasant contact but which do not
become the object of phobias. Children fall off bicycles daily, but they do not develop bike
phobias.
Ohman and Mineka (2001) suggest that phobia learning shares features with other
forms of prepared learning. (For example, song learning in birds). The brain has evolved to
respond selectively to certain stimuli, particularly those that evolution has determined to
be potential threats or dangers. Also, fear is relatively unaffected by cognitive operations.
Consciously realizing that this snake is not harmful does not help you control your fear.
Preparedness can be supported by evidence for selective conditioning of fear-relevant
CSs. The rationale for these studies is the same as for demonstrating selective condition-
ing of tastes and illness. In laboratory studies of preparedness, we compare learning
about fear-relevant CSs, such as pictures of snakes, to fear-irrelevant CSs, such as pic-
tures of flowers. When the pictures are paired with an aversive US, we learn fears to
snakes quicker than fears to flowers (Ohman et al., 1976).
Another example of this research strategy comes from studies of fear in monkeys.
Cook and Mineka (1990) studied monkeys that had never been exposed to snakes. These
monkeys were not fearful of snakes at first exposure, indicating that the fears were not
instinctive. Young rhesus monkeys watched a video in which footage of phobia-relevant
stimuli (snakes) or unprepared stimuli (flowers) were spliced together with footage of
other monkeys showing fear and fright. That is, the participant monkeys saw another
animal apparently exhibiting fear to a snake or to some daisies. In conditioning terminol-
ogy, the snake and flowers are the CSs, and seeing the fearful monkey on film is the US.
The participant animals were later tested for fear of snakes by requiring them to reach
over or go around a toy snake to get a food treat. Those monkeys who saw the snake film
were more fearful of the snakes than those who saw flowers. Those who saw flowers did
not acquire a fear of the flowers.
Alternative explanations of preparedness have been offered. Maybe we are predis-
posed to fear snakes (and other potentially phobic stimuli) because of exposure to nega-
tive information about snakes during our lives. This is a learned associative bias. When
college students show up to participate in a psychology experiment, most already have
some fear of snakes, but not flowers. These preexisting biases can then be magnified
through aversive conditioning. In one experiment on fear relevant conditioning, human
76 Classical Conditioning

participants were told they would receive shocks. From this moment on, these partici-
pants had a higher expectation that shocks would follow the prepared CSs (snakes and
spiders) than the unprepared CSs (flowers and cats), even if no shocks were actually
given (see Davey, 1995).
There is also research indicating attention is biased towards potential threats that
begins in the first few months of life. For example, enlargement of the pupil of the eye
can indicate arousal. Six-month old infants had a larger pupillary response to simple draw-
ings of spiders than to flowers; and a larger reaction to snakes than to fish (Hoehl, Hellmer,
Johansson, & Gredebäck, 2017). Infants will associate snake or spider images with threat
if paired with pictures of fearful faces (LoBue, Rakison, & DeLoache, 2010). This might be
due to early learning, or to some innate template that identifies threatening stimuli.
The preparedness hypothesis does not account for all cases of phobia learning (see
McNally, 1987). People develop fears of non-evolutionarily prepared stimuli, such as den-
tal anxiety. People become afraid to drive after a car accident. Some categories of human
fears tend to be more age-related than evolutionarily determined; for example, 3- and
4-year-old children develop animal fears, whereas 13- to 18-year-old adolescents have
social fears (Miller, Barrett, & Hampe, 1974).

Overview of the Conditioning Theory of Phobias

A conditioning theory suggests that an initially neutral stimulus becomes phobic when it
has been paired with a traumatic or aversive event. Not all fears are learned through direct
experience. Phobias can be learned through observation of others’ traumatic experiences
with the phobic stimulus. Also, not everyone develops a phobia from adverse experi-
ences. Prior safe experiences can prevent acquiring a fear. For instance, many of us have
been bitten by a dog, but few develop a phobia of dogs (some fear or trepidation maybe,
but not a phobia). Children with a history of safe play with dogs are less likely to develop
a fear of dog. Finally, individual differences in personality or temperament could predis-
pose some people towards acquiring a phobia. For instance, anxious individuals or those
with PTSD condition more readily in laboratory experiments.

Extinction as Therapy
If fears can be conditioned, it is reasonable to suppose that they could be unlearned.
Extinction, which is the presentation of the CS alone, should reduce conditioned respond-
ing. For example, if you have a phobia of snakes, spiders, or heights, we might expose
you to pictures or videos of snake-, spider-, or height-related stimuli. Your fear should
decline as the exposures continue and nothing unpleasant happens (that is, the phobic
CS occurs without an aversive US). Extinction occurs.
Extinction does not eliminate the CS-to-US association. Thus, the appropriately titled
“Pavlovian associations are forever” (Baeyens, Eelen, & Crombez, 1995). Extinction is
effective in reducing or suppressing conditioned responding. The problem is, the response
keeps coming back.
There are several circumstances in which fears that have been “eliminated” by extinc-
tion can be revived. The primary means is by spontaneous recovery. After extinction, the
next occurrence of the phobic stimulus or situation will likely elicit a CR. The extinction
 Summary 77

that occurred during a session with the therapist could be followed by spontaneous
recovery at the start of the next session. Extinguished responses return in other ways. If the
US is presented alone sometime after extinction, the CR will return. This is called US rein-
statement; the lost response is reinstated. Also, extinction is context specific. If extinction
occurs in a context different from the conditioning context, the extinction will not carry back
to the conditioning context. This is called contextual renewal. A phobia was learned in one
situation (at work, at school), but extinction occurs in a different context (maybe a therapist’s
office). Upon return to the initial context, there can be a return of the extinguished CR.

How can extinction be made more effective and enduring? There are some general prin-
ciples that will reduce the return of extinguished responses.

Spacing extinction trials. Spacing CS alone trials during extinction leads to less sponta-
neous recovery and less contextual renewal (Urcelay, Wheeler, & Miller, 2009).
Gradual extinction. Gradually phasing out the CS-US pairings and phasing-in the CS
only trials leads to less recovery of fear (Gershman et al., 2013).
Overtraining. Continue to give extinction trials even after the response appears to have
been extinguished. Continued extinction training reduces recovery.
Extinguish in multiple contexts. Presenting the fear CS alone in multiple contexts may
broaden the generalization that the CS no longer signals fear.
Social facilitation. In one study extinction was conducted vicariously. Subjects who had
received fear conditioning in the lab then watched a video of someone else under-
going extinction. Observing the feared CS with no aversive outcome blocked the
return of fear (Golkar et al., 2013).

SUMMARY
Classical conditioning, also called Pavlovian conditioning, is not the simple form of reflex
learning portrayed in popular stereotypes. It is a flexible and adaptive form of associative
learning.

The Definition of Classical Conditioning

An initially neutral conditioned stimulus, or CS, is presented with a biologically significant
unconditioned stimulus, or US. The US elicits an unconditioned response, the UR. After a
number of pairings, the CS elicits a conditioned response, the CR. Contemporary exam-
ples of conditioning allow for indirect measures of learning other than the traditional CR.

Methods of Studying Classical Conditioning

A variety of specific procedures are used in studying conditioning, including salivary con-
ditioning, eyeblink conditioning, and skin conductance responses (SCR). Tones and lights
serve as CSs, and the USs are significant stimuli such as food, an airpuff, or electric
shock. The learned CR often resembles the UR, such as salivation or an eyeblink.
In fear conditioning, a CS is paired with an aversive US such as shock. In magazine
approach conditioning, the CS is paired with food. Other methods of conditioning include
78 Classical Conditioning

taste-aversion learning. The CR in these cases is indirectly assessed, through aversion or

avoidance of the CS, for example.
Various kinds of stimuli can serve as CSs, including external and interoceptive stimuli,
and contextual and temporal stimuli. The unconditioned stimuli are often reflex-eliciting,
but stimuli with acquired values are sometimes used.

Basic Phenomena of Conditioning

Acquisition refers to the development of a CR across pairings of the CS and US. Control
procedures, such as unpaired CS and US presentations and randomly scheduled CS and
US presentations, are needed to evaluate non-conditioning sources of responding.
Extinction refers to the presentation of the CS alone after conditioning, and to the
decline in responding to the CS that then occurs. The CR spontaneously recovers after a
period of time without stimulation, indicating the CS–US association has been suppressed
but is still present. Through generalization stimuli that are similar to the CS produce con-
ditioned responses. Discrimination between CSs can be trained by presenting one CS
with the US and the other CS without the US.

The Role of Contiguity

Conditioning is affected by the temporal contiguity, or spacing, between the CS and US.
Forward pairings, in the sequence CS then US, produce more conditioning than do the
simultaneous presentations of CS and US, or backward pairings of US followed by CS.
There are optimal CS-to-US intervals, but the exact time interval varies, and conditioning
effectively occurs within a range around this optimal interval.

Other Factors Affecting Conditioning

Usually, a number of potential CSs are available on any trial. Through stimulus selection,
only certain of these stimuli become associated with the US. In compound conditioning,
two or more CSs are presented together and paired with the US. Conditioning is divided
among the stimuli; and a more intense CS will overshadow a weaker CS.
Blocking of conditioning to a CS occurs by presenting it simultaneously with another
CS that has already been trained with the US. Blocking shows that the occurrence of the
US must be surprising for conditioning to occur.
The Rescorla-Wagner model describes the trial-by-trial acquisition of conditioning
with multiple stimuli. The increase in conditioning on each trial is a function of the strengths
of the CSs present at the start of a trial. In the blocking procedure, previous training with
one CS blocks conditioning to an added CS, because there is less room left for new learn-
ing due to the high associative strength after pre-training.
Conditioning is affected by the relevance of the CS to the US, also referred to as CS
and US belongingness. A taste CS readily conditions with certain types of USs, such as
poison and illness, but taste conditions slowly if at all with other USs, such as electric
shock. Correspondingly, exteroceptive CSs such as tones and lights quickly condition with
shock USs, and poorly if at all with the illness US. This was shown in Garcia and Koelling’s
experiment using a bright-noisy-tasty CS. Taste-aversion learning may occur readily
 Summary 79

because organisms are biologically prepared to form associations between certain classes
of stimuli. Alternatively, there may be a learned bias acquired during the organism’s life-
time, to form these associations.
Conditioned inhibition occurs when there is a negative correlation between the CS
and the US: The US is more likely to occur without the CS than with it. Inhibition is not
simply the absence of conditioning; inhibition must be actively trained.

What Is Learned in Classical Conditioning?

Stimulus–response (S–R) theory says that the CS comes to elicit the unconditioned
response or some portion of the UR. However, an overt UR is not necessary for condition-
ing to occur, and sometimes the CR is not the same as the UR.
The stimulus–stimulus (S–S) theory states that conditioning produces an association
between the learned, internal representations of the CS and the US. This theory specifies
that knowledge of the US is learned, not a specific conditioned response. S–S theory is
supported by studies in which the size of the CR after conditioning changes in reaction to
devaluation of the US.
Neuroscience research indicates that one form of conditioning, the eyeblink, occurs
through a circuit in the cerebellum of the brain. Not all classical conditioning follows this
circuit, but other brain areas are involved with other forms of conditioning.
Experimental participants are often aware of and can verbally report the CS–US con-
tingency that they have experienced. However, conditioning still occurs if conscious
awareness is minimized by using distractor tasks, simpler organisms, or response sys-
tems beyond awareness.

Extensions of Conditioning
In evaluative conditioning an affectively neutral stimulus, one that is neither particularly
liked nor disliked, is presented along with another stimulus that already evokes a strong
affective evaluation.
Research on drug conditioning shows that drugs can act as USs; that contextual
stimuli can act as CSs; and that the conditioned response may be opposite to the UR (a
compensatory response). Repeated exposures to a drug can lead to the development of
tolerance: The drug seemingly loses its effectiveness. Thus, tolerance may develop as a
compensating CR that counteracts the UR.

Applications of Conditioning
A classical conditioning theory for the origin of fears and phobias says that an initially
neutral stimulus becomes conditioned (phobic) because it has been paired with fear, pain,
or trauma. In Watson and Rayner’s study, a child became fearful of a rat (used as the CS)
that was paired with a loud noise (the US). The conditioning model has been updated by
the addition of vicarious learning (learning fears through observation of other’s fears) and
preparedness (evolution has prepared us to readily acquire certain fears that have high
survival value, such as fear of darkness, heights, or snakes).
80 Classical Conditioning

An alternative learning theory suggests that previous experiences bias our expecta-
tions that certain stimuli (such as snakes) are dangerous.
If fears can be conditioned, it should be possible to unlearn them. The safe exposure
to a phobic stimulus, or extinction, does weaken the phobia. But the extinguished
response returns under a number of conditions: through spontaneous recovery, US rein-
statement, and contextual renewal.
 CHAPTER

4
Instrumental Conditioning: Reward

CONTENTS

Definition and History 82 Is Reinforcement Necessary

Thorndike and Trial-and-Error for Learning? 96
Learning 82 Awareness in Human Instrumental
Skinner and Operant Learning 83 Learning 98
Elements of Instrumental Criticisms of the Use
Conditioning 84 of Reinforcement 98
Methods of Study 85 Response Learning 99
Positive Reinforcement 86 Shaping 99
Reinforcement Variables Affecting Limitations of Response Learning 100
Acquisition 87 Discriminative Stimulus Control 102
Amount of Reinforcement 87 Generalization and Discrimination 102
Drive 87 Summary of Response Learning
Schedules of Reinforcement 89 and Stimulus Learning 104
Delay of Reinforcement 90 What Is Learned in Instrumental
Secondary Reinforcement 91 Conditioning? 104
Social Reinforcement 91 Response–Reinforcer Learning 105
Theories of Reinforcement 92 Stimulus–Response Learning 105
Drive Reduction 92 Stimulus–Reinforcer Learning 106
Incentive Motivation 93 What Is Learned? Stimulus–
The Neural Basis of Reinforcement 93 Response–Reinforcer 106
Reinforcers as Behaviors 94 Applications 107
Reinforcers as Strengtheners 95 Habits 107
Reinforcers as Information 95 Behavior Modification 109
So, What Is Reinforcement? 96 Summary 111

Many of us engage in various behaviors we wish we didn’t: vices such as smoking, vap-
ing, or drinking too much; dangerous actions such as speeding; and unconscious habits
of nail biting, hair pulling, and teeth grinding. These bad habits are seemingly automatic,
pervasive, and beyond remediation. Why can’t we stop? There are other behaviors we
wish we did perform habitually: studying, exercising, and controlling our diets. Why are
they so difficult to start and maintain?

DOI: 10.4324/9781003227090-4
82 Instrumental Conditioning: Reward

As a starting point, our unwanted behaviors are usually rewarded by some positive
outcome. Maybe nail biting is calming. Speeding has the thrill of fast driving. Vaping is
rewarded with peer approval and maybe a little rush from the nicotine. Habitual behaviors
are also associated with stimuli in the environment that trigger the habit. Smoking
becomes attached to too many eliciting stimuli: after eating, after class, after a meeting,
while socializing. Nail biting or hair pulling are triggered by nervous situations: scary mov-
ies, talking to instructors or bosses, working out conflicts with roommates, and ruminat-
ing about tomorrow’s assignment.
On the other hand, desirable behaviors are not often immediately reinforced. Studying
is daily, but the opportunity to earn a good exam grade happens only occasionally in the
semester. Exercising has long-term consequences, but little in the way of immediate
gratification for all that expenditure of energy.
The remedies are, first, change the consequences of the behavior. Penalize yourself
for smoking, speeding, or cursing. Schedule positive rewards for studying or exercising.
Next, establish set stimulus conditions to be associated with the desired behaviors: a
time and place exclusively devoted to exercise or study. Such prescriptions sound like
exercises in self-control, which is what we believe we are lacking to begin with. After all,
if you had the willpower, you would stop snacking and start exercising. The learning per-
spective shies away from explanations based on personal weaknesses, and instead
replaces them with eliciting stimuli and reinforcing consequences, objective conditions
that can be manipulated and that do influence behavior.
The topic of this chapter is instrumental conditioning. Instrumental conditioning is
learning the connection between a behavior and its consequence. The behaviors of indi-
vidual organisms are causal in producing various outcomes, some positive and some
unwanted. This chapter deals more broadly with behavior than just our bad habits. Our
intention is to derive some systematic principles that apply to eliciting stimuli, instrumen-
tal behaviors, and the consequences arranged.

DEFINITION AND HISTORY

Thorndike and Trial-and-Error Learning
Learning evolved as a means for organisms to adapt to changing environments. The
underlying mechanisms need to be universal and applicable across the phylogenetic
scale. Edward Lee Thorndike studied the development of adaptive behaviors to determine
the principles involved (Thorndike, 1898, 1911).
Thorndike is best known for training “cats in a puzzle box.” Cats were placed in a
wooden crate having a hinged door in the front and a trip mechanism somewhere in the
box. For example, pushing a pole that sticks up through the floor or pulling a loop of wire
hanging in the back of the box would open the door, allowing the cat to escape. Thorndike
chose responses that were not already in the cats’ repertoire to study how the response
developed with practice. Learning was measured by the time required to escape the box
across trials.
Thorndike observed and named a number of characteristics of what we now call
instrumental learning. He described the animal’s efforts to escape as trial and error
 Definition and History 83

learning. Trial-and-error involves trying various responses, sometimes at random, until

one produces the desired outcome. The cat at first claws and scratches at the door. Over
trials the ineffective responses drop out. Another behavior, the one that immediately
precedes the door opening, becomes more frequent. If the cat was brushing against the
pole, the animal does this more frequently.
Thorndike said learning was governed by the law of effect: “Of several responses
made to the same situation, those which are accompanied or closely followed by satisfac-
tion to the animal will, other things being equal,… be more likely to recur ” (Thorndike,
1911, p. 244). The law of effect is a statement of the principle of reinforcement: Behavior
is modified by the consequences of the behavior.
The escape response becomes associated with the stimuli present at the time the
response occurs, in this case, the stimuli of the puzzle box. Learning is the formation of a
stimulus–response (or S–R) connection, from the environment’s stimuli to the escape
response. Reinforcement is what conditions or strengthens this S–R connection. The
reward stamps in the S–R association. Thorndike thought that reward exerted its effect
automatically and without conscious thought or reasoning.

Skinner and Operant Learning

Beginning in the 1930s, B. F. Skinner began to develop his own techniques, terminology,
and principles of learning by reinforcement. Skinner’s system of conditioning is called
operant learning. Skinner constructed a small experimental chamber in which to condi-
tion animals such as rats or pigeons (Skinner, 1938). This “operant conditioning chamber”
(or Skinner box) allows precise experimental control over the presentation of discrimina-
tive stimuli and reinforcers, and the recording of responses (see Figure 4.1). A contin-
gency is arranged between a response, such as pressing a handle or bar, and a reinforcer,
usually a small round food pellet delivered through a chute. Skinner coined the label

Figure 4.1
An Operant Conditioning Chamber (a.k.a. the “Skinner Box”). (left) The right wall of the chamber has two
levers on either side of the pellet chute, and small round lights that can be used as discriminative stimuli. A
round magazine canister contains the food pellets. (right) A rodent touch screen could replace one wall, allowing
images to be projected as stimuli. Instead of bar pressing, the animal would touch the image.
Source: Courtesy of Lafayette Instrument Company, Lafayette, Indiana.
84 Instrumental Conditioning: Reward

operant response to indicate that the subject’s response operates on the environment to
produce a certain outcome. The bar-press-to-food contingency leads to an increase in bar
pressing, known technically as positive reinforcement and informally as reward training.
Once conditioned, operant responses can be extinguished. In extinction, the reinforcer is
withheld, which should lead to a decrease in the frequency of responding.
The conditioning chamber shown in Figure 4.1 has two bars, or levers, on either side
of the pellet chute. A round canister, called the magazine, contains the food pellets. Small
round lights are on the upper right panel, and a speaker for sound stimuli can be mounted.
The picture on the right shows a rodent touch screen which would replace one of the end
walls. Images can be projected as stimuli. Instead of pressing a bar the animal would
touch the image.
The distinction between instrumental conditioning and operant learning is significant
among researchers of each, but is less obvious to outsiders. One difference is that in
instrumental learning the subject is given discrete trials during which the response may
be performed, for example, a trial in a maze or puzzle box. In operant studies, the subject
is allowed more-or-less continuous availability to the response. The rat is placed in the
Skinner box for 50 minutes and can perform the response whenever.
A more important distinction is that the instrumental approach tends to adopt a par-
ticular form of theory that hypothesizes psychological factors that operate between the
stimulus and response to explain learning. For instance, Edward Tolman said that rats
develop “cognitive maps” of mazes. These hypothesized maps are not directly observed
but must be inferred from behavior. Skinner eschewed this form of theorizing and opted
instead for using strictly observable variables: The frequency of responding is a function
of the amount of reinforcement, or its delay, or its frequency, and so on.

Elements of Instrumental Conditioning

In Pavlovian conditioning, we had the stimulus elements of CS and US, and the response
elements of CR and UR. There are parallel elements in instrumental learning.
There is a discriminative stimulus (or SD), a cue that signals the availability of the
reinforcer. The stimulus could be the context in which reinforcers occur, such as the
Skinner box or the puzzle box. Or the stimulus could be a tone that cues when reward is
available in the Skinner box.
The instrumental response is the specific response or behavior we are attempting
to condition. This is the to-be-learned behavior.
There is reinforcement. This term is used in the sense of strengthening a response.
Reinforcement would often be the presentation of an appetitive stimulus such as food or
water, or the omission of an aversive stimulus such as shock. The category of possible
reinforcers has broadened considerably to encompass many more outcomes that serve
as reinforcing events. The word reinforcement is frequently replaced by the consequences
of responding, and instrumental learning is called goal driven learning.
The sequence of the three elements could be diagrammed as:

SD → Response → Reinforcement
 Methods of Study 85

METHODS OF STUDY
Instrumental learning is studied through a variety of methods. Rats and pigeons are fre-
quently trained in Skinner boxes. With pigeons, a round Plexiglas disk placed at eye level
can be pecked. This response is called key pecking and is reinforced with pieces of grain
or seeds.
Mazes are once again popular now that spatial learning and memory are topics of
renewed interest. The simplest mazes are T-shaped, with a start alley leading to a choice
point and left and right goal boxes. Entry into the correct goal leads to food or some other
reinforcer. (Maze learning is discussed in detail in Chapter 11). Another instrumental task
is the runway. It is simply a long alley with a start compartment at one end and a goal
compartment at the other end. Learning is measured by the speed of running, which
increases across rewarded trials.
Several interesting methods have been developed to study instrumental learning in
human infants. Turning the head to the left or right can be reinforced by offering milk or
juice that can be sucked from a bottle. Trials can be initiated by a discriminative stimulus,
such as a tone, which signals the availability of milk to the right (or left, whichever is
selected as the target response). Somewhat older infants learn to move their legs in order
to shake an overhead mobile tied to the limb via a ribbon, as illustrated in Figure 4.2.

Figure 4.2
Conditioning of Leg Movements. The left panel shows the experimental arrangement for infant conditioning
with movement of the mobile as the reinforcer. The data in the right panel shows the increase in leg movements
during acquisition; a decrease during extinction when the mobile is untied from the infant’s leg; and re-acquisition
and re-extinction.
Sources: (left panel) Rovee-Collier et al. (1980). Photo courtesy of Carolyn Rovee-Collier. (right panel) From “Conjugate
Reinforcement of Infant Exploratory Behavior,” by C. K. Rovee and D. T. Rovee, 1969, Journal of Experimental Child Psychology,
8, p. 36. Copyright 1969 by Academic Press. Reprinted with permission.
86 Instrumental Conditioning: Reward

During acquisition training, leg movements increase when they are rewarded. After unty-
ing the ribbon, extinction occurs: The flexions decrease when no longer rewarded by
movement of the mobile (Rovee-Collier, Sullivan, Enright, Lucas, & Fagan, 1980).
With older children or college students, instrumental learning can be embedded in a
computer game. For instance, students may be told that they are playing an investment
game; each press of the space bar invests some of their money; the computer displays
the profit or interest earned as the “reward” (Reed, 2001). The amount or frequency of
wins can be systematically varied. Students learn which scenarios (stimuli) are associated
with a profit, and which responses to choose.

POSITIVE REINFORCEMENT
What we call reward in everyday language is a somewhat imprecise description of posi-
tive reinforcement. Reinforcement is defined by the presence of a response-to-reinforcer
contingency. (A contingency is essentially a rule). In positive reinforcement, the rein-
forcer is delivered contingent on performance of the instrumental response. The response
increases in frequency because it leads to a reinforcing consequence. Who establishes
and maintains the contingency? It may be the experimenter, but it could also be a parent
or teacher, or even society.
The notion of contingency is important to defining instrumental conditioning: The
reinforcer is contingent on, or dependent on, the occurrence of a response. Control con-
ditions are required in an experiment to assure that the responding we observe is due to
the contingency and is not incidental to some other aspect of the experiment. If I attempt
to reward my dog’s tail wagging with a food treat, tail wagging will surely increase. But
this is not due to wagging-leading-to-food, and in fact it is probably the reverse (giving
food causes wagging!). A non-contingent control condition is sometimes used, in which
the rewards are programmed to occur randomly, independent of the subjects’ behavior.
That is, the control reinforcers do not require a bar press or keystroke. Figure 4.3 dia-
grams response-contingent and response-noncontingent reinforcement.

Contingent

Action
Outcome

Time

Noncontingent

Action
Outcome

Time

Figure 4.3
The Contingent Relationship Between Action and Outcome. In the contingent relationship, each instance of
the response is followed by the outcome (reward). In the noncontingent relationship, the outcomes occur
independently of the responses.
 Reinforcement Variables Affecting Acquisition 87

When college students play the investment game, an instrumental condition arranges
for a profit to occur after every so many presses of the space bar. A noncontingent control
group receives the same profits, but they are dispensed randomly by a computer program
and are not dependent (are not contingent) on bar presses. The response-contingent
reward group makes many more presses than the noncontingent group. If you question
the students, the reward subjects will believe their responses lead to reward; the control
subjects will not perceive a connection between response and reward (Reed, 2001;
Shanks & Dickinson, 1991).

REINFORCEMENT VARIABLES AFFECTING ACQUISITION

Positive reinforcers can be manipulated along several dimensions, each of which affects
how quickly a response is acquired and the level of performance that is attained. These
variables include the amount and delay of reinforcement, and the consistency with which
reinforcement is administered.

Amount of Reinforcement
As a general principle, larger rewards produce faster learning and better performance
than do smaller rewards. For instance, rats run faster in a straight alley for larger rewards
than for smaller rewards. Illustrative data are shown in the top left panel of Figure 4.4. The
reinforcer was either 1 gram of food or 0.05 gram. The large-reward animals ran faster
than did the small-reward animals. Children offered points for completing school work
finish more when more points are offered (e.g., Wolf, Giles, & Hall, 1968). The points are
later exchanged for a toy, stickers, play time, and so on.
Similarly, reinforcers that are more highly preferred produce better performance. Rats
will respond more for water having a higher concentration of sucrose than a lower
concentration.

Drive
Drive can be described as a motivational need or desire for a given reinforcer. The level
of an appetitive drive is usually manipulated by depriving the subject of access to the
reinforcer for some period of time. For example, depriving a rat of food or water over-
night increases the hunger and thirst drives, respectively. Drive increases responding to
reinforcers that are relevant to the drive. The data from the top right panel of Figure 4.4
are from a study of rats that were food deprived for 4 hours or 22 hours before training
sessions. The rats trained under a state of longer deprivation ran faster during the acqui-
sition trials.
Do the variables of drive and amount of reward affect what is learned, or do they
instead affect the motivation to respond? When a rat runs slowly for a small reward does
this mean that smaller rewards produce less conditioning, or less motivation? This illus-
trates the learning-versus-performance distinction introduced in Chapter 1: reinforcement
could affect learning the response and/or the motivation to respond.
Clark Hull (1949) acknowledged the roles of both learning and motivation in his the-
ory. Learning was represented by the factor of Habit Strength (H), which was determined
 88 Instrumental Conditioning: Reward

.50 .9
1/T Speed of Response (Mean of 3- Trial Mean)

Response Speed: Reciprocal of Latency (sec)

.8 22-Hour Deprivation
1.0 g
.40 .7

.6
.30
.5
0.05 g
.4
.20
4-Hour Deprivation
.3

.10 .2

.1

0 0
3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 1 2 3 4 5 6 7 8 9 10 11 121314151617181920
Blocks of Three Trials Blocks of Four Trials

45 Day 1 Day 2 Day 3

40

35
10/mdn Latency (sec)

30
Group I (1 sec)
25

20

15

10
Group II (10 sec)
5

0
4 10 16 22 28 34 40 46 52 58 64 70 76 82 88
Trials

Figure 4.4
Effects of Three Reward Variables on Running Speed by Rats. (top left) Amount of food reward, 1 gram
versus 0.05 gram; (top right) number of hours of food deprivation, 4 hours versus 22 hours; and (bottom) delay
between response and reinforcer, 1 second or 10 seconds.
Sources: (top left) and (bottom) from Behavior Theory and Conditioning (pp. 131 and 157), by K. Spence, 1956, New Haven, CT:
Yale University Press. Copyright © 1956 by Yale University Press. Reprinted with permission. (top right) from “Performance in
Selective Learning as a Function of Hunger,” by C. K. Ramond, 1954, Journal of Experimental Psychology, 48, pp. 265–270.
 Reinforcement Variables Affecting Acquisition 89

by the number of reinforced training trials (and other variables). Motivation was separately
represented by the theoretical variables of Drive (e.g., the number of hours of food dep-
rivation) and Incentive (e.g., reinforcer amount). The actual degree of responding was
determined by the combination of Habit Strength (H), Drive (D), and Incentive (I), as:

Response strength  H  D  I

Put informally, the rat runs quickly in the straight alley because these three factors
take on high values: the rat has received prior reinforcement for running (H); the rat is
hungry now (D); and the reinforcer is really appealing (I). Reinforcers play multiple roles in
instrumental learning. (Hull actually used another letter for incentive because I was already
used for inhibition).

Schedules of Reinforcement
A schedule of reinforcement refers to the specific rule that relates the timing or fre-
quency of responding on the one hand and the delivery of reward on the other. In general
terms, reinforcement can be given for each occurrence of the instrumental response or
for only some. Reinforcing each instance of the response would be continuous reinforce-
ment (i.e., a 100 percent schedule of reinforcement). In partial reinforcement, a defined
percentage or number of the responses are reinforced (e.g., a 50 percent schedule).
During the initial training of a behavior, continuous reinforcement generally produces
more rapid conditioning than does partial reinforcement. A continuous schedule pro-
vides more accurate information about the contingency between response and out-
come. If one considers instrumental learning to be a form of trial-and-error learning, then
the occasional nonrewards of the partial schedule may lead the subject to try other
responses in an attempt to be more successful. Stevenson and Zigler (1958) compared
children in a button-pushing task. Only one of the three buttons produced reward (a light
flashed and points accumulated), and for different groups that button was reinforced
100, 66, or 33 percent of the time. The other two buttons had no effect whatsoever. The
results nicely showed a schedule effect: There was a higher frequency of pushing the
correct button when reinforcement was given on a continuous rather than on a partial
schedule.
Given that a partial reinforcement schedule is used, there are various ways of arrang-
ing response–reward contingencies. In a fixed-ratio schedule, reinforcement occurs after
a fixed number of responses (e.g., every fifth or seventh response). Fixed-ratio schedules
lead to very high response rates. By contrast, on a fixed-interval schedule, reinforcement
is given for the first response that occurs after a set interval of time. For instance, during
a 30-second fixed-interval schedule, 30 seconds must pass since the last reward before
a response will be reinforced.
Reinforcement can be made less predictable by varying the number of responses
required for a reinforcer, a variable-ratio schedule, or varying the time interval between
rewards, a variable-interval schedule. Thus, reinforcement on a variable-ratio 5 schedule
occurs after different numbers of responses that averages to 5 over trials. On a
90 Instrumental Conditioning: Reward

Table 4.1 Examples of Schedules of Reinforcement

Fixed ratio: Being paid for each piece manufactured, each envelope stuffed, or each assignment
completed.
Fixed interval: Making the response of checking to see if the coffee is ready. No reward is
provided until a certain amount of perk time has passed.
Variable ratio: Sales associates’ attempts to help customers are sometimes rewarded with sales.
Which customer will buy may be unpredictable, but more attempts should produce more sales.
Variable interval: Checking e-­mail or texts when the annoying “ding” is in silent mode. The
messages arrive unpredictably, but the recipient won’t know unless he or she checks.

variable-interval 30-second schedule, reinforcement occurs following the first response

after intervals that average 30 seconds. Variable schedules produce relatively constant
rates of responding because the occurrence of reinforcement is less predictable. Some
everyday examples of schedules are described in Table 4.1.
The slot machine is a textbook example of intermittent reinforcement. The designers
use several behavioral principles to induce high rates of responding and persistence in
playing (Rivlin, 2004). Reward is given on a variable schedule. It occurs intermittently and
unpredictably to induce persistence in responding. Frequent small rewards maintain play-
ing between the rare big payouts. The lights and sounds act as secondary reinforcers (stim-
uli associated with reinforcement and thus act as rewards). Even though some machines
now issue winnings as paper tickets or credits, the machines’ speakers make the sound of
coins dropping down the chute when players win, again acting as a secondary reinforcer.
Schedules are of interest because they demonstrate that the rate and patterning of
responding is sensitive to the specific reinforcement contingencies. Critics of a scientific
psychology argued that behavior is too complex to be predictable. Skinner demonstrated
how behavior can, in some situations, be precisely predicted and controlled (1953).

Delay of Reinforcement
Reinforcement that occurs immediately after the correct response is generally more
effective than reinforcement that is delayed for some interval of time. The lower panel of
Figure 4.4 shows the results of an experiment with rats trained to run in a straight alley
and given a food reward one second after reaching the goal box, or the food was delayed
for 10 seconds. The short-delay group ran faster.
Delaying reward impedes learning in several ways. Other behaviors occur during the
delay interval, and they might inadvertently become conditioned. For instance, a rat
presses a bar but reward does not immediately occur, so the animal starts turning around
and the food pellet suddenly drops in after the delay times out. From the rat’s perspective,
the reinforcer occurs after turning, not after bar pressing!
Delaying reward can also slow learning because the response has been forgotten by
the time reinforcement occurs. Now what was it that I did that worked? Lieberman,
McIntosh, and Thomas (1979) presented a distinctive stimulus after the response, a
“marking stimulus,” to make the response more memorable. In their experiments, when
the rat made a correct turn in the maze a tone sounded. The tone marked the response in
 Reinforcement Variables Affecting Acquisition 91

memory, making the response memory more distinctive, so it could be connected later
with the outcome.
Instrumental tasks can be expanded to allow several responses, each with different
reinforcement contingencies. For instance, two levers could be available in the Skinner
box. One response (say, the right lever) produces an immediate small reward and the
other lever produces a delayed but larger reward. On a given trial, the subject can only
press one of the levers. This sort of choice defines self-control: the capacity to inhibit
immediate gratification in preference for a larger reward after a delay. Impulsiveness
would be choosing the small immediate reward. Whether self-control or impulsiveness
occurs depends on the exact combinations of amount and delay. Self-control decreases
if the delay is too long or the reinforcer at the end of the delay is not large enough.
Self-control, or tolerating a delay, can be learned. By beginning with a relatively brief
delay for the larger reward, and then gradually lengthening the interval, the more pre-
ferred reward after a longer delay will be chosen.

Secondary Reinforcement
So far in this chapter we have cited a variety of different reinforcers, such as food, accu-
mulating points, or the opportunity to socialize. Some of these can be labeled as primary
reinforcers, or events that are inherently reinforcing. Primary reinforcers reduce biological
needs of the organism, such as food or water, or relief from heat, cold, or pain.
Other stimuli that function as reinforcers are derived from primary reinforcers.
Secondary reinforcers are stimuli that have been paired with primary reinforcers and
acquire the capacity to reinforce on their own. For example, a tone that has been paired
with food can function as a reinforcer for the instrumental response of bar pressing.
Animal trainers often develop secondary reinforcers, such as whistles or clickers, to rein-
force from a distance or to immediately reward correct performance without having to
interrupt the animal’s routine.
A secondary reinforcer sounds like classical conditioning: establishing an association
between the neutral and unconditioned stimuli (e.g., between tone and food). Indeed,
secondary reinforcement is sometimes called conditioned reinforcement. The classical
conditioning comparison also suggests another similarity: The secondary reinforcer will
extinguish if it is not regularly paired with the primary reinforcer. Thus, periodic recondi-
tioning of the secondary reinforcer is necessary.

Social Reinforcement
An especially powerful class of reinforcers for human behavior is social reinforcement.
Praise, attention, physical contact, and facial expressions given by parents, teachers, or
peers can exert considerable control over our behavior. In one example, a young student
remained apart from the other children during recess. This withdrawn behavior received
inadvertent social reinforcement in the form of extra attention from the teacher, who
spent more time with the student. Rearranging the reinforcement contingencies to pro-
vide attention when the student interacted with others, and ignoring the withdrawn
behavior, led to a change in the student’s socialization (Kirby & Shields, 1972).
92 Instrumental Conditioning: Reward

Why do social reinforcers have such power? One theory is that social reinforcers are
primary reinforcers, based upon biological drives inherent in humans and other animal
species (Harlow, 1959). Another theory treats social reinforcement as secondary rein-
forcement. For instance, approval by others is reinforcing because it is paired with primary
reinforcers such as food or protection from danger (Miller & Dollard, 1941; Skinner, 1953).
Social reinforcers have several advantages over natural reinforcers such as food in
behavior modification. Praise can be given immediately and does not usually disrupt
ongoing behavior. Classroom social reinforcers not only improve academic behavior, but
often generally improve attentiveness to potential social reinforcement for other behav-
iors, and decrease disruptiveness that leads to social disapproval (Kazdin, 1994).

Summary of Reinforcement Variables

Positive reinforcers can be primary reinforcers, such as food or water; secondary reinforc-
ers, such as a tone, token, or gold star; and social reinforcers, such as praise. A positive
reinforcer is more effective in conditioning and maintaining instrumental responding when
the reinforcer is large; when the organism has a drive or need for the reinforcer; when the
reinforcement is given immediately; and when it is given after each response. There are
exceptions to these general rules. In choice settings, the characteristics of delay, amount,
or quality of the reinforcers can be opposed. For example, one response produces a large
but delayed reinforcer, whereas another response produces a small immediate reinforcer.

THEORIES OF REINFORCEMENT
What makes reinforcers reinforce? A variety of things function as reinforcers: food or water
for animals, shaking a mobile for infants, tokens and stickers for young children, likes and
dislikes for social media users, and accumulating points for button pushing by college stu-
dents. Do these events have something in common that makes them act as reinforcers?
Thorndike defined a reinforcer as that which produces a satisfying state of affairs, but this
is a vague and subjective description and does not offer much guidance. Skinner adopted
a very pragmatic definition: A reinforcer is whatever works to increase the frequency of the
operant response. However, this does not explain why a reinforcer reinforces.
The explanation of reinforcement has been a central concern to learning theorists and
over the years a number of theories have been offered. The major categories describe
reinforcers as stimuli (things such as food, tokens, and points); reinforcers as activities
(e.g., consuming, exploring); and reinforcers as information (whether the response made
was correct versus incorrect). To anticipate the conclusion derived from the following
discussion, reinforcers work because of several reasons. A single explanation may not
subsume all instances of reinforcement.

Drive Reduction
One characteristic of primary reinforcers is that they reduce biological drives. Food
reduces hunger, water reduces thirst, and so on. Clark Hull (1943) postulated that, at the
most basic level, drive reduction was the basis for primary reinforcement. Drive-reduction
is consistent with the effect of reinforcement variables mentioned earlier. Thus, larger
(rather than smaller) reward reduces the hunger drive more; immediate reward reduces
drive sooner than delayed reward; etc.
 Theories of Reinforcement 93

But what about reinforcement that has no obvious relationship to drive reduction,
such as tokens, points, or even money? Here, secondary reinforcement enters in. Stimuli
acquire secondary reinforcing properties through their association with primary, or biolog-
ical, reinforcers. The many stimuli that reinforce human behavior may be said to have
acquired reinforcing properties.
Hull’s drive-reduction theory was enormously influential as it nicely tied psychological
explanations of learning to biological ones. Unfortunately for the theory, numerous exam-
ples of learning without apparent drive reduction began to accumulate. For instance, rats
learn instrumental responses for saccharine reinforcers (Sheffield & Roby, 1950). Saccharine,
a nonnutritive substance, does not satisfy hunger but is still highly reinforcing. In another
example, when juvenile monkeys were made fearful they sought out a soft and comforta-
ble manikin to cling to, rather than one a rigid metal manikin that had previously provided
food reinforcement (Harlow, 1959). The latter should have been a secondary reinforcer. In
the face of these contrary results, alternative theories of reinforcement were devised.

Incentive Motivation
Instead of reducing drive, maybe reinforcers actually increase drive. Saccharin is reinforc-
ing because it stimulates a desire for more of the reinforcer. The ideas of Sheffield (the
saccharin study) and Tolman (latent learning) evolved into the theory of incentive motiva-
tion: Reinforcers are incentives that elicit responding. This corresponds to the way we
talk of rewards in everyday language: We perform to get the reward. The difference
between drive reduction and incentive is the difference between push and pull: Drives
(such as hunger) push us into action, whereas incentives (such as chocolate) pull us on to
obtain them. To suggest an everyday example: You are sitting quietly until you notice the
smell of fresh baked cookies. You are not driven to investigate by a hunger drive; your
searching (and wanting) is elicited by the aroma.
Incentive motivation is illustrated by a manipulation known as reinforcer priming.
Simply giving a free sample of the reinforcer, one that need not be earned, stimulates
instrumental responding to obtain more of the reinforcer. Imagine giving a rat a small
portion of the food reward before it enters the maze. Drive-reduction theory would pre-
dict the rat would run slower because there was less hunger. Instead, the priming rein-
forcer increases running speed (Terry, 1983). The priming reward initiates a “need” that
wasn’t there before.
One difficulty with an incentive theory of reinforcement is that instrumental respond-
ing will sometimes persist even after the incentive value is gone. As will be noted later,
satiated rats will persist in responding even though they no longer consume the accumu-
lating food pellets.

The Neural Basis of Reinforcement

Brain Stimulation
A third approach to defining reinforcers seeks the underlying physiological basis of rein-
forcement. Possibly there is a common area of the brain that is activated by those stimuli
that work as reinforcers. James Olds and Peter Milner (1954) discovered that stimulation
of an area in the brain called the reticular formation was reinforcing. Rats were trained to
make bar-press responses which were followed by brief bursts of electrical stimulation.
94 Instrumental Conditioning: Reward

Stimulation worked well as reinforcement. Subsequent research has discovered brain

areas involved with other reinforcers, such as those for the opiate drugs or alcohol; and
for social media stimuli.

Dopamine
Nerve cells communicate by releasing a neurotransmitter chemical that stimulates adja-
cent cells. Serotonin is one familiar transmitter, being the basis for many antidepressant
drugs. The primary neurotransmitter found for reinforcement is dopamine. Dopaminergic
neurons use the transmitter dopamine. When an unexpected reward occurs (one that is
unpredicted), there is a sudden release of dopamine. The dopamine strengthens the con-
nections among circuits that are active at that time, including the neurons activated by the
stimuli and responses that preceded the reward. Over a series of conditioning trials, the
activation of these cells comes to predict the reinforcer. The dopamine cells that initially
reacted to the reinforcer itself, now react less to the reward, but dopamine increases in
the cells predicting reward, the tone and bar pressing. Dopaminergic cells are reacting
not simply to reward, but rather to unexpected reward and to novel stimuli. The unex-
pected triggers dopamine, and therefore, new learning.
Dopamine neurons receive input from many areas of the brain. This includes informa-
tion about present drive states (e.g., hunger), and past experience with reinforcers (e.g.,
magnitude, frequency, probability). A prediction about a given reward includes details of
the value of that reward: its magnitude, timing, probability, etc. These are all characteris-
tics of reinforcers that affect speed of learning: hungry animals learn food-related behav-
iors quickly, they learn better with larger, immediate, and frequent food rewards, etc.
Given a choice between two responses associated with different rewards, a predicted
value of each can be compared.

Reinforcers as Behaviors
As an alternative to characterizing reinforcers as stimuli, we could think about reinforcers
as activities or behaviors. That is, it is not food that is a reward, but rather the activity of
eating that is reinforcing. This at first seems just a matter of semantics, but the altered
perspective expands the category of positive reinforcers to include all sorts of other activ-
ities that we know function as such.
The major proponent of the reinforcement-as-activity approach is David Premack
(e.g., 1962). His notion, generally speaking, is that behaviors can be ranked in terms of
their preference or value to an individual. Some activities are highly preferred, such as
going to the movies or eating ice cream, and other activities are less preferred, such as
studying or mowing the lawn. These preferences can be established by observing the
relative probability with which the various behaviors occur. According to what is now
called the Premack principle, a higher-probability activity will reinforce a lower-probability
activity. Thus, going to the movies will reinforce studying, but not vice versa. That is, you
will study if rewarded with a trip to the movies.
A simple statement of the Premack principle glosses over some important technical
details. Preferences have to be determined individually, ideally under conditions allowing
unrestricted opportunity to engage in all of the relevant activities in order to determine
 Theories of Reinforcement 95

their baseline frequency. The probabilities of different behaviors vary over time because
the activities are subject to deprivation or satiation (Timberlake & Allison, 1974). Thus,
after watching all of The Lord of the Rings movies, watching videos temporarily lose their
reinforcing property.
The Premack principle was neatly demonstrated in a study of two behaviors of chil-
dren who were allowed access to candy and a pinball machine. In the first phase of the
study, individual preferences for each were assessed. In the next phase, each behavior
was used either as an instrumental response or a reinforcer for different subsets of chil-
dren. The children who preferred pinball now increased their working for candy (used here
as the response) in order to get access to pinball (used here as the reinforcer). Children
who initially preferred candy now played pinball (the response) to gain candy (the rein-
forcer) (Premack, 1965).

Reinforcers as Strengtheners
In everyday speech, we refer to reinforcers as rewards or incentives, but there is an alter-
native sense of the word reinforce: to strengthen. This is the original sense used by many
learning theorists. The reinforcer strengthens the association between a discriminative
stimulus and an instrumental response. For instance, when a stimulus light is followed by
a bar-press response, the food reinforcer strengthens the association between the light
and the bar press. Contemporary research has validated this strengthening role.
The potential strengthening effects of a reinforcer are usually confounded with its
reward or incentive effects, either of which can lead to improved performance. To test the
strengthening hypothesis, the “reward” aspect must be removed.
A learning experience produces activation within certain areas of the nervous sys-
tem. Reinforcers maintain excitement in those neural units, allowing additional opportu-
nity for permanent changes to occur which encode the learning experience (Landauer,
1969). Reinforcers do not have to be rewards like food or water to strengthen learning.
The critical feature is that the reinforcer enhances excitation of those neurons involved in
the learning experience. This means that memories or associations can be produced by
means other than reinforcement. The administration of stimulant drugs or hormones
shortly after a learning experience, or electrically stimulating certain areas of the brain
(White & Milner, 1992), also facilitate learning. Posttrial brain stimulation of the hypothal-
amus, administration of glucose or epinephrine, exposure to cold pressor stress are all
ways to incite neural activation, yet none of these is a “reward.” For example, after a
shock-avoidance learning trial, the animal is given food. The food is not a reward in this
situation, and it is not contingent on the instrumental response. However the food does
boost ongoing neural activation (Huston, Mondadori, & Waser, 1974). Similarly, learning is
increased by injections of epinephrine or glucose given after the learning trial (Gold &
Korol, 2012). In a human memory task, cold pressor stress administered after presenta-
tion of the to-be-remembered lists increases recall (Smeets et al., 2008).

Reinforcers as Information
Whether a reinforcer reduces a biological drive, is an incentive, or strengthens memory,
a reinforcer provides information. Was this the correct response? Was the response
96 Instrumental Conditioning: Reward

performed accurately, efficiently, timely? The reinforcer provides information in the form
of feedback: “yes, I did it right” or “no I didn’t.”
Biofeedback is an example of the informational role of reinforcement. A subtle, usu-
ally undetectable bodily reaction, such as muscle tension in the forehead, is monitored by
an electrode attached to the body. In biofeedback, this bodily response can be converted
into an external signal, such as a tone. Changes in muscle tension are converted into
increases or decreases in the volume or pitch of the tone. The goal is to learn to control
the tone, thereby also controlling the muscle tension. The tone provides information
about muscle tension, information that we otherwise did not recognize.

So, What Is Reinforcement?

Several theories of reinforcement have been reviewed. Although one or another explana-
tion can be adopted exclusively, many theorists accept that reinforcement plays multiple
roles. In some cases, a reinforcer is a significant stimulus that the organism needs to
learn about for survival. In other cases a reinforcer can be well described as an incentive
for certain behaviors. In still other cases, reinforcement is information about whether a
response is correct. We can also note that reinforcers are events that elicit affective reac-
tions of pleasure or displeasure. Thus, there may be no one answer to the question.

IS REINFORCEMENT NECESSARY FOR LEARNING?

Much of human learning goes on without explicit reward. Memory occurs without a delib-
erate intention to remember. We have already seen that habituation and perceptual learn-
ing occur without rewards or reinforcers (Chapter 2). Although animal studies routinely
use reinforcement, such as food, water, or relief from shock, is reinforcement even nec-
essary for learning by animals?
The classic test of this question was Tolman and Honzik’s study of latent learning
(1930). Three groups of rats were trained in a maze. One group received food in the goal
box on each trial. Across 11 days of training, these animals learned to run faster and to
enter fewer blind alleys. A second group never received food in the goal box. Their perfor-
mance improved only slightly over days. So far, this looks like a fairly standard study
showing that reinforcement produces learning. But is it likely that the nonrewarded rats
learned nothing about the layout of the maze in 11 days? Maybe they did learn but had no
reason to run promptly into the goal box. After all, it really wasn’t a “goal” box for the
nonrewarded animals. Tolman and Honzik included a third group that was first trained
without reward. On day 11, food was placed in the goal box for the rats to discover when
they eventually got there. On the next opportunity, day 12, the rats in this third group
quickly ran the maze, entering few wrong alleys en route to the goal box (see Figure 4.5,
upper panel). Tolman said that these rats had learned the maze during those first days
without reinforcement but that this knowledge was latent (or hidden) until they were
given a reason to display the learning was present.
In a second experiment, Tolman and Honzik employed a symmetrically opposite manip-
ulation: After training with reward, on the eleventh day, food was omitted in the goal box. On
day 12, these rats suddenly started taking wrong turns (see the lower panel of Figure 4.5).
 Is Reinforcement Necessary for Learning? 97

32
30
28
26
Average Errors X Constant (3)

24
22 Not Rewarded

20
18
16
14
12
Switched from
10 Rewarded Nonrewarded
8 to Rewarded
6
4
2
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Days

32
30
28
26
Average Errors X Constant (3)

Not Rewarded
24
22
20
18
16
14
12 Rewarded
Rewarded
10
Switched to
8 Nonrewarded
6
4
2
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Days

Figure 4.5
Results of the Tolman and Honzik Latent Learning Experiments. Both graphs show the number of maze
errors made over days. The upper panel shows the effect of switching from nonreward to reward on day 11; the
lower panel shows the effect of switching from reward to nonreward on day 11.
Source: From “Introduction and Removal of Reward, and Maze Performance in Rats,” by E. C. Tolman and C. H. Honzik, 1930,
University of California Publications in Psychology, 4, pp. 257–275. Copyright © 1930, The Regents of the University of California.
98 Instrumental Conditioning: Reward

These rats obviously knew the way through the maze. The sudden increase in errors
reflected a deficit in performance, not in learning.
Traditional reinforcement theorists at first disputed Tolman’s interpretation of latent
learning. Possibly after 10 days of acclimation and adaptation to the experimental proce-
dure the single reinforcement on day 11 was sufficient to condition the correct maze
choices. However, even granting the contribution of other factors, the now-accepted con-
clusion is that learning can occur without explicit rewards.

AWARENESS IN HUMAN INSTRUMENTAL LEARNING

In the 1950s and 1960s, claims were made that subliminal perception and learning
occurred: Sub-threshold stimuli would command us what to do and we would uncon-
sciously respond. This possibility raised the specter of manipulation of human behavior by
unscrupulous individuals. In reaction to such fears, the U.S. Congress passed legislation
banning subliminal commercial messages in films (Moore, 1982). For a decade after-
wards, the scientific claims and data went back and forth over the possibility of learning
without awareness. Does conscious awareness of the response–outcome contingency
play a role in learning?
In one of the first attempts to demonstrate unaware conditioning, Greenspoon (1955)
simply told his subjects to say out loud each word that came to mind. Greenspoon would
mutter “umm humm” whenever a plural noun was emitted and make no response to
other words. The students began to use more plural nouns during the course of the ses-
sion. This is verbal conditioning: An instrumental response (here, plural nouns) is followed
by a reinforcer (“umm humm”) that increases the frequency of the response. In post-ex-
perimental interviews, the college student participants said they were not aware of the
contingency between the words they spoke and the experimenter’s mumbling. These
results were taken to suggest that unconscious learning had occurred.
Our contemporary reaction to this claim is both “yes” and “no.” Other research dis-
cussed throughout this text shows that learning occurs without our awareness or recollec-
tion. But does unawares learning occur in verbal conditioning studies? Self-reports are too
often inaccurate measures of awareness. Subjects may be reluctant to report an aware-
ness unless they are quite certain. More detailed questioning in subsequent studies
revealed that many participants were indeed aware of the response–outcome contingency
(Dulany, 1968). Furthermore, the participants who were aware showed conditioning; partic-
ipants who were unaware of the contingency did not increase their use of the target words.
It is also clear that normal instrumental performance can be enhanced through verbal
instruction of the contingencies. We tell children the “rules” relating behavior to conse-
quences. Systems of reward for appropriate behaviors are more effective when the con-
tingencies are first described verbally (Ayllon & Azrin, 1964).

CRITICISMS OF THE USE OF REINFORCEMENT

The application of reinforcement principles to everyday behavior has been a spectacular
success for psychology. Behavior modification is applied in schools, institutions, work-
places, and clinics. However, behavioral technology has been criticized both for its
 Response Learning 99

underlying philosophy and its long-term effectiveness. One objection is a moral one: the
use of reinforcement is manipulative. Granting and withholding reward is a form of con-
trol. Proponents of behavioral psychology reply in return that our behavior is already
controlled by its consequences (Skinner, 1971). Parents, peers, social groups, online
groups, schools, employers, and governments all use rewards or sanctions to control
human behavior.
Another objection is that reinforcement undermines intrinsic motivation, or an inter-
nal desire to engage in a behavior for its own sake. With intrinsic motivation the incentive
to perform comes from the activity itself, in contrast to the extrinsic motivation by
rewards. In one influential study (Lepper, Greene, & Nisbett, 1973), nursery school chil-
dren were first observed while they drew with colored markers, something kids like to do
anyway. Drawing is said to be intrinsically motivated. Then, in one experimental condition,
the children were offered reinforcement for doing more drawing, and in the control con-
dition there was no reinforcement. Later, the children were again allowed to play with the
markers. Those who had been reinforced now colored less than those not previously
reinforced. Reward had provided an explicit reinforcer for the behavior, replacing intrinsic
reasons for using the markers, such as doing it for fun.
The belief that rewards undermine intrinsic motivation is widely accepted. However,
the counterargument by behavior modifiers is that the use of explicit reinforcers is often
intended to be temporary. Once behavior starts to occur with some regularity, other
reinforcers in the individual’s environment may take over. A person may feel satisfaction
in accomplishing a task and doing it well. And some behaviors simply have little intrinsic
motivation to begin with. Not many of us find real satisfaction in doing calculus home-
work. External motivation in the form of rewards may be better than not doing the work
at all (Chance, 1992).

RESPONSE LEARNING
Many of the instrumental responses described so far were chosen by researchers for
their experimental convenience. However, a significant use of instrumental conditioning
is in shaping a particular form of the response. The contingencies we arrange between
the exact response desired and presentation of reinforcement can be used to create new
behaviors and modify existing behaviors precisely.

Shaping
Say we want to train a behavior that is not presently in the organism’s repertoire. How can
we reinforce something that does not occur? With response shaping, a behavior that is
not currently in the repertoire can be gradually shaped by reinforcing a series of closer
approximations to the desired response. Skinner used the analogy of a potter molding
clay. We start by selecting behavior that approximates the desired response, and rein-
force that behavior. Then step by step we reinforce responses closer to the target behav-
ior. For bar-pressing, the rat is first reinforced for staying near the bar, then for touching
the bar, or pawing the bar, then bearing weight on the bar, etc. By the process of succes-
sive approximations, reinforcement is used to create a new response.
100 Instrumental Conditioning: Reward

Gregory Berns, a neuroscientist, wanted to observe the brain activity in unanesthe-

tized dogs. Conducting MRIs on unrestrained dogs presented several challenges: the dog
is in a closed space, the dog must remain motionless for up to a minute, the scanner
makes frighteningly loud noise, etc. The solutions involved a several instrumental learning
procedures (highlighted below). The researchers built a simulated scanner for the several
months of training required. The dogs were habituated to the noise of the scanner using
prerecorded scanner noise and really big speakers. Training involved shaping each
response in a chain. For instance, the dog was shaped to place her head on a chin rest,
sit sphinx-like, and remain motionless. Using a variable interval reward schedule, the
experimenter signaled a food reward. Thus, the dog knew reward was coming but not
exactly when, and so remained still waiting. Hand signs were used to signal food or
no-food trials (S+ and S– ). The animals had to learn to sit still on command, look at stimuli
the experimenters presented, make a response to indicate choice, to receive reward.
Finally, the animals were ready to perform various psychological tasks (e.g., discriminat-
ing pictures of faces) while being scanned (Berns, Brooks, & Spivak, 2012).

Limitations of Response Learning

As powerful as reinforcement is for conditioning behavior, not all behaviors can be modi-
fied through reinforcement. Skinner himself noted that some reflex responses could be
modified only through classical conditioning.
There are also species-specific limitations on what can be modified. Not any arbitrar-
ily selected behavior can be shaped using reward in any given species. For instance, in
research using hamsters, behaviors such as digging and standing were readily condi-
tioned with food reinforcement. Other behaviors, such as face washing and scratching,
did not increase when rewarded with food (Shettleworth, 1975).
Species limitations on instrumental learning were shown in the field of animal training
by two of Skinner’s students, Keller and Marion Breland (Breland & Breland, 1961). They
titled their paper “The Misbehavior of Organisms,” a play on Skinner’s book title The
Behavior of Organisms. The Brelands found that certain behaviors were resistant to mod-
ification with food rewards. For example, they tried to teach a pig to put wooden coins in
a piggy bank. (This is just too cute!) If the pig did this, it received food. Instead, the pig
persisted in rooting the coins, rubbing the coins on the ground with its snout as if digging
them up. Rooting delayed and even prevented reward. Yet, as training continued, rooting
became even more frequent.
Evolutionary preparedness may account for some limitations on response learning.
Animals may have evolved a readiness to learn certain categories of response-to-out-
come associations, such as behaviors that might lead to food or behaviors to escape from
danger. Alternatively, once the coins became associated with food, maybe as conditioned
stimuli, the rooting behavior might have been a classically conditioned response that
competed with the instrumental response of dropping the coins in the bank.
The principles of learning would seem to be obviously applicable to animal training.
However, the scientist’s study of learning by animals is an entirely different field from
animal training, roughly corresponding to the basic-versus-applied distinction.
Nevertheless, science may offer some unique insights into animal training. Some
instances are described in Box 4.1.
 Response Learning 101

BOX 4.1 ANIMAL TRAINING

In Chapter 1, I offered some reasons for using animals in research, reasons which primarily
benefit the science of learning and memory. Here I want to suggest that research on animals
can benefit the animals themselves. Zoo animals and endangered species are examples of
animals whose well-being is our responsibility. Behavioral learning principles are used to reduce
the risk of injury to the animal and increase safety of those humans who work with them.
Sutherland (2006) wonderfully describes the Exotic Animal Training and Management
program at Morelock Community College in California. “Animal training” in this context does
not mean teaching animals to perform cute tricks. Caretaking often requires that the animals
be compliant in medical procedures to monitor their health or treat illness. But how do you
get an untamed monkey to consent to a blood draw, or a tiger to a dental inspection?
Instrumental learning principles are used to train the animals to acquiesce in these proce-
dures. An integral part of the academic program is the study of learning through course
work, labs, and practical training. Habituation (discussed in Chapter 2) would be used to
adapt animals to the presence of the human technicians. The presentation or omission of
reward is used to teach an animal to allow the techs to touch it, or to bring apparatus nearby.
An example of a secondary reinforcer is training an animal to maintain head contact with the
end of a stick held by the technician (who would usually remain safely outside the cage). The
stick can be used to lead the animal to a desired location or position. By the end of the pro-
gram, students will have trained several categories of animals, such as primates, birds, and
reptiles. These would include the range of species in zoos, entertainment parks, wildlife
sanctuaries, etc.
Another example of training animals for their own benefit is to prepare dogs in animal
shelters for adoption. Some pets are in shelters because of behavioral problems: a dog is
uncontrollable or aggressive, or has destructive misbehaviors when left alone. Behavioral
problems discourage adoption or, even sadder, lead to the return of an adopted dog to the
shelter. David Tuber started a program using volunteers (often, his own college students) to
train shelter dogs (Tuber et al., 1999). Using positive reinforcement, shaping, and discrimina-
tive stimuli, dogs learn to sit, enter a crate, or bark less. Shy dogs are rewarded for seeking
contact with people. A simulated living room is in the shelter so the dog can be trained in a
home-like environment. Animals that acquire some basic skills and a few commands are
more readily adopted. Incidentally, this training seems to lower stress in the shelter animals.
This in itself seems to me to justify the training.
Another application of conditioning is in training animals to detect land mines. An organi-
zation known as APOPO (a Dutch acronym for anti-personnel mine detection devices) trains
rats to detect hidden explosives. In many countries wars and insurgencies have left civilian
areas littered with undiscovered mines. In Cambodia, rats were trained to sniff out these
explosives. The rats are given extensive training in non-bomb environments. They learn to
respond to the scent of explosives by lightly scratching the ground to signal their finding. The
rats receive food rewards for correct identification. If you are concerned about the risk to the
rats: The rats are too light-weight to trigger the mines. In fact, they can have a long life per-
forming this task. One rat was recently retired after five years of mine detecting (obviously,
successfully). This rat was awarded a medal which historically has been given to canine
heroes of war. (APOPO web: https://www.apopo.org/, retrieved 2021).
102 Instrumental Conditioning: Reward

DISCRIMINATIVE STIMULUS CONTROL

We have discussed two of the elements of instrumental learning, the response and the
reinforcer. The third critical element is the discriminative stimulus. Learning involves not
only what response to make, but also when to make it. A discriminative stimulus sig-
nals the availability of reinforcement. In the bar-press situation, a tone or a light can be
used as the discriminative stimulus, or S+, signaling that the reinforcement contingency
is in effect. Responses during S+ are reinforced. Responses in the absence of S+ are not
reinforced. Bar presses only lead to food when the S+ tone was present.
Stimulus control refers to conditioning a response to occur in the presence of the
discriminative stimulus. A response is brought under control of a stimulus. Many of
these stimuli are contextual stimuli of time and place. Behaviors that are acceptable at
a frat party are not appropriate in the classroom. Studying may be difficult when sitting
at the desk is also associated with listening to music, talking on the phone, or
mind-wandering.
An interesting example of discriminative control is that of the remote-control rat
(Talwar et al., 2002). Electrodes were implanted in the rat’s brain, in sensory and rein-
forcement areas, that could be stimulated wirelessly from a distance. Left or right
turns could be rewarded by stimulation of the reward area (the medial forebrain bun-
dle). The cue for making a turn was stimulation of the areas of the brain that sensed
touch to the left or right whiskers (sort of like tapping the rat on the snout). After acti-
vation of the left whisker area (the discriminative stimulus), if the rat turned left (the
instrumental response), the experimenter gave a burst of rewarding brain stimulation
(the instrumental reward). After first being trained on the basic routine in an enclosed
maze, the rats were tested in open areas outside. For instance, a rat could be guided
via remote control up a ladder, across a narrow ledge, and through a pipe, a distance
of 1,600 feet. (Okay, why would anyone want remote-control rats? The researchers
suggest that with the addition of a miniature camera, rats could be used in searching
dangerous locations).

Generalization and Discrimination

In Chapter 3 we introduced the idea of generalization of the conditioned response to
stimuli that were similar to the CS. Instrumental learning can also show the effects of
generalization: A response initially trained to a particular discriminative stimulus will be
made during stimuli that are similar to the trained S+. Figure 4.6 shows responding by
four pigeons and four humans to a discriminative stimulus, in this case a particular color;
and generalized responding to test stimuli of colors of different wavelengths (Kalish,
1958). In each case, the subjects were first trained to make a response in the presence
of one stimulus. Testing involves presenting other stimuli and recording the number of
responses to each. For example, one of the pigeons was trained to peck in the presence
of yellow (580 nm). This bird pecked less often to orange (600 nm), and not at all to green
(530 nm). Stimuli that are similar but not identical to the S+ typically elicit fewer responses
than does the trained stimulus. The curves in Figure 4.6 show peak responding to the
trained color, and less responding to colors above and below. The pigeons showed broader
 Discriminative Stimulus Control 103

500

450 Guttman and Kalish S = 600

(Pigeon)
400

350
S = 530 S = 580
300
Responses
S = 550
250

200

150

100

50

Kalish 530
110
(Human)
580 Standard
100

90
560
80

70
Responses

60
550
50

40

30

20

10

0
470 490 510 530 550 570 590 610 630
Wavelength (mµ)

Figure 4.6
Gradients of Generalization to Color Discriminative Stimuli. Frequency of response by four pigeons (upper
panel) and four humans (lower panel) who were reinforced for responding in the presence of colored lights
ranging from 530 nanometers (green) to 600 nm (orange), and then tested with colors above and below the
trained stimulus.
Source: From “The Relationship Between Discriminability and Generalization: A Re-Evaluation,” by H. I. Kalish, 1958, Journal of
Experimental Psychology, 55, p. 637. Copyright American Psychological Association. Reprinted with permission.
104 Instrumental Conditioning: Reward

generalization to adjacent colors, whereas the human subjects had very sharp generaliza-
tion gradients, with very little responding to different colors.
Generalization may be one reason why learned behaviors do not always transfer from
one situation to another. The stimuli are different, or are perceived as different, and there-
fore exert less stimulus control.
The complementary process to generalization is discrimination. In discrimination
training, responses are reinforced in the presence of S+; responses are not reinforced in
the presence of S–, a stimulus signaling that reinforcement is not available. For example,
a pigeon’s key pecks will be rewarded when a red light is on but not during blue light. A
child may be reinforced by praise or attention for doing homework in the presence of one
parent but is not reinforced in the presence of the babysitter.
(Discriminative stimuli initially went by different labels. Responses were reinforced in
the presence of SD, pronounced S-dee; and responses were not reinforced in the pres-
ence of SΔ, pronounced S-delta. Today S+ and S– are more generally used. They also
correspond to usage in classical conditioning of CS+ and CS–).
In an example from speech therapy, the subjects were children who had difficulty
discriminating consonants in speech, such as “ba” and “da.” Via an animated computer
game, syllables were presented that had certain components exaggerated by making
them longer and louder (e.g., “bbbaa” and “dddaa”). The syllables were gradually normal-
ized over trials as the children were able to identify them correctly, leading to nearly
errorless learning (Merzenich et al., 1996).
Generalization and discrimination are converging means of controlling behavior.
Generalization is the spread of learning to other stimuli, while discrimination is the restric-
tion of learned responses to specific stimuli.

Summary of Response Learning and Stimulus Learning

Instrumental learning involves three elements: a discriminative stimulus, a response, and
a reinforcer. By specifying the form of the response that will be followed with reinforce-
ment, the response can be shaped. There may be species-specific limitations on which
response–consequence sequences can be learned, such as digging for food rewards and
running when shock is administered.
A response can be brought under stimulus control, such that a stimulus controls
when the response is to be made. Stimulus control is subject to limitations by generaliza-
tion and discrimination.

WHAT IS LEARNED IN INSTRUMENTAL CONDITIONING?

In the previous chapter on classical conditioning, we asked which elements from among
the CS, US, CR, and UR became associated. We can ask a similar question with respect
to instrumental conditioning. Given the three elements of discriminative stimulus, instru-
mental response, and reinforcer, what are the associations that are acquired? All combi-
nations have been seriously considered.
 What Is Learned in Instrumental Conditioning? 105

Response–Reinforcer Learning
The obvious answer to “what is learned?” is that the response becomes connected to
the reinforcer. Speaking casually, we say that the response is performed to get the reward.
Rats bar-press for food, schoolchildren read books for stickers, and so on. The response–
reinforcer theory says that learning is the formation of a connection between the instru-
mental response and the reinforcer.
Responding often seems to be under exquisite control of the reinforcement condi-
tions: Larger and tastier rewards provoke more vigorous response, delayed rewards
weaken responding, and satiation of drive leads to a reduction in responding.
The theory of response–reinforcer association predicts that changes in the reinforce-
ment conditions should lead to immediate changes in the response. This has been stud-
ied by changing reinforcement expectancies. An example is the earlier-cited latent learning
experiment of Tolman and Honzik. Rats ran faster or slower on trial 12, corresponding to
their changed expectations of what would be found in the goal box. Introducing food in
the maze on trial 11 led to immediate improvement in performance; omitting food led to
an immediate decrement.
Unfortunately, behavior is not always sensitive to changes in outcome. In an early
study, Tolman (1933) first trained rats to run a maze for food. Then he placed the rats in the
goal box and gave them a strong foot shock. What did the rats do the next time they were
in the maze? Tolman reported that the rats “dashed off…just as usual…and bang whack
into the very food compartment in which they had just been shocked.”

Stimulus–Response Learning
Stimulus–response theory says that instrumental learning is the formation of a connec-
tion between the discriminative stimulus and the instrumental response. Theorists such
as Thorndike and Hull were S–R theorists. Reinforcement acts to condition (or consoli-
date, strengthen) this association.
Evidence for stimulus–response conditioning would be the case in which the response
seems to have become separated from its reinforcing consequence, and becoming
instead an automatic reaction to the stimulus. For example, instrumental responding
sometimes persists even though reinforcement is freely available and the response is no
longer needed to obtain reward. Singh (1970) demonstrated the effects of free, or
unearned, rewards in a pair of experiments using rats and children. (Not together; sepa-
rately). He first trained rats to bar-press for food on one end of the Skinner box, and then,
in the next phase, food pellets were simply delivered on the other end. In a parallel study,
6-year-old children learned to button-press for marbles while standing along one side of a
large box (in which the experimenter was concealed, dispensing marbles), and later free
marbles were dispensed from the other side of the box. Both rats and children still choose
to make the response. Rats would press the bar, go over and eat some from the dish, and
then go back and bar-press some more.
Other evidence for the disconnection of response and outcome comes from stud-
ies of habitual behavior in mazes. Well-trained rats will run through a pile of food placed
106 Instrumental Conditioning: Reward

in the middle of a maze alley on their way to the goal box. The rats ran right through
food pellets, slipping and sliding, on their way to the goal box to get…food (Stolz &
Lott, 1964).
A contemporary demonstration of the independence of response and reward is rein-
forcer devaluation. A reinforcer is modified to be less-valued. We then observe instrumen-
tal responding in the face of the changed reward. For example, rats were first trained to
bar-press for sucrose solution. The rats were then given sucrose–poison pairings in a
taste-aversion learning procedure. One result of this second treatment was that the rats
would no longer drink sucrose. Truly, the sucrose was devalued. The second result of was
that the rats continued to bar-press even though they would not drink the sucrose they
earned (Adams, 1982).
What’s wrong here? These outcomes appear contrary to common sense. Behavior
should change in response to changed reinforcement. However, persistence in the old
mode of behavior is comprehensible in light of the S–R habits acquired over many (some-
times hundreds of) trials.

Stimulus–Reinforcer Learning
A typical sequence of events in an instrumental trial is discriminative stimulus, response,
and reinforcement. Because of this, the S+ is incidentally paired with reward. Thus, clas-
sical conditioning can occur between the stimulus and the reinforcer, or in Pavlovian ter-
minology, between the CS and the US. The S+ then elicits conditioned responses in
anticipation of the reinforcer (such as conditioned excitement or fear), which facilitates
performance of the instrumental response.
An example of stimulus–reinforcer learning occurs in the pigeon key-pecking task. In
this situation, illumination of the Plexiglas disk (or key) at eye-level, is followed with grain
presentation. We would normally train the pigeon to peck the key to get the grain. If the
pigeon simply observes light onset being followed by the grain, without being given the
key-pecking option, the pigeon will key-peck when given the opportunity. Conditioning
occurs in the absence of response–reinforcer training, but which followed stimulus–
reinforcer experience.

What Is Learned? Stimulus–Response–Reinforcer

Rather than concluding that any one of the preceding explanations is exclusively cor-
rect, we might instead suggest that each may make a contribution to instrumental
learning. Combining the several associative relationships is possibly the best descrip-
tor of what can be learned in instrumental learning (Rescorla, 1987). As Skinner noted,
the discriminative stimulus sets the occasion for when a response will be reinforced,
but the stimulus does not elicit the response. And as Tolman noted, the subject may
learn a set of expectancies, some being means-to-ends sequences (e.g., bar pressing
to obtain food), others being stimulus–reward expectancies (the goal box signals food
or shock).
 Applications 107

APPLICATIONS
Habits
A habit is a sequence of discriminative stimulus, response, and consequence. Habits are
learned instrumentally, beginning as intentional behaviors that lead to the satisfaction of
a goal. Repetition of the sequence, stimulus–response–goal, is what makes the response
habitual. Habits are primarily elicited by contextual cues of time and place, and in combi-
nation with moods, feelings, and other habitual behaviors.
Habits are described as automatic behaviors. This means that habitual responses are
performed with minimal conscious awareness or control. The behavior is cued and runs off
with little, if any, direction on our part. Yet, habit and conscious intention work hand-in-hand.
Repetitive sequences become habitual and so require less conscious control. An unex-
pected outcome snaps us out of the habit mode and back to a conscious mode of thinking.
Habits are adaptive and useful behaviors. William James (1890) said “habit simplifies
the movements required to achieve a given result” (p. 112) and “diminishes the conscious
attention with which our acts are performed” (p. 113). Thus habits reduce the cognitive
load of repetitive actions by freeing consciousness, attention, short-term memory, and
decision making for other activities.
How long does it take to form a habit? The popular media offers various estimates: Do
something every day for a week, or 10 times, or three weeks, etc. There is no magic number.
The answer depends (as always) on other variables. One study followed habit acquisition in
96 individuals, of whom only about half were successful in acquiring their selected habit
(e.g., drinking a bottle of water with lunch; running 15 minutes before dinner). The number of
days to habit formation ranged from 18 to 254 (Lally, Van Jaarsveld, Potts, & Wardle, 2010).
Many of our daily routines are based on strings of habits. For example, take the morn-
ing routine. Getting up, washing, dressing, maybe eating something, and commuting to
work or school. Much of this occurs with little conscious direction. Standing by the sink
may cue brushing your teeth. Entering the kitchen triggers coffee. Exiting the house trig-
gers a pat-down to see if you have everything – keys? ID? backpack? The completion of
one habit becomes the cue for the next. Thus, habits might be described as an organiza-
tion of hierarchies, e.g., the morning bathroom sequence or the drive to school sequence.
During these routines, your mind is free for other activities such as planning, mind-wan-
dering, or worrying.

Habit Slips
We have already seen that well-conditioned rats will bar-press for food even when food is
freely available; and run through food pellets in a maze on their way to the goal box.
Persistence in habitual behaviors is not restricted to animals in laboratory experiments.
Humans are creatures of habit. Habits may be useful overall but become a problem when
an exception to the rule is desired.
James Reason (1990) collected examples of errors when a new response was
intended but the habitual behavior intruded: putting sugar in your cereal when you meant
to cut back; driving past an intended stop only to realize when you arrive home that you
108 Instrumental Conditioning: Reward

don’t have dinner; and, nearly tragically, a military pilot flying in an air show who momen-
tarily forgot he had live missiles that day and fired at another jet. (The other pilot bailed
out safely). Sometimes we seem to act as mindlessly as our rats who ran through food
pellets on the way to the goal box.
These sorts of errors are called habit slips: the intrusion of a habit when an alterna-
tive behavior had been intended. Reason suggests that habit slips are more likely to occur
when we are in familiar situations, and we are distracted or preoccupied. While driving a
familiar route, your mind wanders elsewhere and so habit takes you through your usual
route to your home. For the fighter pilot in the previous example, the distraction of warn-
ing gauges allowed the habitual response, practiced in hundreds of hours of training, to
occur.
Capture errors are another type of slip (Norman,1981). If two behavior sequences
have identical starting elements, the more habitual behavior may capture the less fre-
quent behavior. Driving to school or to your friend’s may start off on the same route, but
the more frequent or most recent route takes over while your thoughts are elsewhere.
Going to brush your teeth or comb your hair share similar starting behaviors, but you wind
up doing one rather than the other.
Another type of slip occurs when the habit sequence ceases prematurely. For
instance, after withdrawing cash from an ATM, you leave without waiting for your debit
card to be returned. Getting cash satisfied the intended goal. Other examples of prema-
ture termination include making photocopies and then forgetting to take the original; or
failing to replace the gas cap after filling your car. The main goal was achieved (make
copies, getting gas) which generates a false sense of completion, leading to the omission
of the steps that come after (Chung & Byrne, 2008).

Breaking Habits
Breaking a habit requires forceful measures. Informational interventions, such as public
health warnings against smoking, or personal intentions such as deciding to change your
diet, are relatively uninfluential at the moment the habit is triggered (Verplanken & Wood,
2006). The cookie is already in your mouth before you remember you’re cutting back on
sweets.
One suggestion to correct a habit is to take advantage of the absence or disruption of
eliciting stimuli. For example, moving to a new dorm, apartment, or house is accompa-
nied by new contextual stimuli and different daily routines. A change in jobs can involve
different commuting patterns or friends. The old cues are no longer present to trigger the
old habits. Wood, Tam, and Witt (2005) conducted a survey of students who were trans-
ferring to another university. The researchers quantified the strength of good and bad
habits, such as exercising or watching too much television. The researchers also noted
how much the context had changed. For instance, was the TV in the same room of the old
and new houses? Did the student exercise in the same or a different place? Did room-
mates change? The researchers found that the more the surroundings had changed from
before to after the move, the more the habit (good or bad) was disrupted. One lesson is
that a move may be a good time to change your routine and add better habits. At the
same time, you have to be vigilant against the disruption of existing good habits.
 Applications 109

Behavior Modification
The field of behavior modification (or simply behavior mod) applies the principles of
operant learning to changing behaviors in a variety of settings. In an early demonstra-
tion project, Ayllon (1963) was asked to assist with problem patients in a psychiatric
setting. For example, one patient was stealing food from other patients in the cafeteria.
Ayllon used a reward-omission procedure. Each time the woman took food she was
removed from the cafeteria and that meal was forfeited. (This would not be allowed
today because it violates the patient’s rights). The effects of this procedure are shown
in Figure 4.7. Within a week, she ceased stealing food, and (with only an occasional
lapse) did not regress during the year-long follow-up. Her weight also dropped to a
healthier level.
Another problem was towel hoarding. The psychiatrically trained staff attributed this
to childhood deprivation of affection, and the towels were substitute satisfiers for some
inner need. Ayllon treated the unwanted behavior directly. He suggested a novel proce-
dure of “stimulus satiation”: If towels are reinforcing, then a larger number of towels
should produce satiation (analogous to satiating the hunger drive by giving a large amount
of food). He instructed the staff to give the patient towels every day, which the patient at
first appreciated until 600 towels had accumulated in the patient’s room. At that point, the
patient complained about the excessive number of towels and she began to take them
out herself. She eventually reduced the number to just a few towels, a level that remained
stable thereafter.

Figure 4.7
Behavior Modification of Taking Extra Food. (left panel) Instances of taking extra food in the cafeteria
decreased when it resulted in forfeiting that meal. Occasional regression to food stealing is marked by arrows.
Weeks 20–40 are omitted, as no stealing occurred then. (right panel) Reduction in body weight that paralleled
the suppression of taking extra food.
Source: Reprinted from “Intensive Treatment of Psychotic Behavior by Stimulus Satiation and Food Reinforcement,” by T. Ayllon,
1963, Behavior Research and Therapy, 1, pp. 53–61. Copyright © 1963 Elsevier. Reprinted with permission.
110 Instrumental Conditioning: Reward

The methods of instrumental learning can be used to change our own behaviors.
Some examples of self-behavior modification are presented in Box 4.2.
Some unwanted behaviors are more than simply bad habits. Smoking has an addic-
tive component. Nevertheless, self-modification can be used in combination with other
treatments to produce desired changes in addictive behavior.

BOX 4.2 SELF-BEHAVIOR MODIFICATION

Self-behavior modification uses instrumental learning to change our own behavior. One
approach labels the components of behavior modification the ABCs, referring to Antecedents
(those stimuli that control behavior); Behaviors (the habits we want to modify); and
Consequences (reward, extinction, or punishment).
The starting point is to specify the behavior to change. Behaviors need to quantifiable,
such as the number of cigarettes smoked daily. This is readily measured and improvements
can be monitored. “Exercising more” is too vague and needs to be measurable in terms such
as number of reps or workout time.
A baseline measure of the frequency of the behavior should be recorded. How much time
do you spend exercising or studying now? Just tracking the behavior day by day can reduce
bad behaviors. Because they are habitual, we are often unaware we are doing them. You
might recruit a friend to alert you when you are acting habitually.
Try to determine the antecedents that elicit the target behavior (the A of the ABCs). Are
there stimuli or situations that seem to trigger the behavior? Do you mainly smoke when you
are with your friends, at work, or at school? One strategy to decrease the bad habit is to
avoid or eliminate the eliciting stimuli. Maybe stay away from those friends who lead you to
smoke. (Okay, I sound like your mother. Sorry). Can you shut off the phone or certain apps
for a while? Continue to monitor the frequency of the habitual behavior as the change is
implemented.
To change the behavior itself, try to substitute something different in place of the unwanted
behavior (the B of the ABCs). In response substitution, a different behavior is used to replace
the unwanted response. When you want to cut back on drinking soda, stock your fridge with
flavored water drinks as a substitute.
Re-arrange new rewarding or punishing consequences (the C of the ABCs). For instance,
impose a fine for swearing or smoking: You must put money away for each transgression.
(One particularly effective fining procedure is to pay money to your friend or roommate). To
eliminate nail-biting, you could paint your fingernails with a bitter tasting substance (sold in
pharmacies). This gives immediate punishment and gets around the problem of lack of
awareness that you are biting.
An illustration of one self-modification plan was an attempt to reduce swearing. A baseline
measure of the daily frequency of the use of certain predetermined words was established,
as shown in Figure 4.8 (Powers & Osborne, 1976). A treatment phase followed, in which the
subject tried self-fining: putting a certain amount of money in a jar each time she swore. As
can be seen, this had little effect, partially because the subject simply would not pay the fine.
The next treatment phase was more successful. The student worked out an arrangement
with her suite mates that they would immediately turn their backs on her and ignore her
whenever she swore in their presence. As can be seen in the results of this phase, the strat-
egy resulted in an immediate and sustained decline in swearing. Oh, there were occasional
regressions, like the time she had an argument with her @#/*%& boyfriend.
 Summary 111

Figure 4.8
Number of Swear Words Used. The number of swear words recorded per day during a baseline phase
and two treatment phases: the student paid a fine for swearing, or her roommates walked away from
her when she swore. * and ** indicate days when she had arguments with her boyfriend.
Source: Adapted from Fundamentals of Behavior, by R. B. Powers and J. G. Osborne, 1976, St. Paul, MN: West
Publishing.

SUMMARY
Edward Lee Thorndike sought to systematize the principles of learned adaptive behavior.
Acquisition was governed by the Law of Effect, or what later became the principle of
reinforcement: Responses are acquired based on their consequences. Thorndike’s trial
and error learning, or instrumental conditioning, consists of three elements: the discrimi-
native stimulus, the response, and the consequence.
B. F. Skinner used the label operant learning to indicate that the response operates on
the environment to produce a certain outcome. Skinner described extinction of the oper-
ant response and patterned behavior by providing reinforcement according to certain
schedules.

Positive Reinforcement
Reinforcement is defined by the experimental contingency, or rule, that relates perfor-
mance of a target behavior to a particular consequence. A positive reinforcer is one whose
occurrence increases the frequency or vigor of the behavior.
112 Instrumental Conditioning: Reward

Instrumental-response learning is influenced by several reinforcement variables.

Reinforcement is more effective if it occurs immediately after the response rather than if
it is delayed; if reinforcement is administered for each response than if only given after
some of the correct responses; and if larger or more preferred reinforcers are used rather
than smaller or less preferred rewards.
The amount of Drive, manipulated through reinforcer deprivation, increases respond-
ing. Given the learning-versus-performance distinction, variables such as drive and amount
reinforcement increase the motivation to respond, and also learning.
A schedule of reinforcement refers to the specific timing or frequency of responding
before delivery of reinforcement. A partial reinforcement schedule reinforces the response
only some of the time. The basic schedules involve reinforcing a certain number of
responses (ratio schedules) or reinforcing responses after certain amounts of time have
passed (interval schedules); the ratios and intervals may be fixed or varying.
A neutral stimulus, such as a tone, becomes a secondary reinforcer when it has been
paired with a primary reinforcer, such as food. An example is token reinforcers that can be
used to reinforce human behaviors in a variety of situations.
A powerful class of reinforcers for human behavior is social reinforcement in the form
of praise, attention, physical contact, or facial expressions. One theory says that social
stimuli are primary reinforcers; a second theory says social reinforcers are secondary
reinforcers.

Theories of Reinforcement
Skinner pragmatically defined a reinforcer as being whatever works to increase the fre-
quency of the operant (or instrumental) response. But what makes reinforcers reinforce?
According to drive reduction theory, reinforcers were stimuli that reduce biological
needs, such as hunger. However, some reinforcers, such as saccharin, do not reduce any
bodily need. Alternatively, incentive motivation theory says reinforcers are stimuli that
elicit responding through appetitive sensory or affective properties.
Another approach shows that electrical and chemical brain stimulation can serve to
reinforce instrumental responding. This suggests that certain neural processes are the
final common denominator for all reinforcers.
An alternative approach defines reinforcers as activities or behaviors. According to
the Premack principle, an activity that has a higher probability of occurrence will reinforce
a lower-probability activity. Preferences for activities are not fixed, but can vary across
individuals, and can vary within individuals due to deprivation and satiation.
Reinforcers strengthen the association between a stimulus and a response, in the
sense of cementing the connection. Reinforcers are significant stimuli that produce neu-
ral arousal that aids memory consolidation.
Finally, reinforcers convey information about correct performance of a behavior.
Biofeedback, for example, uses an external signal to provide information concerning the
performance of unobservable bodily responses.
Is reinforcement necessary for learning? Tolman and Honzik’s classic study demon-
strated that maze learning by rats occurred in the absence of reward. The learning was latent
until reward was introduced in the goal box, and then motivated rapid maze performance.
 Summary 113

Conscious awareness of the response–reinforcer contingency is a factor in human

instrumental learning. In the verbal conditioning paradigm, an experimenter “reinforces”
(“umm humm”) the use of a target word, causing an increase in the frequency of the
target word. However, conditioning seems to be limited to those participants who are
aware of and can report the contingency. Awareness benefits learning but is probably not
necessary for conditioning.
The use of reinforcement in the real world has been criticized as being manipulative
and controlling. It supplants intrinsic motivation with an external, material motivation for
performance. And, the effects of reinforcement sometimes dissipate when it is
discontinued.

Response Learning
A significant use of instrumental conditioning is in shaping particular forms of the
response. By the use of differential reinforcement and successive approximations, new
behaviors are created. In response chaining, a sequence of behaviors is constructed and
reinforcement is given after the final response.
Not all behavior can be altered through reinforcement. There are species-specific lim-
itations. Hamsters can be reinforced with food for standing but not for washing. These
limitations may be due to an evolutionarily preparedness to learn certain response–to–
reinforcer associations.

Discriminative Stimulus Control

Learning involves not only what response to make but also when to make it. A discrimi-
native stimulus, or S+, signals the availability of reinforcement if a response is made.
Responding initially trained in the presence of a particular S+ will generalize to similar
stimuli. Through discrimination training, S– comes to signal that reinforcement is not avail-
able, and so responding is then inhibited.

What Is Learned in Instrumental Conditioning?

One possibility is that an association is learned between the response and the reinforcer.
Evidence for this theory is that the response is sensitive to variations in reinforcement,
and can immediately change in reaction to altered conditions of reinforcement, as shown
in the latent learning experiments.
Thorndike and Hull instead said a connection was learned between the discriminative
stimulus and the response, with the reinforcer serving to strengthen this bond. Evidence
for S+-to-reinforcer learning is that the response persists in the face of altered reinforce-
ment conditions, taking on the character of habitual responding in the presence of certain
stimuli.
Because the S+ is frequently followed by reinforcement, classical conditioning can
occur between this stimulus and the reinforcer.
In conclusion, we acknowledge that associations among any of the three elements of
an instrumental trial are possible, with one dyad or another predominating in a given
situation.
114 Instrumental Conditioning: Reward

Applications
Environmental stimuli elicit well-practiced habits. Habit slips are unconscious intrusions
of a habit when an alternative behavior had been consciously intended. Slips occur when
we are in familiar places, and we are distracted. If two behavior sequences have identical
starting elements, the more frequent behavior may capture the less frequent behavior.
Habits are difficult to change because they occur unconsciously and are attached to
many eliciting stimuli. It may be easier to break a habit when your environment changes,
such as when change your housing location, jobs, or schools, which take you away from
the old stimuli.
Behavior modification is a field that applies the principles of operant learning to chang-
ing behavior in a variety of settings, such as schools, institutions, workplaces, and every-
day life. For example, Ayllon used reward omission to eliminate food stealing by a patient
in a psychiatric institution. He then used satiation to decrease towel hoarding.
 CHAPTER

5
Instrumental Conditioning
Nonreward, Punishment, and Avoidance

CONTENTS

Defining the Contingencies: Prediction, Control, and

Nonreward, Punishment, and Helplessness 133
Avoidance 116 Extensions of the Learned-
Extinction 117 Helplessness Concept 135
The Partial Reinforcement Extinction Summary of Learned
Effect 118 Helplessness 137
Punishment 120 Neuroscience and Aversive Learning 137
When Does Punishment Work? 121 The Amygdala and Aversive
Side Effects of Punishment 124 Learning 137
Punishment or Nonreward? 125 Avoidance Conditioning 138
Persistence 127 Applications of Aversive-Learning
Avoidance Learning 128 Contingencies 140
Theories of Avoidance Learning 129 Pet Containment Systems 140
Approach–Avoidance Conflict 130 Treatment of Obsessive-Compulsive
Approach or Avoidance as a Coping Disorder 140
Response 131 Summary 141

The previous chapter described the use of rewarding consequences to change behavior.
This chapter will concentrate on the use of aversive consequences. The outcomes of
nonreward and punishment seem obvious: You simply stop doing whatever it is that is not
rewarded or is punished. Do we need a whole chapter on this?
The effects of aversive consequences are not always so obvious. Imagine visiting a
research lab and seeing a cat pressing a lever and receiving unpleasant blasts of air to its
face each time. Bar press. Air blast. Where did this masochistic behavior come from? Is
punishment paradoxically sustaining lever pressing instead of suppressing it?
There is an old phrase that a parent would repeat just before spanking their child:
“This hurts me more than it hurts you.” There may be some truth to this. The one who
administers punishment may do so even though there is a cost for doing so. The parent’s
hurt may be more emotional, such as disappointment or regret.
Inventors are known for their persistence. They try something new, but it doesn’t work.
What do the inventors do then? They try new versions, continuing until the outcome is

DOI: 10.4324/9781003227090-5
116 Instrumental Conditioning

right. The Wright Brothers tested 200 different curvatures of a wing; Edison said he was
not exaggerating that he had tested 3,000 theories for the electric light. In cases such as
these, the lack of initial success, what we would call nonreinforcement, did not lead to
extinction. How do we reconcile the persistence of nonrewarded behavior with extinc-
tion, which is the expected outcome of nonreward?
Oddities of behavior that appear to be contrary to the laws of learning might be attrib-
uted to quirky personality traits. Some people are born persistent, or masochistic. An
alternative explanation is that the principles of learning can account for these paradoxical
behaviors.

DEFINING THE CONTINGENCIES: NONREWARD, PUNISHMENT,

AND AVOIDANCE
Instrumental conditioning is defined as the contingency arranged between a response
and an outcome. In the case of positive reinforcement, the instrumental response leads
to a rewarding outcome. Response–outcome contingencies also define the three aversive-
outcome learning procedures that are the topics of the current chapter.
In two nonreward contingencies, the target response is not followed by a positive
reinforcer. In extinction, reward is omitted after responses that once had produced pos-
itive reinforcement. Another nonreinforcement procedure is omission, in which a
selected response prevents a positive reinforcer from occurring. An omission contingency
usually implies that doing something else will lead to positive reinforcement. The inten-
tion with the two nonreward contingencies is to decrease the frequency of the response.
In punishment, a response is followed by an aversive stimulus which should decrease
the frequency of the response. Punishment is more than just withholding reward. It is
instead the application of an aversive event. If you misbehave and your allowance is with-
held, that is omission. If you misbehave and you are spanked, that is punishment. (The
term punishment is sometimes broadly used to include procedures in which reward is
omitted or withheld. Here, the meaning is restricted to the active application of an aver-
sive consequence).
In avoidance learning, an instrumental response prevents the aversive stimulus. The
intention is to increase the frequency of the response. Avoidance is also called negative
reinforcement: reinforcement because it strengthens the instrumental response; the
response increases in frequency, just as in positive reinforcement; and negative because
the response removes or prevents the (aversive) reinforcer. The term negative reinforce-
ment is often misused, frequently being misapplied to punishment. Psychologists who
study learning often use the more descriptive phrase, avoidance learning, as a synonym.
The basic instrumental contingencies of reward, omission, punishment, and avoid-
ance learning are summarized in Figure 5.1.
Reward and omission are often used together. Among several potential behaviors,
one response is rewarded (positive reinforcement) but another response is not (omission).
In the behavior modification technique of “praise and ignore,” a teacher praises certain
appropriate classroom responses and ignores (or tries to ignore) inappropriate responses.
Punishment and avoidance sometimes also pair up. A certain behavior leads to pun-
ishment, and so this response should decrease. An alternative behavior prevents the
punishment, and so that response should increase.
 Extinction 117

Reinforcing Consequence
Pleasant, Appetitive Unpleasant, Aversive

The Reward Training Punishment

response (positive reinforcement)
produces
the reinforcing (response increases) (response decreases)
outcome

The Omission, Extinction, Escape,

response Time-Out Avoidance
prevents (negative reinforcement)
the reinforcing
outcome (response decreases) (response increases)

Figure 5.1
The Four Basic Instrumental Conditioning Contingencies. The columns list the outcome, or reinforcing
event, as appetitive or aversive. The rows designate the response–reinforcer contingency: The response is or is
not followed by the reinforcer.

Sometimes a response could be described by either of two contingencies. Do you

study to get high grades (positive reinforcement) or to prevent low grades (avoidance
learning)?

EXTINCTION
If a behavior is maintained by reward, then the elimination of reward should lead to a
decrease in the response. In extinction, a reinforcer is withheld following each occur-
rence of the instrumental behavior, with the expectation that the response will extinguish.
In a study of infants, leg-movements that once shook the mobile extinguish when the
mobile stops moving. (Those results were shown in the previous chapter’s Figure 4.2).
Practitioners of behavior modification advocate extinction as an alternative to punish-
ment. However, extinction can have side effects that discourage its use. First, extinction
can produce unpleasant emotional effects, primarily frustration. The subject is frustrated
because the expected reinforcer does not occur. Frustration is evidenced by increased
activity or aggression. Second, to make matters worse, extinction can temporarily
increase the nonreinforced behavior. This is called an extinction burst. An example is
shown in Figure 5.2, with data from infants conditioned to make arm movements to
shake an overhead mobile (Alessandri, Sullivan, & Lewis, 1990). During extinction, the
mobile did not move. Movements actually increased during extinction instead of decreas-
ing due to frustration. If you put your money in the soda machine and no soda comes out,
what do you do? You push the button several more times, an example of extinction burst.
(Kicking the machine is frustration-elicited aggression). Eventually, however, the extinc-
tion procedure should cause the response to cease. (The control group in Figure 5.2
never had control over mobile shaking).
Then comes the third side effect, spontaneous recovery. After a delay interval fol-
lowing extinction, the response recovers. This is the same phenomenon that occurs in
classical conditioning. Several extinctions are often necessary to thoroughly suppress a
response. Extinction really does work. But given extinction bursts, frustration behavior,
and spontaneous recovery, you can see why some practitioners might be put off.
118 Instrumental Conditioning

Experimental group (N = 48)

Control group (N = 32)

Baseline Learning I Extinction Learning II

12
11

Frequency of Arm Pulling

10
9
8
7
6
5
4
3
2
1
1 2 3 4 5 6 7 8 9 10
Minutes

Figure 5.2
Extinction Burst. Mean rate of arm movements. During the learning phase, arm pulling activated the mobile,
producing an increase in movements from the baseline phase. During extinction, when arm movements no
longer affected the mobile, responding paradoxically increased.
Source: From “Violation of Expectancy and Frustration in Early Infancy,” by S. M. Alessandri, M. W. Sullivan, and M. Lewis, 1990,
Developmental Psychology, 26, p. 740. Copyright © 1990 by the American Psychological Association.

Extinction can have beneficial effects by eliciting new behaviors to adapt to changed
circumstances. When the old response no longer works, new behaviors are tried to
restore reward. In one demonstration, Neuringer, Kornell, and Olafs (2001) first trained
rats to press three levers (of five levers present) in a particular sequence. Pressing the
three in correct sequence was rewarded with food. Then food was withheld during the
extinction phase. The animals tried the old three-lever sequence, but also tried other
sequences and included the other two levers. This makes obvious sense in terms of real-
world adaptation: If food (or safety) is not obtained with the old behavior, the organism
must select a new behavior. Whereas one might think of conditioning as inducing inflexi-
ble, habit-like behavior, in some cases it leads to new behaviors.

The Partial Reinforcement Extinction Effect

During extinction the instrumental response will show varying degrees of persistence.
The phrase resistance to extinction refers to how persistent a response is during extinc-
tion. This can be measured by the number of times the unrewarded response is made.
Resistance to extinction is especially affected by partial reinforcement during acquisition.
As it turns out, other reward variables that slow learning of the instrumental response
actually lead to more sustained responding during extinction.
The partial reinforcement extinction effect (the PREE) is the most studied of the
factors that slow extinction. Consider an example of instrumental training under a partial
reinforcement schedule, say, reward is given for a random half of the correct responses
instead of reinforcing all of them. The data in Figure 5.3 show data on running speeds by
rats in a runway (Weinstock, 1954). During acquisition (the left panel), the group rewarded
 Extinction 119

Acquisition Extinction

100%
100%
Speed (1/Time)

Speed (1/Time)
50%
50%

1 5 10 15 1 5 10 15 20
Blocks of Five Trials Trials

Figure 5.3
Demonstration of the PREE. Speed of running by rats. During acquisition, reinforcement occurred either every
trial (100 percent) or on half of the trials (50 percent). No reward occurred during extinction, yet the previously
partially reinforced group ran faster.
Source: From “Resistance to Extinction of a Running Response Following Partial Reinforcement under Widely Spaced Trials,” by
S. Weinstock, 1954, Journal of Comparative and Physiological Psychology, 48, p. 319 (left) and p. 320 (right). Copyright © 1954
by the American Psychological Association.

with food in the goal on each trial (100 percent reward schedule) ran faster than the group
rewarded only on half the trials (50 percent reward). During extinction (right panel), when
there were no rewards, the (previously) partially reinforced animals ran faster. This is the
PREE. Similar effects are found in human tasks in which button pressing is only some-
times rewarded. We can ask our human participants to estimate their perceived likeli-
hood of being “correct” (i.e., reinforced) on the next trial. During extinction, the partial
reinforcement group maintains higher expectations than does a continuous reward group
(Lewis & Duncan, 1958).
(Students are often confused by the PREE phrasing. The PR (partial reinforcement)
refers to what happened during the acquisition phase. This is the left panel of Figure 5.3.
The EE (extinction effect) refers to what happens during the extinction phase. The labels
100 percent and 50 percent in the right panel refer to the reward schedule during acquisi-
tion. There are no rewards for anybody during the extinction phase).
The PREE at first seemed paradoxical to learning researchers. Conditioning was obvi-
ously stronger after 100 percent reinforcement and therefore should be more persistent.
The added strength of conditioning should carry through longer before dissipating.
The major explanations focus on the sequence of rewarded and nonrewarded trials
during the training phase. The transition from one trial that is not rewarded to the next trial
that is rewarded, may teach the subject that nonreward is (eventually) followed by reward.
According to the frustration hypothesis (Amsel, 1962), frustration is experienced after non-
reward on one trial. It is followed by reward on a subsequent trial. Frustration thus becomes
a discriminative stimulus for reward. The subject who is reinforced on every trial during
training does not experience frustration and so does not come to associate it with even-
tual reward. During extinction, the continuously rewarded subject suddenly experiences
nonreward, becomes frustrated, and stops running. The partially rewarded subject experi-
ences nonreward and is energized, expecting there will be reward next time.
120 Instrumental Conditioning

A different version of the theory says it is not frustration but rather memory of nonre-
ward that becomes a predictor of reward on a later trial. According to the sequential
hypothesis (Capaldi, 1971), at the start of a new trial, the participant remembers the out-
come of the previous trial and associates it with the outcome of the current trial. Though
similar to the frustration hypothesis, the sequential hypothesis’s reliance on memory
makes some different assumptions. During acquisition the trials can be spaced farther
apart because the memory of nonreward is presumed to last longer than the frustration
produced by nonreward. The frustration from the soda machine failure lasts a few min-
utes. The memory will be recalled the next day when you think twice about trying that
machine again.
The frustration and sequential hypotheses for the PREE say something interesting
about the effects of nonreward. Whereas our initial reaction to being nonrewarded might
be to give up, the absence of reward can instead become a cue for continued effort in the
hope that it eventually will be reinforced.

Extinction: An Overview
The frequency of a given behavior should decrease when that behavior is not followed by
a positive reinforcer. This expected decrease is interrupted by frustration, extinction bursts
of responding, and spontaneous recovery. The course of extinction is affected by the
history of reinforcement and nonreinforcement during acquisition. Responding during
extinction persists if reward during training was intermittent rather than continuous. The
partial reinforcement in extinction effect (PREE) has been attributed to initial learning that
nonreward is eventually followed by reward, and so nonreward becomes a cue or signal
for reward.

PUNISHMENT
In punishment, a designated response is followed by the presentation of an aversive
stimulus. In laboratory research with animal subjects, electric shock is typically used as
the punishing stimulus, mainly because the timing and intensity of shock are readily con-
trollable. Usually, an animal is first trained to perform some response, such as bar-pressing
for food. Punishment is then introduced and responding is measured again. With human
subjects, mild shocks are sometimes used, but more often loud noises, penalties, or fines
are applied.
In the present context, punishment refers to the application of a noxious outcome,
such as spanking or verbal abuse. In contrast, withholding a positive outcome, such as
your allowance, would be nonreward and thus an instance of reward omission, time-out,
or extinction. Punishment and omission have similar intended outcomes, which is to
decrease the frequency of the punished behavior. In our usage here, punishment is sep-
arate from extinction or omission.
Opponents of punishment often claim that punishment does not work, or at least
does not work as well as extinction. The evidence goes back to some early experiments
in Skinner’s lab. Estes (1944), in his Ph.D. dissertation under Skinner, compared two
methods to reduce bar pressing in rats: by extinction or by punishing the response with
 Punishment 121

125 Extinction

100

Responses/Session
75

50 Punishment

25

1 2 3 4
Punishment
Session Extinction Sessions

Figure 5.4
A Comparison of Two Means of Suppressing Bar Pressing by Rats. Both groups had been trained to bar
press for food. The animals in one condition were then punished with shock for each bar press; the animals in
the other condition experienced extinction. Extinction was used with both conditions on the next four days.
Source: Adapted from “An Experimental Study of Punishment,” by W. K. Estes, 1944, Psychological Monographs, 57(3), Figure 1,
p. 4. Copyright American Psychological Association. Reprinted with by permission.

electric shock. Notice the distinction between nonreward and punishment. The rats were
first trained to bar-press for food reward. Then, bar pressing was no longer rewarded (i.e.,
the extinction group), or bar pressing was followed by shock (i.e., the punishment group).
The results are shown in Figure 5.4. Food reward was not given during any of the ses-
sions shown in this figure. During the single treatment session, the shocked rats did
indeed bar-press less than did the nonrewarded rats. This is shown as the first point in
Figure 5.4. But on succeeding test days, when shock was no longer given, bar pressing
gradually returned. In the short run, punishment suppressed bar pressing more than did
extinction; in the long run, it was no more effective than extinction. This finding led to the
conclusion that the same end result can be accomplished by simply withholding reward.
Staddon (1995), another of Skinner’s students, notes that contemporary research
discounts this conclusion. Punishment is indeed quite effective in altering behavior, if it is
applied correctly (see what follows). In laboratory studies, the most persistent respond-
ing is often motivated by punishing consequences.

When Does Punishment Work?

Response-Contingent Punishment
Punishment is defined by a response-to-consequence rule. An effective punishment is
one that is administered contingent upon the response. A specific response is punished.
It is important to demonstrate that response suppression is truly due to the punish-
ment contingency and not to nonspecific effects of punishment. Once again, a control
group is needed to assess the effect of presenting punishment in no particular relation-
ship to the response, versus the effect of punishment that is delivered contingent on a
specific response.
122 Instrumental Conditioning

Intensity
A more intense punisher is more effective than a less intense punisher. Numerous stud-
ies have shown that a stronger shock will suppress bar pressing more quickly and more
permanently than will weaker shocks.
This immediately poses an ethical quandary, because our tendency is to start with
mild punishment and gradually increase the intensity if the weaker punishment doesn’t
work. Unfortunately, this actually decreases the effectiveness of the intense punishment.
J. S. Brown (1969) introduced shock in a maze that the rats ran through on their way to
the goal box. By gradually increasing the shock level from 0 to 40 volts across days, run-
ning was essentially unaffected. Beginning with the intense shock would have immedi-
ately stopped the running. Adaptation to punishment occurs when it gradually increases
in intensity.

Delay of Punishment
Punishment is more effective if it is applied immediately after the target behavior, and
decreases in effectiveness the longer it is delayed. Solomon, Turner, and Lessac (1968)
punished dogs by hitting them with a rolled newspaper for eating food placed next to the
experimenter. Punishment was given immediately after the dog began to eat, after
5 seconds, or after 15 seconds. The dogs all learned not to eat when swatted under each
of these delay conditions. The challenging test of the effectiveness of punishment was a
series of trials in which the dogs were returned to the room, with the experimenter
absent, to see how long they would resist eating. The 15-second-delay animals lasted
about 3 minutes before eating. The 5-second-delay dogs avoided the food for seven test
days before breaking down. The dogs who had received immediate punishment resisted
eating for two weeks.
In many circumstances, punishment is necessarily delayed. Corporal punishment in
schools, trials in the legal system, and waiting in your room until your parent comes home
all involve delayed punishment.

Schedule of Punishment
Punishing each instance of a response is generally more effective in suppressing the
behavior than punishing the response only some of the time.

Incompatible Responses
The punishing stimulus can elicit behaviors that are incompatible with the desired out-
come. Have you ever seen a harried parent trying to quiet a crying child by yelling at
them? Similarly, punishing a child for aggression may provoke more aggression in return.

Concurrent Reinforcement
The effect of punishment can be neutralized if positive reinforcement of the unwanted behav-
ior occurs along with punishment. In one sense, this happened in Solomon’s study cited
earlier. Dogs that received delayed punishment also received more food before being hit.
Before punishment is used, possible alternative sources of reinforcement in the situ-
ation should be considered. For instance, punishment of a student by a teacher may be
counteracted by the positive social reinforcement from the student’s peers.
 Punishment 123

Providing a Verbal Rationale

With children, we can provide verbal instruction to supplement punishment. But will pun-
ishment be made more effective by explaining the behavior–punishment contingency? In
one demonstration, less severe punishment was made more potent by adding instruc-
tion. Cheyne, Goyeche, and Walters (1969) arranged a situation in which children were
tempted to reach over and touch a desirable toy. This behavior was punished by a sudden
loud blast of noise. In the absence of any other instructions, the loudest buzzer was most
effective in suppressing future attempts to touch the toy. However, when a verbal admo-
nition not to play with the toy followed the buzzer, a lower-intensity noise was most
effective. As shown in Figure 5.5, without instruction the kids attempted another reach
within 20 seconds, even after punishment with the loud noise. When a verbal admonition
was added, the latency to reach was literally off the graph, approaching 300 seconds. The
kids did not reach for the toy, and the experimenters basically quit waiting.
This experiment demonstrates the benefit of using a verbal explanation to supple-
ment punishment. It also shows that a strong punisher may actually inhibit learning.
Cheyne et al. (1969) ask us to consider the distracting and disruptive effects that punish-
ment may have on learning. The child punished with the loudest noise may have been too
upset to attend to and remember the verbal admonition.

300

250

100
Latency (sec)

80

60

40

20

0
None Moderate Loud None Moderate Loud
No verbal demand Plus verbal “don’t touch”

Figure 5.5
Effect of Verbal Reprimand on Punishment Effectiveness. Mean number of seconds before child reached
toward a desirable toy following a moderate or loud noise punishment or no punishment. The first three groups
received no verbal reprimand. The other three groups received a verbal admonition following the noise.
Source: Based on Cheyne, Goyeche, & Walters (1969).
124 Instrumental Conditioning

Individual Differences
There are individual differences in susceptibility to punishment’s effects. For instance,
individual differences have been demonstrated in comparing breeds of dogs. Freedman
(1958) used the Soloman procedure in which dogs were slapped with a newspaper and
told “no” when offered a bowl of food in the laboratory room. Shetland sheepdogs
refused to eat during subsequent tests, whereas basenjis and beagles ate in spite of
punishment.

Side Effects of Punishment

Whatever desired benefit punishment holds for changing behavior, it also produces
unwanted side effects. These include fear, aggression, and avoidance of situations asso-
ciated with punishment.

Conditioned Fear and Avoidance

The punishment situation is exactly that used to condition fear through classical condi-
tioning: A neutral CS is paired with an aversive US. The place in which punishment occurs
or the person doing the punishing becomes a conditioned stimulus that evokes condi-
tioned fear. For example, if a child is punished in school, the school or the teacher could
come to elicit fear. The child’s fear could interfere with academic performance, or the child
may try to avoid school.

Aggression
Punishment can elicit aggression. In early studies of the punishment–aggression link,
pairs of rats placed in a conditioning chamber were given electric shocks. The rats began
to nip at one another, what is called shock-elicited aggression (Miller, 1948). The parallel
to the human case is obvious: Physical punishment can provoke aggression. If the pun-
ishment is for aggression in the first place, then we have perversely designed a method
to increase, rather than decrease, the undesired behavior.
Matters get worse. The aggressive behavior might be inadvertently reinforced. Each
time the shocked rats attacked one another, Miller turned off the shock, thereby reinforcing
biting. (This is negative reinforcement. The response of biting increases because it causes
shock to be terminated). A child’s aggression against siblings could be reinforced by a
decrease in their annoying or disturbing behavior (Snyder, Schrepferman, & St. Peter, 1997).
The punishment-aggression hypothesis has been tested in human studies. In some
cases, college students are exposed to mild loud noise as the aversive stimulus. The
students are then asked to perform some other task in which aggression might be dis-
played, such as a cooperative game with others who are present, to earn points or
money. The prior aversive conditions evoke a tendency to “punish” others by withhold-
ing points or imposing loud noises on the other players for poor performance (Bushman
& Baumeister, 1998).

Paradoxical Rewarding Effects of Punishment

Pairing a punishing stimulus with a positive reinforcer can convert the punisher into a
secondary reinforcer. Punishment could then inadvertently reinforce behavior rather than
suppress it. The work of psychiatrist Jules Masserman was mentioned in the introduction
 Punishment 125

to this chapter. Masserman (1943) trained cats to lever press for unpleasant blasts of air
to the face. Where did this perverse behavior come from? If we had visited the lab earlier,
we would have seen Masserman first training the cat to lever-press for food. Then the air
blasts were added, only occasionally and mild enough so as not to disrupt bar-pressing.
By gradually increasing exposure to the air blasts, and tapering off the food rewards, he
eventually had the cat bar pressing for punishment. Here, the air blast became a condi-
tioned reinforcer by virtue of its pairing with a positive reinforcer (food). Gradually increas-
ing the intensity of punishment minimized its power to suppress behavior. Knowing the
learning history of an organism can sometimes explain behavior that appears irrational.

Punishment or Nonreward?
Modern advocates of behavioral modification advise withholding reward as an alternative
to punishment (again, the latter defined as the application of an aversive stimulus).
Extinction or reward omission are each alternatives to punishment. But are nonreward
and punishment all that different in their side effects? Both function similarly in several
ways. The frustration provoked by nonreward can elicit aggression, just as punishment
does. Animals and people try to escape from stimuli associated with nonreward, just as
they attempt to escape from stimuli that have been paired with punishment (Wagner,
1969). Each can increase the persistence of behavior, sometimes the unwanted behavior.
Maybe the deciding difference between the two is that withholding reward is categori-
cally different from the application of aversive stimuli as used in punishment.

Spanking
Surveys of American adults have consistently found 60–80 percent agreement that some
physical punishment of children is acceptable. One poll found that 70 percent of parents
agreed with the statement that “a good, hard spanking is sometimes necessary to disci-
pline a child” (Cuddy & Reeves, 2014). In a more positive vein, a survey repeated over a
25-year span found a decline in the number of parents who said they actually spanked their
children, from 50 percent in 1993 to 35 percent in 2017 (Mehus & Patrick, 2020).
What are the long-term effects of physical punishment on children? There have been
hundreds of studies on children’s emotional and cognitive outcomes, making spanking
one of the most studied parental behavior. Gershoff and Grogan-Kaylor (2016) published
an influential meta-analysis covering 50 years of research. A meta-analysis is a statistical
combination of the results of numerous individual studies. Spanking was found to have
long-term associations with negative psychological outcomes in 13 categories. There
were increases in childhood aggression, behavioral and mental health problems,
decreases in cognitive ability, and antisocial behavior in adulthood. Interestingly, adults
who reported that they had been spanked often supported the use of physical punish-
ment. Not all research found negative effects, but rarely were any positive benefits of
punishment found.
Individual studies can be challenged for definitional and methodological limitations
(e.g., Hicks-Pass, 2009). How do you distinguish “customary” or “typical parental usage”
from harsh or abusive physical punishment? Since these studies are correlational, they
cannot prove causation and do not control for preexisting differences between spanked
126 Instrumental Conditioning

and non-spanked children. Retrospective evaluations of childhood punishment (“Were

you punished often? Severely? Unfairly?”) are subjective judgments.
A survey of dog owners’ use of physical punishment reached the same conclusion as
the human studies: Namely, there is a correlation between greater use of punishment
and an increased number of problem behaviors (Hiby, Rooney, & Bradshaw, 2004). Do
problem dogs simply misbehave more often than nonproblem dogs and thus need more
punishment? Or does punishment actually produce more misbehavior? The assessment
of cause and effect shows the ambiguity in studying punishment in the real world.

Should Punishment Be Used?

The answer here requires a consideration of both efficacy (i.e., does punishment work?)
and ethics (should it be used?). In the first case, we can ask whether punishment in a
particular case will be applied according to parameters that are effective. Punishment is
effective when it is intense, immediate, consistently applied, etc.
The answer in the second case is that punishment involves moral and ethical issues
as much as it involves scientific ones. As a behavior modifier, you could decide not to use
punishment as a moral decision, no matter how well it works. Or you might abstain from
punishment on scientific grounds if you conclude that the conditions under which it is
being administered are known to be ineffective.
The pragmatic and ethical questions have come up often in the past have spurred the
development of alternative methods of controlling unwanted behaviors. Some of these
techniques are described in Box 5.1.

BOX 5.1 ALTERNATIVES TO PHYSICAL PUNISHMENT

The usually cited alternative to physical punishment is withholding reward. Screen time, des-
sert, trips, or whatever are taken away or forfeited. However, there are rules for how to best
omit reward just as there are rules for administering reward and punishment. The omission
should be contingent on the unwanted response, immediate, large enough, and consistently
withheld. In addition, there are some procedural variations for withholding rewards to sup-
press unwanted behaviors.
Misbehavior is maintained by some sort of reinforcement. In extinction, the reinforcer is
removed and the frequency of misbehavior should decrease. If the positive reinforcer is
attention, peer approval, or just the pleasure of aggravating your parents, then remove the
reward.
Time-out is another alternative. Time out is contingent on certain misbehaviors and the
child is removed from the situation (e.g., the play group, the classroom). This prevents obtain-
ing rewards for other behaviors that are missed during the time-out period. This can include
rewards for misbehavior, such as peer approval, and rewards for good behavior.
Response cost is essentially a means of fining or taking-back rewards. The target behavior
triggers loss of some accumulated reward. If the child is earning points, screen time, money,
etc., misbehavior causes some amount of this reward to be taken back.
In differential reinforcement of other behavior some other behaviors are selected that are
explicitly rewarded. The idea is that the subject will engage more with the rewarded behav-
iors and less with the unrewarded behavior.
 Persistence 127

Response cost (RC) and differential reinforcement of other (DRO) are flip-sides of each
other. In RC, the participant starts out with some number of tokens, and loses reward for
inappropriate behavior. In DRO, the participant starts with no tokens and earns them for
appropriate behavior. The penalty for misbehavior is a missed opportunity to earn reward.
DRO and response cost were compared in a study using young school children (Hirst,
Dozier, & Payne, 2016). In the differential reinforcement of other behavior, the children earned
tokens for working on certain tasks. Being off-task and not working received no points. In the
response cost condition, the children started off with certain number of tokens, and then lost
points for going off-task. On average, each procedure was successful in maintaining on-task
behaviors, and about equally well. For example, in one comparison, the children earned about
10 tokens as rewards in the DRO condition. In the RC condition, the children ended up with
about 10 tokens after penalties. Although approximately equally effective, individual children
had different preferences for one or the other: When given a choice, some kids chose DRO
and others chose response cost schedules for future tasks.
Alan Kazdin, an early and leading proponent in applied behavior analysis, listed some
advantages of omitting rewards as an alternative to physical punishment. Methods such as
DRO and response cost have been effective in suppressing many behaviors, including smok-
ing and overeating; the methods have worked with different populations, from children to
adults, and in normal and clinical populations. Suppression of the unrewarded behaviors is
lasting, and does not spontaneously recover after omission ceased. (Remember the Estes
and Skinner finding? As soon as shock stopped, the rats went back to bar pressing). Finally,
there are fewer side effects than with physical punishment. Obviously there is no pain or risk
of injury. Emotional reactions are milder: There may be frustration at losing rewards, but not
the fear or aggression seen with physical punishments (Kazdin, 1972).

PERSISTENCE
Extinction and punishment are used to eliminate behaviors. However, there are certain
behaviors we wish would persist, even in the face of unrewarding consequences.
Persistence is the continued performance of an instrumental response even though the
outcome is not rewarded or is punished.
We admire the inventors, artists, scientists, or entrepreneurs who work persistently
with little reward. As noted in the introduction to this chapter, Thomas Edison claimed to have
tried 3,000 different materials for the element (the part that glows and produces the light) of
his electric light bulb (Evans, 2004). Wouldn’t anyone else have given up at, oh say, a thou-
sand, maybe two thousand, failures? The question here is why do some individuals persist in
the face of failure? Principles of learning might explain such counterintuitive behavior.
We already know that the use of partial reinforcement during acquisition can make a
response more resistant to extinction (e.g., Figure 5.3). The partial reinforcement extinction
effect (the PREE) is an instance of a more general principle: Experience with any of several
aversive outcomes can induce persistence. Experience during the acquisition phase with
delayed rather than immediate rewards, or with small rather than larger rewards, also
increases the resistance of the response to extinction. That is, persistence is generalized.
(The expected outcomes in generalized persistence are illustrated in Table 5.1). Notice that
these are opposite of the effects of these same variables during acquisition: Initially learn-
ing a response proceeds faster with continuous, immediate, and larger rewards.
128 Instrumental Conditioning

Table 5.1 Generalized Persistence: The Relationship between Acquisition Conditions and
Subsequent Persistence during Extinction

Partial Reinforcement
Phase I: Acquisition Phase II: Extinction
100%: Rewarded for each response
50%: Rewarded for some responses The 50% group is more persistent
Partial Punishment
Phase I: Acquisition Phase II: Extinction
Food: Food given after each response
Food + shock: Food after each response, Food + shock is more persistent
and shock after half the responses
Delayed Reinforcement
Phase I: Acquisition Phase II: Extinction
Immediate: Reward immediately after each response
Delay: Reward given after a delay Delayed reward is more persistent

The fact that punishment, delayed reward, or nonreward can increase persistent sug-
gests they have a common underlying factor. Possibly transfer occurs across aversive
consequences that share similar negative emotional effects. Negative emotional reac-
tions are produced by delaying reward, giving a smaller reward, only occasionally giving
reward, and punishment. Each of these outcomes may differ in the details of the emotion
(frustration, disappointment, fear), but they all fit the category of negative reactions.
Persistence can also generalize from one situation to another. Generalized persistence
was demonstrated in a study that tested persistence across different tasks (Eisenberger,
Heerdt, Hamdin, Zimet, & Bruckmeir, 1979). Depressed in-patient residents were asked to
perform some ward chores, such as picking up, putting things away, making coffee, etc. In
the continuous-reward condition, each single chore was rewarded with a “Thank you.” In
the partial-reward condition, three or four chores were performed before the patient was
thanked. Later, a different person asked for some help in sorting computer punch cards.
Patients in the partial-reward group were more persistent in this task as measured by the
number of cards sorted and the amount of time spent sorting. Persistence generalized
beyond the initial task of ward chores and extended to requests from another person.
Why is generalized persistence important? For two obvious reasons: bad habits are
often persistent, even in the face of unpleasant consequences; and good habits are ones
we would like to persist in spite of unpleasant consequences.

AVOIDANCE LEARNING
In avoidance learning, an instrumental response prevents an aversive outcome. As a
result the response should increase in frequency or strength. You click your seat belt
before the reminder buzzer, to prevent its sounding. Studying can be a response to avoid
a bad grade. A familiar example of failure to avoid is the scene in every horror movie when
someone is confronted with a door. We all know “Don’t open the door!” Not opening it
would be avoidance, although avoidance rarely occurs in films.
 Avoidance Learning 129

A typical avoidance learning task begins with a warning signal (WS), which is followed
a few seconds later by an aversive stimulus. The performance of a selected instrumental
response during the WS prevents the aversive stimulus and terminates the trial. A stand-
ard task in animal experiments uses a two-compartment, or shuttle-box, apparatus. When
the WS sounds, usually a tone, the animal has to move from one compartment to the
other to avoid being shocked through the metal bars on the floor of the apparatus. Another
trial begins shortly; the WS sounds and the animal hops back to the first compartment to
avoid shock.
Animal subjects often figure out ways to avoid shock, showing that they are at least
as clever as their experimenters. The walls of conditioning chambers are made of smooth
sheet metal or Plexiglas. Otherwise, rats and mice jump up and cling to a seam, ledge, or
even a screw head until the WS goes off. Some rats discover that by standing on their
hind legs on one floor grid they could interrupt the electrical circuit (Broadhurst, 1963).
One rat even rolled over onto his back, using his fur as an insulator while continuing
bar-pressing for food! (Schwartz, 1978).

Theories of Avoidance Learning

Conditioning Theory
A seminal and influential theory of avoidance learning is the two-process, or Watson-
Mowrer, theory (Mowrer, 1947). The pairing of the WS with shock conditions fear to the
signal via classical conditioning. This is exactly the mode of fear and phobia learning pre-
sented in Chapter 3. Little Albert learned to fear the white rat after it was paired with a
loud noise, and is the Watson part of the two-process theory. Once the WS is conditioned,
escape from the signal is reinforced by fear reduction. Albert crawled away from the rat,
and so reduced his level of fear. Escape from the WS is instrumental conditioning, the
second process, added by Hobart Mowrer. Note that in two-process theory, the subject
is motivated to escape the WS, rather than avoid what the signal signals. That is, Albert is
mainly motivated to escape from the rat (the WS) rather than from the loud noise that had
been paired with the rat.
Two-process theory suggests that termination of the WS after a correct response is
critical for avoidance learning. After all, escape from the fear aroused by WS is what is
important. Albert is trying to get away from the rat and not the noise. The data are con-
sistent with this prediction. If WS termination does not occur promptly, avoidance condi-
tioning is markedly impaired (Kamin, 1956).
There are aspects of avoidance that a conditioning theory does not explain. One of
the remarkable features of avoidance behavior is its persistence. Well-trained animal sub-
jects will make the avoidance response for hundreds of trials. Phobic individuals will avoid
their feared target for years. On all these successful avoidance occasions, the aversive
outcome does not occur. Why doesn’t fear of the WS extinguish? The WS occurs repeat-
edly and is not followed by an aversive outcome.

Cognitive Theory
It seems obvious to us that avoidance persists because you have acquired a pair of expec-
tations: Responding prevents an aversive outcome, and if you do not respond, you can
expect to be punished. The first expectation is continuously verified: If I respond, I don’t
130 Instrumental Conditioning

get shocked. The second expectation is never subjected to a reality check: not responding
to see what happens now. This explanation is the basis of a cognitive theory of avoidance
learning (Bolles, 1972; Seligman & Johnson, 1973). Eliminating avoidance behavior requires
modifying both expectations. That is, the subject must learn that the warning stimulus no
longer signals danger, and that not responding will not lead to punishment. This is the goal
of behavioral therapy for anxiety disorders that include avoidance responses.

The Functional Approach

In our laboratory studies, some avoidance responses prove difficult to train. Rats will
readily learn to bar-press for food but not-so readily to bar-press to avoid shock. A func-
tional approach to avoidance learning studies the natural behavior of the organism’s reac-
tion to threats in the wild. These reactions contribute to the ease or difficulty of learning
certain avoidance responses.
Robert Bolles labelled these natural behaviors species-specific defense responses
(SSDRs; Bolles, 1970). SSDRs are innate responses that are primed in a fearful or threat-
ening situation, and so could readily be “learned” in a laboratory study. (To say we have
conditioned the response is not really accurate; we have simply increased the frequency
of what a scared animal does anyway). As one example of an SSDR, mice in a fearful
situation readily learn to remain immobile (or freeze). It makes sense for a mouse in the
wild to remain motionless to avoid detection. A threatened rat may stand up on its hind
legs to defend itself with both paws and mouth. In parallel, the lab rat learns to stand on
its hind legs and rotate a wheel with its front paws to avoid shock. It is easier to teach rats
to wheel-turn than to bar-press to avoid shock.
The functional approach says that avoidance learning is affected by principles of learn-
ing and by principles of natural behavior.

Summary of Avoidance Learning

In avoidance learning, a response is acquired that prevents an aversive outcome. This is
also called negative reinforcement: An instrumental response escapes or prevents an
aversive outcome (“negative”) and that response should then increase in frequency
(“reinforcement”). According to the two-process theory, the warning signal first becomes
associated with the aversive outcome. Thereafter, termination of this signal is reinforced
by fear reduction. The cognitive theory of avoidance learning states that expectancies are
learned: the belief responding prevents the aversive stimulus. A functional approach stud-
ies the contribution of species-specific defense responses, the natural responses that are
primed in a fearful situation, to learning.

Approach–Avoidance Conflict
Why would anyone persist in behavior that is punished? The sensible thing to do is to
cease and desist. However, a punished behavior might persist if it has other sources of
motivation. Perhaps it was previously reinforced, as in the case of Masserman’s cats,
and reinforcement may still be available. Children misbehave because the misbehavior
is fun.
 Avoidance Learning 131

When a behavior has opposing outcomes, positive and aversive, psychological con-
flict results. A child is drawn to a parent who provides nurturance and security, but the
child also fears outbreaks of anger. A dog that has been punished is simultaneously drawn
by anticipation of food or petting and repelled by the threat of further punishment.
Although either approach or avoidance may predominate, there can be a point where the
strength of the approach tendency is about equivalent to the strength of the avoidance
tendency. The opposing motives produce vacillation of behavior, a pattern of called
approach–avoidance conflict (Miller, 1959).
Given that a behavior has been both rewarded and punished, what determines
whether approach or avoidance occurs? One factor is the relative intensities of the out-
comes: the magnitude of the reward versus the severity of the punishment. A second
factor is the proximity to the consequence, physically or temporally. The tendency to avoid
is stronger the closer you get to the previously rewarded/punished goal.

Approach or Avoidance as a Coping Response

The distinction between approach and avoidance has been extended to characterize indi-
vidual styles of coping with stressful situations. In dealing with traumatic stressors, such
as the after-effects of a natural disaster or an accident, some people avoid thinking about
the situation whereas others confront the stressor. The avoidant coping style has been
variously labeled as blunting, selective inattention, and denial (see Roth & Cohen, 1986).
For example, when confronted with a frightening medical diagnosis, one could avoid
talking about the illness, deny its severity, or engage in distracting activities to block
thinking about the illness. The approach-coping style has been labeled as monitoring,
selective attention, and sensitization. The approach orientation might lead one to seek out
more information about the illness, talk about it, or join a support group.
Approach and avoidance styles of coping were demonstrated in college students who
participated in a shock-avoidance experiment (Averill & Rosenn, 1972). The students lis-
tened to an audiotape that had two tracks: One track played music and the other track
would sound a tone warning of an upcoming shock. The subjects could switch between
tracks. Whereas some students switched back and forth between the music and the warn-
ing, other subjects chose to listen to the music exclusively and ignore the warning signal.
Sometimes the anticipation of unpleasantness is worse than the actual aversive stimulus.
Is one coping style better than another? The answer partially depends on whether the
aversive outcome is controllable or not. In one study, students who were anticipating
midterm exams engaged in an approach coping style by actively preparing for the exams.
After the exams, but before grades were posted, the students used distancing and avoid-
ance coping strategies. At that point, the stressor was beyond their control and the out-
come was unknown (Folkman & Lazarus, 1985).
Imagine having to return to the scene of an accident. To approach a stressor provokes
more immediate fear; avoiding the situation minimizes fear. However, approach allows
habituation, counterconditioning, or extinction of the fear, and so may lead to more long-
term reduction of anxiety.
Coping reactions take other forms. Maybe it’s best just to not remember. Forgetting,
suppression, and repression are all forms coping with unwanted memories (see Box 5.2).
132 Instrumental Conditioning

BOX 5.2 REPRESSION AND AVOIDANCE

Our thoughts and memories sometimes make us anxious. Think of your most embarrassing
moment or an instance of danger in your life. To prevent distress, we try not to think about
these things. Repression is an extreme version of “not thinking” about something. In
Sigmund Freud’s theory, unpleasant or unacceptable thoughts or ideas were repressed into
the unconscious. Learning theorists later translated repression into avoidance learning.
Repression, or avoiding certain thoughts, is reinforced by a reduction in fear (Dollard &
Miller, 1950).
Repetitive negative thinking characterizes several psychological disorders, including
depression and anxiety. Repetitive thinking may be an instance of habit learning. Just as
habits become automatic through repetition, unwanted thoughts could similarly develop.
One solution to behavioral habits is to practice not-performing them. Possibly the same idea
would work with repetitive thinking.
M. C. Anderson has developed a method to study not-thinking about newly learned asso-
ciations. Anderson and Green (2001) first had their subjects learn 40 word-pairs (i.e.,
paired-associate learning). The subjects were then given additional practice with some of the
associations, by showing the first item and requesting verbal recall of the second. For other
word pairs, the subjects were instructed not to reply with, or even think about, the response
to the cue word. That is, the subjects practiced not remembering those associations.
Anderson calls this suppression; suppressing the associated responses. After 16 practice
suppression trials, the subjects were asked to recall the associated items. The subjects
recalled fewer of the suppressed word associations. Anderson says that we can learn to
block memories from entering consciousness. When encountering a cue to an unwanted
memory, repression keeps the associate from being retrieved.
“Not thinking” the response word is analogous to not making a habitual movement: you
start to respond but then catch yourself. For example, I start to dip my spoon in the sugar
bowl…but then stop. This is response override (Anderson & Levy, 2009). With repetition,
inhibition replaces making the response. Similarly, you can learn to inhibit the word associa-
tion. The same neural mechanisms may be involved in suppressing memory retrieval as in
suppressing physical responses.
Everyday memory lapses are sometimes attributed to a motivated desire to forget. You
“forget” (in air quotes) an appointment that you didn’t want to keep. Forgetting is not the
problem. In fact, “People are not likely to forget unpleasant intentions; they may, in fact, think
about them obsessively…but they may very well not carry them out when the time comes”
(Kvavilashvili, 1992, p. 514).
As an alternative to avoidance of remembering unpleasant events, you might be able to
psychologically distance yourself when you do remember. Self-distancing views the experi-
ence from the vantage point of an observer, someone watching the event unfold, rather than
from point of view of yourself looking out. Self-distancing arouses less negative emotion in
remembering the traumatic event, and produces a smaller rise in blood pressure (Ayduk &
Kross, 2008). Studies of post-traumatic stress disorder (PTSD) in Vietnam War veterans show
that the self-distancing perspective spontaneously occurs in some cases (Kenny et al., 2009;
McIsaac & Eich, 2004).
Why are some traumas repeatedly remembered, and others simply repressed? Why do
some people experience post-traumatic stress and others do not? These are some of the
challenging questions awaiting research answers.
 Prediction, Control, and Helplessness 133

PREDICTION, CONTROL, AND HELPLESSNESS

There are two independent dimensions that contribute to our understanding of aversive
events. One is predictability: Is an aversive event predicted and anticipated in advance of its
occurrence? The S+ predicts that the instrumental response will produce (in punishment)
or prevent (in avoidance) the aversive consequence. The second dimension is controllabil-
ity: Can the aversive outcome be controlled by an instrumental response? In punishment
and avoidance, the outcome is controllable: there is a response that if withheld, prevents
punishment; and there is a response that if made, avoids the aversive outcome.
But what if the aversive stimulus is not controllable? What if there is no response that
either produces or prevents the punishing stimulus? Learned helplessness is learning
that there is an explicit lack of contingency between responses and an aversive outcome.
Experience with uncontrollable stressors can lead to passivity in the face of subsequent
stressors.
Learned helplessness is demonstrated using a two-phase experimental design: one
phase to induce helplessness and a second phase to assess the effects of helplessness
training. Table 5.2 shows the design of a helplessness conditioning study using rats.
During the first phase some rats are given tail shocks. The escapable-shock group is given
shocks that can be escaped by performing an instrumental response, such as turning the
wheel to terminate shock. The inescapable group is given unavoidable shocks. These two
groups receive the same duration and sequence of shocks, the only difference being
whether there is a response that can terminate the shocks. A third control group is left
untreated in this phase of the study. In the next phase, all three groups are trained on a
new avoidance response. Typically, animals are conditioned to shuttle between two com-
partments in reaction to a warning signal of impending shock.
The results of a study of learned helplessness are shown in Figure 5.6. The measure of
learning is the latency, or amount of time it takes to run to the safe compartment on suc-
cessive trials during Phase 2. The animals which were not shocked and those that received
escapable shock during Phase 1 rapidly learned the avoidance response in Phase 2. That is,
they responded more quickly over successive trials. The animals given inescapable shock
in the first phase showed no learning across trials, and their latencies remained at the
60 seconds maximum allowed in each trial (Maier, Seligman, & Solomon, 1969).
What are the rats learning in this experiment? At the beginning of the first phase,
both shocked groups of rats have the same experiences and reactions. Shocks occur,
which elicit unconditioned defensive reactions. This includes activity in the search for
escape or avoidance of the shock, but also or eventually freezing and inactivity. These are
natural reactions to threats in the wild. However, the escapable shock animals stumble on

Table 5.2 Design of the Learned-Helplessness Experiment

Phase I Phase II

Control group No training Avoidance training

Escapable group Escapable shock training Avoidance training
Inescapable group Inescapable shock Avoidance training
134 Instrumental Conditioning

60

50 Inescapable

Median Latency (sec)

40

30

Escapable Shock
20

Control
10

0
1 2 3 4 5 6 7 8 9 10
Trials

Figure 5.6
Latency to Escape After Inescapable Shock Preexposure. Average latency to escape shock during the second
phase, after previous experience with escapable shock, no shock, or inescapable shock. The maximum allowed
latency of 60 seconds was reached by the helpless animals.
Source: From “Pavlovian Fear Conditioning and Learned Helplessness,” by S. F. Maier, M. E. P. Seligman, and R. L. Solomon, in
Punishment and Aversive Behavior (p. 328), edited by B. A. Campbell and R. M. Church, 1969, New York: Appleton-Century-
Crofts. Reprinted by permission.

a response that turns the shock off. At this point the two groups diverge in what they
learn. The inescapable shock group acquires freezing and passivity as a conditioned
response. The escapable shock group learns an active instrumental response to terminate
shock.
Most early theorizing emphasized the negative aspects of helplessness, the lack of
control over outcomes. However, learning by the escapable shock condition emphasizes
the dimension of controllability. Fear, passivity, and inactivity are unconditioned reactions
to a stressor. However, avoidance learning, the belief in controllability, and active attempts
to escape indicate learned behaviors.
Helplessness-like phenomena occur in humans in simple laboratory tasks. For exam-
ple, college student participants are exposed to loud tone pulses that are “unpleasant but
not harmful.” The students are seated in front of a small box with a push button on it. The
students are told “when the tone comes on, there is something you can do to stop it.”
Actually, only those in the escapable-noise group could turn off the tone with a button
press. For the inescapable-noise participants, the button did not work. After 30 tones, the
subjects are escorted to a different apparatus, a version of a shuttle box. This is a box with
a knob on top that obviously slides from side to side. The participants are again told there
is something they can do to terminate the tones. This is the crucial task, the one used to
measure the effects of helplessness training. Prior exposure to inescapable tones in the
push-button phase of the study lead to slower reactions in the second phase when the
tones were now avoidable. Students exposed to escapable loud tones responded more
quickly in Phase 2 (Hiroto, 1974).
 Prediction, Control, and Helplessness 135

What are the reactions to helplessness training? Seligman (1975) suggests that the
experience with an uncontrollable stressor produces emotional, motivational, and cogni-
tive deficits. Emotional deficits are shown by various psychosomatic illnesses, such as
the development of ulcers (e.g., Weiss, 1977). Motivational deficits are shown by a lack
of initiative to respond, thus producing the maximal response latencies shown in
Figure 5.6. The subjects do not even try to escape the shock.
Cognitive deficits can be described as the absence of belief in the controllability of
the aversive outcome. No matter what responses are attempted, they will be unsuccess-
ful. With our college students, we can ask them to estimate their expectation for success
on their next attempt at a task. What we’ve learned from hundreds of lab studies with kids
is that each successful solution should increase the expectation that more success will
follow. However, the expectations of students exposed to helplessness conditions are
unchanged by success. They do not believe they will do any better.
The cognitive belief in uncontrollability may be more important than actual experi-
ences with uncontrollable stressors. This was demonstrated in a study of the effects of
uncontrollable noise on psychological performance (Glass & Singer, 1972). College stu-
dents were exposed to aversive noise presented via headphones. The sound was a
100-decibel mix of office machine noises and people speaking in several languages. One
group was told there was a panic button they could hit to briefly turn off the noise, but
were asked to resist using it. Thus, these participants believed they had some control
over the noise, although in fact they did not exercise this control. (We might call them
the “potentially” escapable-noise group). The inescapable-noise participants were not
told about a panic button. After termination of the noise-exposure phase of the study,
both groups were given puzzles to solve. The panic-button students were more persis-
tent in working at puzzles, whereas the inescapable-noise group gave up more quickly.
This study shows that it is the perception of control, acquired in the first phase, that is
important.

Extensions of the Learned-Helplessness Concept

Several extensions of helplessness have particular relevance to human learning, in appli-
cations to psychopathology, causal attributions, and health.

Depression
Seligman (1975) noted the similarities between helplessness and depression in their
symptoms, cause, and treatment. (After completing his degree in animal conditioning,
Seligman moved on to clinical psychology. We could define this field of work as experi-
mental psychopathology). The symptoms of both helplessness and depression include
reductions in activity, aggression, and motivation; and disruptions in eating, sleeping, and
sexual behavior. Helplessness and some depressions have similar etiologies: They can be
caused by traumatic experiences that are uncontrollable.
One remedy for helplessness is to force exposure to the fact that escape is possible.
In one study, dogs that were passive in the presence of shock were physically dragged
from the shocked side of the shuttle box to the safe side (Seligman, Maier, & Geer, 1968).
This procedure of forced exposure to shock escape was successful in teaching the
136 Instrumental Conditioning

avoidance response. One parallel therapy for human depression is assertiveness training.
Depressed individuals are forced into action, both to demonstrate that responding leads
to reinforcing outcomes (the behavioral component) and to alter the belief that all efforts
are useless (the cognitive component).

Causal Attribution Theory

Abramson, Seligman, and Teasdale (1978) described further the complex role of beliefs
in the production of helplessness. Their description overlaps with attribution theory, a
concept from social-personality theory in psychology. Causal attributions are the beliefs,
or attributions, we use to explain why things happen. One dimension is labeled the
locus of control (e.g., Rotter, 1990): Are the consequences we experience under our
control or outside of our control? People who have an internal locus of control believe
that they control what happens to them through their own efforts and abilities, or lack
thereof. People with an external locus of control believe that things happen for reasons
outside of themselves, and thus beyond their control: situations, chance, or the whims
of others.
In addition, attributions can be stable or unstable. For example, academic failure could
be attributed to lack of ability, a stable or fixed characteristic, or to lack of studying, a
transient cause that could be reversed. Self-mastery or helplessness depends on the
particular combination of attributions used to explain failure.
Locus of control is often assessed by an inventory measuring agreement with state-
ments such as the following:
1 When I get good grades, it is because of my academic competence.
2 There are some subjects in which I could never do well. Both of these measure inter-
nal attributions; it is the rater’s own ability (#1) or lack-of (#2) that determines whether
they are successful. The cause is also stable. For instance, the person believes they
are math competent or not.
3 When I don’t do as well as expected in school, it probably due to a lack of effort on
my part. This is an internal attribution; taking personal responsibility for poor perfor-
mance. This statement also conveys an unstable cause for poor performance; the
individual could have studied more, and can study harder next time.
4 Some of my good grades are because these courses were easier. This is an external
attribution; the cause of good grades was not so much ability but the fact that the
courses were easy (or the teacher graded easy). This attribution is also unstable,
because you do not expect to do so well in difficult courses.
In one of the laboratory experiments on helplessness mentioned earlier (Hiroto, 1974),
college students could try to terminate a loud noise. The subjects had completed a
locus-of-control inventory as part of the study. External locus-of-control individuals
performed like students who had received helplessness training. If an aversive-noise
stimulus occurs, it is perceived as being beyond their control, and so it seems that
they do not try as hard to escape. By contrast, internal locus-of-control individuals act
more like the group trained with escapable noise. Internals responded quicker than did
externals.
 Neuroscience and Aversive Learning 137

Physical Health
Helplessness is associated with emotional reactions of the body to stress. Correlational
findings from human research indicate that a cluster of negative attitudes, which includes
pessimism, cynicism, helplessness, and depression, have been linked to a variety of
physical health problems, from the common cold to cancer (Scheier & Carver, 1993). It
seems logical to assume that an opposite, positive set of traits would lead to better
health. Indeed, an optimistic orientation does seem to convey psychological and physical
health advantages. Aspinwall and Taylor (1992) studied the adjustment of new freshmen
to college. At the beginning of their first semester, the students completed an optimism
inventory. Three months later, the more optimistic students were experiencing less dis-
tress, even after equating academic performance.
How else do optimists and pessimists differ? The optimistic individuals are more
likely to confront and deal with stressful situations, which is the approach mode of cop-
ing. Pessimists tend to deny stressors or try to avoid them, which is the avoidance mode
of coping (Scheier & Carver, 1993).
Correlations between traits of optimism or helplessness and health have to be inter-
preted cautiously. Does optimism lead to better health or does good health promote
optimism? Do illness and disease lead to pessimism about the outcome of future illness?

Summary of Learned Helplessness

An older psychiatric literature refers to the notion of hopelessness, or of psychologically
giving up in the face of stress or adversity. The experimental work on helplessness gave
a firm scientific basis to this idea. Helplessness is learning that there is a lack of contin-
gency between responding and an aversive outcome. Helplessness produces behavioral,
motivational, and cognitive deficits. A behavioral approach to learning emphasizes the
first two categories, such as acquisition of passive behaviors or reduced motivation as the
sources of helplessness. A cognitive approach emphasizes the beliefs and expectancies
that are acquired, such as a belief that aversive stimuli are uncontrollable or the expecta-
tion that responding is useless. Learned helplessness has merged with other ideas in the
areas of health, educational, and clinical psychology to help explain the origins of illnesses,
school failure, and depression. In one sense, helplessness has come full circle, from
clinical observations through the laboratory and back to the applied areas.

NEUROSCIENCE AND AVERSIVE LEARNING

Aversive learning is not a singular trait, and so neuroscientists will not be announcing the
discovery of the “aversive learning” area of the brain. Undoubtedly a number of neural
circuits will be found to underlie punishment, avoidance, and other forms of aversion
learning.

The Amygdala and Aversive Learning

One step in our understanding of aversive learning is to map the brain circuits involved in
simple fear conditioning. Several researchers, using a variety of methods and species,
have documented the role of the amygdala in the classical conditioning of fear (LeDoux,
138 Instrumental Conditioning

2000; Davis, 2006). For example, the amygdala of rats can be injected with a chemical
that temporarily suppresses its action. The subjects are given tone-shock pairings in a
standard conditioning protocol, but the infused rats do not condition to the tone. The
amygdala must be functioning for this form of learning to occur.
In addition, normal activity in the amygdala is important after the conditioning training
to consolidate the newly formed knowledge. In this type of study, the amygdala is chem-
ically inhibited shortly after the tone-shock training. When the rats are tested later, there
is no fear to the tone. Thus, normal activity in the amygdala is necessary for the initial
registration of the fear conditioning, and for the consolidation of the fear learning
(Wilensky, Schafe, Kristensen, & LeDoux, 2006).
The neurologist Antonio Damasio and his students studied individuals who have dam-
aged amygdalae—injury on both sides of the brain (Bechara et al., 1995). The patients are
tested in a classical fear-conditioning procedure. When a certain color of light is flashed on
the computer monitor (say, blue), an extremely loud noise burst is presented through the
subject’s headphones. Other colors are not followed by anything. Those of us with normal
amygdalae would learn a fear response to the blue light: skin conductance would jump,
as would heart rate. The amygdala-damaged individuals do not show a conditioned fear
reaction. They show a normal spike in the skin-conductance response to the loud noise,
so they perceive the noise as aversive. However, without a functioning amygdala, they do
not learn to fear the signaling noise. Interestingly, the amygdala-damaged patients can
verbally report the blue light–noise connection, so they “know” these two go together.

Avoidance Conditioning
There is not a “punishment” or “avoidance” area of the brain. Psychological functions are
better viewed as circuits, or a series of interconnected locations. Avoidance learning
involves a series of steps, involving a sequence of circuits. LeDoux has described the
circuits involved (LeDoux, Moscarello, Sears, & Campese, 2017).
First, the initial pairings of the warning signal and shock are classical conditioning: CS
paired with US. This conditioning is mediated through the amygdala; the latter triggers
defensive responses such as freezing in place, which is the initial conditioned response.
The second stage is instrumental learning of a response to avoid shock. This also begins
with amygdala activation, but another area, the prefrontal cortex, overrides the freezing
response. A connection between the amygdala and the striatum directs the avoidance
response. After extended training, the avoidance response becomes habitual through
repetition. At this point, the response becomes independent of the amygdala, and shifts
to control by habit-learning areas of the brain (see Figure 5.7).
In a classical conditioning arrangement, a signal (the CS) is followed by an unavoida-
ble shock (the US). In an instrumental avoidance task, there is a warning signal and a
response that can prevent shock. The comparison of amygdala activity in both forms of
conditioning was investigated within the same study (Delgado, Jou, LeDoux, & Phelps,
2009). The participants, students this time rather than rats, were exposed to both classi-
cal conditioning and avoidance learning, with trials intermixed. The participants were clas-
sically conditioned to discriminate between colored squares that were presented on the
computer screen. For instance, a blue square was paired with a shock to the wrist
 Neuroscience and Aversive Learning 139

Figure 5.7
The Transition from Threat Conditioning to Avoidance Learning. Sections of the amygdala are needed for
threat (or fear) learning. As the animal learns to avoid shock, guidance by other neural areas, including the
prefrontal cortex, take over.
Source: Adapted from LeDoux, Moscarello, Sears, and Campese (2017), Figure 1.

(defining a CS+), and a green square was not followed with shock (CS−). In addition, a
yellow square was also followed by shock, but a button response would prevent shock.
So, the yellow square was a warning signal (WS) for avoidance responding.
The researchers were particularly interested in which brain areas became more active
during the CS+ and the WS trials. We know the amygdala is involved during classical
conditioning with aversive USs; and an area known as the striatum is involved in instru-
mental conditioning. On CS-trials there was little amygdala activity. For the two stimuli
paired with shock, the CS+ and the WS, there were different patterns of activity. During
CS+ trials (classical conditioning), there was more amygdala activity; during WS trials
(instrumental learning), there was more striatal activity. So, for two stimuli paired with
shock, classical and instrumental conditioning produced slightly different patterns of
activity.

Social Learning
In most studies of conditioning, there is direct experience with aversive outcomes. Fear
can also be learned indirectly, through observation of another’s experiences or through
verbal communications. Does indirect fear conditioning also involve the amygdala?
In one series of experiments (Olsson & Phelps, 2004, 2007), participants received
one of three different versions of classical conditioning. In the direct experience group,
subjects were shown faces on the computer screen. One face (the CS+) was paired with
shock and another face that was not. The observational experience group watched a video
of someone receiving the direct-experience conditioning. These subjects watched some-
one else receive the face-shocks pairings. In the instructed experience condition, sub-
jects were told about the direct experience procedure of the first group. These subjects
were shown and told which face was paired with shock and which face was not.
The direct experience subjects showed conditioned fear to the CS+ face (as meas-
ured by skin conductance responses). No surprise here. The observational experience
140 Instrumental Conditioning

subjects also conditioned to the CS+ face. We could imagine that watching someone get
shocked would arouse fear, even if we did not experience any pain. Interestingly, the
instructed experience group showed elevated fear to the CS+ face. Explicit conditioning,
social learning, and instructional learning: each produced fear learning.
The amount of activity varies in different brain regions when someone receives a
shock versus watching someone getting shocked. Yet in all three conditions, there was
increased activation in the amygdala in the presence of the CS+. The amygdala that is
essential to direct (experienced) fear conditioning is also involved in learning fears through
indirect experiences. The only real difference among the groups was that the instructed
fear group had elevated activity mainly in the left side of the brain. This would correspond
to verbal or aware learning.

APPLICATIONS OF AVERSIVE-LEARNING CONTINGENCIES

Pet Containment Systems
Electronic fences are available to confine dogs to the yard around the home. A wire is
buried around the perimeter of the yard. The dog wears a collar that tracks the location of
the dog. When the dog gets close to the perimeter, a tone turns on; if the dog continues
approaching the wire, a shock is triggered. (The shocks are said to be not painful, compa-
rable to that generated by a 9-volt battery). The idea is the dog will learn to avoid triggering
the tone. This is a direct application of the Watson-Mowrer theory of conditioning.
Maybe we should be concerned about shock-elicited aggression. It is possible that
when the dog is shocked the animal will become aggressive. Polsky (2000) found several
legal cases which alleged that dogs crossed the barrier and attacked someone. The shocks
may have intensified the attacks. The instances cited are anecdotal; there may have been
other circumstances that contributed to the attacks. Yet, we have some expectation
based on laboratory research that electric shock can elicit aggression.
Could the tone elicit aggression? In laboratory research stimuli paired with shock can
themselves become elicitors of aggression (Miller, 1948). The tone might trigger aggres-
sion while the dog is within the yard.

Treatment of Obsessive-Compulsive Disorder

Contemporary psychotherapy employs both cognitive and behavioral approaches. Each
offers fruitful application to certain disorders. In obsessive-compulsive disorder (OCD), an
individual has an excessive fear of something and/or compulsive behaviors that need to
be performed. An obsessive fear of germs on objects that others have handled leads to
compulsive behavior such as hand washing, wearing gloves, or disinfecting objects.
Cognitive therapy might examine the irrationality of the thoughts and beliefs behind the
obsession. Behavioral therapy would try to recondition the compulsive behavior.
The behavioral approach to avoidance learning suggests there should be two steps in
eliminating obsessive-compulsive behavior: (1) provide safe exposure to the feared object
to extinguish the fear and (2) prevent the compulsive behavior to demonstrate there are
no adverse consequences when the response is blocked. A parallel is seen in the labora-
tory experiments on eliminating an avoidance response: (1) extinguish the fear to the
 Summary 141

warning signal or (2) prevent the avoidance response. Studies that have assessed the
separate contributions of extinction versus response prevention have sometimes shown
a desynchronization between the two. That is, fear to the warning signal is extinguished
after prolonged exposure. Yet, when given the opportunity, the subject still performs the
avoidance response during the warning signal. On the other hand, repeated experience
of blocking the avoidance response reduces the need to perform the response. Yet the
warning signal is still feared (see Mineka, 1979).
The same disconnect happens in humans, so the two methods may need to be used
together to reduce obsessive-compulsive behavior. For example, Foa, Steketee, Grayson,
Turner, and Latimer (1984) studied individuals who have an intense fear of dirt and germs
and used excessive hand washing as a means of decontamination. Foa et al. found that
therapy involving just exposure (i.e., extinction) to contaminants indeed reduced fear to
dirt, garbage, or public surfaces. Alternatively, preventing hand washing (response pre-
vention) subsequently reduced anxiety and the need to engage in the obsessive behavior.
However, neither alone was as effective as the combination of treating both the obses-
sive fear of contaminants and the compulsive hand washing behavior.
Exposure and response prevention therapy, or ERP, has been cited as a first line, evi-
denced-based treatment for OCD. Evidence suggests that cognitive-behavioral therapy is
as effective as the drugs often prescribed e.g., serotonin reuptake inhibitors (Foa et al.,
2005). In addition, the effects of behavioral therapy are maintained after therapy ceases,
whereas the drug effects dissipate once medication ends (Law & Boisseau, 2019).

SUMMARY
Instrumental conditioning is defined by the contingency arranged between a particular
response and an outcome. In addition to positive reinforcement, instrumental learning
includes three aversive-outcome procedures, which are the content of this chapter: non-
reinforcement, punishment, and avoidance.

Nonreward
One instrumental learning contingency withholds a reward. In extinction, the reinforcer is
omitted after those responses that once produced positive reinforcement. In omission, a
selected response prevents a positive reinforcer from occurring.
Extinction produces side effects that sometimes discourage its use: unpleasant emo-
tions such as frustration; an extinction burst of the now-nonrewarded response; and
spontaneous recovery of the extinguished response after a delay interval.
The specific conditions of positive reinforcement present during acquisition affect the
persistence of responding during extinction. In the partial reinforcement extinction effect
(PREE), extinction is slower after training with a partial reinforcement schedule rather
than a continuous schedule. The origin of the PREE occurs during acquisition, when frus-
tration becomes associated with reward on a subsequent trial, or the memory of nonre-
ward is followed by reward on a subsequent trial. Other reward manipulations that slow
or retard acquisition, such as small rewards or delayed reinforcers, can also lead to persis-
tent responding during extinction.
142 Instrumental Conditioning

Punishment
In punishment, a response is followed by an aversive consequence, which acts to
decrease responding. There have been claims that punishment only temporarily sup-
presses responding (Estes and Skinner). However, punishment is effective under certain
conditions: punishment is given contingent on a particular response, and is intense,
immediate, and consistently applied after each response. The punishing stimuli should
not elicit responses that are incompatible with the desired outcome.
Punishment produces side effects of learned (or conditioned) fear, and avoidance of
the punisher or the punished situation. Punishment can also provoke aggression, which
may be inadvertently reinforced and thus strengthened. The punishment (or the punished
behavior) may become a secondary reinforcer through association with a positive
reinforcer.
Nonreward and punishment share some common features: They both produce nega-
tive emotional reactions, and behavior may transfer between these two types of
outcomes.
Advocates of behavior modification advise withholding reward as an alternative to
punishment (again, the latter defined as the application of an aversive stimulus).
Punishment involves moral and ethical issues, in addition to questions about its efficacy.
Research on spanking has shown that it is associated with long-term psychological
consequences.

Persistence
Persistence refers to continued responding during extinction or punishment. Response
persistence can be enhanced by previous exposure to partial reinforcement, small and
delayed rewards, or even mild and occasional shock during acquisition. Persistence can
generalize from one set of adverse consequences to another, e.g., from punishment to
nonreward, and vice versa. The study of response persistence is significant because bad
habits are often so persistent and because good habits are not persistent enough.

Avoidance Learning
In avoidance learning, a response prevents the occurrence of the aversive stimulus.
Avoidance learning is an example of negative reinforcement. The frequency of the
response increases (i.e., “reinforcement”) which leads to omission of the aversive conse-
quence (i.e., “negative”).
How do we explain the fact that avoidance responding increases even when the
response is followed by “nothing” as the consequence? Mowrer said the pairing of a
warning signal with shock conditions fear to the signal via classical conditioning. Escape
from the warning signal is reinforced by fear reduction, which is instrumental condition-
ing. One difficulty for the two-process theory is the fact that the warning signal does not
seem to extinguish (lose its fear-motivating property) even though it is not followed by
shock on successful avoidance trials.
A cognitive theory of avoidance states two sorts of expectancies are acquired: a
stimulus–outcome expectancy (e.g., tone is followed by shock) and a response–outcome
 Summary 143

sequence (e.g., the avoidance response is not followed by shock). Successful avoidance
responding requires the acquisition of two expectancies: Tone means shock, and a
response will prevent the shock. Eliminating avoidance behavior requires modifying both
expectations.
A functional approach to avoidance studies the contribution of an organism’s natural,
evolved response to fear and pain to the acquisition of protective behavior. Some
responses may be easier to learn than others, such as behavioral freezing by a mouse and
or a rat standing and rotating a wheel.
When a given behavior has two opposing outcomes, one pleasant and one aversive,
approach–avoidance conflict results. Approach or avoidance can also characterize styles
of coping in stressful situations. Some people try to avoid thinking about a stressful situ-
ation, whereas others confront the stressors. Avoidance coping leads to reinforcement
through immediate fear reduction. Approach coping provokes more immediate fear, but
may lead to more adaptive learning in the long run for controllable stressors.

Prediction, Control, and Helplessness

There are two independent dimensions that contribute to our understanding of aversive
events. One is predictability: Is an aversive event predicted? The second dimension is
controllability: Can the aversive outcome be controlled by an instrumental response?
Learned helplessness is learning that there is an explicit lack of contingency between
response and an aversive outcome: There is no response that is causing punishment, nor
is there one that prevents it. Helplessness is demonstrated using a three-group design.
During the first phase of an experiment, one condition is exposed to escapable shocks, a
second condition is exposed to inescapable shocks, and the third condition receives no
shocks. In the second phase of a study, the inescapable-shock subjects fail to learn an
avoidance response.
Uncontrollable stressors produce emotional, motivational, and cognitive deficits.
Emotional deficits are shown by various psychosomatic illnesses, such as ulcers.
Motivational deficits are shown by a lack of initiative to respond. Cognitive deficits are
shown by beliefs that whatever response made will be unsuccessful. Helplessness is ulti-
mately determined by the belief that behavior is ineffectual in affecting what happens to us.
Learned helplessness has been extended and applied to explaining depression, fail-
ure in classroom learning, and proneness to stress-related illnesses; and to an opposite
disposition, learned optimism.
Helplessness has also been integrated into causal attribution theory. Causal attribu-
tions are the beliefs, or attributions, we use to explain why things happen. People who
have an internal locus of control believe that they control what happens through their own
efforts and abilities. People with an external locus of control believe that things happen
for reasons that are beyond their control.

Neuroscience and Aversive Learning

The amygdala of the brain is important for learning fear. In classical conditioning, suppres-
sion of this area before or shortly after conditioning blocks the appearance of conditioned
fear.
144 Instrumental Conditioning

Activity in the amygdala was studied during a CS in classical conditioning, and during
a warning signal in instrumental avoidance learning. Both stimuli showed more amygdala
activity than during stimuli that signaled shock.
Amygdala activation is also found during observational learning of fear (watching
someone else go through a classical fear conditioning procedure); and after instructed
conditioning (listening to the description of the fear conditioning procedure).

Applications
Pet containment systems employ principles that correspond to the Watson-Mowrer the-
ory of avoidance learning. A warning tone sounds as the dog approaches the perimeter of
the yard, and shock occurs if approach continues. Backing away from the area turns off
the tone and prevents shock. Shock-elicited aggression is a potential side effect.
In obsessive-compulsive disorders, individuals often have an excessive fear or phobia,
and compulsive behaviors that need to be performed. Research on avoidance learning
suggests that eliminating obsessive-compulsive behavior requires (1) safe exposure to
the feared object to extinguish the fear and (2) prevention of the compulsive behavior to
prove that no adverse consequences follow. An effective behavioral treatment uses both
stimulus-extinction and response-prevention components of the two-process theory.
 CHAPTER

6
Verbal Learning

CONTENTS

The Ebbinghaus Legacy 146 Recognition: Remembering Versus

Serial Learning 148 Knowing 160
Serial Position 148 Accessible Memories 161
The Ubiquitous Serial-Position Curve 148 Recall Versus Recognition Versus
Remote Associations 150 Relearning 162
Serial Learning: A Summary 151 Implicit Learning 163
Paired-Associate Learning 151 Relationships Among the
Analysis of Paired-Associate Verbal-Learning Tasks 164
Learning 152 Statistical Learning 164
Direction of Associations 154 Propositional Learning 165
Paired-Associate Learning: Statistical Language Learning 165
A Summary 154 Applications 167
Free Recall 156 Assessment of Cognitive
Serial-Position Effects 156 Impairment 167
Rehearsal 157 Mnemonics 167
Organization 158 Summary 170
Free Recall: A Summary 159 Recognition Memory 171
Recognition Memory 160

The previous chapters described the methods of classical and instrumental conditioning,
areas of learning that were heavily influenced by the behavioral approach to psychology.
Later chapters will focus on the area of memory, which is more influenced by the cogni-
tive approach. The current chapter is transitional between behavioral psychology, with its
emphasis on association learning; and cognitive psychology, with its emphasis on mental
strategies. Verbal learning refers to the study of factors that affect acquisition, retention,
and recall of verbal items (usually words). Work on verbal learning follows in a tradition
begun by Herman Ebbinghaus, a German psychologist working in the 1880s. Although
verbal learning as a term now sounds old-fashioned and not as contemporary as cognitive

DOI: 10.4324/9781003227090-6
146 Verbal Learning

psychology, the topics studied are still very much in vogue and have a decidedly applied
nature, for example the spacing of study repetitions, the importance of repeated testing,
or detecting unrecallable memory.
Two caveats are in order. First, verbal learning is a misnomer because often we study
learning of nonverbal material: faces, pictures, objects, locations, odors, and action
sequences are just a few examples. Second, rote learning implies a passive, uninvolved
subject (S) who is attempting to memorize information. Our subjects are rarely passive.
“The image of the subject in a verbal-learning experiment as being a tabula rasa upon
which the investigator simply chisels associations, and quite against the S’s wishes, is
archaic. The S is far from passive and the tablet has already impressed upon it an immense
network of verbal habits. A more accurate description of the verbal-learning experiment
is one in which “the subject actively ‘calls upon’ all the repertoire of habits and skills to
outwit the investigator” (Underwood, 1964, p. 52). Therefore, cognitive variables are con-
sidered in our studies: rehearsal, imagery, and organization. In verbal-learning experi-
ments, we may not be dealing with “raw” learning; we study new learning influenced by
knowledge the subject already possesses.

THE EBBINGHAUS LEGACY

In 1879, a twenty-something German scholar named Herman Ebbinghaus began a
remarkable series of experiments on memory. Ebbinghaus had still not yet decided on his
life’s work. Although six years beyond his Ph.D., he worked as a tutor for a wealthy family.
There was a long history of the memory arts, but until this time, memory was not thought
to be amenable to the methods of science. Ebbinghaus became convinced that a quanti-
tative approach to learning was possible. His research was summarized in the now classic
On Memory, published in 1885 (Dover reprint edition, 1964).
Ebbinghaus’s procedure involved serial learning, or memorizing lists in sequence until
they could be recalled perfectly. The learning materials were three-letter syllables, each
composed of consonant-vowel-consonant, referred to in English as nonsense syllables.
Most of the syllables were not words (e.g., NOZ or BUH), but some were. This material
was used to produce sequences of syllables that would not be meaningful prose.
Ebbinghaus’s greater contributions were his methods for objectively measuring learning
and retention. Learning (or acquisition) could be measured by the number of study trials
that were necessary to repeat the list back without error. For instance, a list of 10 sylla-
bles might be memorized after four study trials, whereas a list of 15 syllables might
require eight study trials.
Ebbinghaus then tried to recall the list after a delay. Failing perfect recall, he would
relearn the list. His measure of retention was the number of trials needed to relearn the
list to a perfect recitation. By comparing the number of trials to first learn the list with the
number of trials to relearn the list, Ebbinghaus derived a measure of savings: how many
trials were saved in relearning. If a list took 10 repetitions to learn but could be relearned
in four repetitions the next day, there was a savings of six repetitions. This could be con-
verted to a percentage, in this case a savings of 60 percent.
Suppose you are studying for finals, and you run across some material that is unfamil-
iar. You have no recollection of it, even though you must have studied it for an earlier
 The Ebbinghaus Legacy 147

exam. Savings would be evidenced if you can now learn it faster than you had originally
learned it. The significance of the savings measure is that it allows the detection and
measurement of memory even in the absence of the ability to recall or recollect an
experience.
Another of Ebbinghaus’s contributions was his description of the curve of forget-
ting. Ebbinghaus tested his retention at various intervals after learning: immediately,
hours later, days later, a month later. In the first minutes and hours after learning, the
amount retained dropped dramatically. Figure 6.1 shows Ebbinghaus’s data. The measure
of savings shows a drop from 100 percent recall immediately to 40 percent savings after
60 minutes. At longer intervals, the rate of loss decreases. The change from one day
(30 percent savings) to 31 days later (about 20 percent) was pretty small. The Ebbinghaus
curve shows that most forgetting occurs shortly after the initial learning. Whatever per-
sists past this early phase is better retained and is forgotten at a slower rate.
Ebbinghaus’s work converted the study of memory from philosophical speculation to
a true science. His contributions included devising methods for performing controlled
learning experiments; means of quantifying the results; the measurement of unrecalled
memories; and a description of the effects of several variables on learning.
This chapter will begin with consideration of three basic verbal-learning methods.
Serial learning requires recall of a list in the same sequence in which the items were
presented. Free recall also refers to remembering lists, but the items can be recalled in
any order. The method of paired associates refers to learning lists of pairs of items, in
which the first item is used as a stimulus for recall of the second.

Figure 6.1
The Curve of Forgetting across an Interval of 31 Days. Derived from Ebbinghaus’s own recall data, retention
is plotted in terms of the percentage savings in relearning.
Source: Data from Memory: A Contribution to Experimental Psychology (p. 76), by H. Ebbinghaus, 1885/1964, New York: Dover;
and Grundzüge der Psychologie (Figure 51, p. 646), by H. Ebbinghaus, 1902. Copyright © 1964 by Dover Publications, Inc.
Reprinted with permission.
148 Verbal Learning

SERIAL LEARNING
In serial learning, a list of items is learned and reproduced in the same sequence in
which the items were presented. Serial learning occurs frequently outside the laboratory.
Everyday examples include learning the alphabet by young children, poetry by middle
schoolers, and statistical formulas by college students (“Sum and then square, or is it
square and then sum?”). There are many alpha-numeric codes we need to memorize,
such as ATM or computer passwords.
In a serial-learning task, study trials in which the list is presented are alternated with
test trials in which the subject attempts to recall the list. Learning is quantified by count-
ing the number of recalled items in the correct serial position. Alternate scoring proce-
dures count the number of words remembered in the right order, even if some words
in-between are forgotten.

Serial Position
Probably the most significant factor affecting serial learning is the location of an item
within the list. Words at the beginning and end of the list are typically learned faster,
whereas middle items are learned with more difficulty. Figure 6.2 shows a typical serial
position curve. In this study, a list of nonsense syllables was presented and tested until it
could be reproduced in sequence without error (Robinson & Brown, 1926). The figure
shows the average number of correct syllables recalled at each position during learning.
There is a U-shaped curve: More correct responses occur in learning syllables in the first
and last serial positions than in the middle positions.

The Ubiquitous Serial-Position Curve

The classic U-shaped curve in the recall of serial information is seemingly ubiquitous:
Serial-position curves are found in many situations, both in laboratory and naturalistic
research, in both animals and humans. Some other examples of the U-shaped curve are
the following.
Long-Term Personal Memory. Sehulster (1989) attended 260 performances of the
New York Metropolitan Opera over 25 years. He tested his own recollection of details
such as the date of the performance or the lead vocalists. Sehulster obtained U-shaped
curves, with more accurate recall of details for the first five-years and the last five years.
Long-Term Skill Retention. Gymnasts were asked to reproduce a 12-step floor rou-
tine, either immediately or after a delay of up to a week. U-shaped reproduction curves
were seen. Expert gymnasts could recall much further into the list before forgetting a
step; that is, they could recall the first six, seven, or eight steps before making an error
(Tenenbaum et al., 1999).
Infant Memory. Six-month-olds learned to kick their legs to shake each of three
mobiles that were suspended overhead. (This method was described in Chapter 4).
Testing a week later showed a U-shaped curve: The infants reacted more to the first-
learned and the last-learned mobiles than to the one learned in-between (Gulya, Galluccio,
Wilk, & Rovee-Collier, 2001).
Spatial Memory. Eight- and 12-year-old children were taken for a walk across a univer-
sity campus. The children were then asked to lead the way back. Errors were more likely
 Serial Learning 149

100

80

TR- 17
Mean Percent Correct

TR- 13

60

TR- 5

40
TR- 1

20

0
2 3 4 5 6 7 8 9 10
Serial Position

Figure 6.2
The Serial Position Curve in Serial Learning. Number of nonsense syllables recalled in sequence after 1, 5 13,
and 17 presentations.
Sources: Drawn using data derived from Robinson and Brown (1926).

to be made in the middle of the walk; the children more accurately remembered direc-
tions at the beginning and near the end of the walk (Cornell et al., 1996).
Several hypotheses have been proposed for position effects in a serial learning. The
anchoring hypothesis says that subjects latch onto the end items in the list to serve as
anchors from which the rest of the list is attached. Middle items in the list are too distant
to be firmly attached to either end (Bower & Hilgard, 1981). An analogous finding is our
tendency to organize personal memories around various temporal anchors. Students, not
surprisingly, anchor their memories with respect to the start and finish of the academic
year. In recalling personal experiences, students have more recollections from early and
late in the school year and fewer from the middle (Kurbat, Shevell, & Rips, 1998).
An interference hypothesis argues that learning some list items can interfere with
learning others. Proactive interference occurs when learning the items at the beginning
of the list interferes with learning items in the middle and end of the list. Retroactive
interference occurs when learning the end of the list interferes with learning the
150 Verbal Learning

beginning and middle of the list. Given the list DOG-CAT-TREE, remembering DOG could
proactively interfere with learning CAT and TREE. Remembering TREE can retroactively
interfere with recall of DOG and CAT. The middle items (CAT) suffer maximal interference,
both proactive from the earlier list items and retroactive from the items that follow.

Remote Associations
One explanation of serial list learning is in terms of item-to-item associations. Each item
in the list serves as a cue for the next. In learning the alphabet, A is a cue for B, B is the
stimulus for C, and so on.
Ebbinghaus supplemented item-by-item associations with the hypothesis that
remote associations are learned among nonadjacent items. The letter A can be associ-
ated with C and D, although with less strength than A’s association to B. Ebbinghaus
tested for “connections at a distance” by first learning one list and then deriving new lists
in which successive items had been two, three, or more positions away in the original
lists. For example, the list A–B–C–D– E could be transformed to A–C–E–B–D, in which
every other item from the first list now follows in the transformed list. This defines a
second-degree transformation. A third-degree list would have every third item now follow
one another. Learning the derived lists were easier than learning derived lists in which the
syllables were randomly rearranged. The amount of savings was a function of the number
of steps removed from the original lists.
In attempting to recall material such as prose, poetry, or songs, we may forget some
lines and phrases, yet we can recall portions that come later and continue to the end
(Rubin, 1977). Remote associations may account for our ability to continue later in the
sequence even though some previous sections are forgotten.
In an influential article, Karl Lashley (1951) pointed out that item-to-item associations
are much too slow to accommodate quick, skilled, and unified behaviors. He noted that
multiple items are seemingly grouped, or unitized. An accomplished pianist does not play
one note at a time; rather, whole groups of notes are played as if they were a single unit.
Listen to children recite the alphabet and hear the groupings: the letters H–I–J–K form
one group, and L–M–N–O–P form another.
An alternative to item-to-item associations is the theory of hierarchical associa-
tions: Lists or sequences are divided into sections, and those into subsections, and so
on (Estes, 1972; Lee & Estes, 1977). For example, at the top of the hierarchy might be
the code for “alphabet,” the next level codes for segments within the list (e.g., “begin-
ning,” “middle,” “end”), and the lowest level contains the items within the segment (A
B C). Carrying out a sequential activity involves activating the higher-order code, which
then activates the lower-order segments in sequence. The idea is that each single item
is not activated via association from the immediately preceding item, but instead all
the items contained within a segment are activated. Items within a unit are primed
and run off relatively quickly (ABCD), whereas there may be pauses between units
(between D and the next segment, E F G). The hierarchical theory also accounts for
the distinctive pattern of misrecalls that occur. Errors occur when whole segments are
forgotten (EFG).
 Paired-Associate Learning 151

Serial Learning: A Summary

Serial tasks involve learning and recalling items in sequence. Serial-learning phenomena
have generated considerable interest, both empirically, for example, in describing the
shape of the serial position curve, and theoretically, in terms of explanations of serial
position effects, such as interference theory. In addition, the starting point in other
research on items that occur in sequence, from television commercials to memory for
personal events, has been to plot serial memory.

PAIRED-ASSOCIATE LEARNING
The method of paired associates was described in the 1890s by Mary Whiton Calkins, a
student of William James, and subsequently president of the American Psychological
Association (Madigan & O’Hara, 1992). In paired-associate learning, two items are pre-
sented for study: one is labeled as the stimulus and the other is the response (abbrevi-
ated S and R). Study trials in which S and R are presented alternate with test trials in
which the stimulus alone is presented and the subject attempts to recall the response.
For example, I could pair the stimulus word TABLE with the response AARDVARK; when
you next hear TABLE, you should respond with what? AARDVARK, of course. Paired-
associate learning sounds like classical conditioning, each having an emphasis on
stimulus–response pairings and association learning.
Examples of paired-associate learning abound in our everyday lives. Learning the
vocabulary in another language is heavily dependent at first on paired-associate learning:
Spanish to English, English to Russian, and so on. So is learning the names of people (see
Box 6.1). And don’t forget those many hours of your youth spent memorizing multiplica-
tion tables.

BOX 6.1 NAME LEARNING

Forgetting names is about the most common complaint in surveys of everyday memory
problems. Why are names so difficult? To begin with, names may not receive the sustained
attention that other aspects of people receive. A name is spoken, it fades quickly from mem-
ory. A face can be continuously inspected while we converse with someone.
Also, names as words are less familiar, meaningful, or imageable than occupations labels.
Interestingly, even if the same words are used as names and occupations, names are still
poorly learned. In the Baker/baker paradox, we more readily remember that a person is a
baker than that their name is Baker. In a similar fashion, nursing graduate students could
recall that a fictitious person had Hodgkin’s disease or Bell’s palsy but could not remember
that the person’s name was Hodgkin or Bell (Terry, 1995).
Names are particularly difficult to remember in comparison to other details about people,
such as their occupations or interests. Cohen and Faulkner (1986) described a series of ficti-
tious individuals to a sample of adult subjects (e.g., “In Glasgow a policeman named James
Gibson recently won a prize for ballroom dancing.”). Each sentence included a name, place,
hobby, and occupation. Names were the least recalled category of facts.
152 Verbal Learning

Names are only one of a number of features that go into identifying a familiar person.
Bruce and Young (1986) developed a model of person identification. Face recognition is
usually the first step, which includes all the neural apparatus required for processing and
remembering faces. (The person identity node can be independently accessed by a per-
son’s voice, or their body shape and movements). Once a face is recognized as familiar, the
person identity node can be accessed. This contains the network of connections in seman-
tic memory: their biographical info, or their relationship to us, where we know them from,
their occupation, etc. Note: At this point we still do not recall the person’s name. Some
episodic memories might be recalled: a recent meeting with them, for example. Only once
a person’s identity is activated in memory can access to the name become possible. So you
can recognize their face and remember all of the other details about someone, and not
recall the name. The reverse case never happens: you remember their name but nothing
else.
As important as encoding strategies are for name learning, name retrieval also needs to
be practiced. One mnemonic method, the “name game,” encourages frequent retrieval of
newly learned names. Given a small group of people, each person in turn announces their
name. The next person gives their name and repeats the previous person’s name. Each suc-
ceeding person must also repeat the name of everyone who went before in the introduc-
tions. After the final introduction, the cycle repeats, but now the first person must repeat the
names of everyone else. An experimental test of the name game found significant retention
even after 11 months (Morris & Fritz, 2000).
One final suggestion as to why names are so difficult: There is a sort of cognitive helpless-
ness when it comes to remembering names. Because we expect to forget them, we just
don’t try.

In the typical paired-associate learning experiment several S–R pairs are presented,
often 20 or more pairs to study acquisition. Many kinds of materials have been used as
stimuli and responses: nonsense syllables, letter or number strings, words, pictures, move-
ments, and even smells.

Analysis of Paired-Associate Learning

One useful approach to understanding paired-associate learning is to partition it into a
series of stages or tasks: stimulus discrimination, response learning, and S–R associating
(McGuire, 1961). The approach then assesses the contributions of certain variables to
paired-associate learning.

Stimulus Discrimination
The stimuli used in a paired-associate learning task can vary in their similarity to one
another. High similarity can impair paired-associate learning. How can you give the cor-
rect response if you can’t tell the stimuli apart? In learning a new language, for example,
some words may be alike in spelling and pronunciation. The Latin phrases in vitro (“in
glass”) and in vivo (“in life”) are confusable. In art class, there is the discrimination
between the French painters Manet and Monet.
 Paired-Associate Learning 153

Figure 6.3
Example of Letters Given Distinctive Features.
Source: From “Making Letters Distinctive,” by G. R. Lockhead and W. B. Crist, 1980, Journal of Educational Psychology, 72,
p. 485. Copyright © 1980 by the American Psychological Association. Reprinted with permission.

Stimulus discrimination can be facilitated by finding ways to differentiate among the

items. Young children first learning to read have a difficult time with similar looking letters:
the pairs b and d, p and q, and m and n. Lockhead and Crist’s (1980) starting point was
that certain typefaces are easier to read than others. Some styles of type are especially
elaborated with serifs: a line or a bar at the end of the main strokes of a letter. Serifs
contribute to readability by making the letters distinctive. Lockhead and Crist translated
this notion into a test of children’s ability to discriminate between pairs of similar stimuli.
They gave 5-year-olds sets of cards to sort into piles of b’s and d’s, or p’s and q’s. In some
decks, a distinctive element was added to the letters—for example, a line might pass
through the descending arm of the q, or a dot might be placed within the b’s lower loop
(see Figure 6.3). The children were able to sort these letters more quickly and more accu-
rately than those without distinctive markings (i.e., sans serif). Adding distinctive ele-
ments might facilitate learning the names of letters, and once learned, the serifs could be
gradually faded out.

Response Learning
The ease or difficulty in learning paired-associate response items can also vary. Again,
using our example of learning a new language vocabulary, responding in your native lan-
guage to foreign word stimuli is easier than learning the foreign language words as the
responses. When definitions are to be learned, it is easier to learn the word as a response
than to memorize the (usually longer) definition as a response.
One factor that affects response learning is meaningfulness. Meaningful responses,
items that already have many associations, are familiar, or are encountered frequently in
language, are learned more easily.

Stimulus–Response Associating
The third stage in paired-associate learning is connecting the stimuli and responses (the
term most often used is associated). The challenge is to find ways to connect the two
items of a pair.
Pre-existing knowledge of the S and R items might offer ways of connecting them.
For example, learning a list that contained the pairs TABLE–KITCHEN and WHISTLE–
TRAIN would be facilitated by the associations that already exist between the words in
154 Verbal Learning

each pair. Prior knowledge could inhibit learning if the pairings were TABLE–TRAIN and
WHISTLE–KITCHEN.
S–R associating can also be facilitated by cognitive elaboration. Additional information
can help link the stimulus and response terms. In a study by Pressley et al. (1987), stim-
ulus–response pairs were presented in the form of sentences, such as “the large man
read the sign.” The to-be-learned associations are between the different men and their
actions. A cognitive elaboration group received an additional phrase to suggest why a
particular man and action are related. Thus, the large man read the sign “warning about
thin ice.” More of the person–action sequences were recalled by participants who studied
sentences with the elaborations. Cognitive elaboration is a case in which added informa-
tion is beneficial to remembering.
There are individual differences in paired-associate learning. Individuals with larger
vocabularies or more knowledge will be able to retrieve more potential mediators. This
was demonstrated in a study using U.S. Air Force recruits in basic training (Kyllonen,
Tirre, & Christal, 1991). (The military is interested in the study of learning and so sponsors
much research). The researchers found that a good vocabulary predicted ease of learning
across a range of study conditions. Other research found that vocabulary scores corre-
lated with ease of learning a foreign language (Cooper, 1964). This shows one way in
which individual differences affect paired-associate learning.

Direction of Associations
In everyday use, many of our associations between words are in one direction. Given the
stimulus word BUTTER, a frequent response is FLY. However, when tested in the reverse
sequence, FLY rarely elicits BUTTER.
Does paired-associate learning also produce a unidirectional association? Typically,
backward testing using the response item to elicit recall of the stimulus is much less
successful. Some of this may be simply lack of retrieval practice, of using the response
to cue the stimulus.
This has implications for academic learning. Studying language vocabulary in one
direction, for example, from Spanish to English, does not guarantee recall in the opposite
direction, from English to Spanish. Practicing multiplication tables, 8 x 6 = 48, does not
ensure that reverse sequences will be available during a division drill, for example, 8 goes
into 48 how many times?

Paired-Associate Learning: A Summary

Paired-associate learning has been described here as a series of steps: discriminating
among the stimuli, learning the responses, and associating the responses to the stimuli.
Learning is facilitated by creating meaningful S–R relationships and by cognitive elabora-
tion. Recall of the stimulus, given the response, is sometimes poor, but backward recall
benefits from practice in recalling the stimulus items.
Paired-associate learning has served as a model for acquiring vocabulary in a new
language. Generations of psychologists and educators have studied the effects of numer-
ous variables on learning languages from Morse Code to Swahili. Suggestions for
 Paired-Associate Learning 155

improving students’ study habits have been offered as examples of “best practices”:
translating basic research into applications. Some of the suggestions are presented in
Box 6.2.

BOX 6.2 LEARNING A NEW LANGUAGE VOCABULARY

Learners want learning to be easier and more rewarding. There are study techniques that
lead to successful recall during acquisition or shortly thereafter. But immediate recall is a
misleading indicator of how well the material has been learned. “Easier to learn” might also
mean “easier to forget.” Students who study using more effortful conditions remember bet-
ter when tested days or weeks later. “Creating desirable difficulties” is the phrase most
used in reference to effortful encoding conditions that improve long term learning (Bjork,
1994; Bjork & Kroll, 2015). There are several best practice guidelines to improving language
acquisition.
One variable is the direction of initial learning: from the foreign words as stimuli (e.g., the
French CHIEN and CHAT) to their English equivalents (DOG and CAT), or the reverse, from
English to French. You can already guess that the French-to-English direction is easier to
learn. The English words are already known, and the student needs only to recognize the
French words, not actually retrieve them. “CHIEN means what?” versus “How do you say
CAT?”
The easier task during initial study, learning to produce the English translations to the for-
eign words, does not produce the best retention (Schneider, Healy, & Bourne, 2002).
Another variable is whether the words are grouped by theme (travel, foods, clothing) or
are intermixed. Language courses typically arrange categories of words each week. One
week there are school-related words, another week on food, and then travel, etc. Acquisition
and recall seem better when the items are grouped like this. However the limited pool of
possible items narrows the possible responses and facilitates guessing. If the topic is school,
there will be words for classroom, texts, distinguished professor (okay, so maybe just “Prof”).
Intermixing words from different categories, or intermixing study tasks such as vocabulary
and grammar exercises, is more challenging to recall. You need to retrieve more from long-
term memory. However, intermixing produces better long-term recall (Schneider et al., 2002).
Massed practice, or immediate repetition of the same items, can lead to a feeling of
knowing: “I got it!” However, the old adage “spaced practice is better than massed” still
holds. Separating repetitions of items, and spacing repetitions of a given set of material,
produces better long-term retention.
Dividing material into smaller subsets to study seems to make each easier to learn. You
could parse your 20 vocabulary words into sets of five. Study those five; after just a few
repetitions you are able to recall them; and then move to the next set. However, this is a
variant of the previous recommendations, grouping and massed practice. Learning 20 items
all together is much more difficult. Yet the more difficult arrangement produces better long-
term retention.
Repeated studying needs to be supplemented by self-testing. You need to test yourself to
see whether you really know it. (And not immediate testing; testing should occur after some
delay. After all, the in-class quiz is not now, but tomorrow). In fact, research indicates that
more testing is more beneficial than more studying (Karpicke, & Roediger, 2008; Roediger &
Smith, 2012).
156 Verbal Learning

Table 6.1 Summary of Some Differences Between Recalling Foreign Language Vocabulary dur-
ing Initial Training Versus on a Delayed Test

Easier to Learn in the Short-term More Difficult But Better Long-term Recall

Foreign word as stimulus, English as response English word as stimulus, Foreign as response
Massed practice Spaced practice
Immediate testing of learning Delayed testing of retention
Blocking similar material of tasks Intermixing different material or tasks
Just study Study and self-test

FREE RECALL
In free recall a list of items is presented and the subject attempts to recall the items. Unlike
serial learning the items can be recalled in any order, which is why recall is called free
(meaning unconstrained). In a single-trial free recall task, a list is presented once and recall
of the items is tested. This process is repeated with several lists, each presented once.
In multi-trial free recall, the same list is presented and tested across several trials.
This method is more often used to assess learning. Multi-trial free recall is used in some
standardized tests of learning ability, such as the California Verbal Learning Test (Delis,
Kramer, Kaplan, & Ober, 1987). A 15-word list is presented, and recall tested, five times in
succession. The number of words recalled can be derived for various groups, such as age
or educational level. Figure 6.4 shows the number of words recalled in multi-trial free
recall as a function of age groupings (Davis & Bernstein, 1992). College students recalled
nearly nine words on the first trial, whereas the oldest participants recalled fewer than six
words. By the fifth presentation, the students remembered more than 13 words and the
older participants recalled about 10. Both of these age groups had comparable increases
in recall across trials. A delayed test was given 30 minutes later; forgetting occurred by
the older subjects but not by the younger participants.
Free recall is one of the most widely used research tasks because it is a useful tool
to study many experimental variables that can be manipulated, and participant activities
that occur during learning. By way of illustration, we can describe three widely studied
factors: serial position, rehearsal, and organization.

Serial-Position Effects
Even though participants are free to recall words in any order, we can tabulate recall as a
function of each word’s position in the list during presentation. Just as with serial-learning
tasks, a serial-position effect is found. That is, more words are recalled from the begin-
ning and the end of the list, and fewer words are recalled from the middle.
This serial-position effect in free recall is one of the most investigated phenomena in
the field of human learning. Much of this research has endeavored to dissociate the two
end points. The enhanced recall of the first items in the list is called the primacy effect.
The enhanced recall of the last items in the list is called the recency effect. So, the seri-
al-position effect actually can encompass two sub-effects, primacy and recency.
 Free Recall 157

15

14
Twenties
13 Thirties
Forties

12
Mean Number of Words Recalled

Fifties
11
Sixties
10

9 Seventies

8 Eighties

0 1 2 3 4 5 Delay
Trial

Figure 6.4
Age Effects on Multi-Trial Free Recall. Mean number of words recalled of 15-word list over five presentation
trials and on a final test after a brief delay.
Source: From “Age-Related Changes in Explicit and Implicit Memory,” by H. P. Davis and P. A. Bernstein, in The Neuropsychology
of Memory, 2nd ed. (p. 252), edited by L. Squire and N. Butters, 1992, New York: Guilford. Copyright © 1992 by Guilford
Publications, Inc. Reprinted with permission.

Some of the theoretical explanations for the serial-position curve in free recall are
similar to those for serial learning discussed earlier, in particular the roles of position stim-
uli and interference. Thus, the first and last items may be more accessible to recall because
of their distinctive positions in the lists. Alternatively, the middle positions are exposed to
more interference from the items before and following in the list (Greene, 1986).

Rehearsal
A third explanation for the serial-position effect is that recall is coming from two separate
memory systems, long-term memory and short-term memory (e.g., Atkinson & Shiffrin,
1968). As a word list is presented, the subjects rehearse the items. The first items receive
more total rehearsal, there being fewer items to compete. This produces the primacy
effect. As the list progresses rehearsal cannot keep up with all the items, and so the
middle words receive less rehearsal. At the end of the list, the final few items are still
158 Verbal Learning

100 12
Percent Recall # Rehearsals
90
10
80

70

Mean # Rehearsals
8
Percentage Recall

60

50 6

40
4
30

20
2
10

0 0
2 4 6 8 10 12 14 16 18 20
Serial Position of Word at Input

Figure 6.5
Word Recall and Number of Rehearsals. Serial position curves for free recall, displaying the percentage of
words recalled (left axis) and the number of rehearsals (right axis) by serial position of the items in the list at
presentation.
Source: Adapted from D. Rundus, 1971, Journal of Experimental Psychology, 89, Figure 1, p. 66.

echoing in memory. Those items are usually recalled first, before they fade from memory.
Primacy occurs because the first items are recalled from long-term memory. Recency
occurs because the final items are recalled from short-term memory. (This dual-memory
interpretation of the serial-position effect is discussed further in Chapter 7).
The relationship between rehearsal and recall is demonstrated in a study in which the
participants were asked to talk out loud as they processed the lists (Rundus, 1971).
Figure 6.5 shows the average number of words recalled at each serial position (scale on
the left axis). The number of rehearsals of each word is shown in a second curve (scale of
the right axis). As can be seen, first words in the list were well recalled, and they were
rehearsed more frequently. Words further into the list were less well recalled and had
fewer rehearsals. The words at the end of the list were rehearsed least, although they
were still well recalled. These last few words, having just been presented, were still in
immediate memory when the signal to recall was given.

Organization
Memory for long lists of word is far from perfect. For example, high school students
presented with lists of 12, 24, and 48 words, remembered about 8, 11, and 15 words,
respectively (Tulving & Pearlstone, 1966). The amount recalled increases dramatically if
the words can be organized in some way. Organization refers to using existing
 Free Recall 159

knowledge to group together items that are similar or are related. For example, lists can
be constructed so that there are several words from the same semantic category, such
as words that name tools, animals, or foods (Mandler, 1967). Organization might also be
based on associations (Jenkins & Russell, 1952). If BLACK, TABLE, WHITE, and CHAIR
are presented sequentially in a list, the words will likely be recalled in a sequence that
pairs the already associated word, such as BLACK–WHITE, TABLE–CHAIR.
Organization might better be labeled reorganization. The experimenter presents
lists of randomly sequenced words, which the subject reorganizes by categories or
associations. Organization can increase recall in several ways. It reduces memory load.
Instead of remembering 15 unrelated words, it is easier to remember five categories,
each containing three related words. Organization also directs memory search during
testing. Category labels are used as retrieval cues. And organization influences the
sequence in which items are recalled. Items that are in the same category are recalled
together.
The power of organization was shown in a study by Bower, Clark, Lesgold, and
Winzenz (1969). They employed a multilevel organizational scheme. There were 112 total
words from four basic categories (PLANTS, MINERALS, etc.), each of which was further
subdivided. For example, MINERALS included STONES and METALS; STONES was fur-
ther subdivided into MASONRY and PRECIOUS STONES. The students who were shown
the hierarchical organization learned and correctly recalled all 112 words after only four
study trials. A control group, not given the organizational scheme, remembered about 70.
These results also make the point that it helps to know what the organization is. The
control subjects likely perceived that the 112 words were related by some categories, but
they had not been shown the complete outline.
What if the lists do not have an organizational scheme? People tend to impose their
own organization on the lists. This subjective organization (Tulving, 1962) varies from per-
son to person. Each sees their own associations or relationships among items. We can
see the unique organization by inspecting output across successive trials. Certain words
are recalled together from trial to trial. For example, a list may include the words APPLE,
BELL, COFFEE, and SCHOOL. One person might recall BELL and SCHOOL together
across several test trials, even though these words are separated in the list. This is that
person’s subjective organization. Another might recall COFFEE and SCHOOL together.
This is called clustering in recall, and is an objective way to detect the participants’ strat-
egies in remembering.

Organization in Animal Memory?

Animals remember related or organized information better, and even cluster similar items
at output. Menzel (1973) tested young chimpanzees in an outdoor enclosure, in which he
first showed them the hidden locations of nine pieces of fruit and nine vegetables. Shortly
after, he released an animal to retrieve the food. The chimps clustered by food type: They
first retrieved the fruit pieces and then went back for the vegetables.

Free Recall: A Summary

The variables that affect free recall of lists show that human learners are not passive
participants in experiments. Serial-position curves indicate the use of strategies in
160 Verbal Learning

learning: rehearsing the first items to increase learning, and quickly reporting the last
items before they fade from memory. Rehearsal, which develops during childhood,
enhances recall and indicates a conscious attempt to remember. Finally, participants
attempt to organize to-be-recalled lists, grouping items on the basis of associations or
meaning to facilitate retention.

RECOGNITION MEMORY
The learning tasks considered so far require the participant to recall the presented items.
That is, the subject has to retrieve the items from memory and speak or write them.
Another means to test learning is to ask the participant to identify to-be-remembered
items from a group of distractor items. In a recognition test, you must recognize which
items were previously presented.
In a study on recognition memory, a list of items (e.g., words, pictures, or objects) is
presented. The lists are much longer than those used in free or serial recall because peo-
ple typically recognize much more than they can recall. Often, the list would be 50 items
or so. A second list is used to test memory. It often contains twice as many items as the
studied list; half the items on the test list are “old” (previously studied), and the others are
“new” (previously unseen distractors).

Recognition: Remembering Versus Knowing

Recognition is rooted in judgment. We decide that something is familiar on the basis of
several types of memory evidence. If college professors are shown a list of names of
former students, the profs might specifically remember certain students. Maybe details
about the students’ level of class participation or personal facts are recalled. On the other
hand, a name might simply look or sound familiar, without any specific memory or image
to back up that feeling. This example illustrates judgments based on remembering ver-
sus knowing.
In a standard recognition test, we simply ask the participants to make an old versus
new choice about each item in the tested pool. In a remember versus know test, we
take this step further by asking the participants to decide whether they specifically
remember that a given item was on the studied list, or they just know it was on the list.
These judgements might be accompanied by a confidence rating (e.g., pretty sure, very
sure).
The basis for a remember judgment is remembering some details of the previous
occurrence of the item. For instance, you remember the word reminded you of some-
thing or you noticed it related to another item of the list.
The basis for a “know” judgment is the item simply engenders a feeling of familiarity;
a feeling that you had a recent prior experience without recall of associated details.
Distinguishing between remember and know judgments is subjective, and the basis
for the decision might vary from subject to subject. There is also potential disagreement
among researchers, and between researchers and their research participants, by what is
intended by these terms (Umanath & Coane, 2020).
 Recognition Memory 161

The feeling of familiarity can lead to misjudgments about memory. This is illustrated
by studies of the false fame effect (Jacoby, Woloshyn, & Kelley, 1989). Fabricated names
(PETER TERRY, for instance) were presented without instructions to remember them.
Later the subjects saw a list of names and were asked to identify which were names of
famous people. The participants sometimes misidentified names encountered earlier as
those of famous people. “Peter Terry? Sounds familiar, but I can’t place him. Must be
famous.” This is a “know” judgment, and it’s also a wrong judgment.
In academic learning, we can have both types of memories. We might specifically
remember a lecture or class discussion. There are also words and phrases in our major
disciplines that we just seem to know, without recalling specifically how, when, or where
we learned them. Conway et al. (1997) found that among their better college students
(i.e., those who later scored well on an end-of-year exam), there was a high frequency of
remembering-type responses shortly after learning. For instance, the students could
remember the context in which a term had been encountered. When tested months later,
more items from the remember category had shifted to the know category. These better
students now just knew this information.

Accessible Memories
The significant contribution of Ebbinghaus’s savings test is that it can show the presence
of memory that is not recalled. The fact that a list was easily relearned indicated there
was still a trace available to build on. Tulving and Pearlstone (1966) elaborated on
Ebbinghaus’s idea with their distinction between available and accessible memories.
Available memories are those present in the memory. They are stored in memory. As we
well know, not everything in memory can be recalled, or fully recalled, or recalled right
now. Accessible memories are those that actually can be recalled or retrieved.
(Available and accessible are easily confused terms, both in sound and in meaning.
Here I will use accessible and its opposite, inaccessible).
This distinction should be readily appreciated by students: It is the difference between
knowing the material for the exam and being able to recall it during the exam (it is acces-
sible). Too often known information is not accessible, but is recalled later, after the test
has been turned in. The distinction also applies to the tip-of-the-tongue experience (Brown
& McNeil, 1966). You are searching for a name or a word, and you know that you know it,
but you just can’t recall it right then. The item is inaccessible. The lost word comes to
mind later, confirming that the word was indeed in memory all along.
How can we probe memory to retrieve currently inaccessible items? A direct method
to facilitate access to unrecalled information is to provide additional retrieval cues to
prompt memory. In cued recall, a cue or prompt is presented to aid recall of a specific
item from memory. For example, in assessing memory for a categorized list of words,
category names can be given to cue recall. In Tulving and Pearlstone’s (1966) study, par-
ticipants first recalled as many of the 48 words from a categorized list as possible. (The
words were initially presented randomly, and the participants did not know about the
categories). During a second recall attempt, the category names were given, and these
helped to retrieve forgotten categories of words. A student might be given the cue CITIES
162 Verbal Learning

and then recalls a few cities from the list. The effect of category prompting indicates that
more was available in memory than was accessed by free recall.

Recall Versus Recognition Versus Relearning

Alternative means of testing may offer more sensitive measures of what has been
learned. A recall test requires the subject to reproduce or recall the studied information.
A recognition test bypasses the need to actually retrieve memory and instead assesses
whether the previously studied item can be identified. In a relearning test, the material
does not need to be recalled or recognized. The initially studied material is relearned and
the amount of savings is computed. Do these different testing formats differ in their
sensitivity to detect learning? We need to be careful in comparing these methods,
because each produces a different type of measurement. Simply put, recalling 50 percent
of the items is not the same as a 50 percent savings (i.e., taking half as many trials to
relearn the list).
These several measures were used in a study in which college professors tried to
remember the names of their former students (Bahrick, 1984). Faculty members were
asked to recall the student’s name when shown the student’s picture; recognize names
of former students; or learn name-face associations, given actual and fabricated pairings.
Other students’ names and pictures were used as controls or distractors. The former
students were from different semesters ranging from courses just completed to some
eight years earlier. Bahrick’s results are shown in Figure 6.6. Professors were pretty poor
at recalling students’ names from their photos. The faculty were much better at recogniz-
ing names or pictures of former students, and in matching names to pictures. In this case,
recognition was superior to recall. Savings in relearning was not as good as recognition,
but still better than recalling names. Bahrick says that more is present in memory than is
accessible with certain forms of testing. It is also of interest that forgetting apparently
occurred within 11 days of the end of the semester!
Nelson (1978) compared the several methods of assessing retention in a laboratory
study. College students first learned lists of number–word pairs, and memory was tested
four weeks later. During the test session, the stimuli were presented to see if the
responses could be recalled. Many were unrecalled, as expected after a month-long
delay. Non-recalled responses were then tested by a matching test of recognition: The
participants attempted to match each response with the correct stimulus. About half of
the items produced incorrect matchings. In a third test, responses that were not matched
were then relearned and compared to learning of new pairs. Nelson found that items that
could not be recalled were sometimes recognized; and items that were neither recalled
nor recognized showed savings during relearning.
These sorts of outcomes suggest that the several tests could be ranked: The relearn-
ing test may be more sensitive than the recognition test, which in turn detects more than
the recall test. One interpretation of the recall–recognition difference is that recognition is
a more sensitive test: recognition can detect weak learning that a recall test does not.
This may be why students commonly believe that multiple-choice tests are easier than
essay exams. Recognizing the correct answer is thought to be easier than recalling it.
 Recognition Memory 163

Name Recognition
Name–Picture Matching
Picture Recognition
100 Savings in Relearning
Picture–Cued Name Recall

80
Percent Correct

60

40

20

0
1 2
Days

49.8
13.6
Mo.

Mo.

Mo.
97
11

Log (Time+1) in Months

Figure 6.6
College Professors’ Memory for Their Students. Name recognition, picture recognition, recall of students’
names from their photos, matching of name and photo, and relearning of names to student photos.
Source: From “Memory for People,” by H. P. Bahrick, in Everyday Memory, Actions, and Absentmindedness (p. 27), edited by J. E.
Harris and P. E. Morris, 1984, New York: Academic Press. Copyright © 1984 by Academic Press. Reprinted with permission.

However, there are other explanations. A recognition test presents old and novel
items. You could be correct either by remembering the correct (old) item, or by rejecting
the incorrect (new) item. There are two ways to get this right. It is also the case that the
recall of any particular item fluctuates from time to time. If the subject had been asked to
recall the list a second time, it is likely that some previously unrecalled words would have
been remembered. And, some words remembered at first would be forgotten on the
second recall test.

Implicit Learning
Implicit learning refers to learning that occurs without awareness or intention to learn.
Often the implicitly acquired knowledge cannot be verbally described. Even if you cannot
remember your credit card number, notice how quickly you can type it after all that prac-
tice. Instead of explicitly requesting recall of the target material from our subjects, implicit
learning might be shown by their faster or more skilled performance of a task. Implicit
testing can present evidence for memory than explicit testing does not.
164 Verbal Learning

Implicit learning of paired-associates can be demonstrated by using a word-fragment

completion task. The subjects are shown a series of word pairs, such as BOOK–FOREST.
A cover story is given so the participants do not know that this is a memory study.
Afterwards, the subjects are given strings of letters and blanks and asked to complete
each with the first word that comes to mind. For example, BOOK–FOR_ _ _. The instruc-
tions make no reference to the previously seen words. Some of the fragments are paired
with the same words they were previously paired with, e.g., BOOK–FOR_ _ _; other
fragments now appear with different words, WINDOW–FOR_ _ _. Implicit learning of the
association is shown when fragments are more often completed with the studied
response word in the same-cue condition than in the different-cue condition. The idea is
that the cue word primes the response word that had earlier been paired with it.
Because college students are likely to notice that the study and fragment tests lists
overlap, this procedure may not be a pure test of implicit learning (Bowers & Schacter,
1990). A stronger case for implicit associations tests amnesic individuals who cannot
recall the study list. Nevertheless, the amnesics complete the fragments with words
from the study list (Graf, Squire, & Mandler, 1984).
Is an implicit test of learning simply a more sensitive test of learning? The indication
instead is that implicit learning represents a different form of learning from the deliberate
forms of learning otherwise presented in this chapter. (The distinction between implicit
learning and explicit learning is presented in detail in Chapters 7 and 11).

RELATIONSHIPS AMONG THE VERBAL-LEARNING TASKS

Do serial learning, paired-associate learning, and free recall all tap the same underlying
processes? Does proficiency at one of these tasks ensure ability in the others?
Underwood, Boruch, and Malmi (1978) tested 200 undergraduates on 28 memory tests,
spread across 10 sessions. Several versions of each verbal-learning procedure were
included. For example, free recall was tested with lists of imageable words, abstract
words, and categorically organized words. The statistical procedure of factor analysis was
used to determine which tests were correlated. Underwood and colleagues found that
paired-associate learning and serial-learning were positively correlated: Individuals tended
to do well on both types of tests, poorly on both, or average on both. This suggests that
paired-associate and serial learning have something in common. One possibility is asso-
ciation learning, between stimulus and response or among items in a list. Free-recall
performance was statistically separate. Compared to paired-associate and serial learning,
free recall allows more flexibility to select encoding and retrieval strategies.
The correlational approach suggests that, even among these simple learning tasks,
different abilities or strategies are being tapped.

STATISTICAL LEARNING
The world is full of uncertainties. Any given stimulus might lead to several outcomes; and
any given outcome might be due to several potential causes. However, the world is also
full of regularities. Through experience we abstract the probability that X leads to Y, or X
never occurs with Y. Statistical learning is the detection of regularities in the
 Statistical Learning 165

environment. Associations, rules, and algorithms are acquired through experience that
reflect actual patterns experienced in the world. Learning occurs often without intention
to learn, without deliberate practice, and often without awareness. Statistical learning is
a factor in several forms of learning, such as causal learning (discussed in Chapter 3),
propositional learning, and language learning.

Propositional Learning
By the end of a conditioning experiment, our human subjects can often describe the
relationships among the stimuli, responses, and outcomes that they have experienced.
These relationships can be stated in the form of propositions, or facts: a tone signals
reward, a key press leads to points. Psychologists have developed several propositional
learning paradigms with which to study conditioning-like phenomena.
The general procedure for investigating propositional learning is to present a series
of statements relating various stimuli and outcomes. This occurs in the form of a game
or set of problems for the subject to discover what leads to what. For instance, in a
food allergy scenario sentences are presented in which a food item (such as straw-
berry) is paired with an allergic reaction (such as hives) or no reaction. Across trials, the
subjects learn which food produces which allergic reaction. The participants rate their
beliefs about whether each food is a cause of each allergic response. There is an obvi-
ous parallel here to classical conditioning, in which a CS (e.g., a tone) is followed by a
US (e.g., food).
The purpose of studying propositional learning studies is not just to avoid running
rats. It is used to answer questions about the nature of propositional learning. In particu-
lar, we can compare associative and cognitive theories (Shanks, 2010). Associative theory
says that across trials there is growth in the strength of the association between the two
stimuli (foods and allergies). Cognitive theory asserts that learning is the acquisition of
propositions or beliefs about the relationship among events. These propositions can be
combined to make inferences, in a form of logic, to derive new beliefs. For instance, if
strawberries and cream are eaten and hives occur as a reaction, we might think either of
the foods is the cause. However, if cream is consumed by itself and there is no reaction,
the inferred proposition is that strawberries caused the hives.

Statistical Language Learning

Language is determined by both nurture and nature. The specific language each of us
uses is learned, and there are biological adaptations of brain and voice that allow speech.
Previous generations of learning psychologists tried to explain language as an instrumen-
tal response: certain sounds made by the infant were reinforced, other sounds were not,
and through the process of shaping coherent speech developed.
One focus of contemporary research is the statistical approach to language develop-
ment in preverbal infants. Regularities occur in language at several levels. On the phono-
logical level, certain sounds follow one another in the formation of words. In English the
sound of the letters CH is often followed by an O or U, but not by an S or F. On a gram-
matical level, certain categories of words regularly precede other categories. Adjectives
precede nouns but not verbs (e.g., LARGE DOG, but not LARGE BARK).
166 Verbal Learning

Analysis of the physical sound spectrum of speech shows that words often run
together. From what sounds like a continuous stream of speech, infants learn to identify
individual sounds. Although there are spaces between written words, there are not silent
spaces between spoken words. In fact, silences often occur within words. In the spoken
word SILENCE, there is a gap in the sounds between si- and -lence. How do infants learn
the individual words from this sort of input?
The regularities of spoken language can be simulated by exposing infants to an artifi-
cial language. In one study, eight-month-old infants were exposed to fabricated three-syl-
lable words such as DAROPI and GOLATU. The words occurred in a continuous stream
lasting several minutes. There were no pauses or gaps between words, and no stress or
intonation in the syllables. To determine whether the infants had learned sequences of
syllables, individual words were played and the listening time was measured. Old words,
such as DAROPI, were compared to new nonwords, such as PIGOLA (the PI from
DAROPI, and the GOLA from GOLATU), which had not occurred together as a word. The
infants listened longer to the novel sound combinations, indicating that the nonwords
were recognized as different from the familiar words. Similarly, the children listened
longer to new syllable pairs that spanned words, such as LADA, than pairs that had been
within words, such as GOLA (Saffran, Aslin, & Newport, 1996).
Similar methods are used to study the learning of regularities at the grammatical
level. In English, the words THE or A are likely to be followed by a noun (DOG) and not by
a verb (WENT). Infants more easily learned nonsense languages in which these grammat-
ical regularities were present than when the absent (Saffran, 2009).

Fast Mapping
There are other factors that affect language learning. How do we learn what individual
words refer to? The notion of association formation suggests that people learn to associ-
ate the names of things with the objects themselves. Infants hear “dog” at the same
time that they see a dog. Word learning is slow at first, but with experience learning
becomes faster. Children develop the capacity to learn the names of objects after a single
pairing. Fast mapping takes learning a step further: if a new word and a new object occur
together, by inference the new word must be the name of the new object. This is fast
mapping: inferring new words are labels of new objects, action, or events.
Although we could cite examples of fast mapping in human infants, there are impres-
sive demonstrations in dogs. Border collies have been taught name-object associations
by having the dogs retrieve the named objects. Learning at first was gradual, spread out
over several years of daily practice. Rico had by then learned 200 object names, and
Chaser knew 1,000 names (Kaminski, Call, & Fischer, 2004; Pilley & Reid, 2011). The dogs
were then tested by placing a new object among a number of already-named objects. The
experimenters asked the dogs to retrieve the unfamiliar object, using a new word (“Bring
me the PIGOLA”). This required the dogs to infer that the new word must refer to the new
object. Both dogs were able to do this on multiple tests using novel words and objects.
After a single presentation and test, the new names were usually forgotten within a day.
As with the previously learned words, practice was necessary for the dog to permanently
retain the new names.
 Applications 167

APPLICATIONS
Assessment of Cognitive Impairment
Versions of the tests described in this chapter are frequently used in educational, coun-
seling, or medical settings to evaluate suspected learning or memory deficits. Tests of
free recall, recognition, or paired-associates are standardized to produce uniform memory
items, instructions, and scoring procedures. The tests are normed and average scores are
collected for different age, education, or diagnostic groupings of people. For instance, the
data in Figure 6.4 showed free recall of word lists by adults of different ages.
Among older adults, poor performance in comparison to age-matched controls could
suggest early signs of memory impairment. Mild Cognitive Impairment (or MCI) is a con-
dition falling between normal cognitive functioning and the severe impairment of demen-
tia. It may be a precursor to Alzheimer’s, although not all individuals diagnosed as MCI
progress to dementia. Individuals with MCI have lower scores on several measures of
free recall: immediate recall of a word list; learning the list across several presentations;
and recalling the list after a delay. When MCI and control subjects completed memory
problem questionnaires, their scores were indistinguishable. It seems everyone thinks
their memory is bad. The laboratory tests were better measures of diminished function-
ing than were simply asking people (Mueller et al., 2015).
The validity of the memory tests to predict future functioning was shown in compar-
isons to neuroimaging test results. Individuals with MCI were followed for several years
as some progressed to Alzheimer’s. Neuroimaging measures (CAT scans and MRIs)
made at the beginning of the study correlated with future memory decline. That is, fea-
tures in the scans predicted those whose memory declined and those who remained
stable. The cognitive test scores also well-predicted who would decline. The strongest
predictor of subsequent Alzheimer’s was a combination of both scans and memory test-
ing (Peters, Villeneuve, & Belleville, 2014).

Mnemonics
Mnemonics are various techniques or strategies that aid encoding and retrieval. Students
often use first-letter and acronym mnemonics: remembering a list of things by their first
letters. For example, the first letter of each of the Great Lakes makes the acronym
HOMES. First-letter mnemonics can encode both items and their order. In the past, a
sentence helped to remember the planets in sequence: “My Very Earnest Mother Just
Served Us Nine Pizzas.” Then astronomers dropped Pluto from the planet list. Prior to the
invention of the printing press, epic tales such as the Iliad were transmitted orally for
hundreds of years. Elaborate mnemonic schemes, often based on imagery, were devised
to organize, store, and retrieve information. “Memory was needed by the entertainer, the
poet, the singer, the physician, the lawyer, and the priest” says the historian Daniel J.
Boorstin (1983, p. 482).

Verbal Mnemonics
An example of a mnemonic technique for remembering lists of words is the narrative
story method (Bower & Clark, 1969). The idea is to make up a story that includes each
168 Verbal Learning

word in sequence. For example, the following student-generated sentence includes six
to-be-remembered words (capitalized): a LUMBERJACK DARTed out of a forest, SKATEd
around a HEDGE past a COLONY of DUCKs. College student participants studied 12 lists
of 10 words each. One group was instructed in the narrative procedure, and the other
group was simply told to memorize the list. The interesting results occurred at the end of
the session, when the students were given only the first word of each list and were asked
to recall the rest in the correct order. The narrative subjects remembered about 93 per-
cent of the words, whereas the control subjects recalled only 13 percent.

Imagery Mnemonics
Most mnemonic systems are based on visual imagery. The method of loci, or locations,
is an ancient mnemonic based on imagery. It is sometimes described as taking a mental
walk. In this method, you first create a fixed series of places that you can readily image in
the same sequence each time. This might be the rooms of a house or locations on your
campus. Later, when you have a list of things to remember, you mentally follow the route,
imagining a to-be-recalled item at each location. When you want to recall the list, you
rewalk the route and look to see what is stored at each location. In European medieval
times, scholars were said to recall thousands of facts by locations spread through
churches, theaters, or cities (Yates, 1966).
Another imagery technique is the peg word system. As with the method of locations,
the first step is to memorize a series of images. In this case, the numbers 1 to 10 and
their rhymes are the pegs on which to-be-recalled items will be imagined. Thus, “one is a
bun, two is a shoe, three is a tree,” and so on. Later, the to-be-remembered items are
imagined interacting with the pegs. As with the method of locations, the peg-word sys-
tem allows recall of the list starting from either end, or for individual items. What was the
sixth item in the list? Well, six is sticks, so what image is stored with sticks?
Our everyday theories about distinctiveness suggest that bizarre or unusual images
would be well remembered. Generally, this is true with the imagery mnemonics. However,
bizarre images are not always so effective: It can be difficult to reinstate what you had
associated with the bizarre stimulus (Einstein, McDaniel, & Lackey, 1989).

Analysis of Imagery Mnemonics

What are the essential components of a mnemonic system such as the peg-word or
location method? Bower (1970) reviewed the research findings with respect to these
two, and noted the following features:
1 There is a known list of cues that need to be present at encoding and at retrieval.
2 Imagery that is unusual, bizarre, or striking is better.
3 The cues used at recall must be the same as those thought of while studying.
4 Multiple items can be stored at each location or with each peg.
5 Surprisingly, the systems are reusable with little interference from trial to trial.
Do mnemonic systems really work? A simple answer is “yes,” they can be quite effective.
In one experimental comparison, three imagery techniques (location, peg-word, and link-
ing items in an image) were used to memorize 20-word lists. The college students recalled
more one day later than did control subjects who were told to simply rehearse each word
 Applications 169

or form a mental image of each (Roediger, 1980). The results of one test are shown in
Figure 6.7. The data here are words recalled in the correct serial position. These results
show that the method of location and the peg-word system produced better recall of
items in sequence.
A more complex answer is that different mnemonic systems may work better for
different memory tasks. For paired-associate learning, linking images was the most effec-
tive. For free-recall tasks, in which order of recall is not important, the narrative story
method was superior. For serial recall of items in sequence, the narrative method and the
method of locations were best (Herrmann, 1987).
Joshua Foer was a recent college graduate, working as a journalist and living in his
parents’ basement. He had written an article about memory competitions, informally
referred to as Memory Olympics. Each participant competes in tasks such as memorizing
poetry, the first and last names of 100 faces, or random strings of digits (the winner
learned 148 numbers in five minutes).
Foer decided to enter himself. He spent a year in full-time training, and consulted with
experts in psychology and mnemonics. The mnemonic systems used by the mental ath-
letes are far more sophisticated than first-letter or narrative chaining techniques. The
methods use imagery, adapted to encode letters or numbers; and locations, to guide the

1.00

Peg
.80 Loci

Link
Probability of Recall

Rehearsal
.60

.40
Imagery

.20

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Serial Position

Figure 6.7
Serial Recall of 20-Word Lists as a Function of Several Mnemonic Strategies. Strategies are rehearsal
(repeat each word), imagery (vividly picture each word), a linking story, the peg-word system, and the method
of loci (locations).
Source: From “The Effectiveness of Four Mnemonics in Ordering Recall,” by H. L. Roediger III, 1980, Journal of Experimental
Psychology: Human Learning and Memory, 6, p. 565. Copyright 1980 by the American Psychological Association. Reprinted with
permission.
170 Verbal Learning

rememberer through the correct sequence. Foer’s Olympic event was memorizing the
sequence of a deck of cards. He used a multi-level imagery system that encoded three
cards at a time. Thus, the title of his book, Moonwalking with Einstein (Foer, 2011) was
one of the images. Foer did in fact win the USA Championships with his record score of
learning the card deck in 1 minute and 40 seconds.
One take-away from this demonstration is that almost anyone can develop an excep-
tional memory skill. The ability is not genetic, not pure intelligence, and not pharmacolog-
ically aided. Another lesson is that the exceptional memory is task specific. Foer says he
is no better at remembering everyday things than he was before training.
So, who really uses mnemonics? Surveys show that most people occasionally use an
external reminder such as notes and lists. The only true mnemonic reported was the
first-letter mnemonic: 73 percent of students had used it recently (e.g., Harris, 1980). For
most things it is simpler to put a list on your phone. Electronic aids such as the smart-
phone have replaced the need for mnemonic systems. A survey of psychologists who
work in the field of memory research found that even they do not use the mnemonic
devices they teach in their classes (Park, Smith, & Cavanaugh, 1990).
Perhaps the contemporary value of mnemonic techniques is in demonstrating the
power of variables that most affect learning: imagery, meaningfulness, organization, and
retrieval cues. We may not use formal mnemonic systems to remember, but we use the
components in smaller ways in everyday remembering.

SUMMARY
Verbal learning refers to research that followed in the Ebbinghaus tradition and that seeks
to determine the effects of certain independent variables on learning verbal items.
Herman Ebbinghaus, in his 1885 book On Memory, demonstrated how to quantify
learning (i.e., the number of repetitions to criterion) and memory (i.e., the number of
repetitions to relearn, or savings). He described the Curve of Forgetting, which showed
most forgetting occurs shortly after learning, with further losses occurring at a slower
rate. Ebbinghaus also devised new learning materials (the nonsense syllable) and intro-
duced methods of experimental control in his research.
Three basic tasks have been developed to study verbal learning: serial learning,
paired-associate learning, and free recall.

Serial Learning
In serial learning, a list of items is presented in a fixed order on each trial, and the subject
is required to reproduce items in order. The serial-position of items within the list influ-
ences learning, as the first and last items in the sequence are learned more quickly than
are the middle items.
Item-to-item associations in a list are learned, and remote associations are acquired
between separated items. The theory of hierarchical associations says that serial lists are
divided into sections, and those into subsections, and so on. Rather than each single item
being activated via association from the immediately preceding item, all the items con-
tained within a segment are activated.
 Recognition Memory 171

Paired-Associate Learning
In paired-associate learning, two items are presented for study, labeled stimulus and
response (abbreviated S and R). The task is to recall the response when presented with
the stimulus alone. Paired-associate learning can be analyzed into the components of
stimulus discrimination, response learning, and S–R associating.
The stimuli used in a paired-associate learning task may be similar, causing interfer-
ence. Learning will be faster if the stimuli are made more discriminable from one another,
for example, by adding distinctive features.
Response learning is affected by the difficulty in learning the response items.
Meaningfulness, familiarity, and prior knowledge will facilitate response learning.
Associating the stimulus with the response is affected by preexisting associations
among items in the to-be-learned set and by cognitive elaboration. For example, generat-
ing a mediator to relate stimulus and response will facilitate association.
Paired-associate connections typically appear stronger in the forward, stimulus-to-
response direction. Remembering the stimulus when given the response, so-called back-
ward association, usually shows poorer recall.

Free Recall
In free recall, a list of items is presented and the subject attempts to recall as many of
them as possible. Unlike serial-learning procedures, ordered recall is not necessary.
One of the most-studied features of free recall is the serial-position effect. Enhanced
recall of the first items in a list, or primacy, is often attributed to the additional rehearsal
that the first items receive. Enhanced recall of the last items in a list, or recency,
occurs because these items are often recalled first before forgetting has had a chance
to occur.
Free recall is influenced by rehearsal or repeating the items until they are tested.
Rehearsal shows a developmental progression in children. Rote repetition is not as effec-
tive as some sort of elaborative rehearsal, which seeks meaning or associations among
the to-be-remembered items.
Free recall improves if the material can be organized, that is, if items can be grouped.
Categorical organization occurs when items from the same semantic category are
grouped together at output. Subjective organization occurs when subjects devise their
own grouping of items and is detected by analyzing the sequence of recalled items across
successive tests.

RECOGNITION MEMORY
Another method used to assess retention learning is to ask the participant to identify
previously seen items from a group of distractors. Recognition is a decision whether
something is familiar on the basis of different types of memory evidence. In a remem-
ber versus know test, the participants decide whether they specifically remember that
a given item was on the studied list, or if the item simply engenders a feeling of
familiarity.
172 Verbal Learning

Accessible Memories
Ebbinghaus showed that something could be present in memory even if it could not be
recalled. Available memories are those present in the memory store, whereas accessible
memories are those that can actually be recalled or retrieved. This distinction is illustrated
by the tip-of-the-tongue phenomenon, in which you know that you know a word or name
but you can’t recall it right now. Presenting additional memory prompts during cued recall
makes more information accessible.
How do tests of recall, recognition, and relearning differ in their sensitivity to retriev-
ing available memories? Items that are not recalled are sometimes recognized, and sav-
ings occurs for items that are neither recalled nor recognized correctly.
Tests of implicit memory, such as word-fragment completion, detect memory in
amnesic individuals who otherwise do poorly on recall and recognition tests of explicit
memory. Recall, recognition, and implicit memory may differ in sensitivity to detecting
memory, or they may be assessing different aspects of what is remembered.
Do serial learning, paired-associate learning, and free recall tap the same underlying
learning processes? Correlational studies suggest that there are different abilities or strat-
egies involved instead of a single verbal-learning ability. Do these laboratory tests predict
everyday remembering? The correlations are generally low, but this may be because labo-
ratory and everyday memory tasks differ in kind, interest levels, and ecological relevance.

Statistical Learning
Statistical learning describes the strength of individual relationships among stimuli,
responses, and outcomes that have been experienced. The relationships can be stated in
the form of propositions or factual statements. Statistical learning applies to causal, prop-
ositional, and language learning.
The statistical approach has been applied to language acquisition in preverbal infants.
It focuses on the regularities that occur in language at several levels: phonological, word,
and grammatical. Studies of learning artificial languages show that regularities at each
level are detected and learned by infants.

Applications
The human learning laboratory tasks are routinely used in assessment of cognitive impair-
ment. Tests of free recall, recognition, and paired-associates are used in educational,
counseling, or medical settings to evaluate suspected learning or memory deficits.
Among older adults, poor performance in comparison to age-matched controls could sug-
gest early signs of memory impairment. The validity of the memory tests to predict future
functioning is shown in comparisons to neuroimaging test results.

Mnemonics
Mnemonic techniques or strategies aid encoding and retrieval, such as using the first let-
ters to form an acronym for the to-be-remembered items. In the narrative chaining method,
a string of unrelated words are connected through a made-up story. In the method of loci,
or locations, items are imaged in certain locations along a fixed mental route.
 Recognition Memory 173

Mnemonics work because there is a list of cues that the to-be-recalled items are
associated with during study and that can be used to guide retrieval.
The various mnemonic systems do indeed work, although some are better for
serial-learning tasks (such as the method of loci) than for paired-associate learning
(linking images) or free recall (narrative chaining). However, mnemonic techniques are
time-consuming to learn and to use. Few students use more than acronyms. Most
people instead use external cues such as sticky notes, or electronic devices such as
smartphones. However, mnemonics are important in demonstrating the principles of
acquisition and retrieval.
 CHAPTER

7
Human Memory Conceptual Approaches

CONTENTS

Partitioning Memory 175 Depth of Processing 187

Components of Memory Approach 177 Artificial Neural Networks 188
Dual-Store Theory: Short-Term and Modeling Configural Learning 189
Long-Term Memories 177 Connectionism and the Other
Divisions of Long-Term Memory: Approaches 191
Episodic and Semantic 178 Applications 191
Divisions of Long-Term Memory: The Study of Abnormal Memory:
Procedural Learning and Priming 180 Amnesia 191
The Organization of Long-Term Types of Amnesias 193
Memory 184 Everyday Forgetting and the Models
Stages of Memory Approach 184 of Memory 197
Dissociating Stages 185 Summary 199
Processing Approaches to Memory 187

Forgetting is the bane of every student’s existence. It is bad enough to forget course
material, but forgetting continues beyond the classroom. Students who kept a diary of
memory lapses reported all too frequently forgetting meal cards and keys, appointments,
classes, and even exams (Terry, 1987). In another survey, students reported more
instances of forgetting than did an elderly sample, even though the popular stereotype is
that the latter are the forgetful ones (Reason, 1993). There is the story of one student who
worked all night to complete a history paper, only to realize the next morning that his
English paper was due that day.
When students ask me why they have trouble remembering, I have to respond that
this is too broad a question for a simple answer. “Memory” is not a single thing. Memory
is better understood when partitioned into separate components. A more helpful answer
is the realization that forgetting is due to a failure of one aspect of memory or another.
This chapter considers several theoretical approaches to memory, each of which par-
titions memory along different dimensions. One approach postulates different types of
memory, such as short-term and long-term memories. Another approach divides memory
into a series of stages: our now familiar forming, retaining, and retrieving memory. A third

DOI: 10.4324/9781003227090-7
 Partitioning Memory 175

Table 7.1 Three Approaches to Memory

Memory Components Stages of Memory Processes of Memory

Short-term memory Encoding into memory Depth of processing

Long-term memory Storage in memory Passive (shallow)
Episodic Retrieval from memory Elaborative (deep)
Semantic
Procedural
Priming

approach emphasizes differences in how information is processed. These three approaches

are outlined in Table 7.1. A fourth approach attempts to model memory in the nervous
system by describing the formation and activation of neural connections.
These theories generally derive from cognitive psychology, but other fields also offer
insights into the nature of memory. The field of neuropsychology, with its emphasis on
brain and behavior, seeks to distinguish those areas of the brain associated with the dif-
ferent components of memories. Case studies of individuals with brain injuries provide
important means of testing our theories and for generating new conceptions of memory.
In addition, artificial neural net models merge biological knowledge of the neuron with
cognitive theories of memory.

PARTITIONING MEMORY
Why do we hypothesize multiple memory systems instead of a single memory? Tulving
(1985) considered several reasons. One reason is that we cannot make many generaliza-
tions about memory as a whole, although statements about particular forms of memory
are valid. Thus, there may be separate principles for long-term memory and for short-term
memory, or for personal memories versus knowledge, but just not the same principles. A
hypothesis of multiple memory systems has heuristic value in simplifying our theories.
The several components, each with different characteristics, may be a simpler model to
construct and to use.
Another reason for postulating multiple memory systems is that the results of our
studies suggest multiple processes are involved. This is part of the research strategy of
uncovering dissociations. Dissociation occurs when an experimental variable has differ-
ent effects on different tasks or measures. For example, are verbal learning and spatial
learning the same memory system or are they separate? Injury to the left hemisphere of
the brain impairs learning of verbal items, such as recalling word lists. Spatial learning of
a maze or a new route is unaffected. Because only verbal learning is affected by left-brain
injury, this could suggest these two types of materials are learned by different brain areas.
A simple difference between verbal and spatial learning could occur for reasons other
than a specialized learning capacity in the left hemisphere. A more convincing demonstra-
tion would be a double dissociation, in which the experimental variable differentially affects
performance on two tasks. If we can also show that right-hemisphere lesions impair spa-
tial memory but not verbal learning, then we would have a double dissociation.
176 Human Memory Conceptual Approaches

Hemisphere Intact
100
Words
Locations

75

Percent Correct

50

25

0
Left Right

Figure 7.1
Effects of Lesions to One Side of the Brain on Recognition Memory for Words or Spatial Locations.
Source: “Learning and Memory in Humans, with an Emphasis on the Role of the Hippocampus,” by R. P. Kesner, R. O. Hopkins,
and A. A. Chiba, in Neuropsychology of Memory (p. 109), edited by L. R. Squire and N. Butters, 1992, New York: Guilford Press.

Figure 7.1 shows the results of one such study. The subjects had to remember, in one
case, whether a test word had been recently presented on the video screen, and in the
other case, whether an X had been presented at a certain location on the screen. The
subjects were individuals who had surgical lesions of one hemisphere of the brain to
control epilepsy. Those with an intact left hemisphere (usually the language side) did
better on word memory than on spatial location remembering. Those subjects with an
intact right hemisphere (usually the nonlanguage side) remembered more spatial loca-
tions than words (Kesner, Hopkins, & Chiba, 1992). This finding suggests that the two
cerebral hemispheres play different roles in verbal and spatial learning.
Dissociations may be produced in several ways: experimental manipulations (e.g.,
the length of the retention interval before testing); tasks (recalling words or spatial loca-
tions); neurological differences (people with injuries to the left or right hemisphere); and
subject populations (children versus older adults).
Each uncovered dissociation needs to be cautiously interpreted. A dissociation does
not automatically mean there are separate memory systems. Two tasks could vary in their
procedures, difficulty, or familiarity (Hintzman, 1990). For example, Hanley-Dunn and
McIntosh (1984) gave college students and elderly adults lists of names to remember.
The students did better when the to-be-remembered names were rock music perform-
ers; the elderly did better when the names were of Big Band musicians. This pattern of
results does not show that there are separate memory systems for different categories
of musicians. In this case, it was the familiarity of the to-be-learned materials that varied
between the two age groups.
 Components of Memory Approach 177

COMPONENTS OF MEMORY APPROACH

One approach to partitioning memory divides memory into different types based on how
long memory is stored. The primary divisions are short-term memory and long-term
memory.

Dual-Store Theory: Short-Term and Long-Term Memories

The emphasis within many theories has been on two memory systems: short-term
memory and long-term memory. Thus, these models are often referred to as dual-storage
theories. The theory of Atkinson and Shiffrin (1968) is the prototype of such theories and
so is sometimes referred to as the “modal model” (or average model). Short-term
memory (STM) and long- term memory (LTM) have different characteristics that distin-
guish the two systems.
First and most obvious, STM is brief. Unless maintained by rehearsal, information in
STM lasts 15 to 30 seconds under laboratory testing conditions. If you look up a new phone
number, you will forget it quickly, maybe before you even have time to dial it. Memories in
LTM are more durable. You won’t likely forget your current phone number overnight.
STM has a limited capacity, holding at most a few items, whereas LTM is assumed to
be virtually limitless in size. Retaining a seven-digit number is plenty for STM; retaining
the many numerical sequences we use in everyday life is possible with LTM.
Forgetting from STM occurs when the contents are displaced (and replaced) by lat-
er-occurring items. Reading a second phone number can displace a first number from
STM. Forgetting does not easily occur from LTM. Forgetting your own phone number
does not usually occur, other than momentary lapses.
Finally, STM serves to transfer information into LTM. Specifically, rehearsal in STM
keeps information available longer for encoding into LTM.
(The preceding items are the modal properties of short- and long-term stores. The
next chapter contains an elaboration of STM and developments since Atkinson and
Shiffrin’s original theory).
What sorts of evidence suggest that there are two memory systems? Short-term and
long-term memories are dissociated in several ways. The serial-position curve that is
obtained in free-recall tasks is differently affected by several independent variables. In
remembering a list of words presented once, recall of the first words, also called pri-
macy, is enhanced if the list is presented at a slower pace (Glanzer & Cunitz, 1966). The
slower pace allows more opportunity for rehearsal in STM, allowing more encoding into
LTM. The results from one study are shown in the left panel of Figure 7.2. As can be seen,
there was better recall after the 9 seconds-per-item rate (slow presentation) that after the
3-second rate (fast presentation). Rate of presentation especially affected primacy. Fast
or slow presentation did not affect recall of the last items of the list.
Enhanced recall of the final list items, called recency, occurs if memory is tested
immediately after list presentation. Recall of the last items is impaired if testing is delayed
(Glanzer & Cunitz, 1966). This is shown in the right panel of Figure 7.2. With a test imme-
diately following list presentation, the subjects usually recall the final items first, while
they are still in STM. After testing was delayed for 30 seconds, recency disappeared but
the primacy effect was still there.
 178 Human Memory Conceptual Approaches

1.00
.70

.60
.75 Immediate
Proportion Recalled

Proportion Recalled
.50

.50 9 sec per item

.40

.25 .30

30-sec delay
3 sec per item .20

.00
1 5 10 15 20 1 5 10 15
Serial Position Serial Position

Figure 7.2
Serial Position Curves. Unordered recall of words as affected by (left) rate of presentation of the items, 9
seconds or 3 seconds per item; and (right) testing immediately after list presentation or after a 30-second delay.
Serial position is the place of each item within the lists at presentation.
Source: Adapted from Glanzer and Cunitz (1966).

Short-term and long-term memories are also dissociated by the patterns of memory
loss by injuries to different areas of the brain. The amnesic subject referred to as H. M.
had damage to the interior of his temporal lobes. Specifically, as a result of experimental
surgery in 1953 for the relief of epilepsy, the hippocampus and other nearby structures on
both sides of his brain were removed. Afterwards, H. M. was unable to form new long-
term memories. He could not learn new facts, vocabulary, places, faces, or mazes.
However, his short-term memory was preserved. He could remember short strings of
letters, digits, or words. H. M.’s pattern of what was remembered and forgotten suggests
a dissociation between short-term memory, which was spared, and the ability to acquire
long-term memories, which was lost to the brain lesions (Scoville & Milner, 1957).
The case of K. F. provides the opposite deficit. K. F. cannot repeat back even the
shortest strings of words. If read two or three items, he would likely recall only one.
However, K. F. can form long-term memories. With repetition he can learn lists of 7 to 10
items (Shallice & Warrington, 1970).
(These descriptions of H. M. and K. F. are overly broad. More detail will be given later
in this and other chapters).

Divisions of Long-Term Memory: Episodic and Semantic

There are several types of information retained in long-term memory, including autobio-
graphical memories, factual knowledge, and various skills and habits, which suggest to
theorists that LTM itself is made up of several memory components. One broad division
of long-term memory is that between episodic memory and semantic memory, a differ-
ence that corresponds to what we might casually label “memory” and “knowledge”
(Tulving, 1985).
 Components of Memory Approach 179

Episodic memory is our personal memory system. Episodic memories contain tem-
poral and contextual information about when and where. They are memories for events
that happened at particular places or times in the past. We can use time of occurrence to
retrieve episodic memories, and recollected memories can be approximately dated. An
example would be remembering moments from last summer’s vacation. In recalling an
episodic memory, there is a conscious awareness of remembering. Episodic memory is
a more-recently evolved memory system (versus classical conditioning, which is one of
the oldest). Episodic memory was believed to be unique to humans, but now research on
other animals is showing evidence for its presence there.
Semantic memory is our store of general knowledge. It is more like dictionary or
encyclopedic knowledge. It includes facts, words, language, and grammar. It has some-
times been referred to as generic memory, or knowledge that most of us have in com-
mon, in distinction to our individual unique and personal episodic memories. Semantic
memory includes knowledge, but not the memory of how or when it was learned. For
example, you have a great deal of generic knowledge about dogs in semantic memory—
they make good pets, they bark, they are related to wolves, and so on. You probably don’t
recall how or when you learned each fact. By contrast, a specific memory of an event in
your life involving a dog at a particular time and place would be episodic recall. Our every-
day phrases “I remember” versus “I know” parallel episodic and semantic memory.
Another term that is used is autobiographical memory, which is a mixture of episodic
and semantic memories. There are semantic memory facts that I know about myself, and
episodic memories of experiences I have had.
Many of the laboratory experiments on human memory presented in this text are
tests of episodic memory: memory for words and pictures that occurred in a particular
context (the lab) at a particular time (often just a few minutes ago). In a free-recall task,
you might be asked to recall the most recent list of words that was presented in the day’s
session. The unremembered words are forgotten from episodic memory.
The episodic–semantic distinction is evident in certain types of amnesias. Schacter
(1983) reports a case of impaired episodic but spared semantic memory in an acquaint-
ance with Alzheimer’s. A person with Alzheimer’s often has trouble with ongoing mem-
ory, or remembering events as they occur, and so are forgetful about the recent past.
Rather than simply reporting data from laboratory tests, Schacter assessed memory dur-
ing a couple of rounds of golf. Schacter’s friend remembered the rules of the game and
knew which club to use and when. For example, when faced with a water hazard, he
selected a short iron to safely lay up before the pond. He correctly used an extensive
vocabulary of technical terms, such as birdie, bogey, and divot. All this is knowledge in
semantic memory. Yet after a brief delay, the man could not remember which direction his
ball had gone or which ball on the fairway was his. After moving on to the next tee, he
could not recount any details of the previous hole. This is failure of episodic memory, an
inability to form ongoing memories of personal life.
A dramatic case of episodic memory loss involved a man who received a head injury
in a motorcycle accident (Tulving, 1989). The young man can remember all sorts of facts
about his life. These generic facts, about where he lived and worked, where he went to
school, or how to play chess, are in semantic memory. Note that not all knowledge about
ourselves is episodic. We have personal semantic knowledge. However, this young man
180 Human Memory Conceptual Approaches

could not remember a specific episode involving himself in any of these known events.
He could not remember having vacationed in the family cabin, or having been at work, or
having ever played chess. This is loss of episodic memory.
Attempts to functionally dissociate episodic and semantic memories in laboratory
experiments have had mixed success. One difficulty is that testing the two hypothesized
forms of memory requires different materials, procedures, or tasks. For example, a list of
words might be presented to subjects in an experiment. Later, half of the subjects are
given a recognition test of episodic memory: A long list is presented from which to select
the words studied earlier. The remaining subjects are given a test of semantic memory.
Strings of letters are briefly flashed on the screen, and the task is to decide which strings
form real words (e.g., CAT) and which do not (e.g., ATC). The problem is we are confound-
ing the type of test with type of memory. Maybe the difference is not episodic versus
semantic, but recognition memory versus perceptual identification.
Some theorists downplay the episodic–semantic distinction. Others defend it for its
heuristic value. The distinction between forms of long-term memory is useful in describ-
ing the development of semantic or generic memories. In children, for example, a child’s
first experience with a cow leads to an episodic memory. Repeated experiences lead to
the formation of a generic memory of cow-related facts and lore. The same logic applies
to our autobiographical memories. Your first college class meetings produce distinct epi-
sodic memories. The day-by-day episodic memories are eventually supplanted by generic
memory of what a typical day in class is like, and forgetting of individual class meetings
(Linton, 1982). This process could be described as the transition from episodic to seman-
tic memory.

Divisions of Long-Term Memory: Procedural Learning and Priming

Explicit versus Implicit
The description of some other types of LTM requires another distinction: explicit and
implicit remembering.
Direct questioning about the contents of memory is an explicit memory task.
Explicit tests tap the episodic and semantic knowledge that we can report on. An implicit
memory task assesses the memory indirectly by measuring performance on some task
that does not require explicit recall. For example, a stimulus is reacted to more quickly the
second time it occurs. Performance improves regardless of whether you recall the first
occurrence. Implicit memory is performance that occurs independently of conscious
recall.
The meaning of the terms explicit and implicit has evolved in their history in the psy-
chology lexicon. Some of the contrasting terms used to distinguish between implicit and
explicit memory are shown in Table 7.2. Declarative memory includes the grouping epi-
sodic and semantic memories: this is memory that is explicitly assessed and available to
conscious recall. Implicit memory systems include procedural learning and priming
(described in the sections that follow). Implicit procedures are hypothesized to be medi-
ated by different neural systems from the explicit memory systems of episodic and
semantic memories.
 Components of Memory Approach 181

Table 7.2 Some of the Contrasting Terms Used to Distinguish Explicit and Implicit Memory

Explicit Memory Implicit Memory

Fact memory Skill memory

Declarative Procedural
Knowing that Knowing how
Conscious recollection Perceptual identification
Memory Habit
Source: Squire (1987, p. 169).

Procedural Learning
Procedural learning is the acquisition of knowledge of how to do things and includes
perceptual skills, motor skills, and cognitive skills. Procedural knowledge has been char-
acterized as “knowing how” rather than episodic and semantic’s “remembering that.”
I know how to skate versus I remember that I once fell (Squire & Cohen, 1984). Procedural
means learning general rules for performing a task or procedure, although this knowledge
may not be accessible to conscious verbal recall.
The procedural learning tasks most commonly used in research have different combi-
nations of skills. For instance, mirror star tracing combines perceptual and motor skills.
Subjects trace the outline of a printed star, but by watching their hand movements through
a mirror reflection. This requires reversing the direction of hand movements from the
direction seen in the mirror. Learning to read text that has been inverted or mirror-re-
versed is another procedural task, one that combines a cognitive skill and a perceptual
skill.

Priming
Priming is the facilitated processing of a stimulus due to prior, recent experience with
that stimulus or with related stimuli. Priming occurs without awareness. For instance,
hearing the word “doctor” primes doctor-associated words so they are subsequently
perceived more readily.
A typical word-priming experiment has two stages. In the priming phase, participants
are shown words, in the context of an incidental task. In the test phase, the words are
presented again in a different context. Priming is demonstrated if test performance is
facilitated by the previous exposure.
For instance, in a perceptual identification task, subjects attempt to read words
flashed on a screen for a fraction of a second. The duration is so short that conscious
recognition is difficult. In the priming phase of the experiment, a number of words are
shown for the subjects to process, such as rating each word for pleasantness. In the next
phase these words and some new words are flashed on a screen for durations ranging
from 33 to 83 milliseconds. The task is to try to identify each word. The results of one
such study are shown in Figure 7.3 (Hamann, Squire, & Schacter, 1995). As can be seen,
words could not be identified from 33-millisecond exposures, but more could be read
after longer exposures. The primed words, those seen earlier in the session, were more
182 Human Memory Conceptual Approaches

1.0

Primed
0.8

Percent Correct
0.6

0.4 Unprimed

0.2

0
33 50 67 83
Exposure Duration (msec)

Figure 7.3
Effects of Priming on Perceptual Word Identification. The subjects attempted to identify words presented for
durations from 33 to 83 milliseconds. Primed words had been exposed earlier in the session. The “primed”
words were more readily identified than unprimed (new words).
Source: From “Perceptual Thresholds and Priming in Amnesia,” by S. B. Hamann, L. R. Squire, and D. L. Schacter, 1995,
Neuropsychology, 9, p. 11. Copyright American Psychological Association. Reprinted with permission.

likely to be identified from brief exposures than were the unprimed control words. Priming
facilitated word identification.
In a word stem completion task, the word EARTH may be one of the words pre-
sented in the first phase. In the test phase the first several letters of a word are given
(EAR_ _) and the subject is asked to complete it with the first word that comes to mind.
Priming occurs when the stem or fragment is completed with words exposed during the
priming phase, as opposed to other words that may fit (such as EARLY). For instance, the
stem EAR_ _was completed with EARTH about 49 percent of the time in the priming
condition as opposed to 30 percent in the control condition (Rajaram & Roediger, 1993).
In some cases, priming leads to better performance than explicit testing. For exam-
ple, college students attempted to explicitly recognize words they had studied seven days
previously, or complete word fragments based on those words. In the latter condition, the
students were not reminded that the words had been presented a week earlier. Only 20
percent of the words were correctly recognized a week later, but 50 percent of fragment
completions produced words that had been presented a week earlier (Tulving, Schacter,
& Stark, 1982).
Priming occurs with pictorial stimuli. Drawings of objects are presented and tested
later through a method of fragmented pictures (see Figure 7.4). There are several versions
of each picture with increasing fragmentation. Subjects attempt to identify the object
from the least amount of detail. Across trials, identification becomes easier from the
more degraded fragments. (This technique was actually one of the first used to show that
individuals with amnesia learn implicitly; Warrington & Weiskrantz, 1968a).
 Components of Memory Approach 183

Phase I: Study either or AIRPLANE

Phase II: Test with either or A__P_A_E

Phase I

0.20 Picture
Words

0.15
Priming

0.10

0.05

0.00
Picture-Fragment Word-Fragment
Naming Completion
Phase II

Figure 7.4
Design and Results of Weldon and Roediger. Studying pictures in Phase I produced more priming in Phase II
than did words on the picture-fragment test; studying words produced more priming on the word-fragment test.
Source: From “Implicit Memory: Retention without Remembering,” by H. L. Roediger III, 1990, American Psychologist, 45,
p. 1051. Copyright American Psychological Association. Reprinted with permission.

Priming tends to be sensory modality specific. That is, a picture prime will not facili-
tate identification of the word that names the picture; and a word prime will not facili-
tate recognition of a picture that represents the word. In a study by Weldon and Roediger
(1987), their college student participants studied lists of both pictures and words in
Phase I. The pictures were simple line drawings like that shown in Figure 7.4 and had
simple one-word names (e.g., airplane). In Phase II, either a picture fragment or a
word-completion test was given. The results showed stimulus specific priming effects.
In the word-completion test, previously studied words were more often completed
than studied pictures. That is, the priming word AIRPLANE facilitated completion of
184 Human Memory Conceptual Approaches

A__ __ P__A__E, but a drawing of an airplane did not aid in filling in the A__ __P__A__E
fragment. In a picture-fragment test, previously viewed pictures were more easily identi-
fied when the prime had been a picture rather than a word.

Dissociating Priming and Explicit Memory

To argue that implicit priming represents a different memory system than that accessed
by explicit tests of episodic or semantic memory, we need to demonstrate that the
two classes of tasks are dissociable. However, the contribution of explicit memory to
an implicit test cannot readily be discounted. College student participants may remem-
ber the words presented in the first phase and the test items and perform the task
explicitly.
There are other comparisons we might use to dissociate implicit and explicit memory
(Tulving & Schacter, 1990). For instance, neurologic disorders might affect explicit and
implicit memory differently. By testing subjects who are amnesic, awareness of the pre-
vious study trial is less of a factor in an implicit memory test. On an explicit test of mem-
ory, amnesic subjects simply cannot recall many of the words studied earlier. Yet they do
as well as non-amnesic individuals in completing word stems with previously seen words
(Shimamura et al., 1987) or identifying pictures from fragments.

The Organization of Long-Term Memory

What is the relationship among the several hypothesized components of long-term mem-
ory? Various organizations of memory have been proposed. In the generally accepted
organization (Squire, 1987; Squire, Knowlton, & Musen, 1993), memory is divided into
two major categories, declarative (which is explicit) and nondeclarative (which is implicit).

Memory

Declarative Nondeclarative

Episodic
i d Semantic
S Procedural Priming Habits
H Conditioning

Declarative memory can be consciously recalled and reported on. That is, we can
“declare” these memories. Instructions that explicitly request recollection from memory
are accessing declarative memory. Declarative LTM includes episodic and semantic
memory.
Nondeclarative memory is defined by exclusion. It is whatever types of memory that
are not declarative. Nondeclarative includes procedural learning, priming, habit learning,
and classical conditioning. Nondeclarative knowledge, as is obvious from its label, is often
assessed implicitly and not by explicitly requesting recall.

STAGES OF MEMORY APPROACH

A second approach to memory separates it into a series of three stages. A memory must
first be formed, then retained, and later retrieved. These stages are labeled encoding,
 Stages of Memory Approach 185

Encoding Storage Retrieval

Acquisition Retention Remember
Learning Long-term Recall
Memory Consolidation Access
formation

Figure 7.5
The Stage Model of Memory. Memory processing goes through a series of stages. Shown are the stages and
some terms associated with each stage.

storage, and retrieval. Forgetting could arise from problems at any one of these stages.
For example, if you cannot remember the answer to an exam question, it may be because
(1) you never really learned the material, (2) you learned it but it has since been lost from
memory, or (3) you learned it, it’s there, but you can’t recall it right now (although you will
probably remember shortly after turning in your exam). The stage approach is not so much
a theory about memory, but rather a set of assumptions or “givens” about memory. The
goal of the stage approach is to identify the effect of certain variables on one stage of
memory or another.
Figure 7.5 shows the stages and some terms associated with each stage.

Dissociating Stages
Experimental Dissociations
An experimental dissociation occurs when a manipulated variable produces differential
effects across the stages of memory. An experimental variable is manipulated at one
stage while holding other factors constant at the other stages.
This can be illustrated in a study of the effects of a physical stressor applied at the
time of encoding, during storage, or before retrieval. The subjects were given lists of
words to remember on Day 1 and were tested for recall on Day 2. A cold pressor test was
used to induce stress: The students were asked to hold one hand in a bucket of ice water
for as long as they could tolerate—up to three minutes. The groups differed in the timing
of the stressor. The encoding-stress group was stressed shortly before the lists were
presented. The storage-stress group were stressed after the lists were presented. That is,
after encoding had occurred and memory was presumably stored. The retrieval-stress
group received their stressor on Day 2 before they attempted to recall the word lists. The
number of words recalled on the Day 2 test are shown in Figure 7.6. The groups are labe-
led by the location of the stressor on Day 1 (before or after encoding) or Day 2 (before
retrieval). Cold stress before encoding (labeled a in the figure) had no effect on learning:
stressed and unstressed controls recalled just over 50% of the words. Stress induced
after encoding (labeled b) increased recall to 65%, significantly above the control group;
and stress applied before retrieval impaired recall, impaired recall to 40%, less than the
controls (Smeets, Otgaar, Candel, & Wolf, 2008). Stress had different effects when it
occurred at different places in the life of a memory.
186 Human Memory Conceptual Approaches

(a) (b) (c)

Encode Store Retrieve

65% –

STRESS
CONTROL

CONTROL
50% –

CONTROL
STRESS

STRESS
40% –

MEMORY STAGES

Figure 7.6
Effects of a Stressor at Each Stage of Memory. Mean percentage of words recalled as a function of stressor
given before or after encoding; or before retrieval.
Source: From Smeets, Otgaar, Candel, and Wolf (2008).

The stages of encoding, storage, and retrieval cannot be completely isolated from
one another. Also, an experimental treatment can influence more than one stage. For
example, mentally grouping to-be-remembered words into categories aids encoding into
memory and facilitates retrieval from memory.

Neuropsychological Dissociations
The neuropsychologist attempts to dissociate stages by finding individuals with impair-
ment at one or another stage. The amnesic H. M., who had the temporal lobe operation
for epilepsy, seemed to have had an encoding deficit: He simply could not form new
memories (or at least, not episodic and semantic memories). H. M. could still retrieve
some of his older memories from before his operation.
Other aspects of amnesics’ performance suggest that their forgetting may be a
retrieval problem. For example, in the typical experiment several lists are presented dur-
ing the course of an experiment. Although few words are recalled from the just-heard list,
words from earlier lists are sometimes recalled. These intrusion errors show that there is
some memory for the other lists (Warrington & Weiskrantz, 1968b).
Tests of implicit memory also show that information is encoded but is not explicitly
recallable. Amnesic participants show priming effects on an implicit test of stem-comple-
tion. If explicitly asked to complete the letter stems with words from the previously
shown list, the subjects perform poorly. (“What lists?” they ask. After all, they are amne-
sic). However, if the stem completion task is presented as a game (“Can you think of a
word that begins with these letters?”), the stems are completed with previously studied
words.
 Processing Approaches to Memory 187

PROCESSING APPROACHES TO MEMORY

Depth of Processing
A third approach to memory suggests that the kind or quality of processing determines
memorability. Depth of processing theory (Craik & Lockhart, 1972) was initially offered as
an alternative to the theory of separate short-term and long-term memories. According to
depth of processing, remembering depends on the amount of cognitive elaboration, or
processing depth, of the to-be-remembered material. Rapid forgetting is due, not to a loss
from a transient short-term storage, but rather to shallow processing. Sustained retention
is due to deeper or more elaborate processing. Remembering and forgetting reflect vari-
ations in the cognitive depth to which information is processed.
The idea of shallow versus deep processing is illustrated by two kinds of rehearsal.
When we try to remember something, we are sometimes aware that we are rehearsing
the to-be-remembered information. Maintenance rehearsal is the passive repetition of
information, repeating something over and over. This is shallow processing. Maintenance
rehearsal might be used to remember a phone number long enough to dial it. Elaborative
rehearsal is a more active form of processing. It involves meaningful analysis and think-
ing about the material, and represents a deeper level of processing. Elaborative rehearsal
of a phone number might be looking for a pattern in the numbers or noticing the similarity
to a familiar number (a date, your ID number, or pin number). Many learning strategies are
instances of elaborative rehearsal: using mnemonic systems, forming mental images,
and relating the to-be-recalled material to existing knowledge. Elaborative rehearsal nor-
mally leads to better long-term retention than does maintenance rehearsal.
According to depth of processing theory, remembering is due more to the quality of
processing than a deliberate intention to learn. Craik and Tulving (1975) demonstrated this
in experiments on incidental learning, in which college student participants were shown
a list of words but without instructions to remember them. Instead, the students answered
a question about each presented word, with different types of questions used to promote
different levels of processing. In one condition, the question asked whether the word
contained the letter e (this requires attention to the surface form or appearance of the
word); in a second condition, a rhyme for the word had to be generated (which requires
attending to the sound of the word); in the final condition, the participant could be asked
whether the word fits a given category, such as “animal” (which requires processing the
meaning of the word). Later, the subjects were given a surprise test of memory for the
words. Words that were processed deeper during presentation were better recalled.
“Animal” judgments led to better retention than “e” judgments. Interestingly, the deep-
est processing condition led to memory as good as that found in an intentional learning
condition, one in which participants were explicitly instructed to remember.
Neuropsychologists have considered whether amnesia is due to deficient process-
ing. Individuals with forms of amnesia like H. M.’s, who have trouble remembering new
information, were tested with the incidental learning procedure of Craik and Tulving
(1975). Like normal-memoried people, amnesics had better retention after categorizing a
word than after rhyming it (Graf et al., 1984). However, the amnesics are still severely
impaired. Getting them to deeply process material does not eliminate their amnesia.
Schacter (1983) reports some observations on his Alzheimer’s golfing partner. While
188 Human Memory Conceptual Approaches

playing a particularly challenging hole, the man was asked to elaborate on his actions: why
he chose the club he did, why this shot was difficult, how it turned out. These elaborations
certainly exemplify deep processing. Still, after moving to the next tee, the man could not
recall any details of the just-elaborated hole.
One criticism of depth-of-processing is that that the term “depth” is unclear, taking
on different meanings (e.g., Baddeley, 1978). Depth-of-processing at first clearly empha-
sized the encoding stage: the elaboration of an item in memory during learning. Later
researchers said that deep processing leaves a more distinctive representation in mem-
ory, thereby facilitating retrieval. We now acknowledge that deep processing can produce
elaboration or distinctiveness, and that both are beneficial to memory. (Encoding and
retrieval are discussed more in Chapters 9 and 10).
The notion of qualitative variations in processing at encoding has now been incorpo-
rated into other approaches to memory. Thus, dual-store theorists distinguish between
the two forms of rehearsal in STM—maintenance and elaborative. Elaborative processing
is accepted as a significant factor that enhances the encoding stage of memory, and
distinctiveness facilitates retrieval.

Summary
The processing approach, exemplified by depth of processing theory, emphasizes the
cognitive processing of information. Remembering and forgetting depend on the nature
of the processing an item receives rather than where it is stored or which stage is involved.
However, the processing approach does not preclude the other approaches presented in
this chapter. For instance, we can combine two approaches and study encoding, storage,
and retrieval in short-term memory. Or, one could study processing depth in episodic
memory, semantic, or procedural memory. The several approaches could be viewed as
complementary rather than as exclusionary.

ARTIFICIAL NEURAL NETWORKS

Research on learning proceeds on several levels of analysis. On the behavioral level,
researchers study the relationships between environmental and behavioral variables (i.e.,
stimuli and responses). On a neural level, researchers study the changes in neurons and
their synapses that make up memory. Work in computer science and artificial intelligence
unites these levels by modeling the behavioral and neural changes that underlie learning
and memory. The abbreviation CNS, which usually stands for the central nervous system,
is reinterpreted as the conceptual nervous system (Hebb, 1955).
Neural networks are computer programmed simulations of memory in networks of
hypothetical neurons. (This approach was previously called connectionism, or connection-
ist modeling). Researchers attempt to model changes in the strengths of connections
from input to output: from the sensory neurons to the neurons that produce a response.
Information is combined and selected as it passes through several levels of processing.
An analogy might be face recognition programs, in which the first level analysis is of light-
and dark-pixels, with subsequent levels identifying facial features and eventually a spe-
cific face.
 Artificial Neural Networks 189

The statistics behind the models are basically formulas for combining correlations of
many input variables to predict a certain outcome variable. For example, to predict suc-
cess in college, an admissions committee could use input variables such as high school
grades, admission test scores, and number of advanced courses. Each of these individu-
ally correlate with college success. The formula can be adjusted to give more weight to
certain variables. Some combination of these variables produces the optimal prediction of
the expected college GPA.
Earlier psychological theories offered simple models of connections in the brain. For
instance, classical conditioning occurred when a tone stimulus activated auditory neu-
rons at the same time that the movement neurons activated a leg flexion. Contemporary
neural network theories offer more sophisticated elaborations. The connectionist approach
makes certain assumptions. These include the following:
1 Each neural unit, or hypothetical neuron, is connected to many other units. This mim-
ics actual nerve cells which are multiply connected.
2 Connections strengthen when neural units are active simultaneously and weaken
when activation of one unit occurs without activation of the other. Across a series of
learning trials, the increase or decrease in strengths is described mathematically by
a learning algorithm, or formula. This formula is often the familiar learning curve.
3 Activation of a neural unit may require stimulation from multiple input units. This is
the concept of a threshold for activation. Several input units need to be active to
report the detection of a stimulus; several output units need to be active to initiate a
response.
4 Once activated, the activity of a neural unit gradually fades back to baseline or resting
level over a brief period of time.
5 Network models include multiple layers of neural units. The outer layers correspond
to input and output. Inner layers, called hidden units, combine activation from units in
the previous layer.
6 Neural units can be combined in virtually all possible combinations, but only certain
combinations lead to an accurate prediction. Across simulated learning trials, the
output from neural units are weighted to give more validity to those that lead to
accurate predictions.
Vision is an example of an actual neural layering of information. The rods and cones in the
retina are the input layer, and they respond to light or dark. The nerve impulses from these
cells project to a second layer of ganglion cells. These cells combine the results of thou-
sands of retinal cells to detect contrast, edges, and movement. These ganglion cells are
like the hidden units in a neural network, combining the results of a prior layer to produce
a different form of information (e.g., the detection of objects or features). Still higher lev-
els code the detection of objects, spatial organization, etc. (Lettvin, Maturana, McCulloch,
& Pitts, 1959).

Modeling Configural Learning

Configural conditioning is a particular form of classical conditioning in which the CS is a
unique combination of individual CSs that signal the US. Thus, a compound CS made up
190 Human Memory Conceptual Approaches

of tone + light is paired with the US (represented here as TL+). The tone and the light each
also occur singly but are not paired with the US (represented as T- and L-). Configurations
are difficult to learn because the elements overlap with the compound. The world is made
up of configurations, in which two (or more) elements are required for an action, or to
prevent an action. Configurations are difficult to model in simple networks, as are some
other relationships, such as if-then and either-or. The hidden layers in a neural network can
accommodate more complex relationships.
Figure 7.7 shows a simple network of neural units for configural conditioning of the
form TL+ in the upper panel, and the separate elements T- and L- in the lower panel.
Reading from left-to-right, first is the input layer. This represents sensory neurons that are
stimulated by the conditioned stimuli (tone, light, or the tone and light compound). The
right-most layer represents the output. These are neurons that represent the predicted
outcome (the US, e.g., food).
The interesting part of the model is the hidden layer. This contains a neural unit that
responds specifically to the T+L compound. The compound is not just the sum of the
elements. There is something distinctive about the two stimuli together making the com-
pound different from the elements and can be represented separately in the brain. As an
analogy, musical notes played individually are different than the notes played as a chord.
The C chord is made up of notes C, E, and G.
In this configural learning experiment, there will be a series of conditioning trials. On
tone+light trials, the compound is paired with the US. On tone alone and light alone trials,

Input Hidden Output

a) Layer Layer Layer

T
T

TL US
L
L

Input Hidden Output

b) Layer Layer Layer

T
T

TL
L
L

Figure 7.7
Neural Network Model of Configural Conditioning.
 Applications 191

there is no US. The strength of the connection between the TL unit and the US increases,
following a learning curve, while the strength of the connections between the T unit and
the L unit and the US are not strengthened.
At the end of each conditioning trial, the model compares the predicted outcome of
the network model with the actual outcome. If the prediction is incorrect, it can be altered
by assigning different weights to the connections. The weight assigned to TL in the for-
mula is increased, and the weights assigned to T and L are reduced. Trial-by-trial the for-
mula will come closer to predicting the actual outcomes, the US on T+L trials, and no US
on T alone and L alone trials.
This model is based on actual conditioning experiments with these stimulus configu-
rations. Neural networks can also simulate findings from the nerve activity in the brain
when learning occurs. Alternative versions of the structure of the network and the formu-
las of its algorithms can be tested by comparing the statistical simulations to actual neural
activity, or to behavioral conditioning experiments. For instance, a different model might
include inhibitory connections from T and L to the TL unit (not present in this model).
What other sorts of knowledge can be represented by connectionist networks?
Neural models have been proposed for semantic memory, spatial learning in the hip-
pocampus, and instrumental response–outcome contingencies.

Connectionism and the Other Approaches

How does connectionism relate to the other approaches described in this chapter? Items
in short- term memory are, in connectionist models, those items that are currently active
in the neural network. Asking you to briefly retain the words TABLE–DOG–GRASS
results in the activation of their neural representations above threshold. These units are
now “in” short-term memory. Activation will gradually decay back to resting or baseline
levels, corresponding to forgetting from short-term memory.
Elaborative rehearsal in short-term memory could be described as the activation of
other units concurrently with those already active. The name JOHN can be elaborated by
activation of connected units, such as the image of John’s face, other knowledge about
John, and so on. In the stage model, encoding is the strengthening of connections
between nodes, whereas retrieval is the activation of the retrieval cue’s neural units,
which in turn activate the units connected to them.

APPLICATIONS
The approaches presented in this chapter are part of the general knowledge about mem-
ory. Popular media descriptions of memory regularly refer to short-term and long-term
memory, or to retrieval failure. The finer details of the several approaches presented here
have more precise implications for memory outside the laboratory.

The Study of Abnormal Memory: Amnesia

Cognitive neuropsychology has contributed substantially to testing theories developed in
the lab. As already noted in this chapter, case studies of brain-injured individuals are used
to dissociate components, stages, or processes of memory. The influence works in the
192 Human Memory Conceptual Approaches

opposite direction also. The understanding of amnesic disorders has benefited from the
laboratory-based theories of memory.
We all have some familiarity with the concept of amnesia. Yet most amnesias do not
match the stereotype depicted in films and popular media. Amnesia is actually too broad
a term, as it can originate from several causes and can take different forms. To appreciate
the give and take between the cognitive psychologists and the neuropsychologists, an
understanding of the nature of amnesia is a useful starting point.

Classification of Amnesias
The amnesias can differ along two major dimensions: one of cause (organic or psycholog-
ical) and the other time (loss of memory about events that preceded or followed trauma;
Kopelman, 1987). Organic amnesia is caused by injury or damage to the brain. This might
be head trauma; a brain infection such as encephalitis or meningitis; stroke or anoxia (the
loss of oxygen); exposure to toxic substances and chemicals, such as lead or solvents; or
prolonged alcohol abuse. The resulting amnesia can vary in severity depending on which
areas of the brain are damaged and the extent of that damage.
Psychogenic amnesia is a result of psychological trauma. Someone experiences a
psychological trauma but then becomes amnesic for the event. Psychogenic amnesias
are the stuff of fiction, yet they are real enough, although the cause is not as simple as
popular media suggests.
The second dimension of amnesia is the dichotomy between forgetting the past ver-
sus an inability to form new memories. Retrograde amnesia is forgetting events that
occurred before the onset of the disorder. This is our everyday conception of amnesia:
you get hit on the head and you forget the past. Retrograde forgetting may be limited to
the few minutes or hours preceding trauma, or it may extend back months or years. For
example, after a head injury, retrograde amnesia can be extensive, but the time span of
forgotten material gradually shrinks, leaving a residual permanent amnesia that is usually
fairly limited. In other disorders, such as Alzheimer’s disease, the opposite pattern occurs.
Retrograde amnesia increases over time as more and more of the past becomes
unavailable.
Forgetting can proceed in the opposite direction, from the moment of trauma for-
ward. Anterograde amnesia is the inability to form new memories or acquire new
knowledge. Anterograde amnesia is less familiar to many of us, but it is a common and
serious outcome of brain trauma. The anterograde amnesic individual cannot report the
events of the previous hours or days because those memories have not been formed or
are not accessible. The inability to keep a running memory of our lives can often be a
more serious problem than that of retrograde amnesia.
These two classes of amnesia can be thought of as forgetting that proceeds in differ-
ent temporal directions from the trauma that provokes forgetting:

Retrograde amnesia  TRAUMA  Anterograde amnesia

At the time of memory testing, amnesics cannot remember some portions of the
past (see Figure 7.8). What they have forgotten from before the onset of the amnesia is
retrograde amnesia. What they cannot remember following the onset of amnesia is anter-
ograde amnesia.
 Applications 193

Retrograde Amnesia Anterograde Memory

Amnesia Onset Amnesia Testing

Figure 7.8
The Temporal Division of Retrograde and Anterograde Amnesia.

Retrograde and anterograde amnesias can occur separately, although commonly

both are present. Tulving’s motorcycle accident patient has episodic memory amnesia
that is both retrograde and anterograde amnesia. He cannot recall episodic memories
from before the accident and he cannot acquire new episodic memories since the time
of the accident (Rosenbaum et al., 2005).

Types of Amnesias
There are several types of amnesic disorders, usually distinguished by the patterns of
memory losses and originating factors.

Amnesic Syndrome
The primary deficit of amnesic syndrome is anterograde amnesia: the inability to form
and retrieve new long-term memories. Amnesic syndrome individuals have difficulty
acquiring new general knowledge, or semantic memory; and in forming new episodic
memories. Amnesic syndrome individuals seem to have forgotten their past, but that is
because at some time in the past they stopped forming new memories.
The case study of H. M. (the brain injured patient referred to several times in this
chapter) well illustrates some of the dimensions of amnesia, is the prototype of amnesic
syndrome. H. M. could not recall the last 55 years of his life due to the anterograde amne-
sia which began with his operation in 1953. His life as a career neuroscience subject is
described in Box 7.1.

BOX 7.1 THE CASE OF H. M.

The study of memory has significantly advanced through case studies of people with unique
brain injuries. H. M. is one of the best-known cases in neuropsychology, and his initials are
arguably better known than are the names of the 100 or so researchers who studied him over
the years. The initial description of his case by Scoville and Milner (1957) is one of the most
cited papers in the scientific literature on the brain and memory.
Henry Molaison, at the age of 28, had portions of the insides of the temporal lobes
removed as a treatment of last resort for epilepsy. (The names of living case-study partici-
pants are usually not disclosed. After H. M. died in 2008, his real name was made public).
Remarkably, little about H. M. changed. His epilepsy was controlled; his IQ went up a bit; his
personality was unchanged. But over time, his doctors realized that his post-surgery memory
problems had become permanent. From the time of the operation in 1953 until his death in
2008, H. M. had been unable to acquire new episodic and semantic memories. Some details
of everyday living and a few major public events sifted into his memory. For instance, he
194 Human Memory Conceptual Approaches

could identify presidents Kennedy and Reagan, who came to fame after H. M. became amne-
sic. In the years following the deaths of his parents, he knew each had died but did not
remember the details. He did not learn new vocabulary (semantic memory), even with
explicit practice, and so did not know phrases like rock ‘n’ roll (never mind hip-hop). His defi-
cits included both verbal and spatial anterograde amnesias, reflecting the hippocampal
lesions in the left and right sides of his brain. However, he could draw the floor plan of the
small house he lived in for many years after the operation. His short-term memory was
spared; he could learn through classical conditioning, and he acquired new motor skills (which
together define procedural learning).
H. M. was not able to live on his own, which is characteristic of other severe amnesics.
He could wander off and not remember how to get home. Similarly, he did not remember he
had just eaten a meal and would eat another if offered. If he turned on the stove or another
appliance, he forgot within a few minutes. He once went out to rake leaves and persisted for
hours, not having a sense of how long he had been at it.
H. M. was a frequent research subject. Each new generation of neurological test or psy-
chological theory (CAT scans, implicit and procedural learning) was tried out on him. Many of
the studies were conducted at the Massachusetts Institute of Technology. He had been there
often, for days at a time. Yet if he awakened at night, he had no idea where he was. Brenda
Milner (of Scoville & Milner, 1957) studied H. M. for 20 years, but he did not recognize her.
Suzanne Corkin (2013) studied him for the next 20 years, and he never recognized her either.
If H. M. could not update his knowledge, was he constantly startled by the new world he
saw? What did he think when he looked in the mirror? When asked his age, he reported
some number in the 30s when he was then actually 60 (Corkin, 2002). H. M. showed no
reaction to seeing this now quite elderly man looking back at him from the mirror. (Remember,
the rest of his knowledge basically stopped at age 28). The absence of surprise here may
have been because there was some residual hippocampal tissue left after the operation, or
perhaps the constant repetition supports the learning of familiarity. By contrast, neurologist
Oliver Sacks (1985) reported a Korsakoff’s patient who panicked at the sight of his 70-year-old
face in the mirror because the man thought he was still only 21.
Since the 1950s the devastating effects of temporal lobe resection on memory have been
well documented. H. M. was one of the last individuals to undergo this procedure. Presurgical
testing today maps out areas of the temporal lobe to spare if at all possible.

In individuals with amnesic syndrome forgetting before the trauma, or retrograde

amnesia, is variable in extent. H. M.’s recollection from before his operation was sketchy
and vague. In the research literature, there are reports of wide-spread retrograde amne-
sia, but also cases of minimal forgetting from just-before the trauma.
Amnesic syndrome is also characterized by what is spared. Short-term memory is
still reasonably normal. Procedural learning is spared. Individuals with amnesic syndrome
learn perceptual, motor, and cognitive skills, although they have no recollection of the
procedural learning experience. Amnesic syndrome individuals acquire classically condi-
tioned responses, but more slowly than non-amnesic individuals (Daum, Channon, &
Canavan, 1989). Finally, they show priming effects on word-completion and picture-
fragment identification tasks.
The hippocampus is usually the damaged brain region in amnesic syndrome. Injury to
just one side of the brain impairs verbal memory (left temporal lobe) or faces and places
 Applications 195

(right temporal lobe). Brain trauma or disease rarely damages a single region of the brain, so
amnesic syndrome individuals may have damage to other areas and additional symptoms.

Korsakoff’s Disease
Korsakoff’s syndrome is named after the Russian psychiatrist who described it. This
disease was initially associated with prolonged alcohol abuse (literally decades long), but
the underlying cause is believed to be thiamine vitamin deficiency. A lack of thiamine
occurs with inadequate diet, whether in alcoholics or in those with nutritional deficien-
cies. With Korsakoff’s, there is damage to the thalamus and mammillary bodies, areas in
the vicinity of the hippocampus but not necessarily the hippocampus itself. The disease
develops gradually over time, so it is difficult to pinpoint the exact onset.
Korsakoff’s is characterized by both anterograde and retrograde amnesia. Tests of
general knowledge, such as tests of famous names or public news events from the past,
show profound amnesia that extends back over decades. Some of this memory loss is
retrograde (forgetting of information that was once known) and some is anterograde
(when the formation of new memories ceased). The neurologist Oliver Sacks described
one 49-year-old man who believed the current year was 1945 (it was 1971). The man was
startled and frightened when shown a picture of the Earth taken from space. How could
that be, in 1945? Another patient was repeatedly distressed seeing the changes in his
home town; what he perceived as buildings going up or coming down overnight actually
happened over a 20-year period.
Implicit memory in the form of priming effects are mostly spared. Thus, Korsakoff’s
individuals perform nearly as well as matched-control subjects on word completion. For
example, they complete three letter stems (e.g., MOT_ _) with words shown earlier; on
an explicit test of memory they will not recognize the word MOTEL as having been shown
earlier.

Electroconvulsive Shock
Electroconvulsive shock therapy (ECT) is used as a treatment for severe depression.
The usual course of treatment is several shocks administered over a few weeks. Patients
are anesthetized during the treatment sessions. The shocks produce seizures in the brain,
although the mechanism for alleviating depression is unknown.
Memory loss is the side effect of ECT that most distresses patients. People report
retrograde amnesia for events that preceded ECT that extends back for months, and
sometimes isolated events from years before are not recalled. The amnesia is not as
profound as that of Korsakoff’s or amnesic syndrome.
Retrograde amnesia for real-life memories was assessed in a seminal study by Janis
and Astrachan (1951). During a pre-ECT interview, the patients were questioned in detail
about their life histories: schools attended, jobs held, places lived, and so on. The purpose
of the pretest was to determine what was recallable before treatment. A second inter-
view occurred one month after the completion of ECT. Every one of the subjects forgot
some facts that they could recall before ECT. For example, one patient had first reported
being unemployed for several months prior to treatment. Afterwards, he was unable to
recall this, and claimed to have worked right up until the time of hospitalization. Some of
the patients forgot events from many years prior, things that certainly had been in
196 Human Memory Conceptual Approaches

long-term memory. Janis and Astrachan suggested that ECT affected retrieval, reducing
access to certain memories.
Squire, Slater, and Miller (1981) tested forgetting of naturalistically acquired knowl-
edge. Their subjects were questioned about public events and TV shows that had aired for
a single season. Forgetting these facts would indicate the precise time periods encom-
passed by the amnesia. In the first week following ECT, the patients had retrograde
amnesia that extended back one to three years. However, a second test given seven
months after ECT showed that much of what had been forgotten on the first post-ECT
test could then be remembered.
If ECT produces retrograde amnesia, would other types of brain seizures also cause
forgetting? The study of epilepsy is relevant here. Milton et al. (2010) found that individuals
with temporal lobe epilepsy did show especially poor memory for their own past (episodic
memory) and some impairment in memory for public events (semantic memory).

Functional Amnesia
Our everyday conception of amnesia is that a psychologically traumatic experience can
cause repression of memory for the event. The term psychogenic amnesia suggests
that amnesia results from psychological trauma or mental stress. The term psychogenic
is being replaced by the more neutral label functional amnesia. “Functional” is a medical
term for disorders of unknown causes, and so here refers to memory that is malfunction-
ing, but the cause or reason is not known.
The defining feature of functional amnesia is retrograde loss of episodic and auto-
biographical memory. Forgetting can be limited to a specific past event or period of
time, as when a person cannot remember a life-threatening experience. Or forgetting
can be more generalized, encompassing all memories over an extended period of
time. In an early report, the French psychiatrist Pierre Janet recounted the case of a
woman who was told her husband had just died. The woman became hysterical. The
report turned out to be a malicious prank, but afterward the wife became amnesic for
this whole episode. Interestingly, memory of the event would recur in nightmares
(Nemiah, 1979).
Anterograde amnesia, or difficulty remembering new information after the trauma, is not
typically impaired. Functional amnesics perform as well as controls on tests of new learning.
A more dramatic form of retrograde amnesia is forgetting one’s entire past life and
identity. In a fugue reaction, the person literally does not remember who they are. For
example, one young man developed fugue after the death of his grandfather, to whom he
had been particularly close. For a week afterwards, the young man was unaware of who
he was, and could not recollect any personal (episodic) memories (Schacter, Wang,
Tulving, & Freedman, 1982). Language and other sorts of general knowledge were
retained. The young man maintained a continuous memory during the fugue period. For
example, he could remember why he was in the hospital and the doctors who were
treating him. The fugue ended after several days. Old memories were restored but the
fugue period was then forgotten.
The common stereotype for psychological amnesia is that a traumatic experience
causes the onset, memory is repressed to handle distress, and eventually memory
returns. Rarely is the world so simple. The majority of retrograde amnesias have neither
clear psychological nor physical causes. More often there is an organic antecedent (e.g.,
 Applications 197

a minor head injury from a fall or accident), a mild-to-moderately stressful experience

(e.g., a family argument), and some pre-existing mental health problems (e.g., anxiety
and depression, hypochondriasis, or dissociative symptoms). The functional amnesias
may therefore reflect a mix or organic and psychological contributions (Fujiwara et al.,
2008; Mangiulli et al., 2021).
Psychological explanations of functional amnesia assert that traumatic memories are
repressed into the unconscious as a way of relieving anxiety. Memories are not lost from
the brain, but rather the amnesic individual experiences temporary retrieval failure.
However, a large proportion of functional amnesics do not recover their memories, even
after many years (Kritvchevsky, Chang, & Squire; 2004; Mangiulli et al., 2021).

Everyday Forgetting and the Models of Memory

Surveys consistently reveal that people are concerned about everyday forgetting. A news
poll showed that 46 percent of respondents worried more about failing memory than
about failing health. British psychiatrists refer to “mnestic hypochondria,” an excessive
concern about memory problems among middle-aged individuals who nevertheless test
fine on memory (Berrios, Markova, & Girala, 2000). These worried-but-well people are
characterized as bright, educated, ambitious, and perfectionistic. One could suppose they
might become anxious about perceived declining memory abilities.
Accurately measuring everyday forgetting in the real world is difficult. Nevertheless,
certain forms of forgetting are frequently mentioned. The following are items from various
questionnaires devised to assess memory lapses (Broadbent, Cooper, FitzGerald, &
Parkes, 1982; Cheyne, Carriere, & Smilek, 2006). The several approaches to memory pre-
sented in this chapter offer explanations for why forgetting occurs and suggestions for
remediation.
1 How often do you forget to say something you were going to mention?
2 How often do you have the “what-am-I-here-for?” feeling when you have forgotten
what it is you came to do?
3 How often do you forget to do something that you were going to do after being
interrupted?
At first glance, these seems to be failures of short-term memory. The intention to say or
do something was displaced during a brief time interval by ongoing events and so was
not remembered. Alternatively, the stage approach suggests that retrieval failure is the
source of the forgetting here. At the moment of action, nothing reminded you of what you
wanted to say, do, or get. The retrieval failure explanation is confirmed when later your
intention is suddenly remembered.
4 How often do you find yourself searching for something that you’ve just put down?
5 Do you misplace frequently used objects (keys, glasses, phone)?
These could be examples of absentmindedness: Your mind was occupied elsewhere and
so you did not deeply or elaboratively process where you placed the object. The interest-
ing next question is: When you find the lost object, do you then recall having mislaid it? If
you still have no recollection, then maybe you had an encoding failure. If you do recall
putting down the object, then you were earlier experiencing retrieval failure.
198 Human Memory Conceptual Approaches

6 Do you forget to complete a task, such as failing to remove the original after making
a photocopy?
7 How often do you fail to do something you were supposed to do a few minutes later
even though it’s there in front of you?
Post-completion errors are omissions of actions required after the completion of a task’s
main goal. For instance, not replacing the gas cap after filling the car, or leaving a card in
the automated teller machine (ATM) after withdrawing cash. Attaining the main goal of an
action, such as making a photocopy or filling the gas tank, gives a false sense of comple-
tion. One solution is to create a forcing function to compel the user to complete the
potentially omitted last step. Some ATMs now have a forcing function that prevents the
user from proceeding with a transaction before the card is withdrawn.
8 How often do you intend to bring something with you and shortly later leave without
it?
9 How often have you put your Stradivarius down and forgotten to pick it up when you
left?
These items indicate a failure of prospective memory: Remember to do something in the
near future. You leave for school or work without something you had intended to bring.
You exit a classroom leaving something behind. You put something down momentarily to
free up your hands, and then forget to pick it back up again.
Being stressed, tired, or in a hurry are causes for leaving something behind on the
plane, at the security checkpoint, in the Uber vehicle, or on the kitchen counter. Also,
there is an absence of timely and active reminders. One way to prevent such errors is to
pause and ask yourself: Am I forgetting anything?
Would someone really forget their Stradivarius? Among several instances of forget-
ting a Stradivarius violin or cello, two were left in taxis, one on a train, and one on the front
porch. The slips occurred after traveling all day, performing an evening concert, and/or
returning home or to a hotel late in the evening. These are factors associated with absent-
mindedness. One solution available now are tracking tags that connect your valuable
instrument to your phone and can alert you if they become too far apart. As long as you
don’t mislay your phone.
Certain other memory distortions fall between the amnesias and everyday forgetting.
Some of these are described in Box 7.2.

BOX 7.2 ANOMALOUS FORGETTING PHENOMENON

The theoretical approaches that are covered in this chapter can also be applied to some
unusual memory experiences, such as cryptomnesia and déjà vu.
There are times when you think you have come up with an original idea, only to realize
later it was actually suggested by someone else. Cryptomnesia is the lack of awareness that
a person’s thought or idea is actually an inadvertent plagiarism arising from another source.
Freud believed he had an original insight into the origin of the neuroses, only to be reminded
that his friend Wilhelm Fliess had suggested the idea a year earlier (Freud, 1901/1960, p. 143).
Musicians have been sued because a song they wrote resembled an earlier song. B. F. Skinner
 Summary 199

described the phenomenon well: “One of the most disheartening experiences is discovering
a point you have just made—so significant, so beautifully expressed—was made by you in
something you published a long time ago” (Hostetler, 1988).
Cryptomnesia may be an example of source amnesia. You know something, but without
the where and the when of how you came to know it (Brown & Murphy, 1989).
Déjà vu, French for “never seen,” is the false sense of familiarity of a present experience.
For instance, you feel like you have been here before, but you know you have not. Déjà vu
has the feeling of remembering but without a memory to validate the feeling. In addition to
the factual elements of episodic memories (the when and where; the sights and sounds),
memories also have an emotional component. There is a feeling that accompanies the recog-
nition of familiarity.
A behavioral theorist might attribute déjà vu to generalization: There is sufficient resem-
blance between the new situation and a previous one to remind you of the old. As an old
memory weakens, it loses resolution and detail and becomes easier to confuse with other
memories. One could make an analogy to a badly faded photograph. Thus, a new experience
may “match” the fuzzy image of an older memory, one that is not actually similar to the
present. (See Brown, 2003, for a review of déjà vu).
Another explanation is that a current event is perceived twice in the present, making the
second one seem to be familiar. Sensory information takes multiple routes through the brain,
much of it unconsciously. These separate processes may become asynchronized, that is, one
perception is ahead of another. When the second perception reaches awareness, there is a
feeling of familiarity, due to the earlier first perception. The memory distortion is not that you
had this experience sometime in a distant past; but you experienced it only milliseconds ago
(Brown, 2003).
A related idea is that a first unknowing glance at a situation is not consciously perceived,
but the subsequent focused attention produces a feeling of having seen this before. Again,
the prior experience is from moments earlier, not years before.
On the other hand, in misidentification disorders, the emotional feeling of familiarity is
absent. An individual recognizes familiar people but does not experience the emotion that
should accompany recognition of someone you know. You can identify the face. In the
absence of any feeling towards the familiar person, the individual with misidentification dis-
order assumes these other people must be imposters (Breen et al., 2000). Curiously, there
was a report of a man who did not recognize himself in the mirror and assumed there was
someone else there, or another person besides himself. Misidentification disorders probably
involve dementia or other neurological problems, and not just memory failures.
One reason for considering these anomalous memory experiences is to show they may
not be so odd after all. Although multiple interpretations have been offered for several of
these phenomena, the explanations all derive from principles of learning and memory. The
usefulness of a science of memory is not just to explain basic research findings, but also to
explain the unusual in everyday life.

SUMMARY
Partitioning Memory
Memory may be better understood when partitioned into separate components, stages,
or processes than when treated as a unitary trait. Multiple memory systems are
200 Human Memory Conceptual Approaches

hypothesized because generalizations about memory as a whole are not always valid; and
because there is heuristic value in simplifying theories of memory.
Dissociations are demonstrated when a variable has different effects on different
tasks, which suggests there are separate memory systems. Dissociations may be pro-
duced by the use of experimental manipulations, different memory tasks, neurological
disorders, or varied subject populations.

Components of Memory Approach

The dual-storage conception of short-term and long-term memory is referred to as the
modal model. Short-term memory has a limited duration and capacity, stores items tem-
porarily and is subject to disruption. Long-term memory retains indefinitely, has virtually
unlimited capacity, and retains information in many forms.
The distinction between two memory stores is suggested by the serial-position
effect. Different experimental variables affect the primacy and recency portions of the
curve. Short-term and long-term memories are also dissociated by the patterns of mem-
ory loss in amnesic individuals. H. M. cannot form new long-term memories, but his
short-term span is normal. K. F. is impaired in short-term verbal span.
Long-term memory can be further divided into episodic and semantic memories.
Episodic memory is our personal or autobiographical memory system. Semantic mem-
ory is our store of general knowledge. It is more like dictionary or encyclopedic knowl-
edge. Although not always separable in laboratory experiments, the distinction is useful
in describing the development of memory from episodic to semantic or generic
memories.
Procedural learning and priming are two other hypothesized divisions of long-term
memory. Procedural learning is knowing how to do things: perceptual, motor, and
cognitive skills. Procedural tasks include mirror star tracing and reading reversed or
inverted text. Priming refers to the facilitated response to a stimulus that has been
recently experienced, or has been “primed” in memory. For example, the identifica-
tion of a word from a few letters, or fragments, is facilitated by recent exposure to that
word.
Explicit and implicit memory are terms used to refer to how we test memory. Direct
questioning about the contents of memory is an explicit task, as in recall and recognition
tests of episodic and semantic memory. Indirect assessment of the effects of prior expe-
rience by measuring performance rather than recall is an implicit task. Procedural learning
and priming effects are both examples of implicit memory, although explicit recall may
influence performance on these types of tasks.

Stages of Memory
Another approach to memory separates it into several stages: encoding, storage, and
retrieval. The stages are experimentally dissociated when a variable primarily affects one
stage or another. For example, a physical stressor primarily impairs retrieval rather than
encoding.
Neuropsychological dissociations are demonstrated by individuals with impairment at
one or another stage. H. M. is believed to have an encoding deficit: He cannot form new
 Summary 201

memories. However, other amnesics misrecall words from supposedly forgotten lists,
which suggests that they have trouble directing retrieval.

Processing Approaches
Remembering and forgetting can be attributed to the type of processing information
received. Depth-of-processing theory hypothesizes only a single memory store, and vari-
ations in the degree of cognitive processing determines whether something is remem-
bered. Rapid forgetting is due to shallow processing, whereas sustained retention is due
to deep processing. The distinction between shallow and deep processing is illustrated by
two kinds of rehearsal. Maintenance rehearsal, or the passive repetition of information, is
shallow processing. Elaborative rehearsal involves meaningful analysis and comprehen-
sion of the material, and represents a deeper level of processing. According to depth-of-
processing theory, robust incidental learning occurs due to the cognitive involvement and
interest in the target material.

Artificial Neural Network Models

Connectionism combines cognitive, neuroscience, and computer modeling approaches
to simulate neural changes that underlie learning and memory. Each neural unit in a model
system has connections to many other units. A neural unit can be activated much as a
neuron is activated, and activation gradually fades back to the baseline or resting level
over a brief period of time. The strength of connections increases when neural units are
simultaneously active. Processing goes through several layers of neural units, which com-
bine and summate activation from units in the previous layer.
The goal of a network model is to use the input (sensory stimulation) to predict the
outcome (consequence or response). Trial-by-trial changes in connection strength are
adjusted to produce closer fits to the actual outcomes. A successful simulation produces
a classic learning curve characterized by relatively large increases in learning in the early
trials and smaller increments in later trials.

Applications
Amnesia is not one thing, but several types of forgetting. The amnesias can be described
along a temporal dimension: loss of memory of events that preceded the onset, or failure
to acquire new memories that followed onset. Most amnesias have a physical or organic
cause due to injury or damage to the brain. Retrograde amnesia is forgetting of events
that occurred before the onset of the disorder. Anterograde amnesia is the inability to
form new memories or acquire new knowledge.
Amnesic syndrome, caused by damage to the hippocampus, is characterized by the
inability to form and retrieve new long-term memories (anterograde amnesia). There may
also be retrograde amnesia, and short-term memory is usually normal. Amnesic syn-
drome individuals, like H. M., are impaired in acquiring semantic and episodic memories.
However, new learning occurs via procedural learning.
Korsakoff’s disease is an amnestic disorder associated with prolonged alcohol abuse
and vitamin deficiency. It is characterized by both anterograde and retrograde amnesia.
202 Human Memory Conceptual Approaches

Tests of general knowledge may show profound amnesia that extends backwards over
decades. Damage to the thalamus in the brain is usually present.
Electroconvulsive shock therapy (ECT) produces some retrograde amnesia for events
that preceded ECT. Forgetting extends back for months, and sometimes isolated events
from years before are not recalled. Testing again months later shows that much of what
had been forgotten on the first post-ECT test could then be remembered.
Functional amnesias, previously labeled psychogenic amnesias, are retrograde: for-
getting the past or some part of the past that preceded the trauma. Functional amnesia
may be limited to a period of time in the past. A fugue reaction would be complete forget-
ting of one’s past life and/or identity. Psychological explanations say that memory is not
lost, but it becomes inaccessible to conscious retrieval. The lost memories sometimes
spontaneously return. However, some functional amnesias seem to be permanent.
Many instances of everyday forgetting can be described within the approaches cov-
ered in this chapter. Understanding where a breakdown occurs offers a better chance for
prevention and remediation. Examples of forgetting from an inventory of memory lapses
include momentary forgetting, prospective forgetting, absentmindedness, and temporary
forgetting of common words and names.
 CHAPTER

8
Short-Term Memory

CONTENTS

Some History 204 Theoretical Overview of Working

Short-Term Memory Tasks 205 Memory 215
The Brown–Peterson Distractor Task 205 Measuring Working Memory 218
Memory Span 207 Executive Functions 219
Characteristics of Verbal Short-Term Working Memory and
Memory 208 Consciousness? 220
Acoustic Encoding 208 Working Memory and Culture 220
Limited Capacity 209 Individual Differences in Working
Limited Duration 210 Memory 221
Forgetting: Short-Term Memory Is Aging and Working Memory 221
Sensitive to Disruption 210 Dementia and Working Memory 221
Transfer to Long-Term Memory 210 Anxiety and Working Memory 222
Control Processes 211 Multitasking 222
Summary of the Features of Is There Really a Separate Short-Term
Short-Term Memory 211 Memory? 225
Other Modalities of Short-Term A Single-Memory Approach? 225
Memory 211 Neuropsychological Dissociations of
Visual Short-Term Memory 212 Two-Memory Systems 226
Spatial Short-Term Memory 212 One Memory or Several? 227
Short-Term Memory for Actions 212 Applications 227
Short-Term Memory for Odors 213 Comprehending and Using
Short-Term Memory in the Hearing Language 228
Impaired 213 Problem Solving 228
Working Memory 215 Summary 229

“When deeply absorbed, we do not hear the clock strike. But our attention may awake
after the striking has ceased, and we may then count off the strokes” (Exner, quoted by
William James, 1890, p. 646). This quote captures the subjective feel of short-term mem-
ory. William James was making a distinction between two kinds of memories. Something
in primary memory has never left consciousness and is part of the psychological present.
Something in secondary memory has been absent from consciousness and therefore
belongs to the psychological past.

DOI: 10.4324/9781003227090-8
204 Short-Term Memory

The idea that there are two memories has long been a part of psychology’s history,
both in cognitive and biological theory. In addition to William James, the German psy-
chologist G. E. Muller hypothesized a transient “perseveratory activity” in the brain that
could eventually consolidate into a permanent long-term memory (see Lechner, Squire, &
Byrne, 1999).
In the previous chapter we discussed the amnesias, which for the most part involve
forgetting from long-term memory. In our everyday lives, forgetting over short time inter-
vals is also troubling. Absentmindedness, a common label for everyday lapses, occurs
when people forget what they were doing, were about to do, had intended to do, and so
on. You go upstairs to get something but become distracted. You wonder what you were
looking for, wander around looking for something that needs getting, and likely return
empty-handed. The fretting about age-related memory decline is often in reference to
something forgotten from just a few minutes earlier.
This chapter is about remembering over brief intervals of time. Short-term memory
(STM) refers to memory that is limited both in its duration and its capacity. As assessed
in laboratory tests, STM retains on the order of five to seven items, for several seconds
to less than a minute in the absence of rehearsal. Short-term memory is distinguished
from long-term memory (LTM), which in the laboratory is typically any memory more than
a few minutes old, and in the real world, the memories of a lifetime.
Remembering over short intervals has important implications for everyday cognitive
functioning. “Without it, you couldn’t understand this sentence, add up a restaurant tab
in your head, or even find your way home” (Wickelgren, 1997, p. 1580.) (Now that I reread
this sentence, I realize you’ll probably use your smartphone to do these tasks). Short-
term memory, and its newer variation of working memory, have been referred to as our
mental blackboard: a temporary, reusable workspace in the mind used for comprehen-
sion, reasoning, and planning.

SOME HISTORY
The concept of a primary memory lay dormant in psychology for many years before
Waugh and Norman (1965) noted that many learning tasks drew on both short-term and
long-term memory. For instance, the primacy and recency effects in free recall were
explained by the two memories. This was followed by the development of a multistore
system of memory by Atkinson and Shiffrin (1968). As noted in the previous chapter, it
became the “modal” model of memory, or the schematic that serves as the outline for
other theories of memory (see Figure 8.1).
Atkinson and Shiffrin’s hypothesis that there are separate short-term and long-term
memory stores provoked markedly different reactions from other psychologists. A few
argued that postulating two kinds of memory was unparsimonious (i.e., complicated). The
other reaction was that two memory stores was not enough! Baddeley and Hitch (1974)
proposed the evolution of a single short-term memory into working memory, which has
separate stores for verbal information and spatial information.
Working memory is the focus of much current research, and the term “working
memory” is used much more frequently in the current literature. However, working mem-
ory and short-term memory are sometimes used interchangeably. What follows in this
 Short-Term Memory Tasks 205

Short-Term Store
(STS)
Temporary Long-Term Store
Working Memory (LTS)
Environmental
Input Control Processes: Permanent
Memory
Rehearsal Store
Coding
Decision
Retrieval Strategies

Response Output

Figure 8.1
The Modal Model of Short-Term Store and Long-Term Store.
Source: From “The Control of Short-Term Memory,” by R. C. Atkinson and R. M. Shiffrin, 1971, Scientific American, pp. 152–161.
Copyright © 1971 by Allen Beechel. Reprinted with permission.

chapter is first a consideration of the characteristics of short-term memory, second of

working memory, and finally some applications of each.

SHORT-TERM MEMORY TASKS

Short-term retention is often assessed by one of two tests. The distractor task attempts
to quantify the duration of immediate memory over brief delay intervals when rehearsal
is prevented. The memory span attempts to quantify the capacity of immediate
memory.

The Brown–Peterson Distractor Task

The distractor task was described by British researcher John Brown in 1958, and
Americans Lloyd and Margaret Peterson in 1959 (but see Pillsbury & Sylvester, 1940). In
the Brown–Peterson task, a few items are presented for retention, usually three letters
or words. These items can be recalled accurately if tested immediately. However, the
subject is instructed to perform a distractor task, such as counting backwards by threes,
until told to recall the target items. Thus, the sequence of events in a trial might be a
visual presentation of the to-be-remembered items (e.g., DOG–CAT–TREE), a number
(e.g., 933), counting backwards (“933, 930, 927 …”), and finally the instruction to recall
the target items. Representative data are shown in Figure 8.2 for the retention of three
consonants or three words. The amount recalled drops off dramatically after as little as 9
seconds of distractor activity.
Rapid forgetting with the distractor technique seemed to demonstrate how brief STM
was in the absence of rehearsal. Memory for the items faded quickly.
206 Short-Term Memory

Murdock 1 Word
100
90
80 Peterson and Peterson

Correct Recall (%)

70 3 Consonants
60
50
Murdock 3 Words
40
30
20 Murdock
3 Consonants
10
0
3 6 9 12 15 18
Retention Interval (sec)

Figure 8.2
Mean Percent Recall of Words or Letters in the Brown–Peterson Distractor Task.
Source: From “Implications of Short-Term Memory for a General Theory of Memory,” by A. W. Melton, 1963, Journal of Verbal
Learning and Verbal Behavior, 2, p. 9. Copyright 1963 by Academic Press.

However, some of the forgetting in the Brown–Peterson task might be due to inter-
ference from earlier trials. Imagine you are participating in one of these experiments. You
are given list after list of three words to remember, for as many as 50 lists. Is it really the
case that each list is forgotten, totally and completely, after 15 seconds? After all, the
assumption is that this is an STM task. Or is it likely that recall of the current list is
affected by misremembering items from the previous lists? This is called proactive inter-
ference: Items presented on earlier trials interfere with recall of the current list of to-be-
remembered items.
Curiously, performance in the Brown–Peterson distractor task is often perfect on the
very first trial: word or consonant triplets are recalled perfectly after delays of 15 seconds.
Forgetting only develops as the number of trials increases, consistent with the notion that
interference is due to previous trials.
Interference from previous lists also increases with the similarity of the to-be-remem-
bered items. If the words across adjacent trials come from the same category (e.g.,
names of fruits) forgetting occurs even more rapidly across trials. Proactive interference
seems to build up. If after several trials the category of target words is shifted, say, to
professions, recall of the changed triplet increases dramatically, a phenomenon labeled
release from proactive interference. This buildup and then release from proactive interfer-
ence is illustrated in the left panel of Figure 8.3 (Wickens, Dalezman, & Eggemeier, 1976).
Recall of word triplets that came from the same semantic category (names of fruits)
declined across the first three trials (the buildup of proactive interference). Changing the
word category on the fourth trial from fruits to professions or flowers led to an increase
in recall (the release from proactive interference) as compared to maintaining the initial
category. Switching from fruits to vegetables did not produce much release.
Another illustration of proactive interference is in memory for television news stories
(Gunter, Berry, & Clifford, 1981). Brief news reports from the same thematic category
(e.g., politics or foreign affairs) were presented as the to-be-remembered items. Recall of
 Short-Term Memory Tasks 207

100 100
Percent Correct Recall

Percent Correct Recall

80 Professions 80
Shift: Foreign Affairs

60 60
Flowers
40 40

Fruits (Control) Domestic

20 20
Vegetables

0 0
1 2 3 4 1 2 3 4
Trial Trial

Figure 8.3
Buildup (across Trials 1 to 3) and Release (Trial 4) of Proactive Interference for Words (left panel) and
News Stories (right panel).
Sources: (left panel) “Some Characteristics of Word Encoding,” by D. D. Wickens, 1973, Memory & Cognition, 1, pp. 485–490;
(right panel) Gunter, Berry, and Clifford, 1981.

story triplets from within a category decreased over trials, consistent with the buildup of
proactive interference. If the category then changed, release occurred and the final triplet
of stories was recalled better (see the right panel of Figure 8.3).

Memory Span
A second measure of STM is its capacity, or how many items can be held in immediate
memory. Memory span is defined as the longest sequence of items that can be remem-
bered in correct order after a single presentation. Recall is immediate, without delay or
distractor. Span is tested by presenting lists of increasing lengths, often five to nine
items, to assess the maximum length than can be recalled correctly. The items are typi-
cally letters, numbers, or words. Our spans seem to be limited to about seven items
(Miller, 1956).
Memory span is relevant to our remembering outside of the laboratory. Is it simply
coincidence that seven digits is the length of a standard telephone number? Conrad
(1958) asked his participants to listen to and then dial eight-digit numbers. Only about 50
percent of the numbers were dialed correctly. When a constant prefix was added (such
as the “9” often needed to get an outside line), recall dropped even further. Memory span
is also included in some intelligence tests. The Wechsler Intelligence scales have a digit
span as one subtest. For more fun, there is the digits-backward portion, in which the
numbers have to be recalled in reverse order.
Memory span seems to be a straightforward measure of short-term memory capac-
ity. However, an individual’s span is not invariant, but is affected by numerous variables.
For example, digits are remembered better than words; and drawings of objects are
remembered better than the names of the objects (Brooks & Watkins, 1990). Span also
increases with practice. In one study, adolescents with severe learning disabilities were
given 10 minutes of daily practice at recalling digit and word strings. After two weeks
there was a significant increase in span length (Hulme & Mackenzie, 1992).
208 Short-Term Memory

An important determinant of memory span is the word-length effect: More items can
be remembered when short words are the to-be-remembered items. Baddeley, Thomson,
and Buchanan (1975) found that college students’ spans were dramatically less for sets
of words that were each five syllables long (such as ASSOCIATION, CONSIDERABLE, or
UNIVERSITY) than for lists of one-syllable words. Five one-syllable words could be
remembered 75 percent of the time, yet barely 30 percent of five five-syllable word
sequences could be recalled.
Baddeley et al. (1975) found that span correlated with articulation rate, which is a
measure of how fast you can pronounce the target words. Longer words take longer to
vocalize than shorter words, and so maybe longer words also take longer to mentally
rehearse. Support for the articulation hypothesis comes from a study of Welsh bilingual
speakers, who could recall more digits in English than in Welsh. English has shorter words
for numbers (Ellis & Hennelly, 1980). Later work extended these findings across a range
of different languages (Chinese, Hebrew, Arabic) and showed a consistent relationship
between memory span and the amount of time it takes to articulate digits in that lan-
guage (Naveh-Benjamin & Ayres, 1986; Stigler, Lee, & Stevenson, 1986).

CHARACTERISTICS OF VERBAL SHORT-TERM MEMORY

Short-term memories seem to have a different character from long-term memories.
Recent memories are rich in sensory quality such as sound, color, and texture. In addition
to these subjective impressions, certain objective features have been attributed to short-
term memory:

Acoustic encoding
Limited capacity
Limited duration
Susceptibility to forgetting
Transfer to long-term memory
Control processes.

As we will see, some of the features may not in fact uniquely discriminate short-term
from long-term memory, thus blurring any ready distinctions between the two hypothe-
sized memory systems.

Acoustic Encoding
In a typical STM task, words, letters, or digits are read or shown to the participant, and
the items are recalled out loud. Not surprisingly, the words are remembered as they
sound, as if they were being verbally rehearsed. Long-term memory, on the other hand,
is usually characterized as involving semantic encoding. That is, we remember the mean-
ing of a word rather than the exact word or its sound. One way of demonstrating this
difference is remembering words that sound similar but that have different meanings
(e.g., MAN, CAN, MAD, CAP, MAP). If the words are encoded by sound, they are confus-
able and errors occur as sound-alike substitutions are remembered in STM tests (Baddeley,
 Characteristics of Verbal Short-Term Memory 209

1966; Conrad, 1964). If the words are encoded by meaning in LTM, they are not confusa-
ble: A MAP is different from a MAN and a CAP. On a delayed test of recall, the errors are
synonyms of the presented words: HAT is remembered instead of CAP.
The verbal–semantic distinction helps explain why we can repeat a just-heard sen-
tence verbatim, but in recalling later, we paraphrase the sentence into somewhat differ-
ent words. Sachs (1967) demonstrated this by having participants listen to short
paragraphs, which were occasionally interrupted to test for memory of recently heard
sentences. If the sentence was one just heard, the subjects could usually identify a ver-
batim copy. If the sentence had occurred two to three sentences back in the passage,
verbatim test sentences were often confused with paraphrased versions that had the
same meaning.
There are exceptions to the acoustic-in-STM versus semantic-in-LTM distinction.
Exact wording is sometimes remembered in long-term memory, as when distinctive
phrasing can be remembered days later (Keenan, McWhinney, & Mayhew, 1977).
The purpose for remembering might determine whether acoustic or semantic encod-
ing is used. When the task is to retain short lists for brief periods of time, subjects may
adopt a simple rehearsal strategy, much like we use to repeat a phone number long
enough to dial it. When the task requires retention of meaningful material over a longer
interval, subjects may use an elaborative rehearsal strategy. Thus, the subject might try
to form associations among list items or relate the items to existing knowledge.

Limited Capacity
Short-term memory has a limited capacity to hold information, in contrast with the virtu-
ally unlimited capacity of long-term memory. On the average people remember seven
items in a memory span test, or “the magical number seven, plus-or-minus two,” the title
of George Miller’s famous article (1956). However, this does not mean that the actual
capacity of STM is seven items. In the immediate recall of a string of words, some of the
items are stored temporarily in long-term memory while additional new items are pro-
cessed in the short-term store (Waugh & Norman, 1965). Short-term capacity needs to be
measured without the long-term component, and in the absence of rehearsal. Under
these conditions, the current corrected estimates of the capacity of STM are on the order
of three to five items, with an average of four (Cowan, 2010). “The magical number four?”
This doesn’t have the same ring to it.
But what exactly is an item? The data from the Brown–Peterson procedure shown in
Figure 8.2 show equivalent retention of three letters or three words, even though the
words contain more letters. So, is an item a letter, a word, an idea? One answer is that an
item is a unit already existing in long-term memory. Letters, digits, and words are each
represented in permanent memory; each is an already known item.
The apparent span can be enlarged by increasing the amount of information con-
tained within each item (Miller, 1956). For example, the number of remembered items can
be increased by parsing and encoding the material in terms of meaningful units, called
chunks. Remembering the string of letters FBITSACIA is more difficult than remembering
FBI–TSA–CIA. Nine letters becomes three chunks (or three items). The chunking strategy
allows remembering strings much longer than seven digits or words. Miller describes one
210 Short-Term Memory

subject, Sydney Smith, who could recall lists of 20 randomly sequenced 1’s and 0’s after
a single presentation. Had Smith enlarged his short-term memory capacity to 20 items?
Not at all. A code was created that translated three-digit strings into letters. Thus, 100 = A,
101 = B, 110 = C, and so on. Smith converted the list into letters and just remembered the
letters. Thus, he would actually remember something like A–D–F–C–C–B. In reproducing
the digit string he decoded the letters back into 1’s and 0’s.
There are two points to emphasize here. One is that STM capacity is indeed limited.
Apparent enlargements of STM are often accomplished by recoding items into more
encompassing units. The second point is that short-term memory uses long-term mem-
ory. We can remember items in STM that are already known in long-term memory.

Limited Duration
Short-term memory is short, although the answer to “How short?” varies. Estimates of
primary memory when rehearsal is limited are on the order of seconds (Waugh & Norman,
1965). As we saw earlier, forgetting occurred after 15 to 30 seconds of distraction in the
Brown–Peterson task. Informal use of the term STM often refers to something on the
order of one to several minutes in everyday life.

Forgetting: Short-Term Memory Is Sensitive to Disruption

In the Atkinson and Shiffrin (1968) model, forgetting from short-term memory is due to
the dis-placement of old items by new items. The capacity limit means that the addition
of a new item requires dropping some item already in STM.
It does not take much to disrupt STM. You look up a phone number, someone asks
you a question, and the number’s gone. A laboratory example of this sort of distraction is
the suffix effect. At the end of each list of to-be-remembered items, a recurring item is
added that need not be remembered. This suffix reduces recall of the last item, and
sometimes the last few items. For instance, the word “zero” presented after a list of nine
items caused more forgetting of the final item in the list than did presenting a buzzer after
the list (Crowder, 1972). Similarly, when lists are presented via sign language to deaf
subjects, a suffix effect occurs with an added sign (Shand & Klima, 1981).
Short-term retention in the real world often takes place in the context of divided
attention. Instead of the quiet background conditions of the laboratory, other sounds
occur simultaneously with the to-be-remembered list (e.g., you try to remember an
access code while the television is on, your roommate is talking,…). There is more inter-
ference with memory span if the background sound is speech than if the background is
white noise (Banbury, Macken, Tremblay, & Jones, 2001).

Transfer to Long-Term Memory

STM has a role in transferring information to LTM. This characteristic has commonsense
appeal. When we decide to remember something in permanent memory, we process
information in a certain way in STM. We rehearse, try to form an image, or devise a mne-
monic to aid later recall. Maintenance in short-term memory allows the opportunity for
information to be copied into long-term memory.
 Other Modalities of Short-Term Memory 211

Several sorts of evidence support this transfer function from STM. If subjects are
asked to rehearse out loud when attempting to remember a list, the words that receive
more rehearsals are generally recalled better (Rundus, 1971). That is, more rehearsals
correlate with better long-term memory for those words. (This was illustrated in Figure 6.5
in Chapter 6).
There is other evidence, however, which shows that remembering in STM is not
necessary for LTM formation. One telling observation comes from brain-injured partici-
pants who are impaired in the immediate recall of auditory information. One such case,
an individual referred to as K. F., had a limited verbal short-term memory. Nevertheless,
he could learn long-term memories at normal rates (Shallice & Warrington, 1970). If a
10-word list was repeated until memorized, K. F. learned just as quickly as normal control
participants. K. F. shows a dissociation between poor STM performance and normal LTM.
It is also the case that maintenance of information in short-term memory does not
guarantee entry into long-term memory. Football players who had just suffered a concus-
sion could remember the events leading up to the head injury. These memories gradually
became less available as time passed, and were essentially gone after four hours. Even
though the information had been in STM, as shown by the players’ reproductions of their
stories during the first several post-traumatic minutes, eventually forgetting occurred
(Lynch & Yarnell, 1973).
Overall, the evidence suggests that processing in short-term memory does aid long-
term learning. Control processes employed within short-term memory, such as rehearsal
and imagery, benefit long-term retention. However, simple passive residence in STM may
not suffice to transfer information to LTM.

Control Processes
We can choose (to a degree) when and how to use STM. There are active control pro-
cesses in STM. We can control where to direct attention, how to code new inputs, when
to rehearse, and which retrieval cues to use. Control processes define STM as an active
rather than passive store. Information in STM can be cognitively manipulated.

Summary of the Features of Short-Term Memory

The just-reviewed characteristics suggest that there are few absolute distinctions
between verbal short-term memory and long-term memory. Acoustic or semantic encod-
ing can occur in either memory. The capacity of STM, although indeed limited, is expand-
able by using coding schemes to draw on what is already stored in LTM. Finally, rehearsal
in STM does not necessarily ensure entry into LTM. Thus, not all of the supposed charac-
teristics of verbal STM neatly dichotomize it from LTM.

OTHER MODALITIES OF SHORT-TERM MEMORY

The research presented so far has focused on memory for verbal material. This bias
derives partly from the use of college students and word lists in our experiments, and
partly from our tendency to equate short-term memory with rehearsal. However, short-
term retention occurs in other sensory modalities.
212 Short-Term Memory

Visual Short-Term Memory

To study visual short-term memory, stimuli are presented as short lists of images, pic-
tures, or objects. The images may be meaningless, such as nonsense shapes, or the
pictures may be meaningful, such as a scenic mountain view. Retention is tested by a
recognition procedure: A test picture is shown and the subject decides whether it was in
the previewed list.
The use of picture stimuli does not preclude verbal encoding of the images. People
can recode pictures into words and then rehearse these descriptions. Verbal encoding
can be reduced by using materials that are not readily described in words. Wright et al.
(1990) compared memory for scenic and kaleidoscope pictures. Lists of 10 images were
presented and the “off” time (or blank screen time) between images was varied between
0.5 and 6 seconds. Longer off times facilitated recall of the scenic pictures, as if subjects
were verbally rehearsing between pictures (“tree, brook, snow…”) during the 6-second
intervals between pictures. Remembering the kaleidoscope images, on the other hand,
did not improve with longer off times between images, suggesting they were not being
verbally rehearsed. Remembering these images thus seemed to depend on a visual
memory.

Spatial Short-Term Memory

Short-term memory for spatial positions can be illustrated using a grid on a computer
screen, something like a tic-tac-toe board. An asterisk is presented in one square at a
time, in a random sequence, to mark locations. After presentation of this list of positions,
the participant points to the locations in the same order in which they had appeared.
Dark and Benbow (1991) tested spatial memory in the preceding manner with sev-
enth graders who were mathematically or verbally advanced. Strings of three to nine
items were presented, using lists of spatial positions, digits, or words. These results are
shown in Figure 8.4. In remembering locations in the spatial grid (the middle panel in the
figure), the mathematically advanced children performed better than the verbally advanced
students. The outcomes with digit and word spans were predictable: the math advanced
recalled more numbers (left panel), and the verbal advanced recalled more words (right
panel). Figure 8.4 also includes data from children who were both mathematically and
verbally advanced.

Short-Term Memory for Actions

STM for body movements has been tested using short lists of hand and arm actions. In a
version of the distractor technique, the experimenter performed movement triplets before
adult subjects. Example actions might include clapping, waving, or pressing a button,
although not all of the movements were so easily described in words. After a delay of
15 seconds, the subjects reproduced the actions. There was little forgetting if the partici-
pants did nothing during the delay interval. There was significant forgetting if the partici-
pants performed interfering movements during the delay, or if they had to name the
actions (Kausler, Wiley, & Lieberwitz, 1992). Just as in the Brown–Peterson task, forget-
ting occurred when there was distracting material.
 Other Modalities of Short-Term Memory 213

Digits Locations Words

1.00 1.00 1.00

MV
M
V
0.75 0.75 0.75
Proportion Correct

0.50 0.50 0.50

0.25 0.25 0.25

0 0 0
5 6 7 8 4 5 6 7 5 6 7 8
List Length List Length List Length

Figure 8.4
Correct Recall in Three Span Tasks. Graphs plot the number of digits, locations, and words recalled by the
mathematically gifted (M), verbally gifted (V), and mathematically and verbally gifted 13-year-olds (MV).
Source: From “Differential Enhancement of Working Memory with Mathematical versus Verbal Precocity,” by V. J. Dark and
C. P. Benbow, 1991, Journal of Educational Psychology, 83, p. 52. Copyright © 1991 by the American Psychological Association.

Short-Term Memory for Odors

Olfactory memory is so robust that it is virtually a contradiction in terms to link “short-
term memory” and “odor memory.” Engen, Kuisma, and Eimas (1973) presented their
participants either a single smell or a list of five odors. These were chosen from a pool of
100 odorants being used. After distractor intervals of from 3 to 30 seconds spent count-
ing backwards, the participants smelled a test odorant and decided whether it was the
same as one just presented (or one of the five presented, in the list version). Overall
accuracy averaged around 80 percent, with no significant forgetting between 3 and
30 seconds. Although a short-term memory procedure was used here, there was little
evidence of short-term forgetting.
Other research has shown some forgetting across intervals of seconds to minutes,
particularly if the odors are unfamiliar and not easily encoded verbally (de Wijk, Schab, &
Cain, 1995). A known smell can be encoded as a word (e.g., talcum powder) and after the
delay interval, the subject decides whether the test stimulus is talcum. However, recog-
nition of unfamiliar odors, which depends more on memory for the sensory attributes,
was still excellent after a delay of 100 seconds.

Short-Term Memory in the Hearing Impaired

Nonverbal STM is also studied among hearing-impaired individuals. What is the nature of
their encoding in STM? Generalizations are complicated by a number of factors across
participants: different levels of hearing impairment, training in oral speech, and alternative
forms of sign language (signing versus spelling words). Hearing-impaired children who
214 Short-Term Memory

have learned some spoken English make phonological confusions, just as do hearing
children. Children without spoken language more often make confusions based on the
similarity of visually shown words, and similarity in the form of manually presented signs
(Conrad, 1970). Bellugi, Klima, and Siple (1974–1975) found that signs remembered in
error were usually signs that were visually similar to the list signs. Errors made by hearing
subjects were misrecalls of similar sounding words. Both hearing and deaf made misre-
calls of words similar in meaning to a list word.
The generality of STM is further extended by research on animal short-term memory.
This work is reviewed in Box 8.1.

BOX 8.1 ANIMAL SHORT-TERM MEMORY

Most experiments on animal short-term memory use a variation of the delayed response
method, first introduced by Walter Hunter in the 1920s. A cue is presented that signals
where food is hidden, but the animal has to wait before being allowed to retrieve it. For
example, the animal watches as food is hidden in one of several dishes that have distinctive
covers. After a delay, the animal is allowed to retrieve the food, to see if the location was
recalled.
Delayed responding can tell us many things about animal memory. Rats and raccoons
could remember where the food was hidden for several seconds, dogs and cats for two to
three minutes, and monkeys and young children for a few minutes. Control procedures are
used to assure that correct choice is due to memory rather than some alternative strategies.
For instance, odor cues from the food reward need to be neutralized or eliminated.
In addition to remembering where food was hidden, do animals remember what was
hidden? Tinklepaugh (1928) sometimes would show lettuce being hidden and other times
pieces of fruit. Monkeys performed well with either food. But occasionally the food was
switched during the delay, without the monkey’s knowledge. Monkeys and chimps quickly
realized that Tinklepaugh had pulled a fast one on them. They refused to eat the changed
food, and they would search the other containers. One chimp hurriedly stuffed the food into
her mouth, but then spat it out, as if realizing this was not the expected taste!
Another version is the delayed matching-to-sample procedure, or DMTS. A sample stimulus
is presented and after a brief delay a pair of choice stimuli are offered. When the subject
chooses the one that matches the sample, reward is given. DMTS is often used with pigeons,
which can remember various shape, pattern, and photographic stimuli over a period of a few
seconds. In a DMTS study of dolphins, the animals matched cylinders or balls that could not be
seen visually but were perceived through echolocation (Roitblat, Penner, & Nachtigall, 1990).
Parallels between human and animal STM have been noted. Serial-position curves for
visual short-term memory by pigeons, monkeys, and humans are shown in Figure 8.5 (Wright
et al., 1985). A common set of procedures was developed in which subjects were shown a
sequence of four pictures on each trial. After a delay interval a single picture was presented
for subjects to identify as having been from the list or as being a different picture. The animals
were more accurate in remembering the first and last pictures from the lists: There were
U-shaped retention curves having both primacy and recency. The shape of the curves is
remarkably similar across species.
Animal research has produced other surprising findings. One was in the study of memory
span in chimpanzees. In the chimp version, the numerals 1 through 9 were displayed
 Working Memory 215

Pigeons Monkeys Humans 100

100

80 80
Percent Correct

60 60

40 40

2 seconds 10 seconds 25 seconds

20 20
1 2 3 4 Diff. 1 2 3 4 Diff. 1 2 3 4 Diff.
Serial Position of Memory Item

Figure 8.5
Serial-Position Curves for Pigeons, Monkeys, and Humans. Participants were shown sequences of
four pictures on each trial. After a delay interval, a single picture was presented for participants to
identify as having been from the list or as being a different picture (the points above “Diff.” in the
graphs).
Source: From “Memory Processing of Serial Lists by Pigeons, Monkeys, and People,” by A. A. Wright et al., 1985,
Science, 229, pp. 287–289. Reprinted with permission from AAAS.

simultaneously on a touch screen. The locations of the items varied randomly from trial to
trial. The animals were taught to touch the digits in numerical sequence. The fact that the
animals could learn to sequence nine symbols (our numbers 1 through 9) is pretty neat in
itself. However, in the next phase of the study, the animals had to remember the random
locations of the numbers. As soon as the chimp touched the first number, the remaining
digits were replaced by white squares. So the chimp had to remember where each number
had been located, and still touch them in the correct sequence. Inoue and Matsuzawa (2007)
report that one well-trained chimp (named Ai) could recall strings of five numbers correctly at
far-above chance levels. Ai inspected each newly presented screen for less than one second,
and then started punching off the locations. The animal was faster and more accurate than
were humans performing the same task.
Studies of animal STM are important for several reasons. They are widely used in compar-
ative cognition, a field that studies the mental abilities of different species from an evolution-
ary perspective. Neuroscientists use animal STM to investigate the effects of brain lesions,
neurotransmitter substances, and brain chemistry. Finally, pharmacologists use animal mem-
ory as a sensitive assay of potential cognitive side effects of new drugs and compounds.

WORKING MEMORY
Theoretical Overview of Working Memory
Although the concept of STM was an important theoretical advance, there are some
limitations to it. Short-term memory is not only a place for the temporary storage of
information. We plan, elaborate, compute, and imagine in STM. The concept of working
memory was devised as an advancement and elaboration of short-term memory.
Specifically, the working-memory model (Baddeley & Hitch, 1974; Baddeley, 2000)
216 Short-Term Memory

Figure 8.6
The Working-Memory Model.
Source: From “The Episodic Buffer,” by A. D. Baddeley, 2000, Trends in Cognitive Sciences, 4, pp. 417–423. With permission
from Elsevier.

proposes four components. The temporary storage of information occurs in either the
phonological loop or the visuospatial sketchpad, for temporary storage of auditory and
visual information. The central executive allocates attention, and the episodic memory
buffer connects working memory to long-term memory (see Figure 8.6).
One reason for advocating a multicomponent model is that people can indeed do two
things at once with little cross-interference, depending on certain factors. Performance in
the Brown–Peterson distractor task shows that counting backward disrupts the memory
for three target words. However, both counting and remembering are verbal. If two tasks
involve different modalities, such as a verbal and a spatial activity, then interference is
reduced. For example, we can drive while carrying on a conversation (sort of).
Another feature of working memory is the emphasis on simultaneously retaining and
manipulating information. For instance, in giving driving directions, we mentally count the
number of blocks or traffic signals before a turn, then the number until the next turn, etc.
In a similar manner, experimental tasks used to assess working memory are more chal-
lenging than the distractor and span tests of STM.
Finally, working memory performance is also a better predictor of general cognitive
capacities, such as verbal ability, reasoning, academic achievement, and aspects of intel-
ligence (e.g., Engle, Tuholski, Laughlin, & Conway, 1999).
Let’s look at each component of the four-part model of working memory.

The Phonological Loop

The phonological loop is comparable to verbal short-term memory. It represents our
brief storage of verbal material which we use in rehearsal, verbal problem solving, arith-
metic, and so on. The phonological store has most of the characteristics described earlier
for the modal model of short-term memory, i.e., it is brief, has a limited capacity, and is
easily disrupted. As noted before, verbal short-term memory is measured by a test of
digit span or word span: the number of items that can be recalled after a single presenta-
tion. Working memory adds a second processing task to the span task. For instance,
 Working Memory 217

intermixed in the presentation of the to-be-remembered words are true–false arithmetic

problems. For example, try to remember this string: DOG–CAT–is 6 + 3 = 7?–TREE… etc.
Remembering DOG–CAT–TREE might be impaired by the arithmetic process.

The Visuospatial Sketchpad

As the name implies, the visuospatial sketchpad retains visual images (e.g., pictures or
imaged versions of words) and spatial information (e.g., the locations of squares on a
checkerboard). In a simple span test of STM, we might ask the subject to remember a
sequence of up to eight red and black blocks on a checkerboard as we point to them. In the
WM version, in addition to remembering the locations, we also ask the subject to judge
the symmetry of the blocks or whether a pattern exists. (Note: Baddeley named this com-
ponent the visuospatial store. Another familiar term is visual-spatial, as in visual-spatial
intelligence or visual-spatial memory. The terms seem to be used interchangeably).
Several lines of research support the independence of the phonological and visuospa-
tial stores. Memory for a string of words is disrupted by simultaneously answering arith-
metic questions. Similarly, remembering a picture or a nonsense shape is impaired by
simultaneously having to perform a visual-tracking task, such as tracking a moving dot of
light with your finger. In both of these cases, a single short-term store, either the verbal
or the visual, is doing two things at once and so interference occurs. If the two tasks use
different stores, however, interference is reduced. Remembering a word list is not
severely affected by a visuospatial tracking task: Each can be performed at the same level
alone as when done together (see review by Gathercole, 1994).

The Central Executive

The central executive is in control of attention. It focuses attention, allocates attention,
or distributes attention across multiple tasks. The difficulty we experience in attempting
two things at once can be due to our inability to focus full attention on each.
Attention is a limited resource. How is it allocated across multiple tasks? One possi-
bility is that attention is divided so each task gets a portion. Alternatively, the central
executive could be rapidly alternating (nearly) full attention from one task to another. So,
as applied to an earlier example, does our subject allocate some attention to remember-
ing DOG–CAT–TREE and some attention to solving the arithmetic? Or does the subject
focus full attention on rehearsing DOG and CAT, then switch full attention to adding 6 + 3,
and then quickly switch back to add TREE to memory?
An additional role of the central executive is to act as a sort of manager between the
two memory stores, and this is the sense of executive. The executive manipulates and
coordinates information stored in the buffers, directs attention, and engages in problem
solving and planning.

The Episodic Buffer

The episodic buffer integrates information across (a) the phonological and visual stores;
(b) the central executive; and (c) encoding and retrieval from long-term memory (Baddeley,
2000).
The several components of the short-term memories necessarily interact with LTM.
For instance, in retaining several items in STM, the size of those items can vary, from
218 Short-Term Memory

digits or letters to words. Some of these items are activated in LTM, and the episodic
buffer acts as a bridge between the markers or tags retained in phonological store and the
active items in LTM. Similarly, arithmetic operations (i.e., addition or subtraction) must be
retrieved from LTM to be applied to the items in STM.
Familiar words are represented in multiple ways in LTM. There are the semantic rep-
resentations: synonyms, associated words, and general knowledge that give meaning to
the word. There are also phonological representations that encode the pronunciation of
the word, and those neurons involved in the mechanics of speaking the word. This pho-
nological knowledge in LTM contributes to working memory (Martin, 2021).

Measuring Working Memory

Working memory is typically measured by the capacity to do two things at once, what is
referred to as dual tasks. For instance, one might ask participants to perform a series of
mental arithmetic problems (e.g., 2 + 3 = ?; 4 × 2 = ?; 9−3 = ?), and remember the
answers to the computations (5, 8, 6). Alternatively, the task might use both the phono-
logical and visuospatial stores, for instance by remembering a string of words while judg-
ing whether pairs of objects are the same or different.
The n-back task was specifically developed to test working memory. As a starting
point, in the simple n-back the participant sees a continuous stream of letters (B P M T
M). As each letter is presented, the subject must decide whether that letter was the same
as the letter immediately before, and push a yes or no button. The letter n refers to the
number of items back (see Figure 8.7). The letter n takes on different values: if n = 1, does
the current letter match the previous letter? If n = 2 does it match the item two-steps
back? In the example in Figure 8.7, if n = 2 the letter T in the final grid matches the T that
occurred two steps before; M in the fourth grid matched the M two steps before it.
In a dual n-back task, two bits of information must be tracked. In Figure 8.7 the letters
and the location vary at each step. The participant must make two responses, deciding
whether the letter item matches a previous letter (regardless of location within the grid),
and/or whether the current location tagged match the a previous location (regardless of
letter). In the dual 2-back level, the task is “Is this letter the same as the one 2 steps back?
Is this location the same as 2 steps back?” For the final T, the answer is yes for both letter

n - back task, n = 2

P T M T
Display
M

Same Same Same

Response location letter location &
2 steps 2 steps letter 2
back back steps back

Figure 8.7
Illustration of a Dual N-Back Task. A new grid is shown every few seconds. For n = 2 the subject must make
separate responses to note whether the current letter appeared two steps back, and whether the same location
was occupied two steps back.
 Working Memory 219

and location. For the final M, the answer is yes for matching the letter 2 steps back, but
not for the location 2 steps back.

Executive Functions
Executive functions (EFs) are higher-level cognitive skills used to control and coordinate
other cognitive abilities and behaviors. The EFs direct a number of mental processes includ-
ing short-term remembering, problem solving, language production, language comprehen-
sion, and decision making. Whereas working memory refers to the structural elements
(e.g., phonological store, episodic buffer), executive functions refer to how and when cogni-
tive functions are used. EFs are effortful and intentional; they are not habitual or automatic.
(A task may become habitual and routine, such as taking restaurant orders or directing
planes from the control tower; but the momentary specifics are constantly changing).
An everyday example of executive functions would be preparing a meal. The several
dishes need to be kept in mind; the steps in preparing each need to be sequenced, and
likely the steps are interleaved; attention switches from one dish to another; cooking start
and stop times computed and remembered; completed steps can be dropped from
memory; distractions during preparation are dealt with. There are a number of mental
activities here: attention, planning, sequencing, short-term remembering, retrieval from
long-term memory, mental math, and tracking sub-goals.
A number of executive functions have been suggested, as well as multiple tests to
measure each. Statistical techniques such as factor analysis are used to group tasks
based on their correlations with one another. For example, scores on word span, n-back,
and the distractor task are highly intercorrelated and seem to measure simple short-term
retention. The analyses have suggested that the primary EFs are updating, inhibition,
shifting, and attention control (Miyake & Friedman, 2012).
Updating means updating the contents of WM, by adding new information and drop-
ping content that is no longer needed. The dual n-back test is a prime exemplar of updat-
ing: keeping track of letters and spatial locations over the last few items. Updating also
applies to the completion of the steps in carrying out a specific task.
Inhibition is necessary to override a dominant, automatic or habitual response. For
example, in the Stroop task the subject must name the color a word is printed in, even
though the word is a conflicting color word (e.g., What color ink is this word printed in:
GREEN. The ink is black but the word Green causes conflict). Inhibition is also relevant to
resisting attention to extraneous thoughts or distracting behaviors.
Shifting is ease of switching rules or strategies. For instance, after sorting blocks by
color, being able to switch to sorting by shape. Flexibility is another term used, to empha-
size readiness to respond to changed goals or priorities.
Attention control is the ability to focus attention, avoid distraction, and switch atten-
tion readily. For instance, one task is to detect a briefly flashed letter to the right of a fix-
ation point. Occasionally a flashing asterisk (*) appears to the left. This grabs attention and
the letter on the right is missed. Attention control is maintaining focus on the letter side
of the screen.
The individual EFs do overlap. Attention control shares the most overlap, and may be
the more basic executive function.
220 Short-Term Memory

The concept of executive functions is a popular, and possibly overused term in related
disciplines outside of psychology. Numerous websites advertise the benefits of develop-
ing executive functions for success in education, business, sports, etc. In some of these
cases, the term EF refers to so-called “higher-level executive functions”: self-control,
planning, problem solving, reasoning, and intelligence (Diamond, 2013). Similarly, EFs are
noted to be important to many aspects of life: mental and physical health, school readi-
ness, and social relationships. So, executive functions can refer to very different things at
different levels: an experimental participant ignoring the flashing asterisk versus a child
resisting temptation to take a cookie.

Working Memory and Consciousness?

The questions “what is mind?” and “what is consciousness?” are technically beyond the
purview of this book. However, we cannot help but speculate on the relationship of work-
ing memory to consciousness. An essential property of the central executive is the con-
trol of attention. This suggests something (or someone) is inside the mind or brain actively
directing and choosing what to attend to. Unattended information is, by definition, outside
of consciousness. Rehearsing words in the phonological store, essentially talking to one’s
self, is an attribute of consciousness. Images can be created and manipulated in the
sketchpad. Who (or what) is doing this? Finally, human working memory is a highly evolved
system. Mind and brain scientists have not reached a consensus, yet many see the need
to discuss consciousness and working memory together (Maia & Cleeremans, 2005).
Another unresolved issue is how the mind integrates information that is represented
in so many different codes. This is the consciousness binding problem. For instance,
verbal and visual information originate in different sensory modalities and are coded dif-
ferently in the brain. Yet the mind translates one to another and combines both. The same
issue applies to the many other codings the brain uses for the other senses, for emo-
tions, etc. On the surface this may not seem like such a difficult problem to those of us
now accustomed to having computers translate from one format to another, or from one
software program to another. Yet most of us do not know the underlying machine lan-
guage by which this is done. A parallel understanding of the brain would seek to discover
the mind’s “machine language.”

Working Memory and Culture

The concept of working memory has been influential beyond psychology. Wynn, an
archaeologist, and Coolidge, a psychologist, proposed that the major characteristics of
working memory—the capacity to temporarily hold and manipulate information—led to
the evolution of our present day cognitive abilities (Coolidge & Wynn, 2009). According to
Wynn and Coolidge, the significant evolutionary leap from the other animals was working
memory. There may be variants of working memory in animals, but the development has
peaked in humans. This allowed humans to plan, develop symbolic language, and create
art. The appearance of carved objects, burial rituals, cave art, and more developed tools
are taken as evidence for the appearance of this increase in cognitive ability. The idea that
working memory led evolution is still untested. After all, working memory did not sud-
denly appear in humans; other animals have working memory to varying degrees.
 Individual Differences in Working Memory 221

However, Wynn and Coolidge’s theory illustrates the wide-ranging influence of a psycho-
logical theory (Balter, 2010).

INDIVIDUAL DIFFERENCES IN WORKING MEMORY

Given there are several components of working memory, the number of dimensions
along which individuals can differ is multiplied. Nor is working memory a fixed capacity,
because individual performance varies with practice and training. We have already
described one category of individual differences in memory span among adolescents
with advanced math or verbal ability. Here we will consider a few other categories of dif-
ferences among individuals in the following sections: working memory in aging, in
dementia, and in the presence of anxiety.

Aging and Working Memory

Memory over short retention intervals, as assessed by the distractor or span tests,
decreases with age. Is this decline due to a reduction in the capacity of short-term mem-
ory? There are other aging variables that adversely affect memory. One factor is slower
processing. For example, on some tasks older adults need more study time to perform as
well as younger adults. Once the stimuli are in memory, the rate of forgetting across
delay intervals is the same in older and younger participants. In another example, partici-
pants performed a reading and an arithmetic task concurrently. The older participants took
longer to respond to questions about what they had read, suggesting that they are slower
at switching from one task to another (Salthouse, 1994).
A second factor associated with aging is the ability to inhibit distracting thoughts.
Hasher and Zacks (1988) studied working memory in language processing: listening,
reading, and comprehension. They believe that older participants have particular trouble
inhibiting extraneous associations that are activated by the target material. These tangen-
tial thoughts can be triggered by the material itself (such as personal associations or
reminiscences) or by the external environment (noticing the room, maybe the lecturer,
but not the lecture). Working memory’s capacity is taxed by this irrelevant content. One
result may be a lack of focused attention: failure to keep up with the conversation or not
being aware of what was just read. Another effect is that unrelated thoughts can foster
weaker encodings of the target material and interference among target and irrelevant
memories.

Dementia and Working Memory

Alzheimer’s disease is a particularly devastating form of cognitive dementia. The severity
of impairment in laboratory studies of STM varies from mild to profound (Baddeley, 1992b;
Kopelman, 1994). One reason for the discrepancy may be whether a single task is being
performed or two tasks. In one study, Alzheimer’s patients were found to be impaired at
doing two tasks at once, even though each alone could be performed competently
(Baddeley, 1992b). The difficulty of each task alone, a visual tracking task and verbal mem-
ory span, was adjusted so that each could be performed well. However, when both tasks
needed to be performed simultaneously (tracking a moving light while remembering a
222 Short-Term Memory

string of words), performance dropped off dramatically. Unimpaired control participants

were able to perform both tasks simultaneously with little cross-interference. The
Alzheimer’s patients had difficulty in alternating between the two tasks. This may reflect
impairment in the coordination of attentional resources by the central executive.

Anxiety and Working Memory

Working memory capacity can be compromised by worry. Attention is diverted to preoc-
cupation over performance, self-criticism, and other forms of negative thinking. Negative
self-talk engages the phonological store.
Worry can impair working memory in several areas related to academic performance.
Individuals with mathematics anxiety do more poorly than less-anxious students on math-
ematics problems. No surprise there. The math-anxious also have difficulty on working
memory tasks that involve numbers and computation, but the deficit is not all related to
math ability. There also seems to be some preoccupation with anxious feelings and neg-
ative thoughts that co-opt resources. Working memory impairment is especially evident
on more difficult problems, for example, those that require mentally adding two-column
numbers in which there is a carry operation (Ashcraft & Krause, 2007).
The after effects of worry and rumination carry over onto subsequent academic tasks.
For instance, students with elevated depression scores (not clinically depressed, just
higher than average scores from a college student sample) were asked to think about their
feelings, physical sensations, and reactions to emotional cue words. What the experi-
menter was doing here was eliciting some controlled worry. In the next phase of the study,
the students were given reading comprehension passages to work on. These students
required more time to read and had lower comprehension scores on those passages. In a
parallel study, elicited worry led to poorer memory of a class lecture (Lyubomirsky, Kasri,
& Zehm, 2003).
Worry also contributes to stereotype threat. When a negative stereotype is activated
in the mind of a person to whom the stereotype might apply, performance can be inhib-
ited. One stereotypical belief is that women perform poorly at math. Reminding women
of the stereotype before a math test, or emphasizing that the test is a math test, leads to
worse performance. Note that the comparison in this case is not a math difference
between women and men. The comparison is between college women who are reminded,
or not, of the stereotype.
Does stereotype threat elicit negative thinking, and does the threat impair working
memory? Beilock, Rydell, and McConnell (2007) found affirmative answers to both ques-
tions. In a challenging task requiring number manipulation and problem solving, the acti-
vation of a stereotype threat in women led to more errors. Also, women in the stereotype
condition reported more worry about their performance. The researchers suggest that the
phonological loop was especially taxed in these tasks: both the arithmetic and the worry-
ing occupy the verbal store.

Multitasking
Multitasking is a perfect example of the complexity of working memory. In many every-
day situations, there are several tasks being conducted simultaneously, each of which has
 Individual Differences in Working Memory 223

several steps. Paul Burgess (2000) provided a nice summary of the components of every-
day multitasking situations.
1 (Obviously) there are several separate tasks being attempted.
2 At any one moment only one task (or one step in a task) can be performed at a time
due to physical or cognitive limitations.
3 The steps involved across tasks are interleaved (start one task, put it on hold, and
start a second task, then back to the first task…etc.).
4 Interruptions and unexpected outcomes lead to delays and re-planning. These
increase the demands placed on short-term memory and the central executive.
5 Remembering to return to a task that has not yet been completed.
In many occupations that require multitasking, the correct performance of each step of
each task is essential. In a medical setting, for example, the nurse has several required
steps in administering a drug. First ensure aseptic conditions (washing hands, wearing
gloves); then matching the prescription with the drug’s label and with the patient’s ID. The
nurse may be alternating these steps with those for a second drug, and maybe a second
patient. This activity defines a working memory task: executive function in organizing the
steps; short-term memory to retain steps completed; and dual-tasking with multiple
drugs or patients. As if this is not difficult enough, there are invariably interruptions, such
as requests made by the patient or someone else. In one study in which the nurses were
observed at work, an interruption increased errors by 12 percent. These errors occurred
even though the nurses knew they were being observed. (In this study, the observers
could intervene if they anticipated a serious error). The point of this example is that even
well-trained professionals are subject to working memory failures under stressful, time
limited, or distracting conditions (Westbrook, Woods, Rob, Dunsmuir, & Day, 2010).
Supertaskers? One area of special concern is the danger of multitasking while driving
a car (or plane, train, or even walking). While there may be agreement that texting while
driving is dangerous, too many believe that talking on the phone is not distracting.
However, studies of college students in driving simulators show impairment even during
hands-free phone conversation (Strayer & Drews, 2007).
Any statistic we cite showing that most people are adversely affected is countered by
the belief that you (or I) are the exceptions. Well, how exceptional are you? Students were
tested in a driving simulator while simultaneously performing a difficult WM task (remem-
bering words and answering arithmetic questions). The WM task was presented over a
hands-free phone while the students were driving. Among 200 participants, the research-
ers found only five, that is, less than 3 percent (of bright, motivated, college students)
who could multitask without impairment on the driving simulator. For 97 percent of the
participants, the WM task produced a significant decline in driving measures such as
braking reaction time (Watson & Strayer, 2010). (It is also the case that the driving
adversely affected WM performance). So, you may think the averages do not apply to you.
But are you really among the top 3 percent?
Can working memory be improved? There is a large and controversial literature on the
effects of training on working memory, and frequent reports in the media of treatments
that will supposedly increase cognitive performance. Some of these questions are
addressed in Box 8.2.
224 Short-Term Memory

BOX 8.2 TRAINING WORKING MEMORY

Is it possible to improve overall cognitive functioning through training?

Cognitive training (CT) is an active area of scientific research and commercial enterprise.
Improving working memory (WM) has been a particular focus of training, since it contributes
to a range of complex cognitive tasks. WM performance correlates with measures of “fluid”
intelligence, i.e., tests of reasoning and problem solving. Improving WM should have bene-
fits to everyday functioning.
Cognitive training involves a set of computer-administered tasks. As performance improves
with practice, the program increases the difficulty of the tasks. For instance, 1-back training
progresses to 2-back and 3-back. In an early successful demonstration, college students
completed between 8 and 19 sessions of dual n-back training (i.e., remembering letters and
their locations in a grid). After training, the students took a test of abstract reasoning (e.g.,
identifying the pattern in a set symbols). Reasoning improved significantly over pretest
scores. In a related study, middle-school children were trained on a spatial n-back task. The
children completed an average of nineteen 15-minute sessions. After training some children
did score better on an abstract reasoning test, but only those who showed a large improve-
ment in n-back performance (Jaeggi et al., 2008, 2011).
A commercial industry for brain training has developed with estimated sales in the hun-
dreds of millions of dollars (Simons et al., 2016). (Brain training is a synonym for CT, even
though brain changes are not measured. But it sounds more scientific). Money and reputa-
tions are at stake on both sides of the question of CT efficacy. The vigorous, and vehement,
debate is reminiscent of the arguments over Covid-19 research, which had much to do with
how good the science was. (The disagreements among CT researchers is well described by
a journalist who interviewed those on both sides of the issue. See Hurley, 2013).
Determining whether cognitive training works first requires defining how success is
measured. Practice on any task often leads to improvement in related tests that use the
same skill. This is called near transfer. So, n-back training leads to improvement in short-term
memory as tested with the digit span or letter span. The more important demonstration is to
show that training improves performance in different cognitive abilities. This is far transfer.
Does practice in working memory improve general problem solving, logical reasoning,
abstract reasoning, or intelligence?
A second issue in judging CT is whether the quality of the research unambiguously
answers the question. Much of the supporting research has been challenged for weak design.
As examples, the use of inadequate control groups, small number of subjects, and interven-
tions of limited duration. A particular problem is that cognitive training is particularly suscep-
tible to psychological bias: Participants know they are in a study and what the goal of the
research is, making them particularly susceptible to bias. Placebo effects have been demon-
strated among control participants who thought they were receiving real training. Participants
who are paying subscribers to an online training program have a stake in a positive outcome.
The experimenters, educators, and parents involved have expectations that could influence
the outcomes.
Some meta-analyses conclude that the evidence for cognitive training is still not there
(Sala & Gobet, 2019; Simons et al., 2016). The reviews find that far transfer is rarely found in
high-quality studies; those which used the strictest designs and analyses. Cognitive training
frequently shows near transfer. Cognitive training benefits are larger in the studies with
weaker methods, whose results are therefore more ambiguous.
 Is There Really a Separate Short-Term Memory? 225

It is also difficult to compare outcomes among experiments that differ in so many varia-
bles, e.g., different WM tests, transfer tasks, subject populations (children, college students,
senior adults), and experimental versus correlational designs. To say it is like comparing
apples to oranges is inadequate; the situation is more like comparing one fruit salad to
another fruit salad. The argument over conclusions is illustrated by an exchange of opposing
statements representing over 200 scientists. “There is no convincing evidence that working
memory training is effective…” versus “There is no convincing evidence that working mem-
ory training is NOT effective…” (Au et al., 2016; Melby-Lervåg & Hulme, 2016).
Working memory training may clearly have benefits to individuals with specific deficits,
such as attention deficit, specific learning disabilities, or brain injuries, as a means to develop
alternative strategies to perform cognitive tasks. It is also clear that much more testing and
development will be needed before we know the true benefits of cognitive training.

IS THERE REALLY A SEPARATE SHORT-TERM MEMORY?

It seems perfectly obvious that some things are forgotten quickly and others slowly, if at
all. This view presumes the physical reality of STM and leads researchers to seek the
neural basis of STM. Other memory theorists have questioned the necessity of hypothe-
sizing a separate short-term memory system.

A Single-Memory Approach?
How can the evidence that favors a separate STM and LTM be parsimoniously accommo-
dated within a single memory system? One possibility is to suggest that there is a single
set of memories, but memories can differ between those that are currently active and
those that are inactive (Lewis, 1979). Suppose I ask you to recall the name of your third-
grade teacher or your favorite pet’s name. What you have just done is taken an inactive
memory and made it active. The inactive memory corresponds to LTM, and its activation
is what we have called STM. The fading of the active neural elements to their resting
levels corresponds to forgetting from STM.
Another single-memory hypothesis reminds us of two facts: Memories are multidi-
mensional, and each dimension could have a different rate of forgetting.
The memory for an item or an event may be encoded in multiple dimensions, such as
sight, sound, touch, emotion, and so on. In a memory-attribute model, memories are
complex representations containing several attributes (Tulving & Watkins, 1975;
Underwood, 1983). When we are introduced to someone, we may later recollect attrib-
utes that are auditory (e.g., the sound of their voice), visual (their appearance), olfactory
(maybe a fragrance), tactile (a handshake), and emotional (tension, if this is your new
boss). Each attribute may have a different rate of forgetting. The acoustic attribute (e.g.,
hearing the word CAP) may fade faster than the semantic attribute (the meaning of CAP;
Wickelgren, 1970). All of the attributes may be forgotten from a single memory but forgot-
ten at faster or slower rates relative to each other (see Figure 8.8).
What otherwise appears to be forgetting from short-term memory may instead be
the rapid forgetting of certain stimulus modalities, such as sound (curve g in Figure 8.8).
226 Short-Term Memory

Trace Strength
b

g f e d
Seconds Minutes Days
Retention Interval

Figure 8.8
Hypothetical Curves of Forgetting for Different Attributes of a Memory. Different dimensions (e.g., auditory,
visual, tactile, emotional) may be forgotten at different rates, corresponding to what is otherwise described as
short-term memory (e.g., attribute f ) or long-term memory (e.g., attribute a).

What appears to be more persistent long-term memory may be the slower forgetting of
other stimulus modalities, such as image or emotion (curve a).

Neuropsychological Dissociations of Two-Memory Systems

Neuropsychological studies of individuals with impairment of auditory-verbal short-term
memory indicate that they have different sites of brain injury than do patients without
STM losses. In the last chapter, we described H. M., who had normal STM. His digit span
had declined somewhat in his later years to about five—not bad for an older gentleman.
K. F. is just the opposite, being severely impaired on memory span and distractor tasks,
particularly for verbally presented items (Shallice & Warrington, 1970). H. M. and other
“amnesic syndrome” patients have damage to the hippocampus and related structures
deep within the temporal lobe. K. F. suffers from injury to a different area, one roughly on
the border between the temporal and the parietal lobes of the left hemisphere. The dis-
covery of other patients like K. F. has led to the naming of a “short-term memory” syn-
drome, referring specifically to impaired auditory-verbal STM (Shallice, 1988).
Studies of animal memory also show dissociation between brain areas necessary for
remembering over the short term versus the long term. For instance, reward learning in
animals, a procedural learning task, is not controlled by the hippocampus. In a short-term
memory task, monkeys with hippocampal lesions tried to remember which of two objects
had been shown most recently. The lesioned monkeys could not do this. In a long-term
memory task, the lesioned animals were able to learn which of two objects was consist-
ently paired with food across trials. In both cases, food was given as a reward for choice
of the correct object and withheld for choice of the incorrect object (Mishkin & Appenzeller,
1987).
 Applications 227

One Memory or Several?

The hypothesis of a single memory system has not received as much acceptance as the
dual-store theory (Healy & McNamara, 1996). The modal model, embodying STM and
LTM, is the basis of much physiological research (Zola-Morgan & Squire, 1993).
Neuroscientists are delineating brain areas that are active during short-term retention,
and which are separate from those involved in LTM. Finally, the concept of STM has pro-
vided valuable insights into learning and memory in general. Any single experiment or
observation does not prove the existence of separate memory systems. A familiar phrase
attributed to the psychologist Kurt Lewin is, “There’s nothing so practical as a good
theory.”
As has been noted frequently, memory is not a single thing. The alternative ways of
partitioning memory often cause confusion in comparing one theory to another. Box 8.3
addresses some alternative senses of “short-term memory.”

APPLICATIONS
What purpose does a short-term or working memory serve? Does it do or allow certain
functions that are not easily accommodated by long-term memory? Among several pos-
sible roles are aiding language comprehension and problem solving.

BOX 8.3 SHORT-TERM MEMORY, WORKING MEMORY, EXECUTIVE FUNCTION:

WHAT’S THE DIFFERENCE?

I often hear people refer to any forgetting of recent events as a short-term memory problem.
What they are usually referring to, however, is forgetting that takes place over hours or days.
Technically this is not STM or working memory as discussed in this chapter. These forgettings
could be described as problems with recent memory, which is a nontechnical term, but
possibly should be added to describe everyday lapses than encompass recent events.
Short-term memory is the verbal rehearsal buffer. It is usually tested with the distractor
task or the memory span task, to measure the retention of strings of three to seven digits,
letters, or words.
Working memory is an expansion of STM. WM includes a verbal store; a visuospatial
store; and control processes to select and allocate resources. WM is assessed with a dual-
task that requires doing two things simultaneously, such as remembering visual and verbal
items, or remembering and mental problem solving.
Executive function is an elaboration of Baddeley’s central executive. The core functions of
the central executive are: focusing attention on a relevant task; resisting distraction from
irrelevant stimuli; retaining information in working memory; inhibiting responses; and flexibly
shifting strategies.
The phrase executive functions is used in a much broader sense in education and neu-
ropsychology. Here the term may encompass aspects of self-control (self-discipline), self-reg-
ulation (monitoring goals and performance), emotional control (reactions to frustration),
planning, and creativity.
228 Short-Term Memory

Comprehending and Using Language

Language comprehension, through reading or listening, or watching signed symbols,
takes time: Words occur one-by-one in reading and hearing. The first words and phrases
of a sentence need to be remembered until the end of the sentence in order to compre-
hend the entire thought the speaker (or writer or signer) is expressing. For example, what
is the meaning of this sentence? HE STRODE ACROSS THE COURT AND PROTESTED
VIGOROUSLY TO THE JUDGE THAT HIS OPPONENT WAS INFRINGING ON THE RULES
BY USING _________. You have probably formed some idea even without the missing last
words, but that meaning could be different from your expectation. One ending is (he
protested) AN ILLEGALLY STRUNG TENNIS RACKET. Another is (he protested)
INADMISSIBLE EVIDENCE. Each ending gives a different meaning to the first part of the
sentence. Comprehension of either ending requires memory for the lengthy early portion
of the sentence. Maybe this is what short-term memory is for: to provide continuity in
reading and listening.
Many studies have demonstrated the correlation between working memory and
reading comprehension (e.g., Daneman & Merikle, 1996). In an early study, Daneman and
Carpenter (1980) used a dual-task procedure in which students read a series of unrelated
sentences out loud and also tried to remember the last word in each sentence. This is a
dual-task procedure, in which both tasks (reading and remembering) use a single verbal
memory store. For example, the following sentences might be read:

THE POLICEMAN ATE THE APPLE.

THE GIRL SANG THE SONG.
THE RAIN BEAT ON THE WINDOW PANES.

After a few sentences (the exact number varying over trials), the experimenter requests
recall of the final words. In this example, these are APPLE, SONG, and PANES. Good
readers had a longer memory span (i.e., recalled the final words from longer sets of sen-
tences) than did poor readers. Dual-task performance in older students also correlates
with their verbal SAT scores and standardized reading scores (Daneman & Merikle, 1996).
By contrast, simpler measures of STM capacity, such as the word span, correlate less
well with reading comprehension. Daneman and Carpenter say that a simple test of
memory span assesses only the storage capacity of STM, and not the combination of
capacity and processing that occurs during reading. Consider an analogy of reading com-
prehension to a conveyor belt. As words come down the line, the meaning must be
abstracted and transferred into long-term memory. Some students simply cannot keep
up: Accessing the meaning of some words on the conveyor takes too long, causing other
words on the conveyor to be missed or forgotten (Perfetti & Lesgold, 1979).

Problem Solving
Mental problem solving requires attention, encoding, storage, and manipulation of infor-
mation, all processes for which working memory is well designed. Dark and Benbow
(1990) studied memory for arithmetic word problems in mathematically or verbally preco-
cious seventh graders. An example problem might read: “The entertainment portion of a
 Summary 229

30-minute TV show on Tuesday night lasted four minutes longer than four times the por-
tion devoted to commercials. How many minutes of commercials were there?” Solving
such problems requires retention and manipulation of the numerical facts. These children,
who had longer than average word and digit memory spans (see Figure 8.4), were better
at remembering word problems. The high ability children created more “compact encod-
ings” of the information, which is much like Miller’s idea of chunking to pack more infor-
mation into an item.
Problem solving requires translating the verbal descriptions into mathematical propo-
sitions (the “four minutes longer than four times the…commercials” portion). In the
preceding example, commercial duration could be expressed as x, and the entertainment
duration as 4x + 4, and the sum of these two equals 30 minutes. Or 5x + 4 = 30. Relational
statements pose particular difficulties for students, including college students.
Undergraduates who had longer memory spans accurately recalled more relational state-
ments in math word problems (Cooney & Swanson, 1990).
Efficient use of working memory also means deciding what not to rehearse, such as
the fact that the show in the problem aired on Tuesday evening. Better problem solvers
were able to forget more irrelevant statements (Cooney & Swanson, 1990).
Working memory’s central executive allocates attention and is involved in planning
and decision making. These traits appear to be disturbed with injury to the frontal lobes,
producing a set of symptoms that has been labeled the “dysexecutive syndrome”
(Shallice, 1988). For instance, short-term memory is impaired because the individual is
easily distracted. Perseveration, or persisting in one mode of behavior, is manifested by a
difficulty in switching between tasks (e.g., from sorting a deck of cards by color, then
sorting by suit). Finally, devising or following a plan is disrupted.
Shallice and Burgess (1991) described a naturalistic test of executive dysfunction by
frontal-lobe-injured patients. Their participants were given a list of errands in a shopping
mall. These tasks required memory (e.g., what to buy), planning (sequence errands effi-
ciently), retrieval (remembering when each errand occurs), and information manipulation
(there were limits on spending and on time to complete the errands). The patients made
many more errors than did the control participants. Memory failures often occurred as a
by-product of planning errors: failing to sequence activities efficiently or establish retrieval
cues (e.g., remember to deposit a letter in the mailbox that is just outside the bakery).

SUMMARY
Short-term memory (STM) refers to memory that is both brief and of limited capacity. In
the laboratory STM is measured in seconds, as distinguished from long-term memory,
which is typically measured in minutes, hours, and days.
The distinction between these two memory systems has long been a part of psychol-
ogy’s history. William James distinguished between primary memory (the conscious
present) and secondary memory (what has left immediate consciousness and must be
retrieved). Atkinson and Shiffrin developed what has become the modal (or average) mul-
tistore theory of memory. The next stage in the evolution of STM was Baddeley’s working
memory model, which has separate short-term stores for verbal and visual information,
as well as a central executive to control attention.
230 Short-Term Memory

Short-Term Memory Tasks

Short-term memory is often tested using the distractor technique and the memory-span
task. The Brown–Peterson distractor task is used to measure the duration of STM. To-
be-remembered items are presented, the subject performs a distractor task such as
counting, and then attempts to recall the target items. The amount recalled drops off
dramatically after as little as 9–15 seconds of distraction. STM fades quickly in the absence
of rehearsal. However, forgetting from STM is also caused by proactive interference from
previous test items.
The memory span test measures of the number of items that can be recalled after a
single presentation. This is the capacity of short-term memory. Span is limited to about
seven items (plus or minus two). This span is composed of probably four items actually in
STM, and the others are in LTM. Memory span can be increased by chunking or coding
to increase the amount of information contained within each item.
Span is also determined by word length: More two-syllable words can be remem-
bered than four-syllable words. Memory span may be governed by articulation rate, as
longer words take longer to speak than shorter words.

Characteristics of STM
The verbal short-term store of Atkinson and Shiffrin and the phonological store of
Baddeley’s working memory are characterized by acoustic encoding: Words are remem-
bered as they sound.
STM has a limited capacity, with estimates ranging from two to seven items. Capacity
can be increased by chunking or coding to increase the amount of information in each
item.
STM has a limited duration, lasting from several seconds to less than a minute in the
absence of rehearsal.
Forgetting occurs due to displacement of an item from short-term memory.
Rehearsal of information in STM allows transfer to LTM. Verbal rehearsal may be nec-
essary in forming some kinds of long-term knowledge, such as in vocabulary learning.
However, maintenance of information in short-term memory does not guarantee entry into
long-term memory. Also, brain-injured subjects like K. F., who has a limited verbal short-
term memory, can nevertheless learn and retain long-term memories at normal rates.
Short-term retention can be demonstrated in nonverbal modalities for visual and
olfactory stimuli, for spatial locations, and for actions.

Working Memory
Alan Baddeley’s theory of working-memory (WM) divides STM into several components.
The phonological store is basically verbal short-term memory, having characteristics
already described. The visuospatial store retains visual images and performs spatial
manipulations. The working-memory approach assesses dual-task performance, or doing
two tasks at once. There is more interference if both tasks use the same short-term store
(e.g., both are verbal) than if the two tasks use different stores (e.g., simultaneous verbal
and a visual tasks).
 Summary 231

The central executive acts as a sort of manager to manipulate, organize, and plan the
routing of information. Executive functions are higher-level cognitive skills used to control
and coordinate other cognitive abilities and behaviors. These include updating, inhibition,
shifting, and attention control.
The episodic buffer acts as a bridge between short-term stores and long-term mem-
ory. For instance, knowledge in LTM (such as arithmetic rules) can be applied to the items
in STM (adding or subtracting the numbers there).
Interest in WM extends beyond memory research. WM’s control over attention sug-
gests a connection to consciousness and conscious processing. WM’s capacity to tem-
porarily hold and manipulate information is cited as an evolutionary adaptation that led the
development of language and culture.
There are individual differences in working memory functioning. For instance, work-
ing-memory tasks have been employed to study aging and dementia. Aging is associated
with slower processing, as shown by slower encoding of target items into memory, and
slower switching between tasks. Individuals with Alzheimer’s may be impaired at
dual-tasking, even if each task can be performed competently alone.
Anxiety impairs working memory by limiting attention by the central executive, and
negative self-talk occupies space in the phonological store. Working memory deficits
associated with anxiety have been found in the performance of mathematics and reading
comprehension.
Multitasking is an obvious example of working memory, both when it functions well
and when WM breaks down.

Is There Really a Separate Short-Term Memory?

The necessity of hypothesizing separate STM and LTM remains unproven to some theo-
rists. Can memory be parsimoniously accommodated by a single-memory system?
One alternative hypothesis is that there is a single set of memories, but memories
differ between those that are currently active (STM) and those that are inactive (LTM).
Another hypothesis is that memories are multidimensional and that each dimension has
a different rate of forgetting. What appears to be forgetting from short-term memory is
rapid forgetting of certain stimulus modalities, such as sound. What appears to be long-
term memory is the more enduring retention of other stimulus modalities, such as sight
or smell.
Strong evidence in favor of an STM–LTM distinction comes from neuropsychological
studies. Certain areas of the brain are damaged in individuals with impaired verbal short-
term memory. Studies of animals also show a dissociation between brain areas neces-
sary for remembering over the short term versus the long term.

Applications
What is the purpose of short-term memory? One use is for language comprehension. The
span of working memory correlates with measures of reading comprehension. Dual-task
performance, such as reading sentences and remembering the final word of each sen-
tence, correlates with verbal SAT scores and standardized reading scores.
232 Short-Term Memory

Several distinct working-memory processes contribute to problem solving. Information

must be encoded into working memory, other information must be retrieved from LTM,
steps in problem solution must be sequenced, and irrelevant information must be
excluded. The central executive plays a central role in organizing these activities neces-
sary for problem solving. By contrast, injury to the frontal lobes produces a “dysexecutive
syndrome,” characterized by excessive distractibility, perseveration, and lack of planning.
Finally, numerous theories of long-term learning have hypothesized that STM is a
gateway into long-term memory. Versions of consolidation theory (a biological theory) and
dual-store theory (a cognitive theory) state that interference with short-term processing
can prevent long-term memory formation.
 CHAPTER

9
Encoding

CONTENTS

Separating Encoding from Learner Variables 247

Retrieval 234 Incidental Versus Intentional
Some Basic Variables in Encoding 234 Learning 247
Elaborative Rehearsal 235 Incentives 248
What Exactly Is Elaborative Arousal 249
Processing? 236 Emotions and Encoding 254
Imagery and Memory for Pictures 239 Schemas 257
Meaningfulness 240 Metamemory 258
Presentation Variables 241 Encoding: Summing Up 261
Testing Effects 241 Applications 261
Isolation Effects 243 Academic Learning and Encoding 261
Spacing Effects 245 Summary 262

Encoding refers to the acquisition of information: the formation of a memory trace. A

textbook, such as the one that you are now reading, can cite references to learning under
ideal conditions: simple materials, minimal distractions, and students mentally prepared
to learn. Yet how often will such ideal learning conditions occur in the real world? Recall
your first year of college, taking what seemed like too many courses, each requiring a
different form of studying. Long hours and lots of stress. Or consider having to learn
under particularly trying conditions: the U.S. Marine Corps boot camp at Parris Island,
South Carolina. The physical training and emotional pressure the recruits experience are
well known from film and television. What many of us are not aware of is the academic
learning that occurs over 11 weeks of training. Each recruit attends classes and is issued
a set of textbooks totaling over a thousand pages. Studying occurs at odd times of the
day (or night)—whenever opportunity allows. The recruits are physically fatigued when
they study. They are under intense emotional stress (read: FEAR). The recruits are moti-
vated to learn; but some of the material might appear more like Ebbinghaus’s nonsense
syllables than meaningful prose. How do variables such as time of day, emotional arousal,
incentive, and meaningfulness affect learning? This chapter will consider these and other
more usual laboratory variables that affect encoding.

DOI: 10.4324/9781003227090-9
234 Encoding

SEPARATING ENCODING FROM RETRIEVAL

The division of information processing into stages of encoding and retrieval is in some
ways arbitrary. Successful recall means both stages worked; forgetting does not indicate
which stage failed. One means of separating stages is to experimentally manipulate var-
iables during encoding, while holding constant conditions at retrieval.
Neuroscience methods offer another means of separating stages. Brain activity at
each stage of memory can be assessed with neuroimaging. The differences in activity
between encoding (e.g., while studying a list of words) versus retrieval (e.g., when recog-
nizing previously studied words) may indicate different brain areas are involved at each
stage. In one such imaging study, the brain activity during study of each individual item
was measured. This activity was then correlated with whether those items were recalled
or forgotten. Certain areas in the left hemisphere were found to be more active during the
study of words that were remembered and were less active during study of words that
were later unrecognized (Wagner et al., 1998). Similarly, certain areas of the right hemi-
sphere were more active during encoding of photographs of scenes that were subse-
quently recognized than during the study of photographs that were not recognized
(Brewer, Zhao, Desmond, Glover, & Gabri, 1998).

SOME BASIC VARIABLES IN ENCODING

In the laboratory, encoding into episodic memory is most-often studied. Episodic memory
is remembering information based on its occurrence at a particular time and place. The
free recall of a list of words, sentences, or pictures are tests of episodic memory: recall-
ing the items you just saw, here in the laboratory. A number of factors are reviewed in the
several following sections: rehearsal, imagery, and arousal, just to name a few. These and
other factors are summarized for reference in Box 9.1.

BOX 9.1 SUMMARY OF SOME FACTORS RELATED TO ENCODING AND THEIR

DEFINITIONS

Characteristics of the To-Be-Learned Material

Meaningfulness
Meaningful words are ones having some combination of the following: a high fre-
quency of occurrence in language and print, have many associations to other items,
and are more imageable.

Concrete/Abstract Words
Words that refer to actual, physical things, and that tend to be imageable and have
a number of associations versus words that refer to abstract ideas.

Imagery
The degree to which items can be mentally imaged. Pictures or objects actually
seen are better remembered than their names.
 Some Basic Variables in Encoding 235

Factors in the Presentation of To-Be-Remembered Materials

Testing Effect: Taking a practice test on recently studied material is better than
additional study of the material.
Spacing of Repetitions: If material is repeated, spacing or interleaving the repeti-
tions leads to better retention than if the repetitions are massed.
Isolation Effect: An unusual or distinctive item that stands apart from the other
items in a series.

Cognitive Strategies
Elaborative Rehearsal: Thinking about the material in such a way as to require
more cognitive effort, to activate more associations, or to produce a more
distinctive memory representation.

Subject Factors
Arousal: Levels of mental or physiological arousal, such as those associated with
circadian rhythms, stimulant drugs, or emotional arousal.
Incentives: Offering explicit rewards for learning or remembering. The incentives
themselves do not enhance remembering, but they may lead to cognitive
elaboration that does.
Incidental Learning: Learning occurs without intention to the degree that the
material is cognitively processed.

Elaborative Rehearsal
The role of rehearsal has already been mentioned in previous chapters. Generally, the
number of words that are recalled from a list correlates with the number of rehearsals the
words received (Chapter 6). This is consistent with the notion that rehearsal promotes
better representation in long-term memory (Chapters 7 and 8). However, all rehearsal is
not the same; there are qualitative differences in rehearsal.
One important distinction is between elaborative rehearsal and maintenance
rehearsal. The two terms derive from the levels-of-processing approach of Craik and
Lockhart (1972), which posits that information can be processed to a greater or lesser
extent, and thus by analogy to different “depths,” along a continuum from shallow to
deep processing. Maintenance rehearsal is a form of shallow processing, a recycling of
information in order to keep it available in short-term memory or the phonological store.
We use maintenance rehearsal to temporarily remember a phone number, simply repeat-
ing the number until it is no longer needed. Maintenance rehearsal can be effective for
short-term retention, and our research participants will use maintenance rehearsal if they
know they will be allowed uninterrupted rehearsal until the time of recall (Wixted, 1991).
Elaborative rehearsal is processing in which to-be-remembered material is related
to other information. There is an active or deliberate attempt to cognitively interact with,
reflect on, or use the to-be-remembered information. Elaborative rehearsal represents
deep processing. To enter a phone number into long-term memory, we might look for a
236 Encoding

pattern in the numbers; see if it reminds us of an already known number; or recognize a

familiar area code.
The idea of elaborative rehearsal is a central theme in this chapter: Many of the vari-
ables that affect encoding can be interpreted as instances of elaborative processing.

What Exactly Is Elaborative Processing?

The standard procedure for manipulating processing depth is to present a list of words and
have the subjects answer a question about each word as it occurs. Different types of ques-
tions require different levels of processing of the target items. For instance, given the tar-
get word BONE, the various questions asked might be, “Does it contain the letter e, does
it rhyme with train, is it an animal?” These questions require different kinds of processing:
the surface form or appearance of the word; the sound of the word; or the meaning of the
word. Later, the subjects are given an unexpected test of memory. Words that received
deeper processing during presentation are more often recalled (Craik & Tulving, 1975).
The analogy of elaborative processing to deep processing is helpful, but this does not
specifically define what elaboration is. At least three hypotheses have been offered: elab-
oration, distinctiveness, and effort (Horton & Mills, 1984).
Elaboration refers to expanding a newly forming memory trace. If you have to remem-
ber the word AARDVARK, you might try to think of any associations to the word or any
facts you know about aardvarks. Figure 9.1 illustrates the sets of possible associates that
could be activated by the target words CRAB, FACT, and LAMP (Nelson & Schreiber,
1992). The words CRAB and FACT have larger sets of associations that could be activated
than does the word LAMP. CRAB and LAMP, however, could evoke specific mental pic-
tures or images, whereas FACT does not. Elaborated traces activate more connections or
associations to other memories. This increases the number of potential retrieval cues.
Alternatively, elaborative rehearsal may increase the distinctiveness of the mental
representation. A distinctive memory stands out from other memories, and will suffer
less interference from other memories during retrieval. According to the distinctiveness
hypothesis, shallow processing may sometimes lead to good retention, as long as the
memory representation is distinctive. For instance, words that are unusual looking (such
as LYMPH, KHAKI, or PHLEGM) are recalled better than are orthographically common
words (such as LEAKY, KENNEL, or PRIMATE), even when the words are matched for
other factors, such as meaningfulness and frequency of occurrence (Hunt & Elliot, 1980).
The third possibility is that elaborative processing requires more cognitive effort. The
amount of mental effort expended determines retention. Mnemonic devices exert cogni-
tive effort. Making up a rhyme or a first-letter mnemonic to remember a set of items
takes work.
Cognitive effort can be measured by the impairment in performance on a second task
done simultaneously with the memory task. For example, as the subjects are answering
questions about to-be-remembered words, they must also push a button each time a
tone sounds. An increase in the cognitive effort required to answer the questions leads
to slower reaction times to the tone (Eysenck & Eysenck, 1979). Deeper processing
leaves less capacity available for the tone-detection task. The results of a dual-task proce-
dure are shown in Figure 9.2, which show word recall reaction times as a function of
depth of processing. The left panel shows that more words were recalled when the
 Some Basic Variables in Encoding 237

Figure 9.1
Hypothetical Elaborative Encoding of Three To-Be-Remembered Words. The size of the associated sets
vary. Each word can activate associated words and ideas. In addition, some words may evoke a mental image
of the word’s referent.
Source: From “Word Concreteness and Word Structure as Independent Determinants of Recall,” by D. L. Nelson and T. A.
Schreiber, 1992, Journal of Memory and Language, 31, p. 240. Copyright © 1992 by Academic Press. Reprinted with
permission.
 238 Encoding

Primary Task: Word Recall Secondary Task: RT

1.2 500

400
0.9
No. Recalled

Milliseconds
300
0.6

200

0.3
100

0 0
Deep Shallow Deep Shallow RT Only
Level of Processing Level of Processing
A B

Figure 9.2
Dual-Task Performance. The primary task was to remember the word list. A secondary task was to push a
button whenever a tone sounded. Deep processing of the words produced better recall but longer times to
react to the tones.
Source: Eysenck and Eysenck (1979, p. 479, 481).

participants made a deeper judgment about the word (“Is it an animal?”) rather than a
shallow judgment (“Does it contain a letter e?”). The right panel shows that it took longer
to react to a tone that occurred during the deep judgment. The slowed reaction times are
an indicator of the cognitive effort being expended on the judgment task. (The RT-only
control refers to detecting the tone in the absence of a question).
The three definitions of elaborative processing can be illustrated by studies of mem-
ory for faces. Subjects in Bower and Karlin’s (1974) study were shown high school year-
book photos. In one condition, the subjects made shallow judgments about the faces,
e.g., “Is this person wearing glasses?” In the other condition, they were asked to make
character judgments for each face, e.g., “Does this person look honest?” Character judg-
ments require deeper processing, and led to more correct recognition of faces than did
the shallow judgments. If you want to remember a face, ask yourself, “Would I buy some-
thing online from this person?” Following on Bower and Karlin’s study, Winograd (1981)
had participants study faces for distinctive features. For instance, “Does this person have
a long nose?” The more features that were searched, the better the face was remem-
bered. The focus on facial features increased distinctiveness. What about the role of cog-
nitive effort? With deep-processing instructions, participants made more eye movements
when inspecting the pictures, implying more processing effort (Bloom & Mudd, 1991).
So, why does elaborative processing enhance face memory? There is experimental sup-
port for all three hypotheses of elaboration, distinctiveness, and cognitive effort.
It is also the case that deep-processing subjects report more interest and involve-
ment in the task, whereas the shallow-processing groups find their task boring and unde-
manding (Sporer, 1991).
 Some Basic Variables in Encoding 239

Limitations of Elaborative Processing

In general, elaborative processing leads to better encoding. Are there any negative effects
of elaborative processing? Are there tasks for which elaboration does not help?
Verbal overshadowing is one case in which elaboration may be harmful. The verbal
description of nonverbal information, such as word descriptions of pictures or faces, can
lead to less accurate retention. In essence, we remember the words we used to encode
an item. In describing a face we name features that can be verbalized, such as eye color
or the shape of the lips, and not features that are difficult to articulate (such as the config-
ural arrangement of facial features) that may be more important for recognizing the face
(Schooler, Ryan, & Reder, 1996).
Verbal overshadowing has been shown in other situations, such as remembering the
taste of specific wines. Novice and expert wine tasters were asked to describe several
wines. Later, the tasters were asked to distinguish the tasted from new wines. The
untrained tasters did worse when they had verbally described the earlier wines. Not
having a sophisticated vocabulary for describing wine, the novices tended to remember
what they had said rather than remembering the taste. Experts were not impaired by
verbalization; presumably, their vocabulary matched their subtle taste discriminations
(Melcher & Schooler, 1996).

Conclusions about Elaboration

People will probably learn more if we can find ways of making them work harder at
encoding and if they are encouraged to deal with new information in terms of its
meaning and semantic content. Getting students to simply repeat material or
spend more time looking at it is not the most efficient way to ensure they will
learn and remember the content.
(Kulhavy, Schwartz, & Peterson, 1986, p. 122)

This summary reminds us of several aspects of elaboration: effort, meaningfulness, and

active rehearsal.

Imagery and Memory for Pictures

Visual encoding is another factor in encoding memory. We see this both through the use
of imagery in remembering, and in the recall of pictures versus words.
If words are presented for study, concrete words are typically better recalled than
abstract words. Concrete words refer to things that can be readily imagined or pictured
(such as the word CRAB), whereas abstract words refer more to ideas (such as the word
FACT). Concrete words have a memory advantage over abstract words in a variety of
learning situations (reviewed by Nelson & Schreiber, 1992).
Memory for pictures is often remarkably good. In one study, subjects were shown
over 2,200 photographs presented in 2-hour sessions on four days. During testing, 280
pairs of slides were shown and the participants had to decide which picture in each pair
had been seen before. This is a recognition test of memory. Correct choices were made
over 90 percent of the time (Standing, Conezio, & Haber, 1970). Photographs have much
more detail than a printed word, allowing many cues to discriminate old from new photos.
Still, the number remembered is impressive.
240 Encoding

Objects are remembered even better than pictures. Given stimuli displayed in one of
three ways, actual objects were better recognized than were photographs of the objects,
which in turn were better remembered than the printed name of the objects. This is also
one case in which younger students did as well as older ones: fourth graders were as
good as college students in recalling objects (Bevan & Steger, 1971).
Instructing participants to visually image to-be-remembered items such as words
often increases recall (e.g., Roediger, 1980). We saw further evidence for the beneficial
effects of visual imagery with mnemonic techniques such as the peg-word system and
the method of locations (Chapter 6).
Why is picture memory so much better than word memory? One explanation is dual
coding (Paivio, 1969): Pictures can be remembered with two encodings, one visual and
the other a verbal translation of the picture. Another advantage for pictures is that the
memory is more specific and distinctive than is the memory for an unelaborated word. As
shown in Figure 9.1, the words CRAB and LAMP can take on specific images. For exam-
ple, imagine a simple line drawing of a dog. The drawing cannot be too generic; it must
show some kind of dog. A large dog or small? Pointy or floppy ears? This image may
activate other knowledge about particular dogs that the word alone does not.
Depth of processing applies to pictures as well as to words. Students were shown a
photograph of a living room and given orienting instructions to induce different forms of
processing. Some subjects were told to try and find small hidden x’s within the picture, a
form of shallow processing. Another group was told to think about the kinds of actions
they could perform with the objects in the room. The students were then unexpectedly
asked to recall as many objects from the picture as possible. The x-searching group
remembered only about eight objects, but the action group remembered over 20. What
is neat about this manipulation is that the x-searching subjects did closely attend to and
scan the picture. Even so, their recall was poor (Bransford, Nitsch, & Franks, 1977).

Meaningfulness
Ebbinghaus compared his own retention of poetry versus nonsense syllables. He found
that the meaningful material was remembered better than the nonmeaningful. Okay, no
surprise there. But what exactly is meaningful? And how can we increase it?
Meaningfulness has been operationally defined by those variables that correlate
with ease of learning or remembering. Meaningful words have more associations to other
ideas and knowledge. Associations are determined by a free association test. Given a
word as a stimulus, what words are elicited as responses? Another form of association is
to determine the frequency with which a word is given as an association to other words
(Rubin & Friendly, 1986). The word WATER as a stimulus elicits many associations, and it
is itself given as the response to many stimulus words. A meaningful item can be elabo-
rated in memory by drawing on an extensive network of inter-associations. This network
involves semantic memory (general knowledge) and episodic memory (personal
recollections).
Meaningful words are often more imageable (Rubin & Friendly, 1986). A mental image
or a remembered picture can often be generated for meaningful words (DOG, CRAB). (In
attempting to remember an abstract word such as LIBERTY, you might think of the Statue
of Liberty).
 Presentation Variables 241

Meaningful words are familiar. They have a high frequency of occurrence in the lan-
guage: we hear them, read them, and use them often. Meaningful words are also rated
high in ease of pronounceability. This may be related to familiarity and frequency.
Meaningful words are rated higher in emotionality than less-meaningful words (Rubin
& Friendly, 1986). Meaningful knowledge is significant, interesting, relevant, pleasant,
frightening, etc. (insert your own emotional adjective).
This list of characteristics suggests that meaningfulness has to do with prior knowl-
edge: Meaningful material is stuff you already know about. Generally, learners use knowl-
edge they already possess to relate new items to existing knowledge. The implication is
that the more you know, the more you can learn. This truism has been substantiated in
comparing people who are experts in certain domains and novices in those domains. For
example, when shown arrangements of chess pieces on a board, chess masters will
remember them more accurately than less experienced players. This is particularly true if
the pieces are in a meaningful arrangement (Gobet & Simon, 1996). When given lines of
computer program to memorize, advanced student programmers learned them more
readily than did students of intermediate ability. Again, the advantage for the experts was
dependent on the program lines being in a logical sequence, and the advantage virtually
disappeared when the program lines were scrambled (McKeithan, Reitman, Rueter, &
Hirtle, 1981).
These expert individuals had domain-specific knowledge that was meaningful to them.
There are certain subject areas (or domains) in which we are knowledgeable. We encode
and remember new information easily from areas with which we are familiar. For instance,
students who are knowledgeable about baseball can remember new baseball facts better
than less knowledgeable students. Students knowledgeable about music remember more
new facts about music than about baseball (Kuhara- Kojima & Hatano, 1991).

PRESENTATION VARIABLES
In addition to features of the to-be-remembered material, the manner or form in which
new information is presented affects encoding. For instance, the frequency of presenta-
tion increases retention. Frequency of repetition is a variable so-well known and, maybe
so obvious, that there is not much else to say about it. (Other than to point out that there
are qualifications and limitations in the generality of repetition effects).

Testing Effects
Say you have a big test tomorrow and you are planning your study strategy. After studying
the material once, should you study it again, or should you instead take a practice test on
the material? Students would probably think that more studying is better. After all, it is
called “studying” for a test. However, a significant finding from current research is that
taking a test on the material is often more beneficial to remembering (on tomorrow’s
exam) than is additional study of the material. This is referred to as the testing effect:
there is greater benefit in taking a practice test over additional study.
In the prototypical design for a study of the testing effect the research participants are
exposed to the material two (or more) times, and then take a test on the material. In the
standard condition, students study the material, and then study it again. We can label this
242 Encoding

condition S–S (for study-study). In the study-test condition, the students study the mate-
rial and then take a practice test on the information. This is condition S–T. Later, a final test
is given to both conditions to see who remembers more. We are interested in perfor-
mance on the final test. Thus the design of such studies is as follows:

Standard Condition, S–S: study once…study again…final test

Tested Condition, S–T: study once…practice test…final test

Roediger and Karpicke (2006) used this design to assess learning of information in the
form of textbook-like paragraphs. Some college students studied the material twice;
other students studied once and then wrote as much as they could remember. When the
final test was given two days later, the S–T condition remembered more than did the
study twice condition. Specifically, Group S–T had 68 percent correct on the final test
whereas Group S–S got 54 percent. (The results of this experiment were shown in
Figure 1.3 of Chapter 1).
Roediger and Karpicke (2006) extended the above design to compare a condition in
which students studied four times (S–S–S–S) versus a condition in which the students
studied once and then made three practice attempts at recalling the information (S–T–
T–T). The final test was given one week later to measure long-term learning of the mate-
rial. Correct recall on this final test showed that more testing led to more remembering.
Testing yourself on what you have just studied may help you remember better in the long
run than simply studying more. This may sound counterintuitive, and indeed the students
in this experiment thought so also. When questioned, the students who studied the
material four times gave higher confidence ratings that they would remember more on
the test; students who had studied only once gave lower ratings. However, a week later
the S–T–T–T group recalled more.
Why is testing better than additional studying? Several explanations have been con-
sidered. (You should know by now that there is never just one explanation, the correct
one). One explanation is that taking a practice test is just another version of studying.
After all, the material is presented a second time, although now in the form of questions
and the students’ answers. However, testing does more than simply provide a second
study exposure. Memory in the S–T condition is not simply as good as the S–S condition;
it is significantly better.
A more likely explanation is that the test provides a more challenging restudying of
the material than passively reading it over a second time. Answering test questions
requires more cognitive effort than rereading.
A third possibility is that the practice test mimics what students are required to do on
the final test, that is, retrieve the information from memory. The cognitive activity on the
practice test—retrieval—is a closer match to that on the final test—retrieval. (The testing
effect is now often called retrieval practice effect, acknowledging the importance of
retrieval in the practice test).
The testing effect raises a number of procedural questions about its generality. What
about retention of the material that was not tested on the practice tests? Well, practice
does aid recall of the untested information, although not as well as tested information
(Chan, McDermott, & Roediger, 2006).
 Presentation Variables 243

Testing usually implies feedback: Which questions did you get right or wrong? The
practice test alone, even without feedback, was still sometimes better than restudying a
second time (Kang, McDermott, & Roediger, 2007). The best learning occurred when the
participants took a practice test and received feedback (i.e., the correct answer).
There can be negative effects of testing. On the final test some of the multiple-choice
alternatives on the final test look familiar because, well, they are familiar. They were the
wrong answers on the practice test but are chosen on the final test because they look
familiar.

Isolation Effects
A commonly held belief about memory is that unusual or distinctive material is better
remembered. Helena von Restorff, a Berlin Ph.D. of the 1930s, studied this phenomenon.
In presenting a list of to-be-remembered items, one item in the list was made distinctive.
For example, if the other words were printed in black, one word was printed in red. In a
list of words, a number would be inserted as an item. The distinctive item, distinctive in
the context of the list, was remembered better. This finding was initially called the von
Restorff effect, but the more recent literature refers to it as the isolation effect: In a
series of items, an unusual item or an item presented in a distinctive manner is especially
remembered. A real-world example of the isolation effect was a television commercial
that contained no sound whatsoever; the message was signed by an actor and captioned
at the bottom of the screen. The sudden silence stood out in the midst of the other
high-volume ads.
An example of the isolation effect is a series of studies conducted by Detterman
(1975). A list of 15 words was presented auditorily. The distinctive item in the eighth
position was played much louder than the other words. The results are shown in Figure 9.3,
which plots word recall across serial positions. The graph illustrates the enhanced reten-
tion of the isolated item in the eighth position as compared to the control word (presented
at normal volume) in the same position.
The isolation effect could be attributed to enhanced encoding. The isolated item
receives more attention, more rehearsal, or more elaboration in memory (Waddill &
McDaniel, 1998). Measures of brain waves indicate that subjects orient more to the iso-
lated item (Fabiani & Donchin, 1995). However, the isolated item might be remembered
better because it is more distinctive in memory during retrieval. It stands out against a
blur of memories of many similar items (Hunt, 1995).
An isolation effect can occur in remembering unusual behaviors by friends or family.
Someone acting out of character might be especially remembered. In a study, hypotheti-
cal individuals were described by lists of personality traits or characteristics. For instance,
one person was said to be intelligent, clever, quick, and smart. After this, 20 to-be-remem-
bered behaviors for each hypothetical person were presented. Most were congruent with
the initial personality labels, for example, the person had won a chess tournament or liked
reading history. But importantly, an incongruent behavior was also presented: they failed
math twice, or was always getting lost. During memory testing, the incongruent behavior
was especially recalled. For example, 50 percent of the congruent behaviors were remem-
bered versus 77 percent of the incongruent behaviors (Hastie & Kumar, 1979).
244 Encoding

80

70
Isolation Item List

60

Percent Correct
Control List
50

40

30

20

10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Serial

Figure 9.3
The Isolation Effect in the Recall of a List of 15 Words. A distinctive item occurs at the eighth position in the
experimental list. The control list had no distinctive item. The distinctive item is well recalled, but there is some
forgetting of the items that preceded and followed it in the list.
Source: From “The von Restorff Effect and Induced Amnesia: Production by Manipulation of Sound Intensity,” by D. K.
Detterman, 1975, Journal of Experimental Psychology: Human Learning and Memory, 1, p. 626. Copyright © 1975 by the
American Psychological Society. Reprinted by permission.

Other research on remembering the entire list show that the von Restorff manipula-
tion produces a more complex and interesting pattern of results. Whereas the isolated
item is well recalled, this comes at the expense of memory for items before and after it
in the list. For example, suppose the following partial list is presented: table, dog, tree,
grass, COLUMBUS, school, coffee, soup. The enhanced recall of the distinctive item
(COLUMBUS) impairs recall of the words just before it (GRASS and TREE). This can be
seen in the already described Figure 9.3. In this list the distinctive item was in the eighth
position. The words occurring before the loud item, positions 5–7, were less recalled in
the isolation list than in the control list (the shaded portion before the distinctive item).
In addition, the distinctive item impaired remembering the words that followed, posi-
tions 10–12 as compared with the control list (the shaded portion after the distinctive
item). In our example, COLUMBUS might impair recall of SCHOOL and COFFEE.
Similar effects have been shown with other forms of distinctiveness. For instance, if
an obscenity) is presented as the isolated item, the obscenity is remembered but the
words before and after it in the list are not (Mackay et al., 2004).
 Presentation Variables 245

Spacing Effects
What student has not heard the admonition, “Study every night, don’t wait and cram at
the last minute”? So, you are already familiar with the spacing effect: Spaced practice (or
distributed practice, as it is sometimes called) leads to better learning than does massed
practice. If something is to be studied multiple times, it is better to distribute the repeti-
tions than to mass them close together. Like the serial-position curve, this is an amazingly
general fact about learning (Dempster, 1996).
In a typical spacing experiment, a list of to-be-remembered words (TBR) is presented.
Some words are repeated in the list. The two presentations (P1 and P2) of a word can be
massed, with no other words in between; or P1 and P2 can be separated by other words
on the list. The participants are aware that words can be presented more than once, thus
negating an isolation effect for a repeated item. Across subjects the same words can be
either massed or spaced, thus controlling word differences.
An example of a partial list is shown below.

INSECT HAIRPIN BARREL BARREL VIRTUE SHADOW METAL BARREL FOREST

| | |
P1
1 Massed P 2 Spaced P 2

The repetition of the word BARREL would be massed or spaced in different conditions of
the experiment.
Why does spacing produce better retention? Several theories have been offered
although no one explanation seems to accommodate all of the available data. Yet each
explanation has led to novel insights into the spacing effect.

Hypothesis 1: Attention-Deficit
Is the massed practice deficit due to less learning of the first presentation of the second?
Localizing learning on the two presentations can be determined by tagging them so that
they are somewhat different. For example, the target word can be presented in different
modalities at P1 and P2. The first occurrence might be auditory and the second is visual.
(Or in different colors, or with different backgrounds). By then asking participants to recall
not only the items but also whether the word had been seen or heard, we find that it is
the second presentation less-well remembered with massed repetitions (Hintzman,
Block, & Summers, 1973).
A repetition that occurs too soon evokes less attention. This is the attention-deficit
hypothesis: a massed P2 receives less attention than does a spaced P2. We can see
attention-deficit by recording eye movements during presentations of photographs. The
number of fixations decreases to closely-repeated pictures (Ryan, Althoff, Whitlow, &
Cohen, 2000). If participants are allowed to pace themselves through the list by giving
them the remote control, they skip through immediate P2s but spend more time on
spaced P2s (Shaughnessy, Zimmerman, & Underwood, 1972).

Hypothesis 2: Encoding Variability

Recall is enhanced when there are more retrieval cues, or routes, to the target. When the
two study presentations are more widely spaced, they might be encoded somewhat
246 Encoding

differently into memory. For example, given the list FORK, TABLE, SCHOOL, BOOK,
TABLE, the first TABLE may be encoded as a kitchen item and the second as a classroom
item. Recalling either FORK or BOOK might aid retrieval of the target word TABLE. This
explanation is known as the encoding variability hypothesis. Back-to-back TABLEs are
likely to be encoded similarly, both as kitchen items, for example.
Encoding variability suggests that massed repetitions might not be so bad if P1 and
P2 are forced to be encoded differently, just as they supposedly are when repetitions are
spaced. Different encodings can be simulated by using homographs, or words that have
two meanings. For example, FOOT is first presented in the context of “inch,” and the
second time in the context of “toe.” Words presented in different contexts do not show a
massed-spacing decrement (e.g., Gartman & Johnson, 1972). FOOT is well remembered
even though the two presentations were massed. Okay, so this is a contrived manipula-
tion, and the homographs are actually different words (Martin, 1975). However, it illus-
trates the general idea that the same word might be encoded differently at separate times.

Is There an Optimal Spacing Interval?

How long must an interval be to be considered spaced, and therefore beneficial? The answer
depends on how much time there is between the completion of studying and the test. In
general, the longer the interval until the test occurs, the wider the spacing interval during
study should be. The relationship of the two time intervals can be shown as follows:

P1 — Spacing Interval — P2 — Retention Interval — Test

As we move from retention intervals of minutes, characteristic of many laboratory exper-

iments, to retention intervals of days, weeks, and months, characteristic of academic
learning, the optimal spacing interval between separate study sessions must increase
dramatically. Spaced practice intervals of minutes are perfectly effective if the test will
occur at the end of the hour-long experimental session. When the test is a week away,
spacing the initial presentations a day or so apart is better.
The most effective spacing is best described the ratio of the length of the spacing
interval (P1 to P2) to the length of the retention interval (P2 to Test) (Cepeda, Pashler, Vul,
Wixted, & Rohrer, 2006). For instance, spacing study periods one day apart and testing
seven days later, produces a ratio of 1/7, or about 14 percent. That same spacing of one
day but testing 30 days later produces a ratio of 1/30, or about 3 percent (in round num-
bers). Optimal ratios are said to fall in the range of 10–20 percent. So, the longer the
retention interval, the longer the spacing interval should be to produce a ratio within this
range. Further increases in the spacing interval seem to convey no additional benefit. For
instance, spacing intervals of one day, two days, or four days were about equivalent when
the test occurred 10 days after the final studying (Pashler, Rohrer, Cepeda, & Carpenter,
2007).

Conclusions about Spacing

One idea that has emerged from research on distributed practice is that “a little forgetting
may be good for learning” (Cuddy & Jacoby, 1982; Krug, Davis, & Glover, 1990). A second
presentation given too soon gives a false feeling of knowing when in fact the material is
 Learner Variables 247

not known well at all. After a longer spacing interval the item has been temporarily forgot-
ten, or at least fallen out of short-term memory, which should heighten processing of the
repetition.
There is a significant implication from the results of spacing research. You can arrange
a situation to make learning appear to be easy by using massed repetitions; or you could
arrange to make learning lasting by spacing repetitions (Schmidt & Bjork, 1992).
Although the spacing effect is well known, it is not implemented enough in education
(Dempster, 1988). For example, textbooks could be written that repeat certain ideas from
chapter to chapter, thus ensuring repetition and spacing. Academic calendars could be
revised to spread courses over longer intervals than quarter and semester systems to
allow more widely spaced exposure to the topic’s content (Bahrick, 1979).

LEARNER VARIABLES
The learner brings a number of characteristics to the learning situation. Encoding is influ-
enced by the intention to learn, the use of incentives, and levels of alertness and motiva-
tion. Other learner factors such as age and learning style are discussed in Chapter 12.

Incidental Versus Intentional Learning

A paradox involving everyday memory is that we spontaneously remember things with-
out any deliberate intention to remember. You remember what you had for breakfast this
morning or something you heard earlier in the day. This is incidental learning: learning
that occurs without intention to remember, but incidental to processing the information.
At other times we need to deliberately study in order to remember. This is intentional
learning. How important is intention to remembering?
The general experimental method used to test incidental learning is to present a
series of words to the subjects with instructions to process the words in different ways:
Does the word have an e in it? How many letters are in the word? Is the word pleasant or
unpleasant? (Hyde & Jenkins, 1969). This is the levels of processing idea we discussed
previously. Information can be cognitively processed at varying depths. You need to attend
to the meaning of the word to rate its pleasantness, but not whether it has a letter e. In
the incidental learning conditions, the participants perform the ratings but are not told to
remember the words. In parallel intentional learning conditions, the subjects are told
there will be a memory test. Deeper processing produced better recall (see Table 9.1).
This was true in both intentional and incidental learning conditions. The incidental learning
participants who performed the deepest processing recalled as many words as partici-
pants who were told in advance about the memory test.
Incidental learning is sometimes better than intentional. Medical students and faculty
were given patient charts containing about 20 pieces of information, mostly laboratory
test results. The participants were asked to diagnose the patient or to memorize the case
data. The medical specialists who made a diagnosis recalled more specific facts than did
the students who diagnosed or memorized. An expert’s diagnosis incidentally retrieves
an organization for the material that more-than makes up for the absence of intentional
elaboration (Norman, Brooks, & Allen, 1989).
248 Encoding

Table 9.1 Incidental Memory. Number of words recalled after different orienting instructions,
and incidental or intentional learning instructions

Instruction Incidental Intentional

(Unexpected memory test) (Expected memory test)

Is there an e in the word? 9.4 10.4

How many letters are in the word? 9.9 12.4
Is the word pleasant or unpleasant? 16.3 16.6
No orienting task 16.1
Source: “Differential Effects of Incidental Tasks on the Organization of Recall of a List of Highly Associated
Words,” by T. S. Hyde & J. J. Jenkins, 1969, Journal of Experimental Psychology, 82, p. 475.

Other Incidental Encoding Operations

The following demonstrations all used incidental learning. That is, the participants were
not informed during the encoding stage that memory would be tested afterwards.
The generation effect refers to better memory for words that are generated by the
participant rather than being provided by the experimenter (Slamecka & Graf, 1978). For
example, fill in the blanks with a word that is opposite in meaning: HOT: _ _ _ _; STOP: _ _.
The subject-generated words COLD and GO will be recalled better than words that are
simply read. Generating the words induces elaborative processing.
In the self-referent effect information that is encoded in reference to yourself or your
self-concept is especially well remembered. In an incidental learning task, subjects could
be asked to rate items in reference to themselves. How characteristic is this word of you?
PATIENT; SUSPICIOUS. Words rated for self-reference are especially recalled later.
Survival processing is judging words for their relevance to surviving some catastro-
phe. Imagine you are stranded in a forest. How relevant would these items be in this sit-
uation? Example words were JUICE, SHOES, DOOR. (Nothing so obvious as a compass
or a knife). Words evaluated for survival value are much better remembered than words
rated along other dimensions such as pleasantness, relevance to moving to a new home,
or even self-reference processing (Nairne, Pandeirada, & Thompson, 2008). The hypothe-
sis is that memory evolved as a means of adapting to the challenges of living.

Incentives
Does offering a reward increase performance in a memory task? When subjects are offered
varying amounts of monetary incentives for remembering more words, larger incentives
generally produce no better recall than smaller or no incentive (Nelson, 1976; Nilsson,
1987). The research suggests that once a subject has committed to learn, reward has little
additional effect. When separate groups of college students were offered different amounts
of money for each word remembered, there was no difference in the number of words
recalled. Given that the participants were motivated and were trying to remember to begin
with, as is typical of college students, the amounts offered had no additional effect.
However, when there is a mix of some high-value items and some low-value items,
students devote more processing to the larger incentive items. For instance, if students
are told that recalling one set of items has a higher pay-off than the other items, then
 Learner Variables 249

more of the high-value facts are remembered (Kassam et al., 2009). A rational strategy is
employed: Given limited processing capacity, devote more effort to remembering the
high incentive words than the low value words. This is what students do in juggling their
workload among courses: Devote more time to the important courses (such as this one)
and less to other classes.
What about offering an incentive after learning has already taken place? Telling the
participant after studying does not help remembering the high-value facts (Kassam et al.,
2009). In another demonstration, after studying, the students first attempted to recall as
many words as possible. After this, some participants were offered $1 for each additional
word they could recall. The average number recalled was five, the same as by the partic-
ipants who were simply asked to try and remember some more (Nelson, 1976).
The point is that the promise of an incentive for remembering does not affect mem-
ory itself but rather behaviors that will lead to better encoding, such as rehearsal or cog-
nitive elaboration.

Arousal
We could expect arousal to facilitate learning, and generally it does. But the findings often
must be qualified. One reason is that the term arousal is ambiguous and has varied mean-
ings in psychology. Arousal can refer to a psychological state of increased attention or
alertness; to physiological excitement indicated by heart rate, skin conductance, or
­electroencephalogram (EEG); or to personality traits such as anxiety or impulsiveness.
Another important qualification is that the optimal level of arousal for any given task
varies. The Yerkes-Dodson law states that performance is best at some intermediate
level of arousal, and is less efficient at both lower and higher levels. Performance follows
an inverted U-shaped curve, with peak performance occurring at intermediate levels of
arousal and decreasing performance at higher or lower levels (Yerkes & Dodson, 1908).
As a hypothetical example, remembering facts from a classroom lecture might be poor if
you are either too fatigued or too hyper.

Optimum Performance
low Performance high

Task 1 Task 2

low Level of Arousal high

Figure 9.4
The Yerkes-Dodson law. There is a mathematical relationship between the accuracy of performance and the
level of arousal of the organism. Optimal performance requires some intermediate level of arousal. Performance
declines at levels above or below the optimum. A corollary of the law states that different tasks have different
optimal levels, as shown by the two curves above.
250 Encoding

Figure 9.5
Recall and Stress. Mean number of facts recalled (the means are log-transformed) by 3- and 4-year-old children
as a function hurricane storm severity.
Source: From “The Effects of Stress on Young Children’s Memory for a Natural Disaster,” by L. E. Bahrick, J. F. Parker, R. Fivush,
& M. Levitt, 1998, Journal of Experimental Psychology: Applied, 4, p. 318. Copyright © 1998 by the American Psychological
Association.

A corollary to the Yerkes-Dodson law is that different tasks have different optimal
levels. Taking an exam requires a higher optimal arousal level than does listening to a
classroom lecture. And in each case there can be too little and too much arousal.
An example of the Yerkes-Dodson law is the recall by 3- and 4-year-olds of a hurricane
they had experienced (Bahrick, Parker, Fivush, & Levitt, 1998). The arousal levels the chil-
dren had experienced during the storm were defined by the amount and severity of the
damage: to the house, and days without electricity and running water. For instance, high
arousal was assumed in cases where the house was penetrated, moderate arousal if
there was extensive outside damage, and low arousal if the damage was minimal. (The
three groups of children differed also on the other measures of severity). The number of
facts the children recalled is shown in Figure 9.5. The moderately stressed children
recalled more than did those presumed to have experienced lower or higher levels of
arousal.
The effects of arousal on memory are studied in several ways. Arousal is manipulated
directly through the administration of stimulant drugs and indirectly through correlations
with circadian rhythms. We can also consider the effects of extremely low levels of
arousal, such as sleep (see Box 9.2), and extremely high levels of arousal produced by
strong emotions (discussed separately in a following section).

BOX 9.2 SLEEP LEARNING

Can we learn while we sleep? At one time there was an entire industry that promoted
sleep-learning audio products. Simply listen through a pillow speaker during the night and
wake up the next morning refreshed and fluent in Spanish! But if sleep is a state of low
arousal, can learning actually occur?
 Learner Variables 251

Early studies of sleep learning found conflicting outcomes (see Aarons, 1976). The discrep-
ancy was due to whether the participants were asleep or awake when the audios played.
Learning did not occur if the participants were truly asleep, as verified by recording brain
waves and rapid eye movements. For example, Emmons and Simon (1956) presented a
10-word list 46 times during the night. The list was played only when EEGs indicated the
participant was in deep sleep. The next day the participants correctly identified a hardly
encouraging 25% of the target words on multiple-choice tests, as compared to a chance level
of 20%.
Recent research indicates that, in a sense, learning does occur during sleep. The differ-
ence now is that sleep must occur after studying or practicing. In back-to-back articles pub-
lished in the journal Science, researchers showed that post-encoding sleep enhanced learning
of mazes by rats (Wilson & McNaughton, 1994) and motor skills by humans (Karni et al.,
1994). Learning does occur during sleep—the learning that commenced before sleep.
In terms of the stages of memory, sleep plays an important role in the storage, or consol-
idation, of recent events into permanent memory. During sleep, the neural elements that
were active during recent learning are reactivated. This has been referred to by analogy to
“offline processing” (Stickgold, Hobson, & Fosse, 2001).
While rats explored a new environment, activity in individual cells in the hippocampus
were recorded. Later, when the rats were asleep, those cells displayed a pattern of activity
similar to that of the waking activity (Wilson & McNaughton, 1994). It is as if the neurons in
sleeping rats were replaying the memory trace of the earlier awake experience. Parallel
human studies, measuring brain activity across masses of cells, similarly found that hip-
pocampal activation during learning was replayed during deep (slow wave) sleep. The amount
of replay correlated with the accuracy of recall of the pre-sleep learning.

Targeted Memory Reactivation

Targeted memory reactivation (TMR) is the use of an external cue to selectively activate
and strengthen the memory trace of recently studied material. In a TMR study, partici-
pants are given material to learn, such as pairings of words or pictures. An external stim-
ulus is presented during this learning phase. For instance, while word-pairs are shown, an
odor is present in the background (e.g., a lemon, floral, or mint aroma). The odor is a
contextual stimulus. The participants then sleep. During periods of deep sleep the odor
stimulus is presented. The cue is assumed to reactivate the same brain regions that were
active during learning. When tested the next day, the cued subjects recalled more of the
material associated with the cueing stimulus (Rasch, Buchel, Gais, & Born, 2007; see
review by Diekelmann & Born, 2010).
Targeted memory reactivation has been demonstrated in a number of learning tasks:
learning associations between words and spatial locations; foreign language vocabulary;
spatial learning of a virtual reality maze; and motor skills, such as a sequence of hand
movements.
Beyond the obvious benefits to academic learning, many practical applications of
TMR have been suggested. After psychotherapy sessions, sleep reactivation might
strengthen positive learning, or reinforce the extinction of fears. In behavioral health,
smoking cessation might be facilitated. In rehabilitation, a cueing tone could be paired
with correct motor movements during physical therapy, and then presented during sleep
252 Encoding

(Feld & Diekelmann, 2020). “Neuroscientific findings on memory processing during sleep
have opened up a new world of innovative possibilities for guiding the sleeping brain…”
(Paller, 2017, p. 536).
It is also possible that speculations are outrunning the science. The neural activity
recorded so far is limited to a handful of cells in rats; in humans, an EEG reflects the
action of many neurons and multiple areas of the brain. Finally, the mechanism of tar-
geted reactivation is not known. Although assumed to enhance consolidation, reactiva-
tion may have other effects: TMR might reduce interference, act as a reinforcing stimulus,
or increase arousal.

Stimulant Drugs
Stimulant drugs are so-named because they can increase central nervous system arousal.
Among known stimulant drugs are caffeine and nicotine, amphetamine, cocaine, and
strychnine. The arousal hypothesis would suggest that optimal doses of a stimulant
should improve acquisition in learning and memory tasks.
Stimulant drugs include those used to treat ADHD and other disorders. The purpose
of these drugs is to focus attention, and only indirectly to boost learning. However,
d-amphetamine (d-AMP, or Adderall) and methylphenidate (MPH) are used by other stu-
dents for the purpose of cognitive enhancement, although it is not at all clear that these
drugs can enhance learning. Ilieva, Boland, and Farah (2013) tested the effects of amphet-
amine on a large sample of non-ADHD student volunteers, using 13 measures of cognition.
For example, working memory, immediate and delayed word recall, and face memory were
tested. The design was double-blinded and placebo-controlled. The results showed little
overall benefits of the drug. The exceptions were participants who had low baseline scores
to begin with (poor digit recall, word recall, etc.). They showed some modest gains with the
drug. Interestingly, all the subjects believed they performed better on the drug days.
Although individual studies sometimes show memory enhancement, reviews of the
larger literature find little consistency in positive outcomes. Overall, there may be small
effects on working memory and delayed recall in episodic memory. Again, some effects
are restricted to individuals who had poor performance at baseline, and the drugs even
impaired memory in high performers (Husain, & Mehta, 2011; Smith & Farah, 2011).
There are no simple effects of most stimulants on learning. Caffeine produces com-
plex interactions with other variables such as gender, age, and the type of memory task.
For example, performance on a practice verbal SAT test (a measure of semantic knowl-
edge) showed that caffeine sometimes aided and sometimes inhibited performance,
depending on (a) when the test was taken (morning or evening); (b) the personality of the
subject (introvert or extravert); and (c) the amount of practice (first or second day). The
efficacy of caffeine depended on the combination of all three variables (Revelle,
Humphreys, Simon, & Gilliland, 1980). In other studies, free recall of word lists by women
was impaired by caffeine (Erikson et al., 1985), or enhanced (Arnold, Petros, Beckwith,
Coons, & Gorman, 1987). Caffeine had little consistent effect on men in either study.
The student-user of stimulants just wants to know “Will I remember more on the test
tomorrow?” The scientist wants to know whether a drug affects short-term and/or long-
term memory, encoding or retrieval, by increasing attention or reducing fatigue, etc. There
are as yet no comprehensive answers to either group’s questions.
 Learner Variables 253

Circadian Rhythms and Learning

Arousal varies during the course of the waking day, suggesting the possibility that circa-
dian variations could influence learning. Performance on the digit span test, a measure of
short-term memory, peaked at 10:30 A.M. and declined thereafter into the early evening
hours (Blake, 1967; see right panel of Figure 9.6). Immediate recall of specific facts from
a 1,500-word science article peaked even earlier in the morning: Correct recall was max-
imal at 8 A.M. and steadily declined thereafter, with a slight bump back up after lunch
(Folkard & Monk, 1980; left, Figure 9.6). (These studies used British students. I would be
surprised to see U.S. students peaking at that hour in the morning).
One difficulty in linking arousal to circadian rhythms is the assumption that everyone
has the same cycle. If people differ in their levels of alertness at different times of day
(i.e., so-called morning “larks” and evening “owls”), there may be individual differences
in remembering across the day. Indeed, one study found that morning people were more
accurate in paragraph recall when tested at 9 A.M. than at 8 P.M., whereas evening peo-
ple did better at the later testing time (Petros, Beckwith, & Anderson, 1990).

Some Conclusions about Arousal

An implicit assumption of arousal theories is that an aroused brain simply works better at
encoding information into a stable memory. It is as if once the brain is cranked up, any-
thing thrown in will be remembered better. Apart from any hypothesized neural effects,
we need to consider the broader influence of arousal on cognition. Arousal affects what
we attend to and what is ignored, what and how we encode, and how efficiently we
retrieve. The source of arousal will also have different effects. Arousal elicited by a star-
tling unknown sound will affect learning differently than arousal produced by loud but
familiar background noise.

Figure 9.6
Memory Recall as a Function of Time of Day. Digit-span recall (left panel) and prose-paragraph recall (right
panel) as a function of time of day of testing.
Adapted from “Circadian Rhythms in Human Memory,” by S. Folkard and T. H. Monk, 1980, British Journal of Psychology, 71,
pp. 295–307, published by The British Psychological Society. Reproduced with permission by John Wiley & Sons.
254 Encoding

Emotions and Encoding

Are highly emotional events remembered better than neutral events? This question
seems too easy and the answer seems too obvious. However, getting objective data to
support our subjective impressions is not so easy. The two basic problems are determin-
ing how emotions influence remembering, and whether what we recall is accurate.
Emotions have widespread and contradictory effects on individual stages and processes
of memory.

How Do Emotions Affect Memory?

First, emotional arousal focuses attention on certain aspects of a situation. Given our
limited capacity to divide attention, peripheral details may go unnoticed. In an experimen-
tal simulation of memory for an emotional event, a story was presented via a series of
slides, and, simultaneously, the participants’ eye movements were recorded. When one
slide depicted an emotional outcome to the story (a child hit by a car), there were more
fixations on the central details of the picture. These central details were also recalled
better than central details of a neutral picture. Peripheral details were not recalled as well
in the emotional condition (Christianson, Loftus, Hoffman, & Loftus, 1991).
A second effect of emotion is to produce bodily arousal that could contribute to long-
term memory formation. Epinephrine is released from the adrenal medulla during and
after an emotional stressor, which leads (through a series of intermediate steps) to stim-
ulation of the amygdala in the brain. The amygdala is active during the formation of long-
term memory for certain kinds of emotional events but not for non-emotional memories
(McGaugh, 1991; Thompson & Gluck, 1991).
Third, emotional events are distinctive. Often, they are not everyday experiences.
Emotional events are talked about and thought about after the fact, and so they are well
rehearsed and elaborated in memory (Heuer & Reisberg, 1992). In the extreme example
of post-traumatic stress disorder, the memories replay and intrude into thoughts and
dreams.
Thus, emotions incorporate several themes from this chapter: attention, arousal, dis-
tinctiveness, and elaboration. Emotions also affect retrieval, either to facilitate or inhibit
recall. These latter possibilities are discussed in the next chapter.

Flashbulb Memories
One striking memory phenomenon that most of us have experienced is that of having a
vivid memory for a surprising, emotional, and consequential event. The precipitant is par-
ticular to our own lives, for example, unexpected tragic news or unexpected good news.
Other emotional experiences are shared by a community or a nation, such as the news of
the attacks on September 11, 2001. Brown and Kulick (1977) used the term flashbulb
memories to capture the essence of such memories. It is as if a photograph of the
moment we heard the news is encoded into memory. This image is assumed to be com-
plete, detailed, and accurate. Brown and Kulick proposed the existence of a special mem-
ory mechanism, triggered by surprising and emotional events, that immediately creates a
record of the contents of consciousness at that moment. These memories are immune
 Learner Variables 255

to normal forgetting. While the general nature of flashbulb memories is well accepted, not
all of Brown and Kulick’s assumptions are.
The method introduced by Brown and Kulick to study flashbulb memories was to ask
people if they remembered when they first learned of a significant news event, such as
the assassination of President John F. Kennedy or Martin Luther King, Jr. Even long after-
wards, people reported photolike memories. These included details of where they were
when they heard the news, what they were doing at the time, from whom or how they
heard the news, their immediate affective reactions, and what they did next.
Neisser (1982) criticized the Brown and Kulick procedure on several grounds. What
people claim they remember is not necessarily true, no matter how vivid their memory
seems. The developmental psychologist Jean Piaget remembered being kidnapped as a
young child and often retold the story, although he later learned it had not happened (Loftus
& Ketcham, 1994). What people think they remember years after might already be dis-
torted. Neisser argued that the memories were not formed in a flashbulb instant, but in
the hours or days afterwards, when people told and retold their stories. Indeed, the recol-
lections do have a news-reporting quality to them: when, where, why, who, and what.
Researchers next moved to questioning people shortly after a traumatic experience,
to establish a baseline of what was remembered at the time. This first memory test is
compared to recall tested months later. McClosky, Wible, and Cohen (1988) questioned
people about flashbulb memories for the explosion of the space shuttle Challenger (in
1986) one week afterwards and again nine months later. The participants evidenced
remarkable consistency in their answers to specific questions about where they were,
who told them, and so on. Over time, memories did become more general and contained
fewer details. But nine months later, only about 9 percent of the answers were inconsist-
ent with the one-week recollections. One could be impressed by the robustness of the
memories: These are very good memories. Or, one could focus on the forgetting: These
memories were immune to forgetting.
Much less accuracy was found in a major study of memory of the September 11
attack on the World Trade Center and the Pentagon building in the U. S. Over 3,000 partic-
ipants were surveyed, and follow-ups were conducted out to ten years later (Hirst et al.,
2009, 2015). Some questions addressed objective facts about the events: the number of
planes, the name of the airline, and the order of events. Answers to these questions
showed some forgetting over time, mostly in the first year. Other details, the who-what-
where, produced more inconsistencies from the first questioning to later attempts at
recall. In some cases, a different answer was given in one of the retests (for instance,
who they were with at the time of the attack), and this change was repeated in subse-
quent surveys. One thing that did not change over time: People were very confident in
the accuracy of their recollections. Participants were just as sure of their memories, even
for answers that had changed over time.
Talarico and Rubin (2003) interviewed college students after the September 11
attack. In their comparison the students also answered questions about an everyday
event that had occurred near that same day, which often was a school related or social
event. When the students were interviewed again 32 weeks later, they showed fairly
consistent recall of the September 11 details. Remarkably, the students’ recall of the
256 Encoding

everyday events was just as good. They recalled the same number of details that were
consistent with their previous statements and the same number that were inconsistent.
Their flashbulb and everyday memories did not differ. What did differ was that the stu-
dents had more confidence in the accuracy of their flashbulb memories.
Flashbulb memories are not only caused by public tragedies and catastrophes. We all
have our own flashbulbs, some negative and some positive. One difference is that the
events studied by psychologists have national and historical significance, which may have
a dramatic impact on our processing of the memories. So, is there a separate flashbulb
mechanism? Maybe instead we should view memory as a continuum, with different fac-
tors such as surprise, emotionality, consequentiality, or rehearsal acting to push memo-
ries toward one end point or the other.

Eyewitness Memory and Emotional Arousal

Eyewitness recall offers another source of data on the effects of emotional arousal on
memory. Are witnesses to a crime or an accident more or less likely to accurately recall
the event? Once again, there is no simple answer because recall is determined by a
complex set of interacting variables. Emotional arousal may lead to remembering central
details, and consequently poor recall of peripheral details. An example of this is the weap-
ons focus: Victims or witnesses focus attention on a weapon and not on other aspects of
the situation, such as the face of the perpetrator (Loftus, Loftus, & Messo, 1987). In one
study of weapons focus, participants who thought they were waiting for an experiment
heard an argument in the next room. Someone emerged from the room carrying either a
pen or what appeared to be a bloody letter opener. The participants were then asked to
identify the person in a photo line-up, which was the real purpose of the study. The per-
petrator’s picture was less often identified in the weapon condition than in the pen condi-
tion (cited in Loftus, 1979). An emotional witness and a non-emotional person might
remember equal amounts of information, but different details of the situation.
Research has focused on comparing memory for emotional and non-emotional items.
Given a picture of an emotional item against a neutral background (such as a snake in the
forest) and a non-emotional item (a chipmunk in the forest), there is a trade-off in memory
for the emotional detail versus the background. The emotional item is remembered better
than the neutral item in the same location (that is, remember more snake and less chip-
munk). This pattern reverses for background memory. Less of the background behind the
emotional item is remembered than behind the neutral item (remember less forest in the
snake picture, and remember more forest in the chipmunk picture) (Waring, Payne,
Schacter, & Kensinger, 2010).
A traumatic experience may also produce an isolation or von Restorff effect, causing
forgetting of details due near to a startling event. Loftus and Burns (1982) showed a
2-minute video depicting a bank robbery, a film used by banks to instruct their employees
on how to handle this kind of situation. Toward the end of one version of the film, a bank
teller is shot, but this does not happen in the control version. Participants who viewed the
emotional version of the film did not recall a critical fact shown just before the violent act:
the number on a bystander’s football jersey. Only 4 percent of those seeing the violent
film remembered it, whereas 28 percent of the control participants did. Christianson and
Nilsson (1984) found forgetting on the other side of the traumatic event. Their participants
did not recall a critical detail that followed the arousing stimulus.
 Learner Variables 257

Conclusions about Emotions and Memory

Emotional arousal can facilitate encoding into memory. But knowing this fact will not tell
us what details will be encoded. Remembering a suspect’s face would be helpful to the
police, but the victim’s attention may have been focused on a weapon. Significant details
that preceded or followed a startling moment may be forgotten. Rehearsal may elaborate
some of the memories, but at the same time can introduce distortions. We are therefore
limited in our ability to make broad statements about the effects of emotion on the accu-
racy of memory.

Schemas
Schemas are outlines of general knowledge that are stored in semantic memory.
Schemas, also called scripts, are ways of organizing knowledge. Schemas have a hierar-
chical structure, with packets of information stored at each level. One often-cited example
is the “dining at a restaurant script,” as listed in Table 9.2. The restaurant schema has
categories such as entering, ordering, eating, and exiting. Within each category is general
knowledge about each step: Ordering involves reading the menu and telling the server
what you want; exiting involves paying the bill and leaving a tip.
There are several ways a schema can influence what is remembered in a given situa-
tion (Alba & Hasher, 1983).
1 Selection. Schemas guide selection of what is to be encoded, usually details that are
relevant to the schema.
2 Storage. A schema organizes memory for events, providing an outline of where each
new piece fits.
3 Abstraction. Common features across a number of similar experiences are
abstracted and stored in the schema; specific details from any one event need not
be retained.
4 Retrieval. A schema provides retrieval cues to guide and direct memory search.
5 Normalization. The schema may lead to memory distortion, in that we remember
what usually happens in the schema, rather than what actually happened.

Table 9.2 Restaurant Schema

1 Entering
Customer enters
Looks for/shown to table Sits down
2 Ordering
Looks at menu
Signals server
Orders food
3 Eating
Server brings food from kitchen
Multiple courses (appetizer, dessert) Customer eats
4 Leaving
Server brings check
Customer pays
258 Encoding

An example of schema influence is remembering the objects in a room.

Rooms that have different purposes and contain different items. Research partici-
pants were shown one of two rooms: a graduate student office and a preschool class-
room. Each contained schema-consistent objects (books, a beer bottle, and a desk; or
toys, games, and blocks); and schema-inconsistent objects (simply exchanging some
things between the office and preschool room). The results of this study make several
important points about what is and is not remembered:
Items consistent with the schema are remembered, for example, books and a chair
in an office.
Unusual or unexpected items were recalled even better, for example, a toy truck in a
graduate student’s office.
Objects that are neither schema-relevant nor unexpected seem to get lost. There was
a clock on the desk in the office. We might not expect it to be there, yet its presence is
not surprising either. So it is not particularly memorable (Pezdek et al., 1989).
(I was once surprised to see a playpen in an associate dean’s office. He had been
babysitting his grandchild that day. This is a schema-inconsistent item and was thus mem-
orable. Who expects grandpa to be babysitting at work? Oh, so you thought it was the
playpen that was inconsistent? That too).
Schemas unfortunately can lead to mis-encoding. When students in a speech-and-
hearing course were shown a list of symptoms that led them to make a diagnosis of a
specific hearing disorder, they later falsely recalled some symptoms that had not been
presented but that were consistent with the diagnosis. They remembered what is typical
for this disorder. On the other hand, making the diagnosis improved recall in one way: on
a recognition test, the participants were able to correctly reject symptoms unrelated to
the diagnosis that had not been seen before. There was a trade-off between the memory-
improving and memory-distorting effects of the schema (Arkes & Harkness, 1980).
As repeated events become schematized, the memory of specific details is lost.
Linton (1982) kept a diary of personal and professional events in her life over six years.
She periodically tested herself by to determine what she remembered. She began forget-
ting the details of repeated events. For example, Linton could recall details of the first
meeting of a committee to which she had been appointed. As meetings accumulated,
details of each became blurred and a more generic memory (or schema) developed.
Details that differentiated one meeting from another (where they had lunch one day or
who missed a given meeting) were lost.
Strict adherence to a schema would allow recall of only those events that conform to
the schema. However, exceptions to the rule are remembered, especially if they are unu-
sual or surprising. And schemas undergo change, as new examples are assimilated into
the schema. The restaurant schema presented earlier probably does not fit the schema
you first developed as a child, likely based on Happy Meals.

Metamemory
Metamemory refers to our knowledge about memory. It includes our estimates about
the difficulty of learning certain materials, which strategies will be most effective, moni-
toring progress during learning, and beliefs about how our own memory differs from
 Learner Variables 259

memory in general. (Metamemory also includes knowledge about retrieval, which will be
discussed in the next chapter).
Metamemory is usually assessed by self-report questions, the scores on which can
then be correlated with actual memory performance. For example, students can be asked
to make ease-of-learning judgments, possibly using a 10-point scale, prior to actually
attempting to learn the material. U.S. students should predict that their learning the capi-
tals of South American countries will be more difficult than learning the county seats of a
U.S. state. When students subsequently attempt to learn the facts, they actually spend
more time on the items they had predicted would be more difficult (Nelson & Leonesio,
1988). After studying is completed, students can make judgments of learning (JOL) rat-
ings, indicating how well they think they learned the material. JOL estimates often predict
how much of the material is recalled on an actual test.
What factors guide a student’s decision that they have studied sufficiently or not?
JOL are primarily affected by characteristics intrinsic to the material. In the case of
remembering word lists, the factors of imagery, familiarity, or associations affect JOL
ratings. The conditions of learning, such as increasing the number of repetitions or spac-
ing the repetitions, usually do not affect JOL ratings. If students actually study under
these conditions (such as spaced practice), the students remember more but this does
not affect their belief that more was learned (Koriat, 1997).
Metamemory beliefs can wrongly predict that a particular factor will inhibit learning.
In studies of the disfluency effect, to-be-remembered words are presented in difficult to
read formats. Some words might be written in script or cursive or the words are blurred;
or the font may be ornate, irregular in shape, with close letter spacing. In another case,
some words were presented in an unusually large font, as compared to the normal font
and size of the control words (see Figure 9.7). Students rate their ease of learning and

Instructions: Estimate your chance of recalling each word below. For example, if
you think you have a 50% chance of recalling a word, you would type in 50.

Rated More Likely Rated Less Likely Actually

To Be Recalled To Be Recalled Recalled

LEARNING LEARNING Difficult > easy to read

Disfluency disfluency cursive > print type

Larger Smaller smaller = larger

Figure 9.7
Examples of Easy and Difficult to Read Fonts. In the disfluency effect, words in difficult-to-read fonts are
remembered better than normal print, even though students predict the opposite. Larger versus smaller font
size does not affect recall.
260 Encoding

then are tested for recall of the rated words. Students predict that the difficult-to-read
words will be less recalled, or that words in LARGE typed will be better recalled. Neither
turns out that way. Difficult to read words are often better recalled, and type size has no
effect on recall (e.g., Geller, Still, Dark, & Carpenter, 2018).
People believe they know how good their own memories are. This is referred to as
memory self-efficacy: judgments about how well we think our memory will function in a
particular situation. Self-efficacy can be assessed through self-report inventories. Some
example items are listed in Box 9.3.

BOX 9.3 EXAMPLE METAMEMORY AND SELF-EFFICACY ITEMS FROM THE

MOTIVATED STRATEGIES FOR LEARNING QUESTIONNAIRE

This is a self-report inventory illustrating aspects of metamemory. Students were instructed to

respond to these items using a 7-point scale (1 = not at all true of me, to 7 = very true of me).

Cognitive Strategy Use

• When I study, I put important ideas into my own words.
• When I study for a test, I practice saying the important facts over and over to myself.
• When reading, I try to connect the things I am reading about with what I already know.

Self-Regulation
• I ask myself questions to make sure I know the material I have been studying.
• I work on practice exercises and answer end-of chapter questions even when I don’t
have to.

Self-Efficacy
• I’m certain I can understand the ideas taught in this class.
• My study skills are excellent compared with other students in this class.
• Compared with other students in this class, I think I know a great deal about the
subject.

Intrinsic Value
• It is important for me to learn what is being taught in this class.
• I prefer class work that is challenging so I can learn new things.
• I think I will be able to use what I learn in this class in other classes.

Test Anxiety
• I worry a great deal about tests.
• I have an uneasy, upset feeling when I take a test.
• When I take a test, I think about how poorly I am doing.

Self-efficacy was assessed in samples of younger and older adults (Ryan, 1992). Participants
aged 25 to 85 rated self-perceived memory for such things as conversations or names, and
the incidence of absentmindedness. Both older and younger participants believed that, in
general, memory declines with aging. However, most people, young or old, did not think their
memories were as bad as those of others their age. Personal self-efficacy was rated higher
than the perceived effectiveness of memory for an individual’s own age group.
 Applications 261

The direction of cause and effect between self-efficacy beliefs and actual performance is
debatable. Certainly, low expectations will limit effort devoted to trying to learn certain types
of material (foreign languages or statistics, as examples). On the other hand, self-perceptions
may be accurate representations based on our actual experiences. If I constantly forget
names, I should correctly believe that I have a name–memory problem.
Individuals differ in the ability to evaluate their own learning and performance. Research
on the “unskilled-and-unaware” phenomena, or the Kruger-Dunning Effect (Kruger & Dunning,
1999), shows that individuals who score the lowest on various tests (e.g., reading compre-
hension, grammar evaluation) believed their competence was comparable to the average.
Dunning and Kruger have shown this in studies using college students, across several
domains (e.g., academic or social). For instance, after completing a course exam, students
who scored in the 25th percentile (the lowest quarter of the class) rated their mastery of the
material and their expected test scores in the 60th percentile (Dunning, Johnson, Ehrlinger,
& Kruger, 2003). Overall, the bottom half of the class thought they were among the top 60–70
percent.

Encoding: Summing Up
To understand how new learning occurs, we need to consider several factors: the mate-
rial to be learned, the context in which learning occurs, the cognitive resources available,
the task demands, the existing knowledge base of the subject, the strategies selected,
and our beliefs about how learning occurs. The point of this lengthy list is to remind you,
once again, of the complexity of obtaining a scientific understanding of the learning
process.

APPLICATIONS
Academic Learning and Encoding
Elaboration
Consider some learning techniques used by students in light of elaborative processing.
Underlining in textbooks, taking notes during lectures, and writing summaries are often
used but may contribute little to learning (Snowman, 1986). Although each could be done
in a way that promotes elaborative processing, students instead engage in passive under-
lining and transcribing. In one study, college students extensively practiced taking dicta-
tion while reading unrelated material simultaneously. The students became proficient at
doing both tasks at once, with little cross-interference. This suggests that students could
take lecture notes without attending to the content of the lecture (Hirst, Spelke, Reaves,
Caharack, & Neisser, 1980). Other research found that students who took notes recalled
no more than did students who simply listened to the lecture (Kiewra et al., 1991).
Other study techniques also promote shallow processing. Students tend to highlight
too much when reading or include too much verbatim copying while summarizing.
Highlighting and summarizing improve recall when students are limited in the amount of
each: for example, only three lines of underlining per page or summaries restricted to
three sentences (Kulhavy, Dyer, & Silver, 1975).
262 Encoding

One means of getting students to think about facts they are asked to remember is to
prompt them with a question: Why would this be so? These questions force elaboration
of the to-be-learned material. In one demonstration of elaborative interrogation, Canadian
college students studied short paragraphs about universities. In the interrogation condi-
tion, they were asked why each fact might be true. A sample fact might be “McGill
University stands on land donated by a fur trader.” Students might connect this fact to
other knowledge of Canadian history as an elaboration. These students recalled more
facts than did control participants who simply read the paragraphs (Woloshyn, Willoughby,
Wood, & Pressley, 1990).

Meaningfulness
Meaningfulness includes familiarity with the material, existing knowledge, and organiza-
tion. Meaning is enhanced in textbooks that use text adjuncts. Headings or titles can
increase recall of prose passages by activating existing knowledge or schemas that can
guide encoding and subsequent retrieval. Advance organizers are longer introductory
statements that act as a bridge between what the student knows and the new material.
Analogies activate prior knowledge to draw parallels with the new material that is being
presented. Each of these adjuncts can produce meaningful learning that increases reten-
tion over longer intervals (Royer, 1986).

Self-Efficacy
Teaching students a study strategy will not be of much benefit unless the students are
convinced it will make a difference in their learning. In the standard intervention design,
someone is first taught a mnemonic technique; the method is tested once to show it
works; and then you hope the students will continue to use it. Pressley, Levin, and Ghatala
(1988) added another step after teaching an imagery mnemonic for foreign-language
acquisition. The students studied some vocabulary words using their own methods,
whatever those were. When participants saw the benefit of the imagery strategy over
their own methods, they continued to use the mnemonic for subsequent trials in the
study. Without the explicit comparison, students simply dropped the mnemonic when
given the opportunity.

Circadian Rhythms
Does time of day affect academic learning and retention? N. F. Skinner (1985) found that
course grades among college students at his university were higher in afternoon and
evening sections than in morning sections. We should also expect individual differences
at different times of day, the morning types and the evening types. Students who rated
themselves high on a Morningness scale had higher grades in their 8 A.M. classes than
later in the day, whereas low-morningness people had their lowest grades in 8 A.M.
classes (Guthrie et al., 1995).

SUMMARY
This chapter focuses on variables that affect the encoding of information into memory. It
is important to remember that the encoding stage cannot always be isolated from
 Summary 263

storage and retrieval stages. Most laboratory studies of encoding use episodic learning
tasks such as free recall or recognition.

Some Basic Variables in Encoding

Some basic variables affecting encoding are rehearsal, imagery, and meaningfulness.
Elaborative rehearsal is a form of deep processing in which the to-be-remembered
material is related to other information. Maintenance rehearsal is a form of shallow pro-
cessing in which information is passively repeated. Elaborative rehearsal promotes better
retention through associations to existing knowledge, the formation of distinctive memo-
ries, and the effortful processing that it entails.
Material that is more imageable or concrete is better remembered. Pictures and
objects are remembered better than are words. This may be because of dual coding, in
which memory is encoded as both an image and a word.
Meaningful material is better remembered. Meaningfulness is defined by imageabil-
ity, familiarity, and the number of associations to other items. Meaningful items have a
high frequency of occurrence in language. Basically, meaningful items are ones you
already know about.

Presentation Variables
Encoding is also affected by the way material is presented. This section of the chapter
included several “effects”: the testing effect, the isolation effect, and the spacing effect.
The testing effect refers to the benefit of taking a practice test on recently studied
material, rather than engaging in additional study. Retrieval practice enhances the encod-
ing of information, and its availability in memory.
The isolation effect, also called the von Restorff effect, occurs when an unusual item
is embedded in an otherwise homogeneous list of items. The isolated item is particularly
well remembered due to enhanced rehearsal or to its distinctive memory representation.
If an item is presented two or more times, distributed repetitions produce better
recall than do massed repetitions. The spacing effect has been attributed the reduced
attention to a closely spaced repeated item; and to different encodings of spaced
repetitions.

Learner Variables
A number of learner variables that can influence encoding were considered, including the
intention to learn, incentives to learn, arousal, and emotion.
Retention is often more a function of how information is processed rather than of
explicit attempts to remember. Incidental learning, or remembering without any deliber-
ate intention to do so, is sometimes as effective as intentional remembering. Consistent
with levels of processing theory, deeper processing produced better recall, apart from
any intention to remember.
Offering an explicit incentive to remember, such as money or points, usually does not
improve recall in laboratory experiments. However, trade-offs may occur if both high- and
low-incentive items are intermixed during presentation. Then the high incentive items are
264 Encoding

rehearsed at the expense of the low-incentive items. In sum, incentives encourage cog-
nitive processing which determines encoding.
We expect arousal to facilitate learning, and generally it can. However, there are no
simple effects of arousal on learning. Each statement about daily biorhythms, stimulants,
anxiety, etc., must be qualified by interactions with other variables. According to the
Yerkes-Dodson law, performance is better at an intermediate level of arousal; arousal is
less effective at both lower and higher levels.
Stimulant drugs such as caffeine can facilitate remembering, but the effects interact
with other variables.
Circadian rhythms affect arousal and thus indirectly learning. Some studies have
shown better memory in the morning than later in the day. However, there are individual
differences in circadian rhythms.
Emotional arousal enhances retention, first by focusing attention on certain aspects
of the situation. The weapons focus of crime victims attests to their vivid memory for the
central details of a situation and poor recollection of peripheral details. Second, emotion
produces bodily arousal that contributes to long-term memory formation. Third, emotional
events are distinctive. They are rehearsed and elaborated in memory.
A flashbulb memory is a particularly vivid memory for a surprising, emotional, and
consequential event. Some researchers have emphasized the persistence and apparent
accuracy of flashbulb memories over time. Others have focused on misrecall and inaccu-
racy in these memories. People have high confidence in the accuracy of their flashbulb
memories, even when those memories became altered over time.
Eyewitness memory is affected by all the complexities involved in encoding, arousal,
and emotion in memory. Because of the intense emotional arousal, one could expect
eyewitness memory to be especially accurate. Emotional arousal can facilitate encoding,
yet knowing this fact will not tell us which details are encoded.

Schemas
Encoding is affected by our extensive preexisting knowledge, often summarized in the
form of schemas.
Schemas or scripts are ways of organizing general knowledge in semantic memory.
New events are encoded into memory using the existing schema as an organizing guide.
Schemas affect attention, selection, abstraction, and normalization. Schemas also influ-
ence retrieval, sometimes leading to misrecall. Items that are consistent with a schema
(what is expected) and items that are discrepant with a schema (what is surprising) can
be well recalled, although for different reasons.

Metamemory
Metamemory is our knowledge about learning and remembering. How we learn is
affected by knowledge of which encoding factors work, our beliefs that they will indeed
work, and our monitoring of progress during learning.
There is sometimes a disconnect between metamemory beliefs and actual recall.
Students believed that one variable (the size of the lettering) and not a second (a repetition)
would increase recall. In actuality it was the discredited variable that was more effective.
 Summary 265

Self-efficacy is how effective we believe our memory will be in a given situation. It

correlates with memory performance, but it is not clear whether poor memory in past
situations lowers expectations, or low expectations reduce future memory performance.

Applications
Academic Learning and Encoding
Many student study techniques are ineffective because they do not engage any sort of
elaborative processing. Highlighting, verbatim summarizing, or taking lecture notes with-
out attending to the content of the lecture are all shallow processing activities. Elaborative
interrogation, or asking questions about the material (Why is this so? What would explain
this?) deepen processing.
 CHAPTER

10
Storage and Retrieval

CONTENTS

Storage 267 False Memory 286

Long-Term Memory for Source Memory 287
Naturalistically Learned Material 268 The Effect of Postevent Information 289
The Nature of Storage 270 Debunking Misinformation 290
Retrieval 276 Recovered Memory 291
Retrieval from Episodic Memory 278 Retrieval Versus Reconstruction 292
What Makes a Good Retrieval Cue? 279 Applications 293
Emotional Arousal and Retrieval 283 Strategies for Searching Memory 293
Prospective Memory 284 Context-Specific Learning 295
Metamemory and Partial Retrieval 285 Reconsolidation 295
Feeling of Knowing 285 Summary 295
False Retrieval 286

Why does forgetting happen? One explanation is that memories decay. Decay corre-
sponds to a commonsense intuition that memories spontaneously fade over time. As
obvious as the this sounds, a fundamental problem with this idea is its suggestion that
the passage of time alone somehow “explains” forgetting. McGeoch (1932) offered an
analogy: Iron rusts over time, yet time is not the cause of rust. Similarly, time alone does
not account for forgetting, but something that happens during that time.
In contemporary learning theory, retrieval failure is cited as the primary source of
forgetting. What appears to be forgotten is still in memory but cannot be retrieved. In an
informal survey of individuals with graduate training in psychology, 84 percent agreed
with the statement “everything we learn is permanently stored in the mind, although
sometimes particular details are not accessible” (Loftus & Loftus, 1980). This is the per-
manent-memory hypothesis.
The respondents offered several reasons for their belief in permanent memories. The
most often cited was the belief that forgotten memories can be elicited through electrical
stimulation of the brain. Wilder Penfield, a Canadian neurosurgeon, stimulated areas of
the cortex during surgery for the treatment of epilepsy. The patients were awake during

DOI: 10.4324/9781003227090-10
 Storage 267

this portion of the operation so that the surgeon could map out abnormal tissue.
Stimulation of the temporal lobes sometimes caused patients to re-experience events
from their past. For example, one 26-year-old woman reported: “Yes, I think I heard a
mother calling her little boy somewhere. It seemed to be something that happened years
ago.” When stimulated in another location several minutes later she said, “Yes, I hear
voices. It is late at night—some sort of traveling circus. I just saw lots of big wagons that
they use to haul animals in.” Penfield concluded that our memories are stored perma-
nently, a fact acknowledged in the title of his paper, “A Permanent Record of the Stream
of Consciousness” (1955/1967). Penfield’s notions of memory can be cast into a modern
analogy to a video recording: There is a complete and continuous record of our mental
experiences stored in the cortex. We just need a means of replaying those memories.
Although Penfield’s findings are widely known in psychology, their support for perma-
nent memory can be challenged. First, the majority of stimulated patients did not recover
memories. Second, the mental experiences might not be remembering, but instead
could be instances of imagination, something created at the time of stimulation. Even the
patients themselves sometimes did not recognize the “remembered” events (Squire,
1987).
Other evidence offered for the permanent-memory hypothesis is the supposed
recovery of memories through hypnosis. Hypnotized crime victims or witnesses are said
to be able to recall facts and details that were not reported before. However, hypnotized
witnesses may be more willing to venture a guess about an uncertain memory, and they
may be more susceptible to leading questions. Just as with the memories produced by
brain stimulation, we do not know if they are accurate.
The over-riding difficulty with the evidence for permanent memory is the underlying
logic: The successful recovery of some long-lost memories cannot prove that all of our life
experiences are retained in memory.
This chapter will consider several factors that affect retrieval of information from
memory. Although individual memories may or may not be permanent, retrieval condi-
tions can be arranged to enhance recall. Our first consideration, however, will be the
duration of memory. How long are memories stored and in what form are they stored?

STORAGE
We can begin with an apparent contradiction: the dual observations that we seemingly
forget so quickly and remember so long. According to Ebbinghaus’s curve of forgetting
(shown in Figure 6.1 in Chapter 6), a great deal of forgetting occurs shortly after learning
is completed. (The correctness of this principle is demonstrated if you have forgotten
Ebbinghaus’s curve). Yet our everyday experience suggests that many memories are in
fact long retained. Is Ebbinghaus’s curve wrong, or are we wrong about our own
memories?
What is particularly useful here are controlled studies that quantify retention for nat-
uralistically learned information over long time periods. A number of modern studies do
just that, assessing long-term retention of classmates, school-learned material, and even
TV shows.
268 Storage and Retrieval

Long-Term Memory for Naturalistically Learned Material

Fifty Years of Memories for High School Classmates
Harry Bahrick has studied retention of school-learned materials, ranging from high school
Spanish vocabulary to remembering the grades obtained in those classes. In an early
study, Bahrick, Bahrick, and Wittlinger (1975) assessed memory for high school class-
mates by people who had graduated as long as 48 years earlier. Retention was tested in
different ways: recalling names of classmates; recognition of their names from among a
pool of distractor names; recognition of classmates’ yearbook pictures from distractors;
and name–face matching.
There are many practical difficulties in studying memory under these conditions.
Primarily, there are individual differences among participants that may confound the
results. There are differences in exposure to classmates since graduation. Some people
remain in their hometown, return to class reunions, or take out the old yearbook to remi-
nisce. Others have had little occasion to refresh their memories over the years. Bahrick’s
participants answered detailed questionnaires about such variables so that he could sta-
tistically control for some of the differences from one person to another.
The results of several types of retention tests are shown in Figure 10.1. In one test,
the subjects were asked to name as many members of their graduating class as they
could recall. The numbers ranged from about 15 percent of their classmates for recent
graduates down to less than 7 percent for the oldest graduates. But measures of name
recognition or face recognition, and name–face matching showed basically no loss of
information out to 14 years postgraduation. In general, memory did not severely decline
until 47 years after. Even among the oldest graduates, recognition scores are in the 70 to
80 percent range. This is far better than would be anticipated from laboratory studies of
Ebbinghaus’s curve.
Why is memory so good here? Bahrick notes two factors in particular: naturalistic
learning of classmates involves repeated exposure over several years of schooling; and
spaced repetition of these exposures, e.g., across holiday breaks and summer vacations.

Remembering Knowledge Learned in School

In contrast to the rapid forgetting in Ebbinghaus’s data, several reviews of the literature
suggest that the amount lost is relatively small. Semb, Ellis, and Araujo (1993) assessed
learning 4 and 11 months after completion of a child psychology course. In comparison to
end-of-term test scores, grades fell only about 20 percent on the delayed tests. Multiple-
choice tests produced higher scores than did recall tests, just as they often do in labora-
tory tests of memory using much shorter delay intervals.
How much will be remembered from a course like the one in which you are using this
text? Conway, Cohen, and Stanhope (1991) compared groups of students who had taken a
cognitive psychology course up to 12 years earlier. The largest portion of forgetting occurred
within the first three to four years. After that, knowledge retention stabilized at above
chance levels. The shape of Ebbinghaus’s curve was present (i.e., rapid initial loss followed
by more gradual forgetting), but the time frame is of a different order of magnitude: in this
case years instead of days. Names of theorists and researchers were most likely to be
forgotten, the same material we have trouble with outside the classroom as well.
 Storage 269

100 60

50
80
Name Recognition
Picture Recognition

No. of Free-Recall Responses

Picture Matching 40
60 Name Matching
Percent Correct

30

40
20
Free Recall

20
10

34 Yr, 1 Mo
47 Yr, 7 Mo
14 Yr, 6 Mo

25 Yr, 10 Mo
3 Yr, 10 Mo
1 Yr, 11 Mo

7 Yr, 5 Mo
3.3 Mo

9.3 Mo

0 0

1 2 3
Log Time in Months

Figure 10.1
Memory for High School Classmates. The scale on the left show the percentage of correct choices in name
recognition, picture recognition, name-to-picture matching, and picture-to-name matching. The scale on the right
shows the number of classmates’ names generated in Free Recall. Recall was tested 33 months to over 47
years later in individuals from different graduating classes.
Source: From “Fifty Years of Memory for Names and Faces: A Cross-Sectional Approach,” by H. P. Bahrick, P. O. Bahrick, and
R. P. Wittlinger, 1975, Journal of Experimental Psychology: General, 104, pp. 54–75. Copyright © 1975 by the American
Psychological Association.

Memory for TV Shows

Squire and Slater (1975) tested memory for TV shows that had aired for just one season
and had not been repeated in syndication. People recognized 70 percent of titles of recent
shows (those from a year or two earlier). Interestingly, show titles from 8 to 15 years
earlier were still correctly identified 55 to 60 percent of the time. Control groups of junior
high school students, who would not have been old enough to have seen these shows,
and people who were living out of the country when these shows aired, scored at guess-
ing levels.

Memory for Public Events

Information about newsworthy events and celebrities is assumed to have been available
to most people living at the time. We can use tests of knowledge of public events from
past years and even past decades to get an estimate of the pattern of forgetting across
the lifespan. One example asks participants to identify faces of once famous individuals
(Marslen-Wilson & Teuber, 1975). The people pictured were in the news during different
270 Storage and Retrieval

decades. Older participants were good at identifying faces from the more recent decades
and forgot more of the names from earlier decades.

Long-Term Retention in Animals

Most laboratory studies of animal memory assess retention over short intervals of min-
utes or days. There are anecdotal reports of learned responses that persisted for years in
pigeons (Skinner) or dogs (Pavlov). Vaughn and Greene (1984) demonstrated long-term
retention of visual information by pigeons. Their subjects were trained to discriminate
between 160 nature photos that were cues for food availability and another 160 photos
that signaled no food. That is, the pigeons were trained to key-peck during the positive
pictures but not during the negative ones. The animals were still able to discriminate
between the rewarded and non-rewarded slides after retention intervals ranging from
235 to 730 days.

The Nature of Storage

How are memories stored in long-term memory? Models have been proposed on both
the psychological and biological levels. Psychological models describe the organization of
knowledge, and how facts are connected. Biological theories hypothesize changes in the
activity at the synapses between neurons. Examples of each of these approaches are
described in what follows.

Psychological Models of Semantic Memory

A number of network models have been proposed to illustrate the storage of general
knowledge in semantic memory. Semantic network theories assume that items of
knowledge are interconnected via associations, relationships, or pathways. These con-
nections vary in strength or, in some models, in the distance separating one item from
another. Thus, the representations of TABLE and CHAIR share a stronger connection or
they are stored closer together in memory than are TABLE and AARDVARK.
An example of a semantic network model is illustrated in Figure 10.2. Collins and
Quillian (1969) hypothesized a hierarchical arrangement of superordinate to subordinate
knowledge. For example, the ANIMAL category has subordinate categories of FISH and
BIRD. Knowledge that is general to the entire category is stored at a higher level in the
hierarchy. Information that is restricted or limited is stored at the subordinate levels. Thus,
the ANIMAL node includes information that is general to animals, such as that they
breathe and they move. The BIRD node contains the facts that birds have wings and lay
eggs. The OSTRICH node stores an exception to the generalization about birds—the fact
that ostriches do not fly.
This hypothesized organization is tested by measuring True–False reaction times to
statements that require accessing two elements in memory. Test items are selected to
be at varying distances from one another in the hierarchy. For example, “BIRDS have
WINGS” pairs two items from the same level; “OSTRICHES lay EGGS” pairs items that
are a level apart. Faster True responses to a statement indicate the two items being com-
pared are closer together, whereas slower True responses suggest the two items are
farther apart. For True statements, the farther apart the two items are in the network, the
 Storage 271

Animal

Has Skin
Can Move
Eats
Breathes

Bird Fish

Has Wings Has Fins

Has Feathers Can Swim
Lays Eggs Has Gills

Canary Ostrich Shark Salmon

Can Sing Has Long Legs Can Bite Is Pink

Is Yellow Is Tall Is Dangerous Is Edible
Can’t Fly Swims Upstream

Figure 10.2
Hierarchical Network Model of Semantic-Memory.
Source: From “Retrieval Time from Semantic Memory,” by A. M. Collins and M. R. Quillian, 1969, Journal of Verbal Learning and
Verbal Behavior, 8, pp. 240–247. Copyright Elsevier. Reprinted with permission.

longer it took to verify. This suggests that memory search begins with one item and then
trees up or down the hierarchy to the second item in order to verify the truth value of the
statement.
Reaction times to the False statements presented an intriguing finding. Some False
statements contrast two items that are farther apart than any pair of True statements. For
example, “OSTRICHES have GILLS” has two items that are widely separated in the hier-
archy. Such statements took longer to reject than did the longest True statements, sug-
gesting that the participants actually traced through the hierarchy from OSTRICH to GILLS
before responding False. Other False statements, those obviously nonsensical, were
quickly rejected, indicating that not every proposition gets thoroughly checked (e.g.,
“Statistics is a fun course”).
In other models of semantic memory, items are distanced depending on frequency
and contiguity in experience. In depicting such a model, as in Figure 10.3, the distance
between nodes reflects the strength of the connection between items in memory (Collins
& Loftus, 1975). There are also overlapping associations: Red is linked to fire engines,
other colors, and roses.
A common assumption among semantic memory models is that of spreading activa-
tion: Activation of one item in memory spreads to adjacent items, causing related knowl-
edge to be activated. Spreading activation can be demonstrated by employing a priming
manipulation in semantic memory judgments. The task is to rapidly decide whether a
string of letters forms a word. Immediately before presenting the to-be-identified item, a
272 Storage and Retrieval

Street
Vehicle

Car
Bus
Truck

Ambulance
Fire Engine
House

Orange Fire

Red
Yellow Apples

Green
Pears
Cherries

Violets Roses

Flowers Sunsets

Sunrises Clouds

Figure 10.3
Collins and Loftus Network Model of Semantic Memory. The strength of connections is illustrated by the
distance between nodes.
Source: From “A Spreading Activation Theory of Semantic Processing,” by A. M. Collins and E. F. Loftus, 1975, Psychological
Review, 82, pp. 407–428. Copyright © 1975 by the American Psychological Association.

related word is flashed on the screen to “prime” the target. For example, the word DOG
is presented as a priming stimulus a fraction of a second before the test item COLLIE.
Participants are then faster at identifying COLLIE. The priming activation of DOG in seman-
tic memory spread to related items, thus weakly activating COLLIE before it was actually
presented by the experimenter.
There are factors that limit the spread of activation. For instance, activation weakens
as it diffuses further across the network. Activation spreading from an item with many
associations will be divided across those many connections, resulting in weaker activa-
tion of each of them. This is known as the fan effect. An item with fewer associations will
have its activation less divided, and so connected items will be more strongly primed.

Neuropsychological Dissociations
The pattern of knowledge loss found with neurological injuries also says something about
the organization of semantic memory. For example, semantic memory may have sepa-
rate categories for different grammatical classes, such as nouns and verbs. Caramazza
and Hillis (1991) studied two patients who were impaired at verb production. The verbs
would be mispronounced, misread, or go unrecognized as words. The interesting test
 Storage 273

was to present homonyms, two words that sound and are spelled alike but have different
meanings. Thus, one patient could read “there was a CRACK in the mirror,” in which crack
is a noun, but could not read “don’t CRACK the nuts in here,” in which crack is a verb.
Another individual could not define words that named living things but could define
nonliving objects. When the word “dolphin” was read to him, he could not tell what it
was. The man was otherwise articulate and could define much more difficult words that
referred to nonliving things (McCarthy & Warrington, 1988). This finding suggests that
certain categories of words are stored separately in semantic memory.

Biological Substrates
The search for the engram, or the neural representation of memory in the brain, has been
ongoing since the beginnings of the field of learning itself. Researchers have explored
various ways that learning could be physically represented in the brain.
A promising mechanism for memory storage is that change occurs at the synaptic
connections. Neurons, the cells of the nervous system, adjoin one another at junctures
called synapses. Repetitive stimulation of one cell or set of cells triggers activity in adja-
cent cells. Repeated experience allows activation of one cell by another to occur more
readily. In a sense, something like spreading activation occurs from one cell to another.
This corresponds with what we say occurs when one stimulus cues another, as exempli-
fied in classical conditioning, paired-associate learning, or cued recall.
The number of synaptic connections between neurons is not fixed but is affected by
life experiences. For example, rats raised in enriched environments (toy- and play-
mate-filled cages) develop more synaptic connections in certain brain areas than rats
raised in empty laboratory cages (Greenough, 1985).
Eric Kandel and his research group started deciphering the synaptic changes that
underlie learning in Aplysia, a marine snail with a simple nervous system (see Chapter 2).
Later work focused on the mouse hippocampus. Kandel found evidence for two storage
processes, a short-term and a long-term one (Kandel, 2001). He found similar molecular
chemical processes that subserve learning in both species. An intense stimulus leads to
transient chemical changes at the synapses that allow subsequent stimuli to trigger a
nerve impulse. This is the short-term process that lasts for a few hours. Repeated intense
stimulation leads to chemical changes in the synapse, in the nucleus of the cell, and the
growth of connections (synapses) between the cells. It is these later changes that char-
acterize the formation of a durable long-term memory.

Consolidation Theory
Consolidation is the term used to describe the formation of stable and permanent long-
term memories. Dictionary definitions of consolidation include “to strengthen” and
“unite.” Both describe the memory process. Consolidation theory proposes that a
memory is first encoded in a temporary form, such as short-term memory or an unstable
form of long-term memory, and over time it consolidates into a permanent form. Hebb
(1949) introduced the term reverberation to label the persisting activity of recently stimu-
lated nerve cells. The temporary memories allow time for a permanent memory to
develop. Long-term memories are encoded by structural changes that occur at the syn-
apses between neurons. (See McGaugh, 2000, for a historical review of consolidation
theory).
274 Storage and Retrieval

According to the theory, the formation of long-term memory can be prevented by

disrupting the consolidation process. In animal conditioning experiments, the administra-
tion of electroconvulsive shock (ECS) immediately after the learning experience impairs
memory. For example, rats are given a single tone-shock pairing, which is sufficient to
condition fear in one trial. This is followed by ECS immediately or after a delay of up to 30
minutes. When tested for retention a few days later, the delayed ECS group and the
untreated controls showed freezing behavior during the tone. The ECS animals displayed
no fear to the tone (Chorover & Schiller, 1965).
Retrograde forgetting can also be produced by more targeted methods than ECS.
Another blocking treatment is the administration of protein synthesis inhibitors, which are
drugs that inhibit metabolic activity in the brain. These block the normal chemical activity
that consolidates new memory. In a study by Schafe and LeDoux (2000), rats were given
a pairing of a tone and foot shock. One experimental group was then immediately given
anisomycin, a protein-synthesis inhibiting drug. The drug was injected into a section of the
amygdala believed to be involved in fear learning. Memory for the tone-shock association
was tested the next day. The anisomycin treated animals showed no fearful behavior
during the tone. They had forgotten the earlier fear-learning. A second experimental group
also received the protein inhibitor, but it was administered six hours after the tone-shock
training. These animals did remember the conditioning and showed fearful freezing to the
tone. This indicates the protein inhibitor had to be given before consolidation had time to
take place.
Newer versions of consolidation theory posit different details, for example, whether
consolidation occurs over minutes and hours, or months and years. Hypothetical memory
curves for a generic model of consolidation are shown in Figure 10.4. There is the short-
term memory of an experience, which is strong initially but decays quickly. Then there is
the gradually strengthening long- term memory. The gradual consolidation allows time for
chemical or structural changes to develop.

Figure 10.4
Two Types of Memory in Consolidation Theory. Hypothetical strength of the memory trace over time in a
prototypical model of consolidation. In different specific models, the short-term trace ranges from seconds-to-
minutes in duration; consolidation into long-term memory occurs over intervals ranging from minutes and
hours, to days and months.
 Storage 275

Consolidation theory has been challenged for its assumption that various amnestic
treatments (e.g., ECS or protein inhibitors) prevent memories from entering permanent
storage. In several studies, memories that were supposedly blocked from consolidation
were reinstated by reminder treatments (Miller & Springer, 1973). Exposure to either the
tone or the conditioning chamber was sometimes enough to reinstate the conditioned
fear supposedly erased by ECS. Truly forgotten memories could not return if consolidation
had really been prevented. Contemporary experiments include controls to address the
possibility of retrieval failure.

Reconsolidation
Several memory phenomena described later in this chapter will show that memories can
be altered. But how can this happen if the original memory produced a fixed, structural
change in the brain, as consolidation theory suggests? Permanent memory should be
resistant to modification.
It now appears that the activation or retrieval of a consolidated memory can make it
susceptible to modification. The reconsolidation hypothesis asserts that, when a mem-
ory is retrieved, it returns to a changeable state. That memory would normally reconsoli-
date, unchanged, back into long-term memory. However, the retrieved memory is
susceptible to alteration at this time. If the outcome of the event changes, such as the
tone is not followed by shock, the changed memory would be consolidated. It is also
possible that after reactivation of the initial learning, reconsolidation of the retrieved
memory can be blocked using the same means that block consolidation: administer elec-
troconvulsive shock or protein inhibitors (Alberini, 2005). This means a previously consol-
idated memory can be erased.
An updated version of the stages in consolidation are shown in Figure 10.5. A newly
encoded memory is stored in a temporary form of LTM. Over time, the trace consolidates
into a permanent representation. Later, retrieval cues reactivate the memory. The acti-
vated memory is now susceptible to modification. At this time, different experiences can
modify the memory. The memory for the foot shock previously encoded in LTM might be
blocked from reconsolidation by now presenting ECS. The altered memory is reconsoli-
dated back into permanent LTM.

Figure 10.5
Consolidation of Long-Term Memory. Updated flow chart of the steps in the consolidation of long-term
memory. At the time of retrieval or reactivation, memory can be altered and then re-entered back into
consolidated memory.
276 Storage and Retrieval

An example of reconsolidation blocking was shown in a series of experiments in

which an amnestic treatment was used to eliminate a previously consolidated memory.
Nadar, Schafe, and LeDoux (2000) first conditioned a tone-shock association, and then
reactivated the memory.
On Day 1, rats were given a single pairing of a tone and foot shock. This was sufficient
to condition fear of the tone, leading to freezing behavior. Thus:

Day 1 fear conditioning : tone → shock

The next day, the animals received a reminder treatment. The tone stimulus alone was
presented. This reminded the rats of the previous learning, and the rats exhibited freezing
(immobility) during the tone. The animals were then given anisomycin, a protein-synthesis
inhibiting drug. The anisomycin was injected into the amygdala in the brain, an area that is
active in fear learning. The drug was hypothesized to prevent neural activity to block
reconsolidation of the retrieved trace back into permanent memory. Thus, on Day 2:

Day 2 memory reactivation : tone  freezing  Amnestic Drug

Day 3 test fear memory : tone  fear absent

Memory for the tone-shock association was tested on Day 3. The anisomycin treated
animals, who had remembered what the tone meant the day before, now showed no
fearful freezing behavior. They have forgotten the earlier fear-learning. That is, conditioning
on Day 1 was reactivated on Day 2 and then erased, revealing no trace on Day 3 (Nadar,
Schafe, & LeDoux, 2000). (Control groups established that without anisomycin on Day 2,
fear would have been present on Day 3).
Reconsolidation is analogous to updating a document on your computer. The docu-
ment is retrieved from storage on a hard drive. Then, while it is active on your desktop you
can make modifications. When you hit “Save,” the new document replaces the old one on
the hard drive. Updating is an important adaptive function of memory. For an animal, a
place that was once safe is now dangerous; where food was once found, it is not present
later. When the remembered does not match the present, this should set the condition
for altering memory.

RETRIEVAL
One explanation of forgetting is that the memories are simply not there. Memory has
faded or decayed; it was not consolidated; it has been replaced by a new memory through
reconsolidation. A recurring theme in this book is that much more is available in memory
than is typically recalled. Retrieval failure is an alternative explanation for forgetting. In
most research on retrieval, subjects are specifically asked to recall some knowledge or a
memory. We saw this in the long-term memory studies described in the section on Storage;
Explicit requests to recall are also used in the episodic memory studies reviewed below.
However, remembering also just happens, without intention, direction, or searching. These
spontaneous memories can also teach us about the conditions for retrieval (see Box 10.1).
 Retrieval 277

BOX 10.1 SPONTANEOUS RECALL

Sometimes a memory just comes to mind without an intention to recall. We are likely think-
ing or doing something else entirely, when from out of nowhere comes a memory from
childhood, or last year, or last week. Why did I just remember that? Involuntary autobiograph-
ical memories are memories of events and experiences from one’s personal past. Research
and theory have only recently begun to address uninvited remembering.
This chapter emphasizes the importance of retrieval cues to access memories. Involuntary
memories are no different. It is probable that the current context and our current thoughts act
as the cues for the involuntary memory. Berntsen (2010) offers an example: A student notices
the sound of wind and thunder outside the house, and then remembers an earlier event in
which she calmed a young child who was afraid of thunder. The stimuli that retrieved involun-
tary remembering include context (time and place), internal feelings (affect or emotion),
actions, thoughts, and associations. The specific combination of several dimensions evoke a
particular memory.
Data on involuntary memories can be obtained through surveys, or by asking people to
track spontaneous memories in diary studies. Involuntary autobiographical memories occur
at least several times a day (Berntsen, 1998), although this may be an underestimate.
Spontaneous memories are often emotional: positive and pleasant, or negative and unpleas-
ant. (As mentioned in Chapter 9, emotional events are encoded better, are more distinctive,
than nonemotional events). Negative emotional memories can be a source of distress, as is
the case of flashbacks in PTSD.
Spontaneous recall may be like the free association test psychologists once used. “What’s
the first thing that comes to mind when I say: TABLE? SCHOOL?” The associations that are
produced are not really free and unconstrained. The word may prompt idiosyncratic associa-
tions, personal memories, or ideas connected to the present context.
Kvavilashvili and Mandler (2004) studied another class of spontaneous memories: invol-
untary semantic memories. This can be a word, an image, or a tune that comes to mind,
but it is not a memory of a personal experience. These semantic memories do not have the
emotional feeling that autobiographical memories have. Semantic long-term memory rep-
resents our generalized knowledge, so an involuntary semantic memory might simply be
the triggering of a bit of knowledge. The stimuli for involuntary semantic memories are
sound-alike cues (e.g., you recently heard a similar sounding word or tune); or semantic
cues (e.g., hearing or seeing the word DOG might lead to an image of a COLLIE, or vice
versa).
Semantic memories may be supplemented by priming effects. A word, image, or idea that
was recently processed is primed and so is easily available in memory. Priming also occurs
through spreading activation in a semantic memory network. A prime activates a network of
words and ideas, which are primed and waiting to come to mind.
Another type of involuntary memory is the experience of having a song stuck in your head.
An “earworm” or “brain bug” is usually a short fragment of music, and the tune or song is
repetitive (it plays over and over). The triggers are similar to the cues for spontaneous mem-
ories: Recent exposure, repeated exposure, and associations (sound, situation, or word)
were most frequently reported (Williamson et al., 2011). Earworms occur with similar fre-
quency to involuntary memories, that is, they occur frequently. Too frequently. (No one has
reported how to get rid of the songs).
278 Storage and Retrieval

An evolutionary perspective suggests that spontaneous memory can serve an adaptive

purpose. Organisms live in a constantly changing environment and memory bridges the tem-
poral intervals across changing events. Being frequently reminded of things in memory gives
us the opportunity to notice relationships that recur (“the same thing happened when…”) or
changed (“that didn’t happen the last time…”).
Spontaneous recall might also be beneficial for memory updating. Updating suggests
replacing one memory with another, e.g., where I parked this morning replaces the memory
of where I parked yesterday. Hintzman (2011) says that a strategy would be to keep both old
and new memories, at least for a while. Spontaneous memory may be the mind’s way of
checking whether an earlier memory is relevant to the present, and which might provoke an
update.

Retrieval from Episodic Memory

What factors affect retrieval? Several variables are important, but three general factors
can be offered: the distinctiveness of the memory, practice at retrieving the memory, and
the presence of effective retrieval cues.

Distinctiveness
Events that are distinctive, or that stand out from the background of other events, are
generally well remembered. In Chapter 9 we saw that flashbulb memories are well
recalled, even if imperfectly, because they represent events that are so different from
other events in our lives. In the isolation effect, an item that differs from the remaining
items in a list is better recalled.
Distinctive events might be retrieved better because their retrieval cues uniquely
target a single memory. The request to recall where you were on 9/11 (September 11,
2001) cues a very particular memory. By contrast, other retrieval cues might be too broad,
retrieving many items but not specifically one item (Roediger, 1973).

Retrieval Practice
Retrieval is facilitated by previous retrieval. For instance, taking a test shortly after study-
ing greatly enhances the amount that can be recalled on a later retest. The general design
of studies of testing effects is to give one or more practice tests shortly after learning,
and then a final test sometime later. It is this later test that interests us. For example, the
number of pictures recalled at the end of a one-week delay nearly doubled when three
other tests of memory had been given, immediately after studying (Wheeler & Roediger,
1992). (Testing effects were described in Chapter 9).
Hypermnesia offers a dramatic demonstration of testing effects. Hypermnesia is the
case in which recall actually improves over successive attempts at remembering, in con-
trast to the forgetting that we expect to occur over time. Erdelyi and Kleinbard (1978)
presented a list of 60 object line drawings for study (a DOG, a CAT, a TREE, etc.). The list
was presented just once for study, but memory was tested more than 20 times over the
following week. The total number of different items recalled across repeated testing is
 Retrieval 279

Figure 10.6
Hypermnesia. The number of different items recalled after a single presentation of a list. The list was tested
more than 20 times over a period of a week.
Source: From “Has Ebbinghaus Decayed with Time? The Growth of Recall (Hypermnesia) over Days,” by M. H. Erdelyi and J.
Kleinbard, 1978, Journal of Experimental Psychology: Human Learning and Memory, 4, pp. 275–289. Copyright © 1978 by the
American Psychological Association.

shown in Figure 10.6. Whereas just over 26 items were recalled on the first test, 38 were
recalled on the final tests. Hypermnesia is the opposite of forgetting, or amnesia.
Hypermnesia occurs with other material, including recall of general information,
foreign-language vocabulary, and names (Bahrick & Hall, 1993). The items that were
recalled varied from test to test, being remembered or not and vice versa. From one test
to another, some items are indeed forgotten. But the number of newly recovered items
on successive tests exceeded the number lost, leaving hypermnesia as a net gain.
Why does more remembering occur across successive tests? One factor is that dur-
ing free recall, people prompt their memories with self-generated cues (Roediger &
Thorpe, 1978). These cues vary across successive tests as different things come to mind.
Different cues retrieve items that were not adequately cued on the earlier tests.
This capacity for additional recall of material across a series of tests has implications
for eyewitness testimony. One assumption too often made is that the first recall is the
most valid, and anything remembered later is suspect. Yet here we are asserting that
additional accurate information can be elicited through repeated recall attempts. In a sim-
ulation study of witnessing, participants watched a film depicting a burglary. Memory
recall was tested four times, and each time additional details were remembered without
an increase in intrusions of incorrect facts (Scrivener & Safer, 1988).
In addition to distinctive memories and practice at retrieval, a third important factor in
retrieval is to have effective retrieval cues.

What Makes a Good Retrieval Cue?

We will consider two answers here: cues that have strong preexisting associations to the
target memory, and cues that were encoded along with the to-be-recalled item when it
entered into memory.
280 Storage and Retrieval

Associations
Effective retrieval cues are those that are connected to the target by strong associations.
Given the word TABLE, you might retrieve CHAIR as the first association that comes to
mind, but not AARDVARK. TABLE is a good retrieval cue for CHAIR but not for AARDVARK.
Memory is dramatically improved when effective cues are given. In a typical free
recall experiment, students might be able to recall 15 out of 48 words presented in a
single list. If a retrieval cue was presented for each word (e.g., was there an ANIMAL in
the list? A FOOD? A PLANT?), the number of words remembered increases to around 36
(Tulving & Pearlstone, 1966).
In a dramatic demonstration of the effectiveness of retrieval cues, Mantyla (1986)
attempted to improve recall of several hundred random words. As each word was pre-
sented, the subjects were asked to generate up to three properties that described the
word. These self-generated (one-word) cues were presented to the subjects during the
test phase. More than 90 percent of the cued words were recalled on immediate testing,
and even after one week, 60 to 65 percent of the words could be recalled. If someone
else’s cues were used, only 20 percent of the words were remembered after a week. In
one massive session, subjects studied 600 words and generated three cues for each, and
then were tested for all 600 words. (The procedure took over six hours). Again, 90 percent
of the words could be retrieved from all three cues, and 60 percent from one cue.

Encoding Specificity
Cues that are effective in aiding retrieval are the cues that were present at the time of
encoding. If you forget where you left an object, you mentally go back to where you were
when you last had it. If you want to recollect your childhood, go back to where you lived
as a child. The cues that were present when the memory was encoded aid retrieval later.
This idea of matching cues is made explicit in Tulving’s encoding specificity princi-
ple: Retrieval is better when the cues present at encoding also occur at the time of
retrieval (Tulving & Thompson, 1973). In one demonstration, the participants studied tar-
get words that were paired with weakly associated words as study aids. Thus, the
to-be-remembered word LIGHT is presented along with its weak associate HEAD. On the
recognition test the target words were presented with the same cue words as before, or
with cue words that had stronger associations to the target (e.g., DARK is presented as
the cue). LIGHT was more often recognized in the context of the weak associate HEAD
than when in the context of the strong associate DARK. This is because LIGHT had initially
been encoded in the context of HEAD LIGHT rather than of DARK (see the results in
Table 10.1). According to the encoding specificity principle, recognition occurred when
the test reactivated the same meaning with which the target words had been encoded.
The experimental arrangement of manipulating the encoding conditions in combina-
tion with the retrieval conditions is referred to as the encoding–retrieval paradigm
(Tulving, 1983). The cues present during encoding can be arranged to be the same or dif-
ferent at the time of retrieval. The cues can be words, as in Tulving and Thompson’s exper-
iments. The cues can also be contextual stimuli of time and place; internal cues of
emotional or drug-induced mental and physical states; or cognitive operations the partic-
ipant performs with the target material.
 Retrieval 281

Table 10.1 Variations in the Encoding–Retrieval Paradigm

A. Encoding-Specific (Tulving & Thompson, 1973, Table 1)

Study Test Percent of Words Recognized

head LIGHT head LIGHT 65
dark LIGHT 23

B. Context-Specific (Godden & Baddeley, 1975)

Study Context Test Context Percent of Words Recalled

Land Land 38
Water 24
Water Land 23
Water 32

C. (Drug-)State-Dependent (Eich, Weingartner, Stillman, & Gillin, 1975)

Study Drug State Test Drug State Number of Words Recalled

Placebo Placebo 11.5

Placebo Marijuana 9.9
Marijuana Placebo 6.7
Marijuana Marijuana 10.5

D. Mood-Dependent Recall (Bower, 1981)

Mood at Study Mood at Test Percent Recall

Happy Happy 78
Happy Sad 46
Sad Happy 46
Sad Sad 81

Contextual Learning
You have probably heard the advice to study in the same room you will take the test in.
There is some validity, with limitations, to this claim. For example, Smith (1979) had stu-
dents study a word list in a college classroom. The students were then tested either in the
same room or in a different room. (Both rooms were made equally familiar in a preliminary
phase). More words were recalled when testing was in the same room in which the word
list had been presented. Smith also found that the different-room deficit was reduced if
the students were instructed to mentally “reinstate” the study room prior to testing. You
don’t actually have to be there; just imagine you are there! This idea has been applied to
a witness interview technique that emphasizes reinstatement of a crime or accident
scene as an aid to retrieval.
282 Storage and Retrieval

State-Dependent Learning
State-dependent learning refers to better recall when testing occurs in the same drug-in-
fluenced state as was present during learning. A variety of drugs, including alcohol and
marijuana, can produce state dependency in humans and animals. A study by Eich et al.
(1975) illustrates the case for marijuana. Participants studied word lists either while under
the influence of marijuana or not and then were tested after exposure to marijuana or a
placebo. (The use of controlled drugs for research purposes is highly regulated. Potential
participants must be screened. They cannot be recruited from the Psych 101 subject pool
for dope studies).
The design of this study and the results are shown in Table 10.1. There was a drug
state-dependency effect: Recall was better in the two groups in which the drug conditions
at testing matched those present while studying. That is, recall was better in the group
who studied and tested after placebos and in the group who studied and tested after the
drug.

Mood-Dependent Recall
Bower (1981) theorized that internal stimuli arising from mood and emotional state
can enter into associations just as do the external stimuli that experimental psycholo-
gists typically use. Bower, Monteiro, and Gilligan (1978) demonstrated mood-depend-
ent recall. Happy or sad moods were induced in hypnotized college student participants.
Recall of the to-be-remembered material was better when students were tested in
the same mood state, whether this was happy or sad, as they were in when learning
had occurred (see Table 10.1). This is sometimes referred to as mood-congruent
memory.
Mood-specific recall has implications for depression and other affective disorders.
Being depressed may lead to the retrieval of mostly sad memories, which only perpetu-
ates the depressive mood. For example, if depressed individuals are given cue words and
asked to recall any personal memory that comes to mind, they more often recall unpleas-
ant experiences. In longitudinal research designs that tested the same individuals at dif-
ferent times, more unpleasant memories were recalled when the individuals were
depressed, and the frequency of pleasant memories increased when they were less
depressed (Clark & Teasdale, 1982; Fogarty & Helmsley, 1983).
The opposite of depression is the excited state of mania. A word-association test was
administered to individuals currently in a manic state. On succeeding days, the manic
individuals were asked to recall the associations generated in that previous free-associa-
tion phase. Recall was better when the mood states matched (Weingartner, Miller, &
Murphy, 1977). According to one anecdotal report, a man in a manic state hid a large sum
of money in his house. After his mania abated, he could not remember where. Six months
later, the man became manic once again and went to hide something else behind a pic-
ture, whereupon he found the lost money. Although he still did not remember hiding the
money, it is interesting that the same location seemed so sensible in the manic phase
(Williams & Markar, 1991).
What do the several manipulations of context, drug-induced state, and mood have in
common? One possibility is that they all reduce to mood-dependent memory. What is
 Retrieval 283

common when encoding and retrieval conditions are equated is “not environmental con-
text dependent memory, but rather experiential context dependence, of the kind custom-
arily associated with alterations of a person’s affective, circadian, or pharmacological
state” (Eich, 1985, p. 769, italics in the original). Eich (1995) found that matching the
self-rated affect (i.e., feelings) between encoding and recall was more important than
matching physical locations. That is, being in the same mood was more important than
being in the same place.
Encoding-retrieval effects are regularly produced in other situations. As examples,
state dependent effects have been shown using background music (Smith, 1985) or
odors (Cann & Ross, 1989) as context stimuli. In animal experiments, manipulations
of body temperature and time-of-day can produce state dependency (Richardson
et al., 1984; Holloway, 1978). Finally, subjects exposed to a list of words while expe-
riencing experimentally induced pain (i.e., immersing their hands in ice water) later
remembered more words when exposed to the pain prior to testing recall of the
words.

Limitations of Encoding–Retrieval Paradigm Effects

We should not overemphasize the effect of matching encoding and retrieval cues on
memory retrieval. State-dependent effects are more often found on recall tests but not
on recognition tests. State-dependent effects are also obtained after minimal amounts of
learning; well learned material becomes independent of context. These facts suggest that
retrieval cues are helpful when memory is weak and the participant therefore has few
self-produced retrieval cues to access the memory.
Also, not all contextual cues, moods, or drug-induced states necessarily become
associated with the target. For instance, if cues in the classroom do not become associ-
ated with the lecture material in some manner, such as inspecting some element of the
room while thinking about a lecture point, the room should have no special retrieval power
(Smith & Vela, 2001).

Emotional Arousal and Retrieval

Emotional arousal can have opposing effects on memory. Emotions are stimuli that can
become associated with other events in memory, and thus act as retrieval cues. And
emotional arousal can impair our ability to retrieve memories.
High levels of an emotion, particularly fear or anxiety, can block retrieval of memories
that otherwise would be recalled. Smeets et al. (2008) tested their subjects’ memory for
word lists. Some participants were given cold pressor stress before attempting retrieval
of the words. (Participants place one hand in a bucket of ice water for as long as they can
tolerate—up to three minutes). Fewer words were recalled by the retrieval stressed sub-
jects in comparison to control groups which received the stressor at other times. (This
study was presented in more detail in Chapter 7).
In one particularly dramatic demonstration of emotional blocking, army recruits were
tested while on a swinging rope bridge over a ravine. Compared to nonfear control groups,
the bridge crossing group recalled significantly less (Capretta & Berkun, 1962). Research
of this nature addresses a significant question about whether emotional arousal could
284 Storage and Retrieval

interfere with the performance of critical duties. Individuals trained in emergency proce-
dures might fail to recall them under the stressful conditions of an emergency.
There is a correlation between the level of stress hormones in the blood and remem-
bering. The Smeets et al. (2008) study cited earlier found significant negative correlations
between measures of stress hormone levels and the number of words recalled. The more
stressed subjects retrieved fewer words. In an animal study, rats were given electric
shocks before being tested in a maze. The shocks were used to induce stress hormones.
The rats remembered less 30 minutes after the shocks, when stress hormones were
highest in the blood. Memory was not impaired 2 minutes after the shocks, before hor-
mone secretions peaked; or 4 hours after the shocks, when the hormone levels had
subsided (deQuervain, Roozendaal, & McGaugh, 1998). Memory was not retrievable
when stress levels were elevated.
Among people exposed to the same unpleasant event, individuals may experience
different degrees of negative emotion. Does the experienced intensity of negative emo-
tion affect memory? In some cases, greater unpleasantness provokes less remembering.
Crying by children would be a good indicator of negative emotional arousal. In one case,
children aged 2 through 6 underwent a catheterization procedure, during which some of
them cried. (Only some cried?) Six months later, the children who had cried recalled less
information and with less accuracy, even when given memory prompts (Salmon, Price, &
Pereira, 2002).
In a student’s life, test anxiety can interfere with retrieval. Benjamin, McKeachie, Lin,
and Holinger (1981) found that test-anxious college students performed poorly on essay
and short-answer questions, which make heavy demands on retrieval of information. The
students performed better on multiple-choice tests, which are less demanding on retrieval.
Other research confirmed that some anxious students benefited from training to calm
themselves in the test-taking situation, and their grades actually increased the following
semester (Naveh-Benjamin, 1991).

Prospective Memory
One of the lessons learned from memory experts is that they plan for retrieval at the time
of encoding (see Chapter 11). That is, while they are rehearsing, elaborating, imaging, and
all the other things people do when they try to remember, the experts are also devising
strategies to retrieve the encoded material later.
Retrieval is not always about recalling the past. Prospective memory is remem-
bering to perform future actions. For instance, if you had the intention to do something
at a future time or place, then a prospective memory failure would be forgetting to
recall at the correct time or place. We can use external cues and electronic devices to
bypass our dependence on memory. Professional bartenders use external cues when
they set out each type of glass needed to fill an order. A programmed alert reminds you
when to act.
In other instances, prospective memory depends on internal cues, or “remembering to
remember.” Remembering to take something out of the oven, to stop at the store on the
way home, and to call someone later that day are examples. Recalling depends on an inter-
nal prompt to remember at the correct time. Prospective remembering requires planning.
 Metamemory and Partial Retrieval 285

One of the lessons learned from studying memory experts is that they plan for retrieval at
the time of encoding (see Chapter 11). That is, while they are rehearsing, imaging, etc., to
encode, the experts are also devising strategies to retrieve the encoded material later.

METAMEMORY AND PARTIAL RETRIEVAL

Retrieval can produce incorrect memories in many ways. Memories might be partial,
incomplete, inaccurate in details, or just plain wrong. Retrieval is also affected by metam-
emory, or our knowledge about memory. We know how to search memory, for example,
by using active strategies to facilitate recall. We also know what is stored in our memo-
ries (or at least, we think we know what’s there), and what is not in memory. This latter
form of knowing has been studied in two related phenomena: feeling of knowing and
tip-of-the-tongue.

Feeling of Knowing
At one time or another, each of us has had the experience of not being able to remember
something even though we are sure that we know it. This feeling of knowing (FOK) is
characterized by an “irritating mixture of surety and bafflement. The individual is convinced
he knows but is frustrated by the inability to demonstrate his knowing” (Reed, 1979,
p. 9). FOK is related to tip-of-the-tongue (TOT). This is a more intense feeling that not only
do we know the sought-after word, but we are so close to recalling it. In Brown and
McNeil’s (1966) classic study, tip-of-the-tongue “states” (as the authors labeled them)
were elicited by reading definitions of unusual words to their college students. In the TOT
state, the participants could accurately report the first letters of the forgotten word, or the
number of syllables it contained. For example, an unremembered street name led to
recall of similar names that were rejected: CONGRESS, CORINTH, and CONCORD. The
actual street name was CORNISH. What is interesting about FOK and TOT is how confi-
dent we are that we know, even though the item cannot be recalled to verify our feelings.
(The tip-of-the-tongue experience seems to be universal, although the linguistic expres-
sion for it varies. Schwartz found evidence in 45 languages that referred to the tongue: On
top of my tongue, lost on my tongue, or the Korean “sparkling at the end of the tongue”
[Schwartz, 1999]).
These partial recalls are also interesting because we can readily reject wrong answers.
As William James said:
Suppose we try to recall a forgotten name. …There is a gap therein…a gap that is
intensely active. …If wrong names are proposed to us, this singularly definite gap
acts immediately to negate them. They do not fit into its mold. And the gap of one
word does not feel like the gap of another, all empty of content as both might
seem…when described as gaps.
(1890, p. 251)

How do we know our feelings of knowing are correct if we cannot recall the item? Hart
(1965) introduced a three-step procedure to assess FOK accuracy. The participants are
first tested with some general information questions. For example, “Which planet is the
286 Storage and Retrieval

largest in the solar system?” Second, for the unanswered questions, the participants rate
their confidence that they really do know the answer. These are the FOK ratings. Third, a
multiple-choice test is given to see whether the unrecalled items are selected. For the
example question given here, the alternatives might be “Pluto, Venus, Saturn, and Jupiter.”
Items that received high FOK scores are more likely to be answered correctly in multi-
ple-choice tests, showing that the initial failure was indeed one of retrieval.
High FOK ratings might suggest that people can assess the strength of unrecalled
knowledge. People know what they know. However, respondents often use other infor-
mation to infer whether they should know the answer. The recall of related facts or famil-
iarity with the topic may mislead you into thinking you will know the answer. In these
cases, the participant makes an educated guess that the answer is indeed known (Nelson
& Gerler, 1984).
It is also the case that high FOK ratings sometimes correspond to flat-out wrong
answers. People give a high FOK rating to a particular question; select the wrong answer
on the multiple choice; and are highly confident in the correctness of their (wrong) choice.

FALSE RETRIEVAL
False Memory
A different sort of retrieval failure is the mistaken recall of some stimulus or event that
had not actually occurred. False recall and false memory are used to label mistaken rec-
ollection. Deese (1959) demonstrated false recall by presenting lists in which the words
were associated with a target word that was not included in the list. For example, a pre-
sented list might include: BED, REST, AWAKE, SNOOZE, SLUMBER, SNORE.
When subsequently tested, many participants believed that the word SLEEP had
been on the list. It was not. Deese found that false recall of the associated word (e.g.,
SLEEP) occurred 30 to 40 percent of the time, and Roediger and McDermott (1995) found
intrusions in as many as 50 percent of their participants. (This task is now referred to as
the Deese-Roediger-McDermott procedure, or the DRM, in the research literature).
The false memory effect is robust. For instance, warning subjects about it decreases
false recall only slightly, and by making participants cautious, the warning decreases cor-
rect recall. Falsely remembered items are forgotten over time at about the same rate as
the presented words. In fact, the associated words were remembered better than the
actual list words on tests given two days to two months later (Seamon et al., 2002).
About the only factor that reduces false recall is to present the list items as pictures (i.e.,
line drawings). During testing, there is more certainty in recall of the specific images
(Israel & Schacter, 1997; see review by Roediger & McDermott, 2000).
Why does false recall occur? According to the notion of spreading activation in seman-
tic memory, activation spreads from the representations of each list item in memory to
associated items in memory, including the unpresented word. The cumulative effect of
activation from several cues is stronger activation of the related word. The more associates
there are in the list, the more likely false retrieval becomes (Robinson & Roediger, 1997).
An important question in explaining false recall is when do they appear in memory: at
encoding or at retrieval? One could imagine at the time of list presentation the lures are
 False Retrieval 287

activated via association in memory. Alternatively, maybe the lures first appear during
testing, as the subject uses some remembered words to retrieve a few more. One way
to test for presence of the target words during study is to have the subjects rehearse out
loud while they are listening to and encoding the list words. In one case, the associated
targets were frequently said aloud by the subjects during the study phase and became
“part of the lists” (Goodwin, Meissner, & Ericsson, 2001).
In some ways, the falsely recalled words are as real in the brain as are the actually
presented words. If event-related potentials are recorded, a form of EEG recorded from
electrodes placed on the scalp, the reaction to falsely recognized words (e.g., SLEEP) is
no different from that to correctly identified words (e.g., SLUMBER), and differs from dis-
tractors presented during testing (e.g., DOG) (Johnson et al., 1997). Thus, calling this phe-
nomenon “false retrieval” may be a misnomer. From the perspective of the neural reaction,
the recognition of actually presented and associatively activated words does not differ.

Generality of False Memory

The DRM procedure has been tested in many experimental variations and in several
types of subject populations. Generally, some degree of false recall of the lures occurs no
matter what the variation. Some examples follow.
Amnesics Show False Recall. Not surprisingly, amnesic individuals remember fewer
of the actual list words, and they remember fewer of the associated lures that were not
presented. But they do recall more lures than expected by chance or guessing (Schacter,
Verfaellie, & Pradere, 1996).
Children Are Susceptible to False Recall. Children remember fewer words than do
adults, either actual list items or the lures. Comparisons of children across ages 6 to 11
show increasing likelihood of false recall, and this continues through adolescence. Getting
older and remembering more actual words is accompanied by more misremembering
(Brainerd, Forrest, Karibian, & Reyna, 2006). The increased general knowledge possessed
by older children might make them more susceptible to false recall.
Even Experts Falsely Remember. The DRM procedure can be demonstrated among
specialist groups or experts who share unique associations to words in their fields.
Undergraduate business students could recall more investment-related words (such as
P-E ratio, IPO, day trade, and midcap) than could nonbusiness students. This was as
expected, given that these words were familiar to the business students. However, the
business students also falsely recalled more business-related words that had not been on
the list but were associated with the list words (Baird, 2003).
False Recall Crosses Languages. Bilingual participants can be presented with word
lists in one language (e.g., French or English), and a recognition test in the other language.
False recognition of the lures still occurs (Cabeza & Lennartson, 2005). Possibly there is
a common lexical representation that is activated by the words of either language.

Source Memory
The false memory phenomenon has parallels in other false retrieval phenomenon. Have
you ever been unable to remember whether you actually said something or had only
intended to say it? This distinction is one of source memory: distinguishing between an
288 Storage and Retrieval

actually experienced event and one that was imagined, thought of, or even dreamed (e.g.,
Johnson & Raye, 1981). A memory’s origin can be external in the words we spoke, actions
we performed, or objects we perceived; or the memory’s origin could be internal in words
thought of, actions planned, and objects imagined.
We distinguish between externally and internally derived memories by making a judg-
ment about the qualitative aspects of the memory: What does the memory feel like?
Johnson and Raye (1981) hypothesize that we look for certain attributes that indicate
whether the memory seems to have originated externally or internally. External memo-
ries are richer in sensory attributes: there are stronger traces of sight, sound, or texture.
The memory is more firmly set in time and place, as events preceding or following the
target are also remembered. In contrast, imagined memories are more schematic and
lacking in sensory detail. There is less recall of what came before and after the remem-
bered event. A judgment is made on the likely origin of a memory based on the number
of internal and external attributes. Some memories will be judged probably internal or
probably external, and some memories we are just not sure of.
Mistaking the source of a memory could lead to the false belief that the event really
occurred. (Remember Piaget’s vivid memory of having been kidnapped as a young child,
when in fact he was only told that he had been kidnapped). Loftus and Pickrell (1995)
were able to implant childhood memories in children and adults by getting family mem-
bers to talk about fabricated events as if they had really happened. In one case, a 14-year-
old boy was told how he had been temporarily lost in a shopping mall several years prior.
Other family members corroborated the story, adding details. After two days the boy
really believed that he had been lost, “remembering” more details, his feelings while lost,
and even the man who eventually rescued him. (I suspect we now have a generation of
young adults who vividly “remember” events from childhood, when in fact what they are
remembering are the videos of themselves that they saw later).
Correctly recalling sources is a problem in collaborative memory tasks. Participants
take turns generating ideas or solutions to problems. Afterwards, the subjects are asked
to recall all of the proposed answers, and who said which. Individuals recall they had
generated more of the ideas then they in fact had. This finding relates to unconscious
plagiarism, in which a person copies a word or an idea without realizing it.

Whose Memory Is It?

A new form of source memory is the memory whose origin is claimed by two people. For
example, one child claims to remember some long-ago event, such as being sent home
from school, but a sibling claims ownership of the memory and the experience. In study-
ing a sample of same-sex twins, about two-thirds had such disputed memories (Sheen,
Kemp, & Rubin, 2001). It is also possible that neither twin actually experienced the event
and both are falsely remembering.

Imagination Inflation
Internally generated memories can seem more real if they are thought about often. In
imagination inflation, frequently imagining or thinking about some event or action leads
to an increase in the belief that the action or event actually happened. Goff and Roediger
(1998) had their subjects perform some simple actions, such as knocking on the table or
breaking a toothpick, and imagine performing other actions. Some imaginings were
 The Effect of Postevent Information 289

repeated. The more frequently an action had been imagined, the more likely it was
remembered as an actual memory.

THE EFFECT OF POSTEVENT INFORMATION

One of the more provocative issues in memory research today is the effect of postevent
information on recall. After to-be-remembered information is presented, other contradict-
ing information occurs. Will the postevent alter the target memory? For example, we
show our participants a short video depicting a minor car accident. Then they were asked
a misleading question. For instance, we ask about a traffic sign that was not present in
the film. (“How fast was the car going when it passed the stop sign?”) Participants will
later recall seeing the sign in the film (Loftus, Miller, & Burns, 1978). This procedure and
its outcome are referred to as the misinformation effect: remembering incorrect infor-
mation instead of remembering the previously presented correct information.
The misinformation effect can occur for several reasons. One possibility is that the
original memory is no longer there. Maybe the first event memory has been replaced,
and so the postevent fact is all that is available for the subject to recall. Another possibility
is that the subject remembers both memories at the time of testing, but only reports the
postevent information because the experimental procedure seems to suggest that is the
correct response.
The misleading questions can also cause misrecall. This may be due to acquiescence
with the questioner’s suggestions, or an assumption that the questioner must be right
about what appears (to the participant) to be an unimportant detail.
The effects about postevent information are related to interference theories of forget-
ting. A persisting concern in laboratory experiments is whether remembering one event
retroactively interferes with retrieval of the earlier memory, or whether it actually alters
the earlier memory. Estes (1997) argues that some of both occur. In an early study by
Zangwill (1937), participants were shown a drawing of an irregular figure and then pro-
duced a series of reproductions (see Figure 10.7). Systematic distortion appeared across
the successive drawings, which became less like the original. However, when a recogni-
tion test was given afterwards, the original figure was more often correctly chosen and
not some figure closer to the participant’s own most recent reproductions. Memory for
the original object was still there even after much potentially distorting interference.
A similar result occurs in recall followed by recognition of spatial information.
Participants are shown dots within a circle. The circle is divided into quarters by + shaped
lines. Immediately after viewing, the students try to reproduce the dot locations. The
reproductions are consistently off, being closer to the center of the quadrant. However,
when then given a recognition test among several options, the students select the cor-
rect one, even over versions that more closely match their own reproductions (Sampaio
& Wang, 2009).
Attempts to correct misinformation can sometimes backfire. When your professor
tells you that such-and-such a theory was proven wrong, all you remember at the time of
the test is that the professor talked about such-and-such theory. It must be a right answer!
An example of this sort of misremembering occurred in an attempt to correct medical
misinformation. A number of hypothetical medical facts were presented, in the context of
correcting medical myths. So some of the facts were given the disclaimer that they were
290 Storage and Retrieval

Figure 10.7
Distorted Reproduction but Correct Recognition. Zangwill (1937) had participants study figures like that
labeled TARGET. The participants then made several drawings from memory, leading to successive distortions.
Finally, the participants were given a multiple-choice recognition test. The participants correctly chose the target
from among distractor figures that were more like the participant’s own reproductions.
Source: From “Processes of Memory Loss, Recovery, and Distortion,” by W. K. Estes, 1997, Psychological Review, 104, p. 159.
Copyright © 1997 by the American Psychological Association. Reprinted with permission.

not true. On later testing, recognition of the facts presented earlier was excellent, but
many more errors were made in remembering which had been labeled false.
Misremembering was worse when the facts had been presented, and labeled false, three
times (Skurnik, Yoon, Park, & Schwarz, 2005).

Debunking Misinformation
Misinformation is a subject of concern and conflict in today’s world. It is a broad topic dealing
with questions such as who disseminates misinformation (and how, and why), who believes
the misinformation (predisposing biases), and how misinformation can be corrected.
Memory researchers focus on the more-restricted role of remembering misinformation.
The primary tactic to counteract misinformation is to present correct information.
Simply inform people of their misconceptions by telling them the facts. As rational and as
logical this technique seems, it rarely works. In research using fabricated and non-contro-
versial topics, corrective “true” information is not often remembered. One reason is the
familiar primacy effect: Misinformation is presented first, and then corrective information
comes along later. Wrong information has already been incorporated into memory,
whereas corrective information requires rewriting memory. Failure of source memory is
another factor that works against corrective information. Misinformation that is recalled
could be misremembered as having come from the authoritative source. (See
Lewandowsky, Ecker, Seifert, Schwarz, & Cook, 2012, in Psychological Science in the
Public Interest).
There is a tendency to accept familiar information as true. In experiments on the false
fame effect (mentioned in Chapter 6), participants are exposed, under a misleading cover
 Recovered Memory 291

story, to a fabricated name (e.g., PETER TERRY). Later, in a different context, the subjects
see a list and try to identify which names were those of famous people. Names encoun-
tered earlier are mis-identified as celebrities. “Peter Terry? Sounds familiar, so it must be
someone famous” (Jacoby, Woloshyn, & Kelley, 1989).
In summary, the research suggests that misinformation is too-often believed; that
debunking strategies have minimal benefits; and misinformation effects are greater than
debunking effects (Chan, Jones, Jamieson, & Albarracin, 2017).

RECOVERED MEMORY
Can memory for abuse be repressed? Can memory of trauma remain hidden and unre-
called? These questions do not have clear and accepted answers. On one hand, research
shows that emotional events are well remembered, that false recollection is readily pro-
duced in laboratory tasks, and that misleading information and suggestion lead to misre-
call. On the other hand, clinical experience shows that memory repression and the sudden
recovery of repressed memory are accepted as genuine psychological phenomena. To
deny recovered memories on the grounds of “insufficient” scientific evidence is
unacceptable.
The debate over false memories versus repressed memories (depending on which
side of the controversy you take) is a public one. (In an effort to be neutral some use the
term recovered memory). The conflict pits psychological experts against one another,
frequently in court settings. Recovered memories of abuse can lead to accusations that a
parent or relative is an abuser, and to countercharges that suggestions made by thera-
pists are the source of the memories.
Arrigo and Pezdek (1997) tried to defuse the controversy surrounding recovered
memories of abuse by reminding us that amnesia occurs after other sorts of traumas.
These include forgetting of accidents, natural disasters, combat, suicide attempts, or vio-
lent crimes.
L. M. Williams (1994) confronted two problems frequently noted in repression
research: Did abuse occur in the past, and did the victim actually forget that the trauma
occurred? Williams first identified documented cases of childhood abuse that had
occurred, on average, 17 years previously. She then interviewed the (now) adult women
for their recollections of the event. Williams found 49 women (or about 38 percent of her
sample) who did not remember the abuse. This was not due to a reluctance to discuss
something painful or embarrassing, because the women were willing to relate intimate
details about other events. This study offers evidence that, indeed, forgetting for a trau-
matic experience can occur.
By what mechanisms could memories be unremembered yet still be in storage for
later recall? Brewin and Andrews (1998) considered several. One explanation is in terms
of repression: the mind suppresses recall of a painful memory. This could be an uncon-
scious defense against the psychological distress produced by the trauma, or the repres-
sion could be conscious and deliberate. One can deliberately avoid thinking about
something. Alternatively, the traumatic memory might involve the implicit memory sys-
tems of the brain rather than the explicit systems. By definition, an implicit memory is
less available for conscious recall. Finally, memory may not be accessible because
292 Storage and Retrieval

adequate retrieval cues are not present. A memory formed during the intense emotional
state of a traumatic experience could remain unrecalled as long as this strong and unique
emotional state does not recur.

RETRIEVAL VERSUS RECONSTRUCTION

One misconception about memory is that it consists of static memory traces. This might
be called a copy theory of memory. Our casual description of flashbulb memories fits the
copy theory. We believe these memories persist unchanged over the years, almost like a
photographic image. (In Chapter 9, we considered the evidence for accuracy of flashbulb
memories). An alternative conception of remembering emphasizes less the retention of
detailed memory images and more the reconstructive process of retrieval. Reconstructive
memory describes retrieval as the reconstruction of what happened on the basis of
stored fragments of information, cues, and knowledge of similar events. Neisser offered
an analogy to the work of a paleontologist who reconstructs the appearance of a dinosaur
from fossil fragments (Neisser, 1967).
In 1932, Sir Frederick Bartlett reported some classic experiments demonstrating
reconstruction. His subjects listened to a Native American folk tale entitled “The War of
the Ghosts” and then attempted to reproduce the story several times later. The story had
many unfamiliar elements, or at least unfamiliar to Bartlett’s subjects who were British
and mostly college students. The story was not organized the way familiar fairy tales are,
containing a sequence of characters, a danger, and a (usually) happy resolution. The sub-
jects’ retellings of the story were marked by omissions and distortions, attempts to make
the flow more consequential, and the use of English paraphrasing. The then-unfamiliar
word canoe was replaced by boat, and hunting seals became fishing. Also, the man who
survived the attack was apparently “feeling none the worse for his experience.” Bartlett
said the listeners were trying to make sense of the story. Bartlett introduced the word
schema to memory theory, suggesting that the students formed a schema in memory, or
a general outline of the story, which was influenced by their knowledge, beliefs, and
expectancies.
This study was key in demonstrating that remembering is not simply the passive
recoding of events. Rather, the rememberer is active in the construction of a memory and
in the reconstruction at retrieval. (Today, Sir Frederick Bartlett, knighted for his contribu-
tions to psychology, is ranked with Ebbinghaus as a pioneer in the field of memory).
Familiar and repetitive events, such as going to the movies or taking a trip, are repre-
sented in memory by schemas (Alba & Hasher, 1983). The use of a schema during retrieval
of a specific event can lead to recall of what typically occurs (and is thus stored in the
schema) rather than what actually occurred in a specific instance (e.g., Hudson, 1990;
Linton, 1982). For example, you believe you bought popcorn on your last visit to the mov-
ies, when in fact you arrived late and so did not buy any snacks. Memory reconstructed
according to the schema is not accurate in this instance.
Instead of an either–or position of memory reproduction versus memory reconstruc-
tion, we could treat these as end points on a continuum. Consider remembering a con-
versation you heard. Occasionally, the exact words are recalled: a particular phrase or the
wording of a joke. More often, we retain the general meaning of the conversation and
 Applications 293

paraphrase when recounting it. When the conversation is reconstructed later, inferences
or interpretations are added to the memory.

APPLICATIONS
Strategies for Searching Memory
What strategies are used in everyday life to aid retrieval? Two frequently reported tech-
niques are retracing one’s steps (mentally or physically) to retrieve mislaid objects or for-
gotten intentions; and alphabetical searching when names or words cannot be recalled.
The search strategies that people use are studied with a thinking-out-loud protocol.
Participants are given general prompts and are asked to think out loud as they attempt to
recollect a specific instance (Reiser, Black, & Kalamarides, 1986). To recall, as examples,
a chance meeting with someone or the date of the first exam this semester, you might
ask questions like: Where was I? What was I doing when this happened? Who was with
me? What preceded or followed the event? Searches for a distant memory might use
temporal landmarks to cue relevant information: where was I living, or going to school, or
working then?
Some cues work better than others. If you tried to recall a course schedule from last
year, what sorts of cues would be most effective? Wittman and Healy (1995) tested their
college students’ recall of each course, who the instructor was, where the course met,
and at what day and time. One of these details was used to prompt the remaining three
facts. For example, “You had a 9:30 class on Tuesday. What course was it, who taught it,
and where did it meet?” The best prompt for retrieving memory was the where cue:
Given the building, students could remember over 80 percent of the remaining details
about that course. The other cues were effective in recounting about 60 percent of the
course information.
Another retrieval strategy is to change your perspective about the to-be-recalled
material. After searching memory fruitlessly, we sometimes realize the mental image of
the lost item was wrong. Changing perspective is illustrated in a study in which students
read a paragraph-long description of a house. For some readers, the story was titled
“Buying a House,” and for others, the same story was titled “Burglary.” Either version led
to about the same number of recalled details. However, if the participants were asked to
think about the story from the perspective of the alternate title, additional details were
recalled on a second test (in comparison to a second test without a changed title)
(Anderson & Pichert, 1978). The alternate perspective suggested different retrieval cues
that evoked otherwise unrecalled information.
Retrieval can be thought of as an example of active problem solving, in which the
person selects plausible cues and evaluates the information each retrieves. Individuals
with frontal-lobe-impairment have trouble searching memory. Memories are haphazardly
recalled, being accurately remembered on one occasion but not on another. There is diffi-
culty generating specific personal memories from cues, such as “Recall an experience
you once had involving a dog” (Baddeley & Wilson, 1986).
The notions of state-dependency and search strategies are also relevant to under-
standing the effect of alcohol on memory. This is discussed in Box 10.2.
294 Storage and Retrieval

BOX 10.2 ALCOHOL AND MEMORY

Alcohol and alcohol intoxication affect memory in several ways. These effects can be organ-
ized in terms of the stage model of memory: encoding, storage, and retrieval.
Alcohol can inhibit memory encoding. Participants, usually volunteer college students,
learn less when under the influence of alcohol than when they study sober. For example,
recall of just-presented word lists was impaired in intoxicated participants (Weingartner,
Adefris, Eich, & Murphy, 1976). By using an immediate test, the participants are still in the
same drug state during input and output, so forgetting was not a state-dependency effect.
An encoding deficit raises an issue with respect to alcoholic amnesias for criminal actions.
Criminal suspects claim they cannot remember their crime because they were intoxicated.
State dependency suggests that reentering the alcohol state could reinstate memory of that
intoxicated state. One attempt to do so did not lead to memory recall (Wolf, 1980). Thus, the
memories may not have been encoded in the first place. Or, the suspects are malingering.
Alcohol seems to be less detrimental to retrieval than to encoding. Introducing intoxica-
tion after studying did not significantly reduce recall of previously studied word lists (e.g.,
Birnbaum, Parker, Hartley, & Noble, 1978). Similarly, Miller et al. (1978) found alcohol affected
encoding, not retrieval, and showed no state dependency.
Alcohol can impair the maintenance of stored memories, at least in extreme cases.
Korsakoff’s syndrome is a neurological disorder associated with prolonged alcohol abuse (see
Chapter 7). A characteristic of Korsakoff’s patients is that memories that were clearly present
in storage earlier are later lost. In one case study, a distinguished scientist had completed his
autobiography shortly before the onset of Korsakoff’s. After onset, he had profound loss of
the details of his professional life (papers written, conferences attended, developments in his
field), going back decades (Butters & Cermak, 1986).
Heavy drinkers are impaired in memory tests, even when tested sober. There is a dose-
impairment continuum ranging from social drinkers, who are not impaired or are mildly-
impaired, to alcohol-dependent, who are more impaired. The cessation of drinking may lead
to recovery of normal cognitive abilities within a few weeks (Goldman, 1983). The cognitive
loss in Korsakoff’s, however, is not reversible.
Comparing non-drinkers to heavy-drinkers confounds long-term effects of alcohol on the
brain with the immediate effects of recent consumption. Jackson et al. (2021) studied individu-
als who reported having had numerous alcoholic memory blackouts (or MBOs). An MBO is a
drinking event in which the person is conscious at the time but will later be amnesic for episodic
memory of the event. Control subjects were also used, and everyone was tested under sober
and alcohol conditions (.06 percent blood alcohol). When given alcohol, both MBO and control
subjects had deficits in several standard tests (free recall, serial recall, and depth of processing),
compared to their sober scores. The MBO participants did not differ from the controls.
The MBO participants were tested one more time, shortly following a blackout event. This
session was scheduled in the afternoon, after subjects had slept 6.5 hours and had 0.0 blood
alcohol at test. These subjects had already shown a decline in recall from the sober to the
alcohol testing conditions. Now, they showed poorer recall the day after an alcohol bout, even
though they were not under the influence.
The adverse effects of alcohol on memory seem to occur at several stages, making it
difficult to isolate alcohol’s effects to any one stage. One review of the literature suggests
that alcohol has such a pervasive effect on information processing in general that a deficit at
any stage of memory is bound to be found (Maylor & Rabbitt, 1993).
 Summary 295

Context-Specific Learning
The possibility that learning is context-specific has important implications for training pro-
grams. If people are trained in one context, will they remember when tested elsewhere?
As a specific instance, Godden and Baddeley noted that students learning scuba diving
are taught emergency procedures in the classroom that might well be unrecallable under
water during an emergency. Godden and Baddeley assessed this possibility by having
dive students study word lists on land or under water. Memory testing then occurred in
the same or the opposite location. Their results, summarized earlier in Table 10.1, showed
context specific retrieval. More words were recalled when study and testing both occurred
on the dock or under water.
How significant are context effects for classroom learning? Researchers at one school
analyzed final exam grades from over 5,000 college students enrolled in five different
kinds of courses. The researchers took advantage of the fact that many finals were given
in other rooms to relieve overcrowding and facilitate monitoring. The students knew this
school policy but did not know which classes would be switched. As it turned out, there
were no significant effects of context change on final exam grades (Saufley, Otaka, &
Bavaresco, 1985). Average grades of those students who took their finals in the course
classroom did not differ from those who took their finals elsewhere.

Reconsolidation
Methods of fear desensitization have been considered in several places in this book. For
example, exposure therapy is used to reduce phobic fear. The research on reconsolidation
suggests there may be ways to enhance such therapies. Recall that in the reconsolidation
procedure, the subject is reminded about a previous learning experience. At this time, as
the reactivated memory is about to be reconsolidated, a new outcome could be intro-
duced that will become the new permanent memory.
Propranolol, a drug that suppresses the fear response, has been used to facilitate
extinction. A fear response was first conditioned in human participants. On the sec-
ond day, the subjects in the experimental group were reminded of the first day’s fear
learning and given propranolol, a fear blocker. When tested on a third day, the meas-
ures showed no fear reactions from the propranolol subjects. Propranolol before
memory activation had weakened the fear response, and so what was reconsolidated
overnight between Days 2 and 3 was a lesser level of fear (Kindt, Soeter, & Vervliet,
2009).

SUMMARY
Long-Term Memory
According to the permanent-memory hypothesis, everything we learn is permanently
stored in the brain. Often-cited support for this hypothesis is the recovery of supposedly
long-lost memories by electrical stimulation of the temporal lobes, or recovery through
hypnosis or free association. However, these mental experiences are not necessarily
memories, nor are they necessarily valid.
296 Storage and Retrieval

Although Ebbinghaus’s curve of forgetting seems to suggest rapid loss, a number of

studies of naturalistic memory show substantial retention after delays of years and even
decades. These include memory for names and faces of high school classmates, content
from high school and college courses, and public events.

The Nature of Storage

How are memories organized in long-term memory? Semantic network theories
assume that items of knowledge are interconnected in memory. Collins and Quillian
used reaction-time measures to verify the hierarchical arrangement of knowledge
organization. Comparisons of more distantly separated facts took longer to verify than
comparisons involving more proximal facts.
Case studies from neuropsychology illustrate the organization of memory, when
whole categories of words (e.g., nouns or “living things”) are forgotten following brain
injury. The possible biological mechanisms of memory include protein synthesis and syn-
aptic change.
Consolidation is the formation in the brain of stable and permanent long-term mem-
ories. Consolidation theory proposes that a memory is first encoded in a temporary form,
such as a short-term memory, and over time consolidates into a permanent form.
Consolidation theory was supported by the finding that electroconvulsive shock seemed
to interfere with the development of long-term memory. Newer studies focus on the
biological mechanisms that underlie consolidation, such as protein synthesis.
The activation or retrieval of a consolidated memory can make it susceptible to mod-
ification. The reconsolidation hypothesis asserts that when a memory is retrieved, it
returns to a changeable state. The retrieved memory is susceptible to alteration at this
time and could subsequently re-reconsolidated into permanent memory.
Forgetting is increasingly being construed as a failure of retrieval of memories that are
potentially available, rather than attributing forgetting to either decay or interference.

Retrieval
Three general factors affect retrieval: the distinctiveness of the memory, practice at
retrieving that memory, and the presence of effective retrieval cues. Distinctive memo-
ries are retrievable because they stand out against a background of otherwise similar
memories.
Retrieval practice, also known as testing effects, produces better retention after
delays than does additional study. At any given moment, more may be available in mem-
ory than can be retrieved then. This is shown in repeated attempts to recall. Hypermnesia
is an increase in recall across successive tests in the absence of additional study. The
hypermnesia curve is the opposite of the forgetting curve.
What makes a good retrieval cue? One theory identifies cues that are strongly asso-
ciated to the target. Thus, TABLE would be a good cue for CHAIR. Alternatively, the
encoding specificity principle states the best retrieval cues are those that were also pres-
ent and encoded with the target. Recall is facilitated when encoding and retrieval condi-
tions are matched, as shown by studies of context-specific (or place-specific) retention;
 Summary 297

drug-state-dependent learning; and mood-dependent memory. However, context-specific

memory is not always found. Contextual cues are sometimes overshadowed by more
explicit retrieval cues.
Retrieval can be impaired by high levels of emotional arousal, as is the case with test
anxiety and momentary forgetting.

Metamemory and Partial Retrieval

Metamemory judgments include the feeling of knowing, the belief that we know some-
thing that cannot be recalled now. Tip-of-the-tongue is a related experience of a feeling of
knowing a word or name even though it cannot be recalled. The accuracy of our unre-
called knowledge can be verified by recognition tests. However, our intuitions are some-
times based on educated guesses about what we think we should know, and sometimes
our intuitions are incorrect.
Prospective memory, or remembering to do something in the future, requires the
establishment of retrieval cues that will prompt the desired behavior at the appropriate
time. Prospective memory can be externally cued by reminders such as alarms or a list of
instructions, or internally cued by intentions to remember.

False Retrieval
Mistakenly remembering something that did not actually occur has been studied with the
false memory method of Deese, Roediger and McDermott (DRM). False recall of a target
word, such as SLEEP, is elicited by presenting a list of associates of the target (e.g., bed,
rest, dream).
Source memory refers to remembering whether an event was actually experienced, or
was imagined, thought of, heard about, or even dreamed of. External- and internal-originating
memories are discriminated partly on the basis of remembered attributes and partly on the
basis of judgment. One could forget the source of information, but correctly remember the
information itself. False memories of childhood experience could be produced through failure
of source memory. Internally generated memories become more real-like through repeated
imaginings. Imagination inflation leads to an increase in the belief that the action or event
actually happened.

Effect of Postevent Information

Witnesses can be led to misrecall an event after exposure to misleading postevent infor-
mation. Although such effects readily occur, their interpretation is still uncertain. Does
postevent information replace the original memory? Or does postevent misinformation
block access to the original memory?

Recovered Memory
There is controversy over whether traumatic memories can be repressed and later recov-
ered, or whether these are actually false memories. There are cases where forgotten
trauma or abuse has occurred, been forgotten for years, and is subsequently remembered.
One explanation is that painful memories can be repressed into an unconscious part of the
298 Storage and Retrieval

mind. Another is that people can learn to not-think about the painful experiences. A third
theory is that the memories might be implicit, and so not available to consciousness.

Retrieval Versus Reconstruction

Frederick Bartlett’s conception of remembering describes retrieval as the reconstruction
of what happened on the basis of stored fragments of information, cues, and knowledge
of similar events. The rememberer is active in the construction of a memory and in the
reconstruction at retrieval. His subjects’ retellings of “The War of the Ghosts” were trans-
lations by English students into a more familiar framework.

Applications
Retrieval can be a directed process, in which we employ strategies to direct the scope of
memory search. One search strategy is to change your perspective or way of thinking
about the to-be- recalled material. Try to remember from another point of view.
Neuropsychological studies suggest the frontal lobes are critical for initiating, direct-
ing, and evaluating retrieval. Frontal lobe injuries can impair search and retrieval of mem-
ories or knowledge that is still there.
Context-specific learning has important implications for training programs. People
trained in one context may have difficulty remembering when tested elsewhere. An
experiment using scuba diving students showed context specificity of retrieval.
Memory reconsolidation has implications for therapy for fears and phobias. Individuals
could be given a reminder of their fear, activating its long-term memory. Extinction train-
ing or some other fear inhibiting treatment could be given shortly after reactivation. The
idea is that the non-fearful memory will be reconsolidated back into long-term memory,
replacing the previous fearful memory.
 CHAPTER

11
Spatial, Motor-Skills, and Implicit Learning

CONTENTS

Spatial Learning 300 Implicit Learning 316

Rats, Mazes, and Psychology 300 Some Implicit-Learning Tasks 317
Routes Versus Cognitive Maps 300 Dissociating Categories of Implicit
Maze Learning and the Brain 305 Learning 320
Schemas in Spatial Memory 307 Expertise 321
The Development of Spatial Memory From Declarative to Procedural 322
in Children 309 Applications 323
Motor-Skills Learning 309 Implicit Learning 323
Practice 311 Spatial Memory 324
Knowledge of Results 314 Developing Memory Skill 326
Long-Term Retention of Skills 316 Summary 327

Current psychological theories distinguish among different forms of long-term memory.

One distinction is between declarative and procedural knowledge. Declarative knowl-
edge refers to memory for verbalizable knowledge. It includes semantic memory and
episodic memory. Procedural knowledge underlies skilled behavior and the ability to
quickly perform various cognitive, perceptual, and motor operations. Whereas declarative
knowledge is shown by recall of information, procedural knowledge is demonstrated by
facilitated performance of behavior.
This chapter includes several forms of learning that correspond, in a general fashion,
to the procedural description: spatial learning—knowing how to get from place to place in
an environment; motor-skill learning—knowing how to perform coordinated bodily move-
ments quickly and accurately; implicit learning—knowing the underlying rules that govern
complex sequences of behaviors; and expertise—expert performance in a specific
domain. All of these refer to fluent, quick, and skilled performance, whether running a
maze or reading an MRI scan. To anticipate what follows in this chapter, consider the fol-
lowing questions: Is spatial information remembered in the form of a route or a map?
What is the role of feedback in learning skilled movements? Is skilled behavior accessible
to conscious description? Is expertise due to talent or to practice?

DOI: 10.4324/9781003227090-11
300 Spatial, Motor-Skills, and Implicit Learning

Procedural learning is often referred to as a form of implicit learning. Explicit and

implicit refer to the awareness and conscious accessibility of the knowledge learned. In
some cases of procedural learning, we can perform but we cannot articulate how we do
it. Some theorists have suggested that procedural learning is indeed unconscious,
because it evolved earlier than conscious forms of knowledge representation (Reber,
1993). One important question we will consider is whether procedural learning is implicit
or whether there is instead some explicit knowledge that can indeed be reported.

SPATIAL LEARNING
Spatial ability in our everyday lives is often taken for granted until we become lost. On
leaving a movie theater, you forget whether to turn left or right. You exit the mall at a dif-
ferent place from which you entered. You forget where you parked the car earlier. (On a
really bad day, you forget all three). What sorts of experiences are necessary for acquiring
spatial knowledge, and what exactly do we learn?

Rats, Mazes, and Psychology

The rat and the maze are two stereotypes that are inextricably linked to psychology’s
image. The use of rats as subjects began early in American psychology laboratories.
Willard S. Small at Clark University first studied rats in mazes in 1900, capitalizing on the
animals’ burrowing and tunneling behavior. John Watson, at the University of Chicago,
began research on the sensory cues used by rats in maze learning. At both schools biol-
ogists were studying rat anatomy and physiology, so it was natural for biologically oriented
psychologists to start studying rat behavior.

Routes Versus Cognitive Maps

An organism moving within a complex environment needs to develop a representation of
that environment. There are two broad conceptions of how spatial knowledge might be
represented. Route knowledge is knowledge of a series of routes, directions, or paths
through a spatial environment. For instance, we could describe maze learning as the
memorization of a sequence of left and right turns in the maze. When you ask directions
to an unfamiliar location, the response is likely to be in the form of a route: a sequence of
distances and turns, with maybe an occasional landmark (“go straight 2 miles, turn left at
the light, then the next right”). In its extreme, a route is characterized by knowledge of
sequential locations but not of general interrelationships. By contrast, a cognitive map is
a more abstract representation of an environment, placing specific routes in context with
the surrounding area. The phrase cognitive map implies that a schematic image is repre-
sented internally (Cohen, 1989). Route maps might have a first person perspective, rep-
resenting the view you saw as you traveled. Cognitive maps might have a third person
perspective, as if you were viewing the area from above. Table 11.1 lists some other
contrasting descriptors of route maps and cognitive maps.
Route and cognitive maps should be considered as endpoints along a continuum
rather than as separate categories of representation. In practice, a route has some ele-
ments of a map. Repeated experience navigating a specific environment can lead to the
 Spatial Learning 301

Table 11.1 Terms Used to Characterize Cognitive Map and Route Knowledge of Spatial Information

Cognitive Map Route Knowledge

Paths in context List of directions
Global Local
Semantic Episodic
Bird’s eye Ground-based
Schematic Concrete
Third person perspective First person perspective
Flexible Habitual

development of more map-like representations. Landmarks would be part of a route

description, indicating locations along the route, or where to make a turn. Yet landmarks
would also be part of a cognitive map, serving as orienting stimuli from which to compute
direction or location.
Given these two possibilities (route versus cognitive map), which form of representa-
tion does spatial knowledge take?

Place Versus Response Studies

According to one tradition in psychology, rats learn to make specific turns in mazes. Clark
Hull (1949) theorized that the stimuli at each choice point in a maze became associated
with a certain response, a left or right turn, for example. Edward Tolman (1948) hypothe-
sized instead that the rats acquired a cognitive map of the maze and the surrounding
environment. The rat learned the place where food was located in the maze and within
the room. The two views can be illustrated by reference to learning in a simple T maze
(see Figure 11.1, left). Rats start at point A and are trained to turn right at the choice point
to get to the goal box (Y). What have the rats learned? Hull said they learned the response
of turning right. Tolman said they learned a cognitive map of the maze, and the place
where food was in the maze.
Tolman proposed an experimental test to distinguish response learning from cogni-
tive maps (Tolman, Ritchie, & Kalish, 1946, 1947). He rotated the maze 180 degrees, so
that now the rat would be starting from a new location in the room (point B in Figure 11.1,
right). Which direction would the rats turn according to the two theories? Hull’s theory
predicts that the rats will make a right turn as they have been trained to do, and now go
away from the food to point X in the diagram. Tolman’s theory predicts the rats will check
their cognitive map and make a left turn to compensate for the change in starting location,
and arrive at point Y. The design of the place versus response experiment nicely contrasts
route versus cognitive map descriptions and the differing predictions of the two
theories.
What did the rats actually do? Most students guess that the rats turned left (after all,
rats are pretty smart). It would be nice to say the data turned out clearly in favor of one
theory or another. In fact, in some experiments the rats turned right as Hull predicted; in
other studies, they went left as Tolman predicted. Stating the results in this manner does
not seem to be helpful (and leaves you wondering what is the correct answer on the exam).
302 Spatial, Motor-Skills, and Implicit Learning

Figure 11.1
Maze Arrangements for Training and Testing Place versus Response Learning.
Source: From “Mazes, Maps, and Memory,” by D. S. Olton, 1978, American Psychologist, 34, p. 590. Copyright © 1978 by the
American Psychological Association.

This simple and elegant experimental manipulation did not, in fact, resolve the response
versus place learning controversy. Instead, the results told us something about the stimuli
used in spatial learning. In the studies done by Hull and his followers, the maze had walls
and the maze itself was surrounded by curtains. There were no visible cues for the rat to
orient itself within the room. This lack of external cues (or “landmarks”) encouraged
response learning. In the experiments by Tolman and his students, the mazes had no walls,
and were elevated several feet off the floor. Features in the room were clearly visible from
the maze. This procedure encouraged cognitive map learning. The place versus response
controversy taught us that either specific responses or cognitive maps can be learned:
rats, and people, are flexible in their use of whatever cues are available (Restle, 1957).

The Radial Maze

A contemporary example of cognitive mapping is performance in the radial maze. The
prototype of this maze has eight arms radiating out from a central platform, as shown in
Figure 11.2. The maze is usually elevated, often without walls. Food is placed in recessed
cups at the end of each arm. The rat is allowed to enter the arms in any sequence. The
optimal strategy for a foraging animal is to retrieve all the pieces of food without repeating
an arm entry. Rats quickly learn this, achieving a mean level of accuracy of choosing 7.6
different maze arms among the first 8 choices (Olton & Samuelson, 1976).
During each trial, the rat must keep track of which arms it has entered and which ones
remain. The rat could do this with a cognitive map of the maze, and checking off each arm
as it is entered. But there are simpler alternative explanations. A rat could use odor stim-
uli, either from its own previous perambulations or from the smell of food, to determine
which arms have been entered and which not. Scents can be ruled out by several manip-
ulations, ranging from surgically making the rats anosmic to dousing the maze in Old
 Spatial Learning 303

Figure 11.2
A Radial Maze. Note the presence of extramaze cues.
Source: Adapted from “Spatial Memory,” by D. S. Olton, 1977, Scientific American, 236.

Spice aftershave to mask any scents. Correct choices are relatively unaffected by such
manipulations (Olton & Samuelson, 1976; Zoladek & Roberts, 1978).
Rats instead use cues outside of the maze, or extramaze cues, to keep track of entered
and unentered arms. This can be shown by rotating the maze within the room between
maze-arm choices. The subject is allowed to make four arm entries and then is removed
from the maze temporarily while the maze is rotated slightly. The animal is then returned
to the maze and allowed to complete its selection of arms. The rat avoids arms in the
directions previously visited, and mistakenly reenters previously chosen arms that are in
the direction not yet visited. The animal is responding to locations within the room and not
to stimuli within the maze. If these extramaze cues are blocked, for example by surround-
ing the maze with curtains, the accuracy of choosing different maze arms declines.
If external cues are not available, rats can use internal, kinesthetic cues to help navi-
gate the maze. Even in total darkness, rats can learn to choose at better than chance
levels (Brown & Moore, 1997). There are multiple types of cues available to navigate
spatially. This is especially the case in long-distance navigation (see Box 11.1).
304 Spatial, Motor-Skills, and Implicit Learning

BOX 11.1 LONG-DISTANCE NAVIGATION BY ANIMALS

A small dog named Sam was left behind in Colorado when his family moved to California. Ten
weeks later the dog showed up at their new home, apparently having traveled 840 miles
across the plains, deserts, and mountains of four states (“Dog Roams,” 1983). The dog, tired
and dirty from his trek, was also described as being nervous and suspicious. “He follows us
everywhere,” his owners say. (Hey, no kidding!) How does a dog find its way from Colorado
to California?
Many feats of animal migration require extraordinary spatial abilities. Studies of animal
navigation indicate that there are several components to wayfinding: some means of deter-
mining direction, sensitivity to different kinds of stimuli, and possibly a cognitive map.
Take the example of homing pigeons. They return home to nests after being released
hundreds of miles away. The first thing the pigeon needs to do is determine direction. You
cannot fly north, for instance, without knowing which way north is. Pigeons and other birds
use the combination of sun position and time of day (i.e., the sun rises in the east and is in
the southern sky at noon).
Pigeons may also possess a magnetic sense, a sort-of built-in compass. This magnetic
sense has been tested in homing situations by using miniature magnets to disrupt orienting.
Cases of lost homing pigeons have been attributed to anomalies in local magnetic fields
associated with mountains, iron fields, or sunspots that disrupt the birds’ sense of direction
(Walcott, 1989).
Granted direction finding, the pigeons need to know where home is. Maybe pigeons
simply retrace the outbound route when released. However, pigeons home even if prevented
from seeing the route taken to the release point. In one study, the pigeons were anesthetized
for the trip to the release point. They still found their way home. Pigeons fitted with opaque
eye cups, which allowed light in but not clear vision, homed to the general vicinity of their
roost. Once there, the birds circled aimlessly and fluttered down anywhere. Apparently,
visual landmarks are important for proximal homing, but not for long-distance travel (see
review in Pearce, 1997).
Wayfinding in other animals has been attributed to an ability to perceive and follow subtle
sensory stimuli. For example, a “nuisance” wolf living near an airport in Alaska was captured
and relocated several hundred miles away. When the wolf returned to the airport, it was
hypothesized that it had followed low-frequency sounds from planes, which carry over great
distances in Arctic regions. The proof was a second relocated wolf that found its way to an
airport, but a different one from which it had been relocated (Rogers, 1989).
The place-learning versus response-learning controversy taught us an important lesson
about spatial learning. Animals (and presumably people) can learn to use different kinds of
stimuli: landmarks, magnetic sense, sun and star positions, olfactory stimuli, and who knows
what else.
Okay, so how did Sam find his family? I really don’t think the dog could smell them 800
miles away. The family speculated that since they had once before made the trip with the
dog, Sam remembered the route and retraced it. This is a little far-fetched. I think the Colorado
people who took Sam in got tired of him, drove overnight to California, and left the dog on his
old family’s new porch. That’s my theory.
 Spatial Learning 305

Morris Water Maze

The Morris water maze is another test of spatial navigation (Morris, 1981). A rat or mouse
is placed in a small swimming pool in which the water is clouded by the addition of pow-
dered milk. (Rats are very good swimmers). There is a hidden platform just under the
surface of the water, and the goal is to learn the location of this platform. From trial to trial
the starting location from which the animal is placed in the pool is varied, so the animal
has to learn the platform location on the basis of cues in the room. At first, the animals
spend considerable time searching for the platform. Over trials, the animals become
faster and follow more direct paths in going to the platform.
Virtual mazes have been developed in which human subjects “move” through a com-
puter presented maze. A Morris maze simulation has distinctive cues, such as patterned
and colored walls outside the maze to provide orientation information. The mazes have a
progressively changing image, as if one were “swimming” (well, maybe wading) through
the pool. The time to find the hidden platform and the directness of the route from differ-
ent starting points can be recorded just as with mice and rats. In one case, college stu-
dents were tested in a series of progressively more difficult virtual mazes. Compared to
mice trained in actual mazes, the humans learned faster (Shore, Stanford, MacInnes,
Klein, & Brown, 2001).

Maze Learning and the Brain

The radial maze and the water maze are particularly useful for testing theories about the
anatomical basis of spatial memory. The radial-maze task differs from the mazes used
earlier by psychologists, in which both food locations and the route through the mazes
remained fixed over trials. The flexible use of working memory in the radial maze has
been contrasted with habit learning by training animals with different rules to follow in the
maze. McDonald and White (1993) produced a double dissociation among tasks and brain
areas. Rats who received hippocampal lesions were impaired in learning the standard
radial maze, which requires choosing all different arms in a trial. Rats with hippocampal
lesions reenter already visited arms, which are now empty of food. The hippocampus is
important in this type of maze. The hippocampus is also essential to episodic memory in
humans.
Other rats were trained to repeat choices of the reinforced arms, the opposite of the
usual radial maze alternation rule. That is, the rat had to return to the alley that was recently
rewarded to receive an additional reward there. Lesions of the striatum (specifically dorsal
striatum) impaired learning the repeat-choice rule. Hippocampal lesions did not inhibit
learning in the repeat-choices task. The striatum is otherwise known to be important in
reinforcement learning and habit learning. The pattern between lesions and impairments
is summarized in Table 11.2.
A larger body of research indicates there are multiple systems in the brain that direct
spatial behavior. Spatial behavior relies on both cognitive mapping and habitual respond-
ing. A given navigation situation may involve more of one or the other, depending on the
task, past training, or current motivation.
As you may recall from Chapter 7, damage to the hippocampus produces amnesia
(Squire, 1992). In humans the left hemisphere exerts more control over verbal memory,
306 Spatial, Motor-Skills, and Implicit Learning

Table 11.2 Dissociation of Two Learning Tasks Conducted in a Radial Maze after Lesions of
Different Brain Regions

Tasks –> Choose Different Arms Repeat Same Arms

Impairing lesion Hippocampus Striatum

Unaffecting lesions Striatum Hippocampus
Source: McDonald & White, 1993.

whereas the right hemisphere exerts more influence over spatial memory. People with
right hippocampal injury are more impaired on spatial learning and memory. This includes
tests of maze learning, remembering the location of objects, and learning to tap a series
of blocks in sequence (Kolb & Whishaw, 1985).
The role of the hippocampus is clearly shown by H. M., the amnesic individual pro-
filed in Chapter 7. To refresh your memory, Henry Molaison had most of the hippocampus
on both sides of his brain removed in an experimental treatment for his epilepsy. As a
result he was severely impaired on spatial tasks, as shown in the following passage
reported by the researchers working with him:
His limitations…are illustrated by the manner in which he attempted to guide us to
his house…when we were driving him back from Boston. After leaving the main
highway, we asked him for help in locating his house. He promptly and courteously
indicated to us several turns, until we arrived at a street which he said was quite
familiar to him. At the same time, he admitted that we were not at the right
address. A phone call to his mother revealed that we were on the street where he
used to live before his operation.
(Milner, Corkin, & Teuber, 1968, as cited in Kolb & Whishaw, 1985, p. 482)

This was eight years after H. M. had moved away from that address. (You have got to
wonder what these two psychologists were thinking. H. M. was lost, he is amnesic, and
you’re asking him to remember how to get to his house).
The function of the hippocampus in spatial navigation was vividly demonstrated in a
series of studies of London taxi drivers. London has stringent requirements to become
licensed to work there. Applicants study for several years and pass multiple exams
assessing knowledge of routes and landmarks. Using magnetic resonance imaging
(MRI) to obtain graphic images of the brain, the drivers were found to have more devel-
opment and greater volume in their hippocampi (the plural of hippocampus) than did
matched control subjects (e.g., Maguire et al., 2000). The researchers suggested that
the extensive use of spatial navigation skills led to their increased development in the
hippocampus.
This is a correlational finding. The researchers addressed the alternative possibility,
that those who become taxi drivers had better spatial skills and larger hippocampi to
begin with. (If you cannot find your way around, you likely won’t last long as a cabbie). In
a follow-up comparison, applicants to the taxi program were followed for three years,
from start-to-finish of training and licensure. At the end, the applicants were divided into
two groups for data analysis: those who successfully completed the program, and those
 Spatial Learning 307

who did not. MRIs showed once again that the newly licensed cabbies had more volume
in the gray matter of the hippocampus. And importantly, their gray matter had increased
from the start of training; at the end they had more gray matter than did the unsuccessful
applicants, and more than non-driver controls subjects (Woollett & Maguire, 2011).

Automatic Spatial Responding

Implicit behavior is characterized as being unconscious and automatic. The response runs
off effortlessly, seemingly without conscious deliberation or direction. Habits have these
same features (see Chapter 4). Spatial actions often also seem to be automatic and
habit-like.
Habits develop after extensive repetition of an unvarying response. Similarly, rats will
develop habit-like responses after extensive practice. In one case, rats received 40 days
of training in a simple maze, receiving food pellet rewards in the goal box. The experi-
menters then spread food pellets in the middle of the maze to see how the animals would
react. Nearly all the rats ran through the food, continuing on to the goal box to receive a
single pellet. Nearly half the rats persisted like this across 22 test trials. This is habit.
Place versus response learning also varies with amount of training. Most rats given a
limited number of trials adopt a place learning strategy: When tested from the opposite
start location of the rotated maze, the rats compensated by turning in the direction of
where food was located (review Figure 11.1). However, with extensive practice at making
the specific turn (e.g., right turn), animals will continue to make that turn (turning right) in
the rotated maze, away from where food had been.
In everyday life, people too often make spatial errors. In a state of absentmindedness,
you leave your house intending to go one direction but instead turn away to follow your
usual route. At a familiar intersection you turn, momentarily forgetting you needed to go
the opposite direction that day.

Landmarks
Landmarks are distinctive elements in an environment that stand out by virtue of their
features (e.g., size or shape) or their meaning (e.g., historical or social). Because such
elements are literally outstanding, landmarks are readily perceived, remembered, and
used as reference points. When we give someone directions, or when asked where we
live, we often start with some nearby landmark and describe a route from there. The
extramaze cues used by rats in a maze, such as a door, window, or light fixture, can be
considered landmarks. According to some theories of spatial knowledge, we first learn
landmarks, then routes around them, and finally we develop cognitive maps. Although
spatial memory declines with aging, elderly participants could still recognize landmarks
after traveling through an unfamiliar hospital building, even if they could not reconstruct
the route. By comparison, college-aged participants could remember more routes and
landmarks (Wilkins, Jones, Koral, Gold, & Manning, 1997).

Schemas in Spatial Memory

Schemas are ways of abstracting, organizing, and storing general knowledge (see
Chapter 9). Spatial knowledge can be organized hierarchically. We may know the basic
geography of the United States, and subsumed under that is the local geography of our
308 Spatial, Motor-Skills, and Implicit Learning

home states, towns, and the layout of our neighborhood. Some of this knowledge was
learned through direct experience with places; some knowledge was learned indirectly
through study; and some knowledge is inferential: We guess this must be so.
Schemas have two prominent effects on spatial memory: they facilitate organization
and they can distort recall.

Distortion in Cognitive Maps

Spatial schemas distort recall due to the averaging, normalizing, or rounding off that
occurs when a generalized map is acquired. For example, we tend to encode all turns as
being 90 degrees, or right angles. Byrne (1979) asked people to draw some familiar road
intersections. The intersections actually deviated from right angles by at least 20 degrees,
yet the angles drawn averaged closer to 90 degrees. This means that if a series of left
turns are consistently greater than 90 degrees, going “around the block” will not return
us to our starting point as we might predict from a schematic memory.
Distortion in spatial schemas also occurs in comparisons involving geographic loca-
tions. For example, which U.S. city is farther to the west, San Diego or Reno? We assume
the correct answer is San Diego because California is farther west than Nevada. Which
city is farther north, Seattle or Montreal? Since Montreal is in Canada, it must also be
farther north than Seattle in the U.S. In fact, Reno and Seattle are the correct answers.
Our wrong guesses derive from schematic spatial knowledge. The locations of the state
or country is used to infer the relative locations of cities (Stevens & Coupe, 1978).
Spatial schemas have a preferred perspective. Much like a physical map, the
picture-like image of a spatial environment has a top, or a “specific orientation” (Sholl,
1987). Aligning the image with the environment facilitates using the cognitive map as a
guide. (Have you ever consulted a map and found yourself turning either the map, or
yourself, to line up with nearby landmarks?) Judgments are easier and more accurate if
forward in the environment corresponds to upward in the cognitive map (just as it is with
a physical map). For example, we could ask college students to point in the direction of
various campus locations. The students could more quickly identify locations in front of
them than locations behind them. When the students were asked to turn around, loca-
tions now in front were more quickly identified than were locations behind (Sholl, 1987).

Organization in Spatial Memory

Just as verbal material is recalled better when its organized, so also spatial memory is
better when organized. Menzel (1973) showed that chimps will organize their food search-
ing route to be more efficient and direct. Juvenile chimps were used as participants in a
study. As one researcher carried the chimp around a one-acre enclosed field, the second
researcher placed pieces of banana or lettuce. The experimenters crisscrossed the field
distributing 18 pieces of food in a random fashion. (This probably mimics your class
schedule: one class here, the next one is across campus, and so on). When later released
to retrieve the food, the chimps employed a principle of least distance in going from one
piece to the next nearest piece (Menzel, 1973). (Incidentally, the animals remembered an
average of 12.5 pieces per trial. Control animals that had not seen the placements but
were simply allowed to search averaged less than one piece). Searching locations was
further organized by type of food. The chimps first retrieved the fruit pieces, and then
 Motor-Skills Learning 309

went back for the vegetables. They would even pass by a vegetable on the way to the
next piece of fruit!

The Development of Spatial Memory in Children

The distinction between learning routes and maps has been demonstrated in the devel-
opment of spatial knowledge in children. Cornell and Heth (1979) arranged a learning task
analogous to the place-versus-response procedure of Tolman and Hull. Infants ranging in
age from 4 to 12 months old were seated in their mothers’ laps. Small projection screens
were placed on either side. Random shapes were projected to one side every 10 sec-
onds, and a constant checkerboard pattern appeared simultaneously on the opposite
side. Infants orient to novel stimuli and look less at repetitive stimuli (Chapter 2), and so
the infants learned to turn and look in the direction of the changing patterns. To test
whether the infants had learned turn responses or a cognitive map, the mothers turned
their chairs around to face in the opposite direction. Just as the starting location was
rotated 180 degrees in Tolman’s studies, the orientation with respect to the novel images
was rotated. On these test trials, the youngest infants continued to turn in the same
direction as before, which indicates response learning. The oldest infants correctly com-
pensated for change of orientation within the room and now turned in the opposite
direction.
Children have been tested in a human-sized radial-arm maze constructed outdoors
(Overman, Pate, Moore, & Peuster, 1996). The eight arms were about 8 feet long and
were enclosed by plastic mesh walls. Optimal performance in the radial maze requires
entry into each arm without repetition. Children as young as 20 months were somewhat
able to do this, with 40 to 50 percent accuracy in selecting different arms over the first
eight choices. By 5 years of age, the children were as good as the adults, who achieved
97 percent accuracy. In parallel to the rat experiments, the provision of distinctive cues for
each arm improved performance, and a delay interval between choices led to forgetting
of the previously visited arms.

MOTOR-SKILLS LEARNING
Spatial knowledge is one component of performance in a spatial task. There is also the
contribution of skilled movement. A well-trained rat, running through a maze at 4 feet per
second, “generally look(s) like a piece of well-oiled machinery, moving smoothly through
the maze with no hesitation or jerkiness, rounding off all corners, banking off the centrif-
ugal walls, and generally moving at full tilt throughout” (Olton, 1979, p. 584). Maze times
decrease over trials because the animal learns the route, and because it also knows how
to run the maze efficiently. To cite a human example, downhill skiers know the path and
have learned precise movements to traverse the route efficiently.
Motor-skills learning can be defined as the acquisition of precise movements in
which the amount, direction, and duration of responding corresponds to variations in the
regulating stimuli (Adams, 1987). The skill of playing tennis, for example, involves precise
and accurate movements in response to momentary changes in stimulus conditions,
often in anticipation of stimulus changes. A motor skill has perceptual, cognitive, and
310 Spatial, Motor-Skills, and Implicit Learning

Figure 11.3
Mirror-Tracing Apparatus (left) and Rotary Pursuit (right).
Source: Courtesy of Lafayette Instrument Company, Lafayette, Indiana.

motor components, the relative contributions of which vary from skill to skill. For instance,
keyboarding requires coordination of eye-(perception)-and-hands (motor) along with read-
ing in advance (cognition) of what is currently being typed (Inhoff & Gordon, 1997).
The combination of perceptual and motor skills is central in tasks such as the pursuit
rotor and mirror drawing (see Figure 11.3). The goal of the pursuit rotor is to keep a stylus
on a fixed point on a rotating disk (the latter being like a phonograph turntable). The speed
can be increased over trials to increase difficulty, and time on target is measured. In mirror
drawing, the subject attempts to follow the outline of a star with a pencil or stylus, but
visual guidance is through a mirror. Eye–hand coordination must compensate for the differ-
ence in actual direction of hand movements from their perceived direction. The time taken
to outline an object and the deviations from the outline can be taken as measures of accu-
racy. Each of these tasks requires eye–hand coordination, or perceptual and motor skill.
Is motor-skill procedural knowledge or declarative knowledge? Motor habits are
sometimes implicit: habitual behavior that requires little conscious control to perform. We
often cannot describe what we know. Try to verbally relay the steps involved in program-
ming your cell phone or performing a computer routine. Manipulations intended to
increase conscious control in mirror drawing, such as written instructions or having the
experimenter verbally guide your drawing, do not improve tracking performance (Borresen
& Klingsporn, 1992). On the other hand, motor learning does possess aspects of declara-
tive learning. Conscious intention to learn, verbal self-guidance, and knowledge of the
goal are indicative of declarative knowledge.
The separation of motor-skills learning and declarative learning is demonstrated by
comparing people with various neurological disorders. For example, mirror tracing occurs
at a normal rate among Alzheimer’s dementia individuals, even if they are severely
impaired in declarative memory tasks such as remembering words (Gabrieli, Corkin,
Mickel, & Growdon, 1993). An opposite pattern can be seen in some individuals with
movement disorders, such as Huntington’s or Parkinson’s diseases. These subjects are
impaired at pursuit rotor learning but not in recall of word lists (Harrington, Haaland, Yeo,
 Motor-Skills Learning 311

& Marder, 1990; Heindel, Butters, & Salmon, 1988). (There are exceptions to the preced-
ing, depending on the stage of the disease).
Although the emphasis in this portion of the chapter is on the motor skill, these tasks
also have a memory component that improves with training. Fendrich, Healy, and Bourne
(1991) had participants enter random-number strings into one of two keyboards that had
different placements of the numbers 1–9. On one touch pad, the numbers ran from 1 to
9, top to bottom; on a different keyboard the numbers ran from 9 to 1. Learning was
tested 1 week later by having the participants enter old versus new number strings, using
the same or different keypads. Old number strings were typed faster than new strings,
indicating memory for the numerical sequences. Typing old strings on the original key-
board was faster still, indicating memory for the motor movements originally used.
Decades of research on motor-skill learning has emphasized two important controlling
factors: practice and feedback.

Practice
Amount of Practice
Motor skills improve with repetition. The relationship between practice and one measure
of skilled behavior, the speed of performance, is well described by the power curve. The
idea of the power curve is much like that of the learning curve, a concept we have seen
before (Chapter 1). Basically, responses become faster with additional practice, but not in
a one-to-one fashion with the number of practice trials. Rather, the idea behind a power
function is that ever greater amounts of practice are required to produce comparable
increments in performance. That is, whereas the first 10 trials might produce a certain
increment in speed, the next increment of that size may require 100 additional trials. Thus,
the first point on the learning curve represents x number of practice trials, the next point
is x2 number of trials, and then x3 number of trials. Plotting speed over the log number of
trials produces a straight-line increase in speed over trials.
You may have noticed something like the power curve in learning a video game or a
new sport. At some point, improvements in playing seem to have diminished. Frustration
or discouragement may prevent the additional (and extensive) practice needed to further
improve performance.

Schedules of Practice
A general principle of learning is that spaced repetitions lead to faster learning than do
massed repetitions. The spaced-practice advantage also applies to motor-skill learning.
For example, mirror star tracing accuracy improves more quickly with one trial per day
than with 10 trials— that is, if one counts trials not days (Hovland, 1951).
Bourne and Archer (1956) compared different spacings between practice trials in the
pursuit rotor task. Each trial lasted 30 seconds. Different conditions had different intervals
of down time between practice trials. As shown in Figure 11.4 (curves on the left side),
there was longer time on target when there were long intervals between trials (30 or 60
seconds) than when the trials were massed one-after-another (0 second interval).
Baddeley and Longman (1978) studied the spacing of practice sessions on learning to
type. The participants (British postal workers) were divided into four groups who received
either one or two training sessions per day, with each session being 1 or 2 hours in
312 Spatial, Motor-Skills, and Implicit Learning

35
60 sec
30 sec
30
0 sec
Percent Time on Target 25

20

15

10

0
1 2 3 4 5 6 7 Rest Test
Blocks of 3 Trials

Figure 11.4
Effects of Trial Spacing on Pursuit Rotor Time-on-Target. Trials were separated by 0, 30, or 60 seconds. After
21 trials, a 5-minute rest interval was given, and then three test trials were given.
Source: Adapted from “Time Continuously on Target as a Function of Distribution of Practice,” by L. E. Bourne and E. J. Archer,
1956, Journal of Experimental Psychology, 51, p. 27. Copyright American Psychological Association. Reprinted with
permission.

length. The most dramatic results came from the two extreme conditions, the group that
practiced just one hour a day, and the group that had two, 2-hour sessions each day (see
Table 11.3). After 60 hours of practice, the subjects that received the most distributed
practice, a single 1-hour session per day, were the best typists. The group that had the
most concentrated training conditions, two 2-hour sessions daily, had the poorest perfor-
mance. The distributed group required fewer hours to learn, typed faster, and made fewer
errors. (The 2 x 2 massed group was given 20 additional hours of training so they could
catch up to the rest of the experimental subjects). The advantage for spacing persisted on
tests given 1, 3, and 9 months later.
Although spaced training produced better performance, training extended over many
more days than it did in the massed conditions. The 1 × 1 group, training five days a week,
took 12 weeks to complete the standard 60 hours of training, whereas the 2 × 2 group
finished in three weeks. (I’ll let you check the math). The postal workers were surveyed
for their reactions to the training regimen. Surprisingly, the spaced groups were less sat-
isfied and said they would not choose to train under those conditions. The extended num-
ber of training days did not appeal to them. As Baddeley and Longman note, if left to the
participants, people would choose to train under the conditions that produced the poorest
learning but required the fewest days.
One way to space trials is to intermix training of different skills within a session. Given
several skills to learn, one could block trials on one skill at a time (massed practice on
each) or alternate trials among the skills (spaced trials). Goettl, Yadrick, Connolly-Gomez,
and Regian (1996) tested a schedule of alternating blocks of practice on the acquisition of
 Motor-Skills Learning 313

Table 11.3 Learning to Type as a Function of the Distribution of Practice. Groups received
1-hour or 2-hour practice sessions, and one or two sessions each day

Groups 1x1 1x2 2x1 2x2

1 hr / day 1 hr / 2x day 2 hr / 1x day 2 hr / 2x day

Hours to learn 35 42 43 50
Letters /min 79 73 71 65
Errors 1.1 1.1 1.4 2.0
Source: From “The Influence of Length and Frequency of Training Sessions on the Rate of Learning to Type,” by
A. D. Baddeley and D. J. A. Longman, 1978, Ergonomics, 21, p. 630. Copyright © 1978 by Taylor & Francis, Inc.
Reprinted with permission.

three very different tasks: “Space Fortress,” an attention-demanding, spatial- and motor-
skills video game; “Phoenix,” a flight-simulator video game but using the keyboard instead
of a joystick; and algebra problems, a cognitively demanding task that is not quite as much
fun as the other two. Participants either practiced one task each day, or alternated the
three tasks within each day. Both schedules provided the same total amount of practice
over the several days of the experiment. The alternating schedule enhanced performance
on two of the tasks, Space Fortress and algebra, and was no worse on the flight simula-
tor. These results have implications for organizational training. Organizations usually want
to schedule training sessions over a day or two, especially if the employees need to be
brought in from the field. The alternating module schedule might offer the benefits of
distributed practice but in the same two or three days a company might otherwise devote
to massed practice.
Massed presentation of practice trials can inhibit performance, apart from its effect
of learning. Massing a large number of trials can lead to fatigue, loss of motivation, or
attentional lapses that lower performance. Spaced trials allow time for recovery from the
“work decrement,” as it was called by Hovland (1951). This is our learning-versus-perfor-
mance distinction again. Performance after a series of massed trials sometimes improves
immediately if the next trial is delayed, presumably allowing opportunity for fatigue or
other inhibiting factors to dissipate (Kientzle, 1949).
The Bourne and Archer (1956) data shown earlier in Figure 11.4 show the effects of
allowing time for recovery from massed practice. At the end of training, all subjects
received a 5-minute rest interval, and then additional practice trials commenced. As can
be seen in Figure 11.4, on the test given after a rest, each spacing condition showed an
increase in performance. The massed practice subjects had the largest increase. Their
performance during the previous practice trials underestimated how much the massed
practice group had actually learned. They would have had the most fatigue, work decre-
ment, or inhibition and so would have benefitted the most from a rest interval.

Practice-Independent Learning
Skill learning improves when the learning session is followed by a period of uninterrupted
sleep (see Sleep Learning, Chapter 9). This is practice-independent learning. It is an
improvement is learning or memory that occurs without additional practice.
314 Spatial, Motor-Skills, and Implicit Learning

For example, one study looked at the speed of typing a sequence of numbers repet-
itively (e.g., 4-1-3-2-4). Students could type this sequence about 22 times in 30 seconds
after a single practice session. Retesting after 12 hours of wakefulness only raised the
score to about 24 times in 30 seconds. A more dramatic effect was observed if sleep
occurred during that 12 hour break: performance increased to about 27 repetitions. The
time spent sleeping was more effective than the time spent awake in improving perfor-
mance. In another experiment, sleep was found to be as effective as a second training
session (Walker et al., 2003). The question for students is: Practice or sleep? There’s a
difficult choice.
The idea is that sleep—or specifically, certain phases of sleep—is important for the
long-term consolidation of recently acquired skills and habits. This is the practice-inde-
pendent component of learning.

Knowledge of Results
Repeated practice will be of little benefit unless you know how well you are performing.
Knowledge of results (KR) is feedback concerning the success or accuracy of a response
that is given after study or practice. This information serves as a basis for corrections on
the next trial. Theorists have considered KR to be analogous to reinforcement in instru-
mental learning, and therefore KR was manipulated in ways that paralleled reinforcement
variations. For example, one could vary (1) the frequency with which feedback was given,
from feedback after every response (a continuous schedule of reinforcement) to feedback
after some responses (a partial reinforcement schedule); or (2) the immediacy of feed-
back, in parallel to delay of reinforcement. The commonsense expectation is that more
frequent and more immediate knowledge of results should enhance learning. Imagine
learning to hit a golf ball. Without feedback after each stroke, how can you know whether
you hit the ball well or not? If feedback is delayed, will you remember the exact move-
ments you made that led to a good or poor shot? A long history of research clearly indi-
cates that frequent, immediate, and detailed feedback leads to faster learning of motor
responses (see Adams, 1987).
However, some researchers have questioned the benefits of using too much feed-
back during training (e.g., Schmidt & Bjork, 1992). “Leaner” (less-frequent) schedules and
less-consistent schedules of feedback may produce better long-term retention of a skill,
and more transfer to altered forms of the skill. We can cite examples involving both the
frequency and the delay of feedback.
In one study, the participants learned a tracking response, using one finger to follow
a curve projected on a screen that changes in speed and direction. Feedback was given
after each trial or in summary form after every 5th or 15th trial. These are equivalent to
reinforcement scheduled for every response, after every 5th response or after every 15th
response. The group that received the highest frequency of feedback made smaller errors
throughout the acquisition phase of the experiment. But on a tests given two days later,
the leanest feedback schedule (that given every 15th trial) had the best performance
(Schmidt, Young, Swinnen, & Shapiro, 1989). Consistent feedback produced better perfor-
mance during initial acquisition of the skill, but a partial feedback schedule led to better
delayed retention of the skill.
 Motor-Skills Learning 315

One explanation for why less-frequent KR leads to better performance is the

self-guidance hypothesis. Feedback guides the learner toward the correct movements.
But consistent feedback may block the participant’s learning to detect their own errors.
As the participants become dependent on external KR, they are less likely to attend to
their own bodily kinesthetic feedback, and do not learn to recognize good and poor
performance. Skilled athletes know immediately whether a movement (e.g., a pitch or a
hit in softball) was good or not, before seeing the actual outcome. Thus, frequent KR has
both positive and negative effects. Optimal performance requires a balance between
the two.
Is there a way to combine the beneficial effect of consistent KR on initial learning with
the beneficial effect of partial KR? Wulf and Schmidt (1989) did so by initially providing
feedback on each trial and then fading (or gradually reducing) feedback to a 50 percent
schedule. In this case, acquisition was as rapid as it was for a 100 percent feedback
group, and as good as a 50 percent feedback group on tests of novel versions of the
required movements.

Delayed KR
In studies similar to those varying frequency of feedback, knowledge of results can be
given immediately after the response, or after a delay interval. As with the intermittent
schedules of feedback, delaying feedback allows the participants to develop their own
error-detection capabilities. Participants for whom feedback was delayed for 3 seconds
after each movement learned just as well during an initial training session. The
delayed-feedback group actually did better on the second day, and this superiority
persisted on a test given two days after that (Swinnen, Schmidt, Nicholson, & Shapiro,
1990).
What is happening during the delay-until-feedback interval? Participants may be
reviewing the response they just made and making a judgment about how accurate it
was. Immediate KR does not give an opportunity for review or rehearsal to occur, interfer-
ing with development of self-guidance (Swinnen, 1990).

Metacognition in Motor Skills

Motor skills are objectively measured by tests of speed or accuracy. We can also question
participants about how well they think they are performing. Actual skill and self-perceived
skill may be different things (much like your friend’s dancing skill). Metacognition deals
with beliefs and self-knowledge about what we have learned. Inaccurate perceptions can
lead to insufficient practice, learning, and performance.
Simon and Bjork (2001) used a task in which their subjects practiced a keypad skill:
typing out three numbers in a specific sequence and timing pattern (e.g., 2–7–_–5, having
a slight delay after the 7). Subjects who practiced one pattern at a time (massed trials on
one pattern, then the second pattern, then the third), performed better than subjects who
practiced the three patterns intermixed. So far, this is an objective measure of perfor-
mance. The participants were brought back to be tested the next day. Given what you
have read so far in this chapter, you know how delayed testing will turn out: The inter-
mixed practice condition will do better than the massed practice condition. The students
were asked how well they thought they would do on the test. The massed practice
316 Spatial, Motor-Skills, and Implicit Learning

subjects judged they would do better; their superior performance the day before misled
them into thinking they were better than they actually were. On the day two test, the
massed practice group did far worse than the intermixed practice group: They were badly
off in reproducing the timing between keypunches in the sequences.

Long-Term Retention of Skills

The goal of training is long-term retention. Performance in a job or occupation often
depends on skills and knowledge that are used infrequently. To take one extreme exam-
ple, for many of us, emergency procedures are not needed every day, yet they will need
to be quickly recalled when an emergency does arise.
A team of researchers at the University of Colorado at Boulder has been especially
interested in long-term retention of skills (see, e.g., Healy et al., 1993). They studied skills
ranging from learning Morse code to performing in an army tank simulator (which is like
the most awesome video game ever!). These researchers verified several guidelines that
apply to the learning of motor skills, cognitive skills, and factual knowledge. Several can
be summarized by the phrase: Optimize the Conditions of Training.
Train or study using spaced practice and interleaving of different tasks. Practice one
component of the skill for awhile, then another, then back to the first. After a skill is
learned, return to it at a later time to refresh the skill.
Feedback is important, especially at first. However, feedback can be tapered off to
allow self-guidance to develop. Use practice or refresher tests.
Over train. “Overlearning” is the name used in conditioning and verbal learning con-
texts. Additional practice, even though it does not lead to further improvement now, leads
to longer and better retention in the future.
Practice/train in different contexts. You may be called upon to demonstrate your skill
in different situations (e.g., home game versus away game). The learning needs to gener-
alize beyond a single place.
Train until retrieval is automatic. Mnemonics, verbal self-guidance, counting the steps
all aid initial learning, but the skill (physical or cognitive) needs to operate automatically.
Motor skills learning overlaps with other forms of learning. In the following sections
of this chapter, more will be said about skill learning’s relationship to implicit learning and
the development of expertise with extended practice.

IMPLICIT LEARNING
Another form of “learning how to” is implicit learning. Implicit learning is the acquisition
of cognitive, motor, or perceptual skills that develop with training. Implicit knowledge is
said to be independent of conscious awareness of specific details of the tasks. Implicit
learning is exemplified in learning our first language. We learn the rules of word order,
word endings, pronunciation, and stress patterns, even though we cannot articulate the
rules. Similarly, in laboratory simulations of implicit learning, participants are exposed to
cognitive tasks governed by abstract sets of rules. Skilled performance develops as a
function of practice, and in such a way that the knowledge is difficult to express verbally.
Learning is therefore said to be implicit.
 Implicit Learning 317

Some Implicit-Learning Tasks

Many implicit-learning tasks are rule-based. One well-studied example is that of learn-
ing an artificial grammar. Human languages are guided by a set of rules, or the gram-
mar. For example, the usual word order in English is subject–verb–object (e.g., dog
bites man). We readily reject ungrammatical strings of words (dog man bites). Reber
(1967) devised an artificial grammar involving a set of letters instead of words, such as
P, S, T, V, and X. The grammar determines which letters can follow which other letters.
Figure 11.5 shows allowable sequences in one grammar. Each circle labeled S repre-
sents a choice point. For instance, the first letter would be T or P and the last letter
would be S or V. The arrows show the permissible transitions from letter to letter; TSXS
is a grammatical string; PSXS is not. Recursive loops allow letters to be repeated; thus
TSSXS.
The grammar is not learned by studying a set of rules as in grammar school.
Participants first memorize short sets of letter strings. The strings derive from the gram-
mar, although the participants are not told there is an underlying set of rules. For instance,
a set of three strings of letters is presented, and recall tested, until the participants can
repeat them back. Then another set of letter strings is learned. Over trials, learning new
strings of letters becomes easier, indicating that the rules that govern grammatical strings
are being abstracted. Participants who memorized grammatical strings learned new let-
ters faster than did participants who studied letter strings that did not correspond to the
grammar. Participants who had memorized grammatical letter strings were able to clas-
sify new sets as grammatical with 60 to 65 percent accuracy, significantly better than a
chance level of 50 percent.

S2 S4
X
T S

S1 X P S6

P V
V
S3 S5

Figure 11.5
Artificial Grammar. Each “state” (S) represents a random choice: a certain proportion of the time the transition
is in one direction (from S1 to S2, producing the letter T) or the other (from S1 to S3, producing the letter P).
Source: From Implicit Learning and Tacit Knowledge: An Essay on the Cognitive Unconscious (p. 28), by A. S. Reber, 1993, New
York: Oxford University Press. Copyright © 1993 by Oxford University Press. Reprinted with permission.
318 Spatial, Motor-Skills, and Implicit Learning

Grammar learning is said to be implicit because it occurs incidentally, not deliberately.

The regularities of the letter strings are abstracted across memorized sets. Deliberate
intention to discover and learn those rules can actually impede learning. Informing the
participants that there are rules leads to more errors during set memorization and more
mistakes in later classifying grammatical and ungrammatical strings (Reber, 1976).
Another implicit-learning task is the serial-reaction-time procedure. A sequence of
items is presented and the participant’s task is to make a particular response to each as
it occurs. For instance, a light appears at one of four locations on a video screen and the
participant pushes a corresponding key for each location. A continuous sequence of loca-
tions is presented. The sequence is seemingly random, although in fact it repeats.
Sequence learning is indicated by reaction times that become faster over practice trials.
The results from a representative study are shown in Figure 11.6 (Nissen & Bullemer,
1987), in which the sequence of locations repeated after every 10 lights. Key-pressing
reactions become faster across 80 trials, even for those participants who said they were
unaware that the sequence repeated. Are people faster because they have learned some-
thing about the sequence, or have they just become practiced at more general aspects of
the task? A control condition received a random sequence of locations, and their speed
of reaction improved little over trials.
Some implicit tasks produce learning for the specific items studied, whereas other
tasks teach a more general skill. Both specific and general learning effects are found in
learning to read transformed text. The letters in words are mirror-reversed (left-right) or
inverted (rotated upside down). Moscovith, Winocur, and McLachlan (1986) recorded the
time to read words composed of inverted letters. One group of their subjects were older,
community-dwelling adults who had some memory-impairment. A comparison group
included unimpaired age-matched controls. Both groups became faster at reading trans-
formed words with practice. This shows learning of the general skill. What is interesting

380 Random

340

300
Reaction Time

260
Repeating

220

180

140

1 2 3 4 5 6 7 8
Block

Figure 11.6
Mean Reaction Times (in milliseconds) to Repeating or Random Sequences of Locations.
Source: From “Attentional Requirements of Learning: Evidence from Performance Measures,” by M. J. Nissen and P. Bullemer,
1987, Cognitive Psychology, 19, p. 8. Copyright © 1987 by Academic Press. Reprinted with permission.
 Implicit Learning 319

60

Control Memory Impaired

50
Mean Reading Time (sec.)

40

30

20
Repeated

New
10

0
1st 2-hr 2-wk 1st 2-hr 2-wk
Trials Trials

Figure 11.7
Mean Times to Read Inverted-Letter Text.
Source: Adapted from “Memory as Assessed by Recognition and Reading Time in Normal and Memory-impaired People with
Alzheimer’s Disease and Other Neurological Disorders,” by M. Moscovith, G. Winocur, and D. McLachlan, 1986, Journal of
Experimental Psychology: General, 115, p. 335. Copyright by the American Psychological Society. Reprinted with permission.

is that this skill was retained when the subjects were tested two hours and two weeks
later. As can be seen in Figure 11.7, reading times continued to decrease from the first
practice session through the two delayed tests. Although the memory impaired read
slower than the matched control subjects, the impaired still showed faster reading with
practice. Some of the words from the first session were repeated on the two hour and
two day tests, and some were new words. The repeated words were read faster than
new words, indicating some memory for the repeated words.
The subjects were also given explicit tests of memory. Words were shown and the
subjects judged whether the words were old (had been seen before) or new. The memo-
ry-impaired were consistent with their label: They were unable to distinguish between old
and new words on the delayed tests. Yet they read old (repeated) words faster than they
read new distorted words, even when those words had last occurred two weeks earlier.
This shows they implicitly remembered words they had read earlier.

Is Implicit Learning Unaware Learning?

The difficulty in answering this question is in determining whether the knowledge is
actually unconscious. One means is to show a dissociation between a measure of
320 Spatial, Motor-Skills, and Implicit Learning

performance and direct verbal report. For instance, maybe the participants can perform
the implicit task but they cannot articulate the rules (Dienes & Berry, 1997).
An example of this strategy is research on the Iowa Gambling Task (or IGT) (Bechara,
Damasio, Tranel, & Damasio, 1997). (The researchers were at the University of Iowa
Medical School at the time). In the IGT, participants make a succession of choices from
four decks of cards. Cards in two decks produce large winnings (e.g., a hypothetical
$100), but also large and more frequent losses. In the long run, these decks produce net
losses. Cards in the other two decks produce smaller winnings ($50) and less frequent
losses. After sufficient training, the subjects learn to choose from the decks with the
modest payoffs. This is explicit knowledge. Early in training, the subjects cannot yet
describe these contingencies. Yet measurement of the skin conductance response, an
indicator of emotional arousal, shows that the participants react more strongly as they
choose the next card from the “bad” decks. The reaction occurs before the outcome—
gain or loss—is revealed. Before consciously knowing the good or bad decks, the sub-
jects appear to be unconsciously aware which ones could produce large losses. The
science writer Malcolm Gladwell (2005) describes this as a gut-level feeling; just a premo-
nition that this is not the best choice. What is the basis for this intuition? Antonio Damasio
says that implicit knowledge produces emotional reactions in the body. Participants do
not know which deck is better. Instead they feel it.
This dissociation of implicit and explicit learning is based on the comparison of bodily
(SCR) reactions to the subjects’ self-reports. But verbal reports are not always accurate
depictions of the subjects’ knowledge. Maia and McClelland (2004) found that their par-
ticipants knew a lot more, and much earlier in their experiment, than Damasio found.
Using an extensive set of questions, they found that participants’ verbal answers revealed
they knew which sets produced more advantageous outcomes even though their choices
did not mirror this.
Even if the subject has explicit knowledge of the rules, when does this knowledge
appear? Awareness could develop before implicit learning occurs, after implicit learning,
or coordinately with the implicit learning. Thus, the degree and influence of explicit knowl-
edge on performance of an implicit learning task is still unknown (Frensch & Runger,
2003).

Dissociating Categories of Implicit Learning

Priming is another form of implicit-memory. In a priming experiment, a list of words is first
presented as part of an incidental task (e.g., rating the words for pleasantness). In a sec-
ond phase, the words are presented in the form of word fragments (letters and blanks).
A priming effect occurs if the fragments are completed with the previously shown words.
Priming, implicit learning, and procedural learning are often categorized together as
one form of memory (Squire, 1987; Tulving & Schacter, 1990; see Chapter 7). However,
priming and procedural learning are separable, as shown by comparisons among individ-
uals with various neurological disorders. Huntington’s disease is characterized by involun-
tary movements of the body. Average age of onset is in the 40s. Alzheimer’s disease is
evidenced by primarily cognitive deficits and has a later age of onset. Individuals with
these diseases have been compared in learning a motor skill, such as the pursuit rotor.
 Expertise 321

The goal in this task is to keep a stylus on a fixed spot on a rotating turntable. The starting
speed is adjusted so that everyone starts with the same time on target. Alzheimer’s
individuals improve across blocks of trials; that is, they become more skilled. Huntington’s
disease participants do not (Heindel et al., 1988). The reverse pattern is found on a prim-
ing task: The Huntington’s patients completed more word fragments with the primed
words than did the Alzheimer’s participants (Shimamura, Salmon, Squire, & Butters,
1987). Note that the Alzheimer’s participants were cognitively capable of completing
word fragments; they just completed them with other words.

EXPERTISE
An expert is someone who has a special skill and/or knowledge of a subject derived from
training or experience. Experts can demonstrate clear and consistent superiority over
their peers in their field. Expertise can be developed in virtually any domain of knowledge
or skill. One can become skilled at video games, programming computers, speaking mul-
tiple languages, or reading brain scans.
Where does expertise come from? What is it that differentiates chess masters from
lesser players, or world-class musicians from others? One assumption is that expertise
requires an inherent talent. An ability might be inherited that allows performance at an
expert level, and not everyone has this talent. This inherent-ability conception is a popular
stereotype about child prodigies, and “savants” who display an exceptional talent in
music, art, or math (see Chapter 12 for further discussion of superior memories).
The notion of inborn talent was challenged by Anders Ericsson (see particularly
Ericsson & Charness, 1994), who argued that what separates the expert from the also-
rans is the sheer amount of practice devoted to their skill. Those who attain expert status
differ in the amount of deliberate training and instruction. Individuals on their way to
expert status maintain schedules of intense and prolonged practice, in some cases
essentially engaging in the activity full time. For example, college students trained to
develop digit spans of 80 to 100 numerals (Ericsson, Chase, & Faloon, 1980). That is, the
students learned to repeat back a random string of 100 numbers after hearing (or seeing)
it once. The students did this through extended practice, learning various coding schemes
and becoming adept at rapidly encoding the numbers into long-term memory. Joshua
Foer spent a year becoming a memory expert, partly by studying with Ericsson, learning
how to memorize the sequence of a shuffled deck of cards. He won a national Memory
Championship (see Chapter 6). Maybe child prodigies get an early start at practicing, and
spend years perfecting their talents.
It may be that expertise is determined by a combination of factors (Oerter, 2003).
Genetics may contribute to a number of individual skills, not just a single talent. For exam-
ple, musical expertise might require absolute pitch, memory for music, and subtle motor
skills. Such individual skills (and their underlying genes) would be distributed throughout
the population, but some individuals would fortuitously inherit the combination of genes
for exceptional musical ability. In addition, advantageous environments might foster
development. A child’s aptitudes may be encouraged by parents or teachers, or the pre-
cocious child seeks out experiences that enhance their skill. Moreover, differences in the
brain’s structure or functioning might underlie exceptional performance. Such brain
322 Spatial, Motor-Skills, and Implicit Learning

differences may be present from birth or result from the early childhood use of musical
abilities. The hippocampus in London taxi drivers was found to be larger, which may have
been due to their practice at navigating.
Hambrick and Meinz (2011) found evidence for talent and practice. They studied the
ability to sight-read music as a function of practice and working memory (the latter is their
proxy for talent or ability). The results showed independent effects for both factors: read-
ing music was better among those with better working memory, and those who had
more total practice. The best performance was by those with both good memory and lots
of experience.
How much practice is necessary to attain expertise? Ericsson, Krampe, and Tesch-
Romer (1993) concluded that it takes about 10 years of full-time preparation, which corre-
sponds to several thousand hours of practice or study. This 10-year rule applies to many
domains of expertise, such as sports, chess, or music. For the most talented in the arts
and sciences, more than 10 years may be required. Ericsson et al.’s (1993) review showed
that those who attained the highest levels of expertise had begun practicing two to five
years before less accomplished experts and spent more time on deliberate practice. Top
violinists at age 20 had spent more than 10,000 hours in practice, twice as much as that
by those at lower levels of accomplishment (who were still experts themselves).
Is 10,000 hours necessary? Among chess masters, the mean number of self-reported
hours of deliberate practice was about 11,000 (Gobet & Campitelli, 2007). However, the
range was great, plus-or-minus 6,000 hours. So, a few (the number is not reported)
achieved master status with far-less than 10,000 hours. And, some players had not
achieved master status after as much as 16,000 hours (and one individual with 20,000
hours) of practice. Very generally, something like the 10,000 hour rule holds, with excep-
tions: Some become expert after fewer hours, and some do not attain expertise with
considerably more hours (Campitelli & Gobet, 2011).
The kind of practice is important also. Simply participating, or playing the game, does
not count towards practice. Ericsson notes that deliberate practice is sustained and repet-
itive practice, particularly on your weaknesses. Playing rounds of golf does not count.
There, you get one chance at each shot, whereas with deliberate practice you would
repeat the poor shot to try and perfect it. Practice in a skilled behavior (video games,
sports) also requires cognitive effort: analyzing good and poor movements, how to
improve, and what to change.
We might argue in response that without sufficient innate intelligence or athletic
ability, no amount of practice could turn each of us into a Bobby Fischer or an Arthur Ashe.
The point Ericsson wants to make is that when we look at the lives of experts, we see
they have devoted an enormous number of hours to their craft. Ten thousand hours of
training can account for a substantial amount of talent. Ericsson does allow a possible
hereditary factor, but it is not a specifically inherited talent. What may be inherited is
motivation: the drive and perseverance necessary to practice.

From Declarative to Procedural

We have mentioned the roles of conscious and unconscious factors in skilled behaviors.
Theories of skill learning have incorporated both by breaking down skill learning into a
 Applications 323

series of steps or stages. First, factual knowledge is acquired: learning the nature of the
task, the rules of the game, and so on. With additional practice, less conscious guidance
is required. And after extensive practice, a skill is performed seemingly automatically. This
succession of steps marks the transition from declarative knowledge to procedural
knowledge.
Anderson’s adaptive control of thought (ACT*) theory (1983) is designed specifically
to describe this transition. Anderson proposes a two-stage model of skill acquisition, with
a transitional step between stages. During the declarative stage of skill learning, informa-
tion is learned. This stage involves conscious processing and attention, so there is heavy
reliance on working memory. For example, you need to learn the different pieces in chess,
their range of moves, and some strategies. During this initial phase, we verbally encode
and rehearse information, or give ourselves feedback and guidance. The declarative stage
is followed by a transition phase in which knowledge application starts to become proce-
duralized. During this transition, groups of rules or operations that are frequently used
together are chunked, or compiled, which increases efficiency. This transitional state is
known as knowledge compilation. An analogy can be made to playing a piece of music.
With practice, individual notes become combined into chunks that are stored in long-term
memory. Playing becomes the activation of groups of notes, rather than playing note by
note. Performance now depends less on conscious control and more on long-term
retrieval, thus freeing up working memory.
The second stage, or procedural stage, is marked by skill refinement. Continued prac-
tice leads to further strengthening of the procedures. Knowledge is retrieved automati-
cally, and separate responses become unitized. For instance, eye movements of chess
players have been monitored as they plan a move. The novice’s eyes focus at one piece,
then at a second location, then to an opponent’s piece at a third location. The skilled player
focuses on the first location but sees the other two and the implication of the move:
putting the King in check. A pattern is perceived with a single eye fixation.

APPLICATIONS
Implicit Learning
If amnesic individuals are, by definition, memory-impaired, how can they learn new skills?
Because implicit-learning is often spared in amnesia, it can be used instead of explic-
it-learning methods in remediation. Cognitive rehabilitation can help individuals who have
amnesia cope with everyday life in the face of memory deficits.
Glisky and Schacter (e.g., 1989) used implicit learning to teach a woman with antero-
grade (or ongoing) amnesia to use a computer. She was first taught technical definitions
using a “vanishing-cue” method. A word and its definition were first presented, and
across repeated trials letters were omitted from the target word, e.g., DISK DRIVE, then
D_ _K DR_ _E, then D_ _ _ D_ _ _ _. Eventually, the letter cues were omitted as the
woman learned to name the word given the definition. “Vanishing instructions” were also
used to teach the steps involved in operating the computer. Gradually, each line of instruc-
tions became abbreviated, then whole steps were eliminated. This woman eventually
learned over 250 codes, symbols, and abbreviations. Through all this, her explicit memory
324 Spatial, Motor-Skills, and Implicit Learning

remained impaired. She was able to perform data-entry and returned to work with her
former company. The vanishing-cue method minimizes retrieval of the wrong information,
which might later be mistakenly recalled as the correct answer. The method at first elicits
correct responses. If errors were made, they, and not the correct words, might be recalled
in future trials (Baddeley, 1992a).

Spatial Memory
Sense of Direction
Surveys of real-world memory often turn up a statistical factor associated with wayfind-
ing, or its opposite, the ease of getting lost (Ryan, 1992). Individuals will readily volunteer
self-estimates about their own sense of direction, and these estimates are accurate.
When students rated their sense of direction on a simple 7-point scale, the estimates
significantly correlated with the accuracy in pointing to (unseen) campus buildings, judg-
ing the distance between pairs of buildings, and pointing in the direction of nearby cities
(Kozlowski & Bryant, 1977). After being led through an underground utility tunnel on cam-
pus, students who rated themselves as having a good sense of direction showed
improved accuracy in knowing where they were.
It is easy to get lost when we travel. After finding your way there, the problem is then
remembering how to get back. The psychologist Colin Ellard (2010) offers a number of
ways to find your way back. Some of his suggestions are:
1 When going forward, occasionally look back. That’s the view you will need to recog-
nize on the way home.
2 In visiting a new area, choose one location as a home base. Learn the relationship of
other locations to this starting point.
3 Make up a story to link together locations or landmarks you pass. This elaboration
makes for a better memory, and the story will be helpful in retracing the outbound
route.
4 Don’t get upset: Whether it is fear or anger over being lost, emotional arousal inter-
feres with retrieval. The result will be poor recall of what you do have in memory.
5 Keep track of time. The novel outbound trip seems to take so long. Over-estimating
time (and distance) on the way back can lead to missed landmarks or turns.
A related issue is whether sense of direction might be affected by the use of electronic
way-finding devices, such as GPSs. Just as our memory for phone numbers seems to
have disappeared, maybe our way finding abilities will become extinct also. This topic is
considered in Box 11.2.

BOX 11.2 GPS AND SPATIAL KNOWLEDGE

Physical maps have been replaced by ground positioning devices (GPSs), turn-by-turn direc-
tions, and location detectors. Is our new knowledge of routes and directions different from
what we would have acquired using a map? A map and a GPS differ in several ways. The GPS
shows only a portion of the route at a time; or you might be listening to turn-by-turn directions
 Applications 325

with no map image at all. The GPS continuously displays the user’s location, whereas map
users need to update their position on the map as they travel. Does less effortful cognitive
processing (or at least, different processing) lead to a decline in spatial cognition, as more of
us off-load wayfinding to electronic devices? Just as we have ceased memorizing phone
numbers (the cell phone does that for us) we might not remember spatial directions.
Studies have compared memory after using a GPS or a printed map to navigate an unfamiliar
city. In one case, Japanese college students traveled several such routes. After reaching the
end point of each, the students who followed GPS directions drew less accurate maps of the
route and made larger errors in estimating the direction back to the starting location. Interestingly,
the GPS users traveled longer distances (Ishikawa, Fujiwara, Imai, & Okabe, 2008).
In another case, American students traveled two routes, one with a map and one with a
GPS. When the physical map was used, the students produced more detailed and accurate
sketches of the routes. There was more information on distance, direction, and turns. There
was no difference between using a map or a GPS in recall of semantic knowledge of the
route: the names of buildings or roads. Places could be named and arranged sequentially
along the route (e.g., a store near the railroad crossing). The differences seemed to be more
in the overall spatial arrangement that was less-well known in the GPS condition (Wessel,
Ziemkiewicz, Chang, & Sauda, 2010).

Inuit
The native Inuit populations of northern North America have long had their own wayfinding
methods to navigate their barren terrains (or, what seem to be barren to the rest of us). The
Inuit use wind direction, patterns of snowdrifts on land, direction of currents in the water, and
celestial patterns as navigational aids. It takes many years to learn to navigate by physical
environmental cues and to build mental knowledge of the area. These methods are disap-
pearing as the younger Inuit generations began using GPS units. Their elders are concerned
that traditional lore and skills will be lost. Parallel concerns have echoed throughout history:
that printed books would replace the need to remember information; that photographs would
replace artists’ renderings; and, among the Inuit, that snowmobiles would replace dog sleds
(Aporta & Higgs, 2005).
The GPS does have some advantages over traditional environmental navigation cues. It
provides a margin of safety for the inexperienced, or when fog and snowstorms distort envi-
ronmental stimuli. Even traditional hunters will use the GPS in certain circumstances.
Experienced hunters who used GPS also knew when not to follow its advice. The GPS points
in straight lines but does not indicate obstacles (chasms or ice) that block the direct path.
Navigating with environmental stimuli requires a high degree of cognitive engagement
and learning requires a long period of apprenticeship. The GPS can be learned quickly. It is as
yet unclear whether the GPS will replace, or simply supplant, traditional Inuit techniques for
wayfinding. In the end, the GPS may follow the same route as the snowmobile, by becoming
an accepted part of Inuit life.

Improved Building Design

The principles of spatial learning can be applied to improving everyday navigation.
Hospitals, large public buildings, shopping malls, and residential developments all tax
user-orientation skills. These are difficult environments to get to know because of size,
infrequent exposure, or lack of external frames of reference.
326 Spatial, Motor-Skills, and Implicit Learning

An aid that could be employed in large buildings is to use a color-coding scheme to

distinguish different sections. Evans, Fellows, Zorn, and Doty (1980) took advantage of
the repainting of a large university building. Students in the “before” condition toured the
four-story building when the interior was painted a monochromatic beige. They were then
asked to find their way back to a specific location (e.g., go to the chem lab). The interior
walls on each floor were subsequently painted with different colors. The “after” painting
group were faster in finding target locations and made fewer errors along the way.
“You-are-here” maps (YAH) are another aid to orientation. However, careful place-
ment and alignment with the environment is necessary for these to be effective. Judging
direction from a map is more accurate if the user is facing the same direction as the map
(Levine, Marchon, & Hanley, 1984). You-are-here maps are sometimes placed on a con-
venient wall, and so are reversed or inverted with respect to the surrounding environ-
ment. When you look at a YAH map, the top should correspond to in-front of you and the
bottom of the map behind-you. A survey of one city’s maps found they were correctly
aligned only 25 percent of the time. One mall had the same map mounted on all four
sides of a square column, which meant that only one was correctly oriented (Montello,
2010). This forces the user to make some cognitive rearrangements and leads to a greater
number of direction errors. Planners should choose the map location first and then design
the map, and different sites may require different versions of the map.

Developing Memory Skill

The study of memory experts offers some learnable skills that will enhance our own
memories. Individuals who are expert at remembering various types of material—num-
bers, maps, or scripts as examples—share a number of remembering techniques.
Skilled memory theory was developed to characterize the skill of those individuals
who learned to remember long strings of random numbers (Chase & Ericsson, 1981).
Three central features are postulated. First, during encoding, existing knowledge is used
to organize and make target items meaningful. For instance, random numbers might be
encoded as running times. Second, experts have well-developed retrieval routines. They
can recall the cues (e.g., a fast mile, or 3:55). Third, with practice, both encoding and
retrieval processes become faster.
Good map learners use several strategies during encoding. First, they partition the
map into smaller sections and focus attention on one area at a time. This is comparable to
grouping a string of numbers into smaller chunks. Good learners use visual imagery to
encode patterns and spatial relationships. Poor learners use more verbal rehearsal and
naming of components. Finally, the good learners engage in more self-testing to assess
whether they know the map yet (Thorndyke & Stasz, 1980).
Acting and Memory. If there is one group of people who should be skilled learners it
would be actors. They obviously need to remember dialogue, but also actions and emo-
tions. Studies by Noice and Noice (1997, 2006) show that actors do not learn lines by brute
memorization. Instead they use many strategies to encode material, such as distinctive-
ness, mood congruency, self-generation, and imaginal elaboration. The lines in a speech
would be organized into the successive reasons for the lines (to flatter, to warn, to reas-
sure). Maybe most importantly, the actors tried to feel what they were expressing.
Emotions, movements, expressions, and dialogue would be integrated into a meaningful
 Summary 327

unit. The behavior might be described less as acting and more as meaning. Students
could be taught this general idea, which Noice and Noice call “active experiencing.”
Undergraduates with no acting experience were trained in active experiencing. They were
able to remember more than students simply given instructions to memorize.

SUMMARY
Spatial Learning
Two types of spatial representation are considered. Route knowledge is knowledge of a
series of routes, directions, or paths in a spatial environment. By contrast, a cognitive
map is a more abstract representation of an environment, placing specific locations in
context with the surrounding area.
Place versus response-learning experiments attempted to determine whether rats
learned cognitive maps (Tolman) or specific turns (Hull) in mazes. The results were incon-
clusive with respect to the theories. Most organisms can learn about local cues to guide
specific turns (i.e., routes), and global cues to guide general orientation (i.e., cognitive
maps). It depends on which sorts of cues are available,
Cognitive mapping is shown by performance in the radial maze and the water maze.
In the radial maze, the participant must enter each goal arm once without repetition. In the
water maze, the subject learns the location of an unseen platform. In both mazes, sub-
jects rely on cues outside of the maze—room cues—to orient themselves with respect to
the goal location. Lesions of the hippocampus impair performance in these mazes, just as
hippocampal damage also impairs human declarative and spatial memory.
Landmarks are elements in an environment that by virtue of their distinctive design
or meaning stand out. Landmarks are central to either route or survey maps. Landmark
memory seems to be neurologically separable from route learning and cognitive
mapping.
Spatial schemas can lead to distortion during recall due to the averaging and normal-
izing that occurs when spatial knowledge becomes general. Incorrect inferences from
schematic spatial knowledge leads to errors in distance estimation, road-intersection
angles, and relative proximity of geographic locations. Spatial information can be organ-
ized to aid retention, just as was the case for verbal memory. For example, chimpanzees
organized a search route to retrieve food from a number of locations.
Cognitive mapping shows a developmental progression. Human infants learn turn
responses before place responses, and radial-maze performance increases from age 2 to
nearly adult levels of accuracy by age 5.

Motor-Skills Learning
Learning motor-skills is the acquisition of control over the amount, direction, and duration
of responding in reaction to regulating stimuli. The pursuit rotor and mirror tracing are
exemplars of laboratory motor tasks.
Not surprisingly, the amount of practice is an important determinant of skill. The
power law frequently describes the relationship between practice and the speed of per-
formance. The learning curve is a linear (straight line) increase in speed of performance,
when the data are plotted as a function of the log of the number of practice trials. That is,
328 Spatial, Motor-Skills, and Implicit Learning

each successive increment in performance requires greater amounts of practice. As with

verbal learning, spacing practice produces better learning than does massing trials.
A second important factor in skill learning is knowledge of results. Outcome informa-
tion or feedback on the success or accuracy of the response can be externally provided
to the participant. Consistent feedback (given after each trial) and immediate feedback
generally produce faster acquisition. However, partial-feedback schedules and delayed
feedback can produce better long-term retention, better self-guided performance, and
transfer of the skill to other motor responses. According to the guidance hypothesis,
occasional omitted or delayed feedback promotes attention to internal kinesthetic feed-
back, and thus learning to recognize good and poor performance.

Implicit Learning
Procedural knowledge is “knowing how” to perform a motor, perceptual, or cognitive act,
rather than the factual “knowing that” of declarative learning. Procedural learning is often
learned implicitly. Examples of implicit-learning tasks include artificial grammar, sequence
learning, and the Iowa Gambling Task. In these tasks knowledge of a complex environ-
ment is acquired largely independent of conscious awareness of specific components of
that environment. Different forms of implicit learning, such as priming and skill learning,
can be dissociated in certain neurological conditions such as Alzheimer’s or Huntington’s
disease.

Expertise
Although popular conceptions assert that innate talent underlies exceptional ability, an
alternative theory emphasizes the role of extensive practice. Expertise in many domains
comes after years of deliberate and intensive training. The 10,000 hour rule is one reliable
guideline for acquiring expertise.
John Anderson’s adaptive control of thought (ACT*) theory states that the acquisition
of a procedural skill is marked by a series of stages: acquiring factual knowledge during
the declarative stage; knowledge compilation, in which procedures are grouped into
units; and proceduralization, in which operations become speeded and automatic.

Applications
Implicit training has been applied to teaching skills to individuals with ongoing amnesia
(amnesic syndrome), persons who otherwise cannot acquire information explicitly. The
vanishing-cue technique is one example of implicit training.
Our knowledge of spatial learning can be applied to understanding sense of direction,
navigating new spatial environments, and in improving the design of buildings.
Long-term retention of skills can be improved by applying several “expert” guidelines:
optimize training conditions and strategies; train until retrieval is automatic; and optimize
retention conditions.
For instance, memory skills can be improved by partitioning material into smaller
units, establishing retrieval cues at the time of encoding, and self-testing to ensure that
learning has occurred.
 CHAPTER

12
Individual Differences in Learning and
Memory

CONTENTS

The Nature of Nurture: The Genetics Gender and Cognitive Abilities 345
of Learning Ability 330 Learning Styles 347
Animal Studies 330 The Visualizer–Verbalizer
Human Studies 332 Dimension 347
Age Differences in Learning and Kolb 348
Memory 333 Sternberg 348
Conditioning 333 Learning Styles? 349
Memory Development in Children 335 Self-Control 349
Aging and Memory 340 Social and Cultural Differences 350
Exceptional Memory: Epic Memories 351
The Mnemonists 342 Experimental Studies 351
Highly Superior Autobiographical Contemporary Cultural Psychology 352
Memory 343 Summary 355

One of the goals of a scientific psychology is to describe general laws. This textbook,
which includes the word principles in its title, attempts to specify some general principles
of learning and memory. There are exceptions to the principles, and individual differences
are one source of variation. The combination of genes, physical constitution, environ-
ment, and life experiences causes each of us to react somewhat differently to what are
seemingly similar situations. The remarkable thing, then, is that there are any general
principles at all.
It would be impossible to describe the laws for each and every individual. Instead,
researchers focus on broad categories of differences. How are children and adults alike or
different in their remembering? Are there personality or cultural differences that interact
with the principles of learning?
Two distinctive approaches to research in psychology are the experimental approach
and the correlational approach (Cronbach, 1957). The experimental approach focuses on
those variables that can be actively manipulated by researchers. The experimental approach
looks for general principles that transcend individual differences. The correlational approach

DOI: 10.4324/9781003227090-12
330 Individual Differences in Learning and Memory

focuses on variables that experimenters cannot control, such as personality, age, or the
environment, but that nevertheless are important determinants of learning in individuals.
The correlational approach notes the limiting conditions of the general principles. A com-
plete theory of learning should provide two kinds of knowledge: the general principles,
and the contributions of individual differences.
In some ways, what we are offering is a comparative psychology. As Robert Yerkes
said, “Comparative psychology in its completeness necessarily deals with the materials
of the psychology of the infant, child, adult, whether the being be human or infra-human;
of animal or plant (!)—of normal and abnormal individuals; of social groups and of civiliza-
tions” (cited by Cronbach, 1957).

THE NATURE OF NURTURE: THE GENETICS OF LEARNING ABILITY

The ability to learn (or the effect of nurture) is affected by the genetic makeup (or nature) of
the individual. Genetic factors could directly affect learning, for example, by affecting the
structure or functioning of the nervous system. Genetic influences also can indirectly affect
learning, because people differ in their sensitivity to various stimuli, their intensity of reac-
tion, or their level of emotionality, all factors that can influence learning and retention.

Animal Studies
The classic experiment on the genetics of learning ability was Tryon’s breeding of rats for
maze learning (e.g., Tryon, 1940). Tryon used the method of selective breeding, in which
animals of similar learning ability were bred together over several generations. This is the
same method used by animal breeders to produce specific physical traits. Tryon first
trained a large number of generic laboratory rats in a 17-turn maze. Some animals learned
quickly; some learned slowly. Even among rats, there are sizable individual differences.
Tryon selected the best maze learners and paired them for breeding. He also selected the
worst learners and allowed them to breed. This selection and breeding continued for
several generations.
After 18 generations of selective breeding, the two lines of descendants had diverged
into the Maze Bright (MB) and Maze Dull (MD) rats. There was almost no overlap in maze
performance between the MB and MD lines of rats. The MB were all good maze learners.
The MD were uniformly poor learners.
Rats, mice, and fruit flies have been bred for differences in specific learning abilities.
For instance, genetic mutations in fruit flies can be induced by exposure to certain chem-
icals. Tim Tully tested mutated flies in a shock-avoidance learning procedure (Tully, 1996).
The flies were trained in a T-maze made of Plexiglas tubes. An odor is introduced into one
arm of the T; after the flies enter that arm, they receive a mild electric shock. (Mild! This is
not a bug zapper). A different odor in the other arm of the T-maze is not shocked. Tulley
found, and subsequently bred, flies that could not learn this odor discrimination (so-called
Dunce flies). He also found flies that could learn but quickly forget (Amnesic flies).
Another method that derives from molecular genetics is to introduce altered DNA
into fertilized cells. Tang et al. (1999) produced strains of transgenic mice. The inserted
DNA altered cells in the hippocampus, an area we have frequently noted is involved in
 The Nature of Nurture: The Genetics of Learning Ability 331

learning. This particular strain of mice, labeled Doogie (after Doogie Howser – MD, the
character on TV who was a precocious child), did indeed learn better than normal mice.
What was especially nice about this study is that the researchers used a battery of learn-
ing tests, not just one task. The transgenic mice were better at detecting novel objects
added to their environment; better at learning the placement of the hidden platform in the
Morris water maze; and faster in acquisition of classical conditioning. This enhancement
of general learning ability is important in ruling out genes that simply increased attention,
emotionality, or activity levels.
To say that genes affect learning seems to imply that ability is inherited. However,
heredity combines with environment so that the outcome is better described as an inter-
action of genes and environment. For example, maze learning by MB and MD rats is
affected by the quality of social and environmental stimulation the animal experiences.
Cooper and Zubek (1958) compared MB and MD rats raised in three different environ-
ments. In the restricted condition, the rats were housed individually in small cages. In the
stimulating environment, several rats were housed together in a multilevel cage filled
with toys and objects to manipulate (like a hamster habitat). The “normal” environment
was something in between: a few rats living together, but with no toys. The animals were
later trained in the maze. The measure of learning was the number of errors in running the
maze, so in Figure 12.1 better performance is shown by fewer errors. The MB rats from
the normal living conditions were superior to the MD from the normal environment (see
Figure 12.1). This is what Tryon found. Rats reared in the deprived environment were very
poor learners, regardless of genetic background (i.e., MB or MD). Rats reared in the stim-
ulating environment were good learners regardless of genetic background. Thus, a bad
environment made all rats dull; a good environment made all rats bright.

The Biological Costs of Learning Ability

If learning conveys such an adaptive advantage, then why haven’t animals (including
humans) become incredibly smarter than they are? The experiments on selective breed-
ing and gene transplants have shown that smarter rats and mice can be created. (Just
what the world needs: smarter rodents). These findings show that there is room for
improvement. So why hasn’t evolution already produced better learners?
One answer is that there are metabolic costs associated with superior learning ability
(Hills & Hertwig, 2011). It takes energy to form new neural connections and to retain
memories in the brain. What evolution may have arrived at is an optimal balance among
several metabolic-intensive activities, such as learning, food seeking, predator avoidance,
disease resistance, etc. (Mery & Kawecki, 2004). Kawecki (2010) demonstrated this trade-
off by selectively breeding fruit flies to be better learners. Flies were subjected to taste
aversion learning. A novel flavor, either pineapple or orange, was paired with bitter tasting
quinine (bitter even to flies). The flies that learned the flavor-quinine association best were
selectively bred across 15 generations to produce better learners. However, superior
learning came at a cost. The smarter flies had fewer offspring that survived to maturity;
were less successful in competition for scarce food resources; and lived shorter lives
(Burger, Kolss, Pont, & Kawecki, 2008). In a parallel study, flies were bred to live longer,
but when tested in the taste aversion paradigm they were not as smart. That last point is
interesting: so far, flies have been bred to be better learners or to live longer, but not both.
332 Individual Differences in Learning and Memory

170

160

150 Dull

Mean Number of Errors

140

130 Bright

120

110

0
Restricted Normal Stimulating
Environment

Figure 12.1
Interaction of Rearing Environment and Genetic Strain. Number of maze errors made by Maze Bright and
Maze Dull rats as a function of the rearing environment: restricted, normal, or stimulating.
Source: From “Effects of Enriched and Restricted Early Environments on the Learning Ability of Bright and Dull Rats,” by R. M.
Cooper and J. P. Zubek, 1958, Canadian Journal of Psychology, 12, pp. 159–164. Copyright © 1958 by the Canadian Psychological
Association. Reprinted with permission.

Human Studies
Looking for genetic differences in learning among humans is more difficult than in animals.
Breeding history and environment are not experimentally controllable. However, genetic
influences are inferred from the results of methods such as the comparison of twins.
In the twins method, we compare sets of identical twins (or monozygotic twins,
referring to the production of two individuals from a single fertilized ovum) and sets of
same-sex fraternal twins (or dizygotic twins, referring to two fertilized ova). Basically, we
are asking whether identical twins are more alike in learning or memory than are fraternal
twins. For example, in one study of 10-year-old twins, the identical pairs were more alike
in the number of names and pictures recalled (the correlation was .43) than were fraternal
pairs (the correlation was .31) (Thompson, Detterman, & Plomin, 1991). At the other end
of the age spectrum, elderly identical twins (mean age of 67 years) were more similar in
the number of words they could recall and in memory for abstract figures (Finkel &
McGue, 1993). Results such as these are taken to suggest that memory has a hereditary
component.
The statistical estimates of the heritability of memory are often relatively small. This
may be because memory is treated as a single ability, instead of multiple types of mem-
ory. A better approach is to compare identical twins and fraternal twins on a variety of
 Age Differences in Learning and Memory 333

memory tests (Thapar, Petrill, & Thompson, 1994). The identical twins were more similar
on tests of associative memory, such as matching names and faces. Identical twins were
no more alike than were fraternals on tests requiring brief retention, such as the digit
span. Similarly, a twin study of episodic and semantic memory suggested that each had
a genetic component, yet episodic and semantic were independent of one another (Volk,
McDermott, Roediger, & Todd, 2006). Thus, rather than concluding that memory, broadly
speaking, is heritable, these results suggest there are genetic influences on certain kinds
of learning or remembering.
Even among genetically identical individuals, a given gene will not make a difference
unless it is expressed. The environment determines whether or when genes turn off or turn
on. Gene expression was demonstrated in a study in which mice were reared in enriched
or deprived environments like that described earlier. The researchers determined that the
enriched environment enhanced the expression of 100 genes (Rampon et al., 2000). We do
not yet know what functions these genes control. The enriched mice and deprived mice
developed differently because of how the environments affected their genes.

AGE DIFFERENCES IN LEARNING AND MEMORY

Changes in learning and memory across the lifetime have been studied experimentally
using the same basic methods otherwise described in this text. Learning, the acquisition
of information, has been studied using classical and instrumental conditioning proce-
dures. Memory, the retention of learned information, has been studied using episodic
procedures such as recognition and recall methods.

Conditioning
Conditioning techniques have been applied across the age spectrum. As a general princi-
ple, the very young and the very old do not condition as well as the age groups in between.

Conditioning in Infants
Research with human infants has demonstrated conditioning of a number of responses
such as eye-blink, head-turn, and sucking responses (see, e.g., Fitzgerald & Brackbill, 1976).
Newborns quickly learn scents associated with their mothers. Such learning is impor-
tant in helping to maintain mother–infant contact, particularly for neonates whose audi-
tory and visual systems function poorly at birth. In one study of newborns, a neutral
odorant was accompanied with stroking their torsos (Sullivan, Taborsky-Barbar, Mendoza,
Ition, & Leon, 1991). The experimental design included control groups of newborns who
were exposed to just the odor, just the stroking, or the two unpaired. Those infants who
had received pairings of the odor with bodily contact turned their heads more often in the
direction of the odor and were more active when the odor was presented.

Aging and Conditioning

At the other end of the age spectrum, classical conditioning begins a decline with
increased age. Solomon and Pendlebury (1992) studied eyeblink conditioning in both rab-
bits and humans, in each case pairing a tone and an airpuff. The rabbits ranged in age from
334 Individual Differences in Learning and Memory

six months to four years; the humans, 25 years to over 70. The younger subjects in each
species reached a higher level of responding to the CS, between 65 and 80 percent,
whereas the oldest subjects responded to only about 20 to 30 percent of the CSs. Human
conditioning, in general, begins to show a decline after the age of 40 or so.

The Generality of Conditioning

Individuals with intellectual or developmental disabilities learn classically conditioned
responses, often at about the same rate as nondisabled participants. In a study of eye-
blink conditioning, developmentally disabled young adults, as defined by IQ scores,
learned a conditioned response at the same rate as a sample of college students. Each
required about 25 CS–US pairings to produce conditioning. The two groups diverged dur-
ing extinction, when the disabled individuals kept responding longer than did the control
subjects (Lobb & Hardwick, 1976).
Down syndrome (DS) is a genetic condition in which there is an extra, or third, chro-
mosome where there should be a pair. The degree of intellectual impairment varies widely
in DS children and adults. Down syndrome is of research interest because the brain
pathology that develops in older DS individuals (in this case, over age 35) is similar to that
which develops in Alzheimer’s disease.
For example, Down syndrome individuals participated in a study of eyeblink condition-
ing (Woodruff-Pak, Papka, & Simon, 1994), in which a tone was paired with an airpuff. The
researchers had a portable computer control system that allowed them to take their labo-
ratory with them. The participants could therefore participate within a familiar environment.
The results of one study are shown in Figure 12.2, which shows the percentage of eyeblink

80

70

60

50
Percent CRs

40

30

20

10

0
<35 <35 DS >35 >35 DS Alzheimer’s
Groups by Age and Disorder

Figure 12.2
Mean Percent Conditioned Eyeblink Responses at the End of Training. Graph shows results for young
adults (under age 35), young adult Down syndrome, older adult and older adult Down syndrome (both over age
35), and Alzheimer’s patients.
Source: Woodruff-Pak et al. (1994, p. 20).
 Age Differences in Learning and Memory 335

CRs over the final block of trials. We can note three interesting findings in these data. First,
the younger subjects (those less than age 35) conditioned better than did the older sub-
jects (in this case over age 35). Second, there was less conditioning in each DS group than
in age-matched control groups (under or over age 35). And third, conditioning in the older
DS individuals did not differ statistically from that of elderly Alzheimer’s individuals.

Memory Development in Children

The capacity to remember develops from infancy, through childhood and adolescence,
and into young adulthood. The interest for us here is not that children seem to remember
less than adults, but why? As Flavel (1971) posed the question: “What is memory devel-
opment the development of?” Several answers have been proposed, ranging from
hypotheses that children have lesser short-term memory capacities to their having less
knowledge of how memory works. What we find is that much develops in children’s
memory.

Encoding Differences
Episodic memory is memory that we can access consciously and (usually) report verbally.
The difficulty with testing preverbal children is that memories might be acquired but the
children cannot tell us about them.
If memories are present then maybe episodic recall in infants can be tested by non-
verbal means. Bauer (e.g., 1996) has developed an elegant set of procedures in which
children as young as 11 months can demonstrate remembering by imitating actions they
observed the experimenter perform. The procedure is called elicited imitation. The exper-
imenter demonstrates an activity for the child, such as putting a teddy bear into a small
bed, pulling the covers up, and reading the bear a bedtime story. The child is then encour-
aged to repeat the sequence. The number of target actions that the child can reproduce
is a measure of immediate memory. Even 11-month-olds correctly imitate sequences of
two actions, and children of 24 to 30 months can repeat five or more actions. After a delay
ranging from a week to, in some cases, eight months, the child returns and is shown the
props used initially. Retention is demonstrated when these children spontaneously imi-
tate the previous sequences. Six-month-old infants remembered a sequence for a day
later; 10-month-olds remembered after three months; and 20-month-olds remembered
for a year (Bauer, 2006). Apparently, quite a bit can be remembered during the early years
of development.

Capacity
Another explanation for what differentiates younger from older children is the capacity of
their immediate memories. That is, maybe the short-term or working-memory capacity is
smaller in younger children.
Memory span, the number of items that can be retained after a single presentation,
is less in younger children. Children in first grade (about 6 to 7 years old) retain three or
four digits; sixth graders have a span of five to six digits; and adults have a span of seven
± two items (Engle & Marshall, 1983). Working-memory capacity, as assessed by requir-
ing simultaneous performance of two tasks, also increases with age.
336 Individual Differences in Learning and Memory

Knowledge
A general rule is that the more you know, the more new information you can learn. Maybe
older children remember more than younger children and infants because of the amount
of their preexisting knowledge.
This hypothesis predicts that younger children might remember more if they are
more knowledgeable about the to-be-remembered material. This was nicely demon-
strated by Chi (1978), who had 10-year-olds and adults remember arrangements of 20
chess pieces on a board or random strings of numbers. As expected, the adults had a
longer memory span for digits. These particular children were chosen for the study
because they were more knowledgeable about chess than were the adults. So, memory
spans completely reversed for chess memory. Here, the children remembered the cor-
rect positions of more chess pieces than did the adults.

Strategies
Retention increases with the use of various mnemonic strategies: rehearsing the to-be-re-
called material, forming mental images, grouping and organizing the items for later recall,
and so on. Younger children tend not use these strategies, and therefore do not recall as
well as older children.
In simple memory tasks, children of different ages are asked to remember a
series of picture triplets. Children aged four or five tend not to rehearse spontane-
ously, whereas older children, usually by age 8 or so, do rehearse (Locke & Fehr,
1971). (Rehearsal was monitored by watching their mouth movements). Teaching the
younger children to rehearse helps them to remember more (Keeney, Cannizzo, &
Flavell, 1967). With still longer lists and somewhat older children, more efficient
rehearsal strategies develop. Younger children repeat one item at a time, whereas
older children intermix rehearsal of several items, a pattern that increases recall (Elliott
et al., 2021).

Metamemory
Metamemory, or knowledge about memory, shows a developmental progression. It is
important in deciding what types of learning tasks will be difficult or when learning
strategies should be used. For example, rehearsal increases as children age, as does
their metamemory belief that rehearsal is beneficial. Older children can better assess
when they are ready to be tested (judgment of learning) than younger children (Kail,
1990). On the other hand, even very young children have accurate perceptions of when
memory might fail. Children were asked how they would remember to bring something
to school with them tomorrow (Kreutzer, Leonard, & Flavell, 1975). The children were
aware they might forget so they suggested using reminders, such as placing the item
near the front door or in their backpack, asking Alexa, or setting a reminder in their
tablet.
Although children are knowledgeable about memory in general, this knowledge is
inaccurate in detail. When asked how many items of a given kind they could recall, chil-
dren of most ages (including college students) overestimate. Practicing a memory test
does not eliminate these overestimates. Pressley and Ghatala (1989), in a study spanning
 Age Differences in Learning and Memory 337

grades 1 to 8, asked students for three estimates of their performance on a vocabulary

test: before a test, after a test, and for a future test. The children were shown a preview
of the vocabulary test. The items were chosen to produce about 50% correct. First, one
set of children were asked how well they thought they would do on the test. The results,
presented in Figure 12.3, show that all grade categories overestimated their subsequent
performance. Before the quiz, 60 to 90 percent of the kids predicted they would do better
than they actually did. (The average number of correct was 15 out of 30 for each age
group).
Second, after taking the test but before receiving any feedback, another set of chil-
dren rated how well they thought they did. Again, there was a fair number who overesti-
mated. But there was a decline from the before ratings, and the oldest grades made a
major correction in their estimates from before, from nearly 100 percent who overesti-
mated to less than 40 percent.
Finally, a third set of children predicted how they would do on another, similar test.
The older children benefited most from their poor first test performance and were able to
lower their estimates of future performance, adjusting their predictions so that fewer
eighth graders overestimated performance.

Percent Who Overestimate Quiz Scores

Before After Next Test

100

80
Percent of Children

60

40

20

0
Grades 1–2 Grades 4–5 Grades 7–8

Figure 12.3
Metamemory and Overestimates of Predicted Performance. Percentages of school children who
overestimated performance before taking a difficult vocabulary quiz, estimating performance on the quiz they
just took, or predicting how they will do on another quiz.
Source: Reprinted from “Metacognitive Benefits of Taking a Test for Children and Young Adolescents,” by M. Pressley and E. S.
Ghatala, 1989, Journal of Experimental Child Psychology, 47, pp. 430–450. Copyright © 1989, with permission from Elsevier.
338 Individual Differences in Learning and Memory

Autobiographical Memory
Autobiographical memory is our personal memory of events and experiences. It
includes episodic memories, and also semantic knowledge about ourselves (e.g.,
where we grew up, went to school), as well as schematic knowledge about some of
our repeated experiences (e.g., the general outline for family camping trips).
Autobiographical memory is most often assessed using verbal reports available from
older children and adults.
Autobiographical memory is thought to be related to the development of a sense of
self, or self-recognition. We cannot definitively say when a child develops a sense of self,
and thus also personal memory. Self-recognition is sometimes tested by the mirror test:
Do children recognize themselves in the mirror? If an ink mark is surreptitiously placed on
their foreheads, will the spot be noticed in the reflection? Two-year-old children usually do
not realize that the spot is on them, whereas by age 4 children do (Howe, 2000).
Language is an important factor in the development of autobiographical memory.
Language provides both words and schemas to organize memories for later retrieval.
Language also allows social interaction with parents and others to elaborate memories.
Talking about the past teaches children the narrative structure adults employ in remem-
bering our life stories. That is, parents’ discussion of the day’s events suggests to the child
the “who, what, where, when” format (Fivush & Nelson, 2004).
Finally, children need to develop a sense of the past and to learn labels for different
times in the past. As children develop, large scale time markers, such as SUMMER, are
supplemented by more precise markers, such as LAST WEEK.

Childhood Amnesia
When adults are asked to describe their earliest memory, episodic memories are conspic-
uously absent. The earliest childhood memories are typically from age 3 or so; occasion-
ally a memory from before this. This phenomenon, called childhood amnesia, has been of
interest since at least the time of Freud and is the subject of many theories. Is it really
possible that the experiences of our first years are not remembered? This phenomena is
discussed more in Box 12.1.

BOX 12.1 CHILDHOOD AMNESIA

When adults and older children are asked to recall their earliest memory, episodic memories
generally appear between the ages of 3 and 4. The number of episodic memories is conspic-
uously sparse. This phenomenon, called childhood amnesia (or infantile amnesia), has been
of interest since at least the time of Freud and is the subject of many theories. How can it
be that the novel, varied, and unexpected experiences of our first years are not
remembered?
Some claims for even earlier memories (events before age 2) are considered to be highly
improbable (Akhtar, Justice, Morrison, & Conway, 2018). People misremember early events,
misdate them, or forget the actual source. Some of these supposed memories are likely
created later from family stories, photographs, and videos. Some memories are merely frag-
ments or a moment in time. Memories are elaborated based on inferences from other
 Age Differences in Learning and Memory 339

knowledge, such as where they were living or what pet they had. When adults and older
children are asked for recollections that can objectively be dated, such as the birth of a sib-
ling, those memories fit a later time frame: generally between ages 3 and 4 (Eacott & Crawley,
1998).
Still, the number of episodic memories is conspicuously sparse. Other forms of memory
are present early in life: classical and instrumental conditioning; habituation and perceptual
learning. Procedural, implicit, and statistical learning occur in the acquisition of motor skills
and language.
A neurological explanation says that key areas of the brain, particularly the hippocampus,
have not yet matured in infants and so episodic memories cannot be formed (Nelson, 1995).
Another explanation is that childhood amnesia is a retrieval problem. The memories of pre-
verbal infants and toddlers are based on sensory and movement dimensions rather than on
words, and these former modes of encoding become inaccessible to retrieval when the
child does become verbal. Older children label the world in a different way. The new verbal
retrieval cues simply do not work in retrieving nonverbally encoded memories (Neisser,
1967). An analogy would be a new software program that cannot open files written in older
software.
Simcock and Hayne (2002) tested this possibility that, as children age, they remember but
cannot put those memories into words. In their study, 27-month-old children were shown a
series of actions performed with a variety of objects. (e.g., a toy is placed in a Magic Shrinking
Box, a crank is turned, a door is opened, and a smaller version of the toy is revealed). The
children’s expressive vocabulary did not contain the words to name and describe these
actions and objects. The children were tested one year later, after their vocabulary had devel-
oped sufficiently that they could have described the event. The children recognized the old
objects and could discriminate them from similar objects; and the children could re-enact the
sequence, a measure of the delayed imitation. Clearly the children remembered. However,
when asked to “describe the game we played…” the children did not use their later-devel-
oped vocabulary to describe the event.
The research on elicited imitation described earlier (Bauer, 2007) suggests that non-verbal
episodic memory does begin before age 2. Abundant research now demonstrates that within
the time-period eventually obscured by childhood amnesia, around ages 2-4, children do form
and retain autobiographical memories. As such, it makes clear that childhood amnesia among
adults cannot be explained by absence of autobiographical memory competence during
these years. In one study, conversations were recorded between mothers and their 3-year-
olds (Bauer & Larkina, 2014). Separate groups of children were then questioned three to six
years later. Children tested at ages 5 to 7 recalled 60 percent of the events discussed at
age 3. Children between ages 8 and 9 remembered less, about 40 percent. This shows that
the 3-year-olds were not amnesic then; children simply forgot memories over time, just like
the rest of us.
Another explanation for why episodic remembering seems to develop has to do with how
episodic memory is structured: as a story, a narrative. According to the social interaction
model (Nelson & Fivush, 2004), when parents go over an event with the child, they suggest
the form that memories might take. Questions point out relevant dimensions (who? where?
when?). Explanations are considered (why?). The episodic event is cast in a story form that
would guide the child in relating the story later, or as children age, how they will rehearse the
story. This social interaction aids encoding and subsequently retrieval.
340 Individual Differences in Learning and Memory

So, What Is It That Develops?

There does not seem to be a single reason why memory improves from infancy to adult-
hood. Several factors, including encoding differences, memory strategies, knowledge,
memory span, and metamemory, contribute. As we will see later (i.e., cross-cultural com-
parisons), age alone is not the only determiner of memory. Schooling also influences
metamemory, as does the use of rehearsal strategies, two factors that affect recall inde-
pendently of age.

Aging and Memory

Memory in older adults, typically referring to those over age 60, is not as good as in
younger adults. In parallel to our question about what develops in children, Perlmutter
(1978) asked: “What is memory aging the aging of?” The explanations for memory decline
are similar to those offered for the poorer memory found in young children. Thus, reduced
memory capacity, metamemory failures, encoding and/or retrieval difficulties are all pos-
sibilities. In addition, motivational factors become important.

Capacity
Tests of working memory show a decline with aging. For instance, there is increased
difficulty in doing two tasks at once, such as mentally adding pairs of numbers and
remembering the sums. This could indicate a reduction in working memory capacity (e.g.,
Charness, 1987). Others have suggested that there is a general cognitive slowing that
accompanies aging (Salthouse, 1994). This slowing would be manifested at each step in
a working-memory task: more time to encode the numbers, add the numbers, rehearse
the accumulating sums, and then recall them. Thus, even if memory capacity was
unchanged in the older adult, working memory still could be overtaxed by cognitive
slowing.

Metamemory
Older adults might have inaccurate beliefs about memory. Having been out of school for
a while, they have forgotten all those strategies they used when they were students.
Maybe the elderly have unrealistic expectations about what can be remembered.
Explanations such as these are not often supported by survey results. Perlmutter (1978)
did not find any age differences in self-reported use of memory strategies, whether these
were external aids such as lists and reminder notes, or internal methods such as rehears-
ing and organizing information. Knowledge about how memory works (e.g., that memory
is better for organized, concrete, or interrelated material) also did not vary with age.
However, the belief that memory declines with age could influence performance.
That is, memory decline could be a self-fulfilling prophecy. In one demonstration of this
age stereotype, both young and elderly female subjects read vignettes (case examples)
of everyday forgetting by people. The cases were accompanied by a picture of the person,
which could be of an older or younger person, as a way of introducing age into the manip-
ulation. The participants were asked to explain why forgetting had occurred. Participants
from both age groups attributed the memory failures in the older persons to declining
 Age Differences in Learning and Memory 341

memory ability. Memory failures in younger person examples were attributed to a lack of
effort: They could have remembered but they just didn’t try (Erber & Rothberg, 1991).

Aging and Self-Efficacy

Memory performance is affected by motivational factors. Some older adults fear that
forgetting indicates the onset of dementia. Instances of absentmindedness only
strengthen their beliefs in their cognitive fragility.
Self-efficacy is an important determinant of cognitive performance. Memory self-
efficacy refers to confidence in one’s ability to learn and perform competently in everyday
life. Self-efficacy is affected by everyday memory failures, as well as stereotypes of aging
and memory. Low self-efficacy can have several negative consequences, such as avoid-
ing tasks that depend on memory.
Touron and Hertzog (see Touron, 2015) demonstrated that elderly subjects working on
a task that could be done from memory would still avoid the memory strategy. The partic-
ipants were given a table of noun pairs (e.g., BLANKET-DOG; CAT-TREE; TABLE-
AARDVARK). Participants, younger and older people, were then shown word pairs and
were required to determine whether the pairings were correct or not. The instructions
stressed speed and accuracy. In the beginning the subjects needed to refer to the chart.
After doing this for a while, many of the pairs are remembered. This task encourages a
transition from the effortful strategy of looking up each new pairing, to the easier strategy
of remembering the answers. The younger subjects transitioned between these strate-
gies; the older adults on average persisted in using the scanning strategy: consult the list
to look up the correct pairing.
Interspersed recall tests of the correct pairings showed that the older participants
had learned them. Yet they persisted in using the looking-up strategy.
This memory avoidance carried over to everyday tasks. The older adults reported
less-frequent use of memory than did the younger adults in things like cooking or
wayfinding.

Noncognitive Sources of Decline

Recall the distinction between learning versus performance. Performance could be poor
even though the underlying learning or memory is not impaired. Aging is associated with
a number of variables that could affect performance on a test of memory. Older individu-
als may suffer poorer health, take more medications that affect cognition, or experience
mood impairments due to depression, grief, or life-style changes. General health prob-
lems are good predictors of everyday memory problems (Cutler & Grams, 1988), whereas
being healthier, educated, and intellectually active all predict better memory performance
(Arbuckle et al., 1992).
Aging is associated with a number of non-cognitive factors that can cause a decline
in memory performance. However, even when uniformly healthy elderly participants are
selected for research purposes, the major factor in describing memory decline is age
(West, Crook, & Barron, 1992). Holding other factors constant still left a deficit that corre-
lated with aging.
342 Individual Differences in Learning and Memory

Table 12.1 Sources of Memory Change With Age

Improves as Children Age Declines as Adult Age

Encoding • use of strategies (rehearsal) develops • less use of strategies
• switch to verbal and imagery coding • slower processing; need more
time to encode
Capacity • short-term memory span increases • working memory capacity
decreases
• multi-tasking declines
Knowledge • semantic memory increases • existing knowledge causes
• schemas are acquired interference
• schematic mis-recall
Meta-memory • inaccurate knowledge about memory • self-efficacy decreases
• aging stereotypes
Neuro- • hippocampus still developing • hippocampus starts to decline

Summary of Age and Memory

Memory in young children and in aging adults is affected by many of the same factors,
such as encoding, strategy use, or metamemory knowledge. A summary of these fac-
tors is presented in Table 12.1. In some cases, a factor has opposite effects. For
instance, toddlers do not have sufficient knowledge to make information meaningful.
Older adults may have knowledge and schemas which interfere with remembering new
experiences.

EXCEPTIONAL MEMORY: THE MNEMONISTS

The psychology literature contains examples of individuals with truly exceptional memory
abilities, some showing talents for remembering numbers or languages, others for mem-
orizing the Bible, the Koran, or the Iliad. Many such remarkable claims fall into the cate-
gory of anecdotes: interesting stories but of questionable scientific validity. However,
other cases include testing under controlled conditions, and here we can ask how good
can memory be, and what causes it to be exceptional?
One of the most remarkable examples of memory is a man referred to as S, studied
by the Russian neuropsychologist Luria (1968). (His full name was revealed later as
Solomon Shereshevsky). S could remember extraordinary amounts of verbal material.
Long lists of names, numbers, and words were presented once, and could be recalled
even years later. As examples, S memorized from a single presentation lines from Dante’s
Divine Comedy in its original Italian, a language with which S was unfamiliar. He also
memorized the following meaningless mathematical formula:

85 2762  86x 2 1624 2

N  d2   2
n b  sv r s
vx n v 264 322
 Exceptional Memory: The Mnemonists 343

When given a surprise test 15 years later (that’s right, 15 years), S was able to reproduce
both the Italian verse and the formula. S worked as a mnemonist in clubs, doing several
shows a night memorizing lists provided by the audiences. (Mnemonist is one name for
people with exceptional memories. Memorist has been offered as an alternative label). If
S had any problem with his memory, it was that he remembered too well. S occasionally
made mistakes by misrecalling lists from an earlier performance that evening. He had to
develop techniques to “erase” lists from his mental blackboard.
How did S remember so much? One factor was that he experienced a stimulus presented
in one sensory modality (visually or auditorily) in multiple sensory modalities. This is called
synesthesia, a condition of cross-modal perception. For S, spoken words evoked visual
images, smells, tastes, and tactile feelings. It is an inborn condition of the brain (Cytowic &
Wood, 1982). The lesson for us is that remembering is better for material presented to multiple
senses. S also used mnemonic strategies, such as the method of loci, or locations: To remem-
ber a list, he placed each object at a location during his mental walk down an imagined street.
Contemporary research has investigated differences between memorists and regular
rememberers (Maguire, Valentine, Wilding, & Kapur, 2003). The memorists had competed
at the top level of the World Memory Championships, which requires players to memo-
rize 57-line poems, columns of 25 unrelated words, and strings of 40 random digits. The
memorists were matched to control subjects on secondary variables. The memorists
were bright but not exceptionally so. The memorists were good at remembering verbal
material, which relates to their memory competition skill, but they were not significantly
better at remembering other things, such as faces or images of individual snowflakes.
(Snowflakes are difficult to encode verbally).
Brain scanning revealed two especially interesting findings. The more active brain
areas were those usually involved in spatial and navigation tasks and not particularly those
involved in verbal memory. Most of the memorists reported they had used some version
of the method of loci mnemonic (just like S). Thus, their world-class verbal memories
were attributed to the use of a spatial strategy.
Another interesting finding was the memory experts had less activity in certain brain
areas, compared to control subjects, that are essential to remembering. It seems that the
experts are so efficient at encoding that they require less cognitive (and neural) effort to
remember (Wagner et al., 2021).

Highly Superior Autobiographical Memory

History records a certain Reverend Dr. Phelps who claimed he could recall any day of the
preceding 60 years of his life. If given the date March 6, 1879, “In a few moments he will
state the day of the week. Then he will give you the weather for that day and describe
some particular thing that happened” (Neisser, 1982, p. 413).
In a contemporary example, a woman with exceptional personal or autobiographical
memory approached three memory psychologists (Parker, Cahill, & McGaugh, 2006). A.
J., in her 30s when the research began, can apparently remember every day of her life
since age 11. Give her a date, and she can tell you what she did that day. When asked to
recollect Easter Sundays, which fall on different dates each year, within a few minutes she
had produced a list of the past 24 years, listing the date and some personal recollection
344 Individual Differences in Learning and Memory

of the day. Two years later, she repeated this feat, noting the same personal events for
each year as noted in the first test. The memories are automatic, and not retrieved by
specific search or reconstructive strategies. Her recollections are personal: what hap-
pened to her, what she did, what she felt. Newsworthy events of the day are recollected
only if the event related to her life story.
On formal memory tests, A. J. scored well above average on some tests. Yet some
of her other memory capacities were below average, such as remembering word lists,
nonsense drawings, and recognizing faces she had seen earlier.
Not all aspects of A. J.’s ability were welcome. Her initial desire to contact the psycholo-
gists was partly due to the distress experienced by constantly remembering her past. She
describes being a prisoner of her memories. One recollection leads to another, then another,
and so on, as if she cannot stop chaining through these linked memories. (A. J. has described
her memory and life in a memoir, The Woman Who Can’t Forget, Price & Davis, 2008).
This exceptional memory ability is now called Highly Superior Autobiographical
Memory, or HSAM. Since A. J., McGaugh and his colleagues have identified several more
individuals who have these same detailed memories of their lives. In one study, eleven
HSAM individuals and matched control subjects completed a battery of memory tests
(LePort et al., 2012). The HSAM were clearly better than the normal memory controls in
personal memory and details of what happened on specific dates. This is basically what
defines HSAM. The HSAM subjects were mostly average on other memory measures:
digit span, paired associates learning, and free recall. Their superior memory ability was
domain specific.
The HSAM individuals differed from the controls on MRI scans. One difference was
in the parietal lobe of the brain. This area is involved in integrating information across
sensory modalities. (The same area that might underlie the synesthesia of the mnemon-
ist S). This area is also involved in binding or joining elements of episodic memory, such
as contextual stimuli of time and place, or the experienced emotion.
A summary of some comparisons of HSAM and World Memory Champions with
matching controls is shown in Table 12.2. The memory experts do not have superior mem-
ories overall. Memory champs remember things that use their practiced skills, such as
recalling strings of numbers or names matched to faces. But they are unexceptional with
memory for other types of information, such as recognizing whether a face was seen
earlier, or the sequence in which faces were shown.

Table 12.2 Standard Memory Tests in Which Memory Experts Remembered More Than Matched
Controls, or Remembered as Well as Controls

Memory experts exceed controls Memory experts same as controls

Digit Span HSAM and WMC Sequence (of names, faces) WMC
Story Recall HSAM and WMC Verbal Paired Associates HSAM
Recall Name-to-Face HSAM Recognizing Faces WMC
Autobiographical Memory HSAM Visual Memory (snowflakes, abstract design)
HSAM and WMC
HSAM = Highly Superior Autobiographical Memory; WMC = World Memory Championship participants.
 Gender and Cognitive Abilities 345

In contrast to the extraordinary memory of the HSAM, there may be people with just
the opposite trait: severely deficient autobiographical memory (SDAM). These individuals
have no personal episodic memories. They cannot recall life events that have the visual
imagery and emotional components of autobiographical memories. The SDAM are not
amnesic. In the real world they “know” what they did last week, they just cannot “remem-
ber” what they did. They reached adulthood without being aware that they were different
from others, or how they were different (Palombo et al., 2015).
The severely deficient memory subjects seem to get by on semantic knowledge. On
our standard laboratory tests of memory, such as recall and recognition, they show nor-
mal memory scores. They can remember a list by using verbal semantic knowledge:
rehearse the words, or organize the list. Their only deficit in standard tests is remember-
ing visual stimuli. Whereas visual is almost always remembered better than auditory stim-
uli by most people, the SDAM subjects remembered auditory and verbal better.

GENDER AND COGNITIVE ABILITIES

Do men and women differ in memory abilities? Most research has focused more broadly
on cognitive ability rather than specifically on memory. As examples, verbal ability is often
defined by performance on a specific test, such as the verbal SAT; spatial ability is often
measured as mental imagery, such as a test of mental rotation of objects (see Figure 12.5).
For many years the prevailing view was that men performed better on spatial tests and
women on verbal tests. However, gender differences in language and mathematics test
scores have been decreasing for several decades (e.g., Feingold, 1988).
More persistent differences have been found on spatial memory. In a study of actual
wayfinding, male and female college students were individually led through a wooded
area by two researchers. Periodically, the students were asked to point in the direction of
the starting location. Men were more accurate in identifying the starting point. At the end
of the trek, the subject was asked to show the way back. The men took shorter paths
back and were judged by the researchers as having selected a more direct route (Silverman
et al., 2000).
In responding to surveys, women reported more difficulties than did men in navigat-
ing, and women expressed higher levels of anxiety about unfamiliar spatial situations. For
example, women described more nervousness about finding their way back after making
a wrong turn, and finding their way out of a complex office building (Lawton, 1994).
Differences in performance do not necessarily mean differences in underlying ability.
Women’s anxiety about getting lost could impair actual spatial performance. Self-fulfilling
expectancies come into play when the instructions for a task emphasize its spatial nature.
For instance, women made more errors on the mental rotation test task (Figure 12.4)
when the standard instructions were given, i.e., “This is a test of your spatial abilities.”
When nonspatial instructions were given, men and women did not differ in their perfor-
mance (Sharps, Welton, & Price, 1993).
Studies of gender and verbal memory have more often shown a female advantage.
One study was able to control for a number of incidental variables by using a large and
representative sample of subjects (Herlitz, Nilsson, & Backman, 1997). The 1,000 partici-
pants were randomly selected from a region in Sweden, sampling adults aged 35 to 80.
346 Individual Differences in Learning and Memory

Figure 12.4
Mental Rotation Test of Spatial Cognitive Ability. A sample three-dimensional drawing of an object is shown
on the left. The task is to identify which two objects from the right are rotated versions of the sample object.
Source: From “Mental Rotation of Three-Dimensional Objects,” by R. N. Shepard & J. Metzler, 1971, Science, 171, pp. 701–703.
Reprinted with permission from AAAS.

Education, intelligence, and general health were assessed to statistically control for these
factors. A battery of verbal episodic memory tests was administered, including tests of
recall and recognition of word lists, sentences, new factual information, and names and
faces. There were also measures of short-term memory, and semantic memory was tested
via general knowledge and vocabulary. The women performed better, on the average, than
men at word recall and recognition, learning new facts, and recognizing newly learned
names and faces. These are all tests of episodic memory, and specifically, episodic verbal
memory. Men and women did not differ in semantic memory or in short-term memory.
A recent meta-analysis of 168 studies found a “small but robust female advantage”
in verbal episodic memory (Hirnstein, Steubs, Moèand, & Hausmann, 2022, p. 12). These
were mainly laboratory studies, but included a variety of specific tests to measure recall
and recognition. For instance, the California Verbal Learning Test measures free recall
learning across five trials, delayed recall, cued recall, and recognition memory.
There are a number of explanations for gender differences in learning and memory. A
social-cultural explanation posits there are differences in life experiences, education,
expectations, and/or stereotypes between boys and girls, and men and women. Such
differences begin early in life.
Gender can also influence how the same material is remembered differently. College
students listened to a list of words labeled either “Grocery List” or “Hardware Store List.”
The same words were in the two lists. Males recalled more of the words when given the
hardware list and women recalled more items from the grocery list (Herrmann, Crawford,
& Holdsworth, 1992). Similarly, in a study of prospective memory, children were asked to
remember to perform an action 30 minutes later. Boys were better at remembering to
check a battery charger (a stereotypically guy thing) and girls were better at remembering
to check the oven. The gender effects were larger among the 14-year-olds than among the
10-year-olds (Ceci & Bronfenbrenner, 1985).
Alternatively, men and women may differ in memory capabilities because of inherent
biological differences. One source of such differences may be hormones, such as estro-
gen or testosterone levels. For instance, adult women showed fluctuations across the
menstrual cycle on tests of spatial ability (Hampson & Kimura, 1988). This study did not
compare how women performed relative to men; only that women’s performance varied
 Learning Styles 347

across their cycle. In another study, women who received estrogen-replacement therapy
following a hysterectomy maintained their preoperative levels of performance on verbal
learning tasks (Sherwin, 1994). The estrogen did not cause an improvement in memory
but rather prevented a decline.
Corresponding research has been conducted using the male hormone testosterone.
In men higher levels of testosterone correlate with better spatial cognition. However,
most of the benefits seemed to be among older men, in which the testosterone is natu-
rally decreasing. The general conclusion seems to be that hormone supplementation
could improve memory in older adults (Hogervorst, DeJager, Budge, & Smith, 2004).
Another perspective on gender differences is an evolutionary approach. Possibly the
division of labor between males and females during primate evolution led to different
adaptive specializations. The spatial abilities required for hunting and stalking prey are
better developed in men (on average). Other spatial abilities required for foraging for edi-
ble plants (possibly analogous to object location) are better developed in women (Silverman
& Eals, 1992). The latter suggestion was confirmed in a study of object recall. As individual
participants arrived for their study, they were left in a waiting room for several minutes.
The participants were then ushered into the experiment room and told the real purpose
of the study: to recall everything they had seen in the waiting room. The women remem-
bered more of the objects than did the men, and the women were more accurate about
the exact locations of the objects.
As we have seen, gender differences can be addressed by theories ranging from
hormones to socialization. The point of mentioning these alternatives is to show the
uncertainty that underlies explanations of gender differences.

LEARNING STYLES
The phrase learning style refers to a way of categorizing individual differences in how
people learn. For instance, style is used to refer to left brain thinkers versus right brain
thinkers. An advertisement for a new car presents a table of facts for the left brain (the
verbal and analytic half) and a picture of a hot car for the right brain (the visual and emo-
tional half). Many theories and inventories of style have been developed but there is little
empirical work to validate the idea that people of different styles learn differently. A few
typologies, however, can be cited as examples of style theory.

The Visualizer–Verbalizer Dimension

One style dichotomy is a preference for verbal learning (via words, books, lectures) versus
visual learning (via pictures, illustration, or videos). Mayer and Massa (2003) gave college
students a battery of tests that were selected to measure both their preferred style, and
their performance on verbal and visual tasks. Visualizers scored higher on spatial ability
than did verbalizers, as measured by tests similar to the mental-rotation test (shown
earlier in Figure 12.5). Visualizers also had a preference for the visual versus verbal pres-
entation of information. In a computer-learning exercise on cloud formation, for example,
visualizers preferred help screens that elaborated on the material with drawings (maybe
water vapor rising to the sky), whereas the verbalizers preferred help screens that pro-
vided textual descriptions.
348 Individual Differences in Learning and Memory

Concrete Experience
Accommodator Diverger
Pragmatists Activists

Business History, English,

Major Psych Majors

Risk Takers Idea Generators

Intuitive Brainstorming
People-Oriented, and Emotional
Active Reflective
Experimentation Observation
Converger Assimilator
Theorists Reflectors

Engineering Major Math, Economics,

Sociology Majors

Quickly Seek Correct Solution Abstract Theories and Ideas

Unemotional Less Application-Oriented
Like Ideas Not People

Abstract Generalization

Figure 12.5
Kolb’s Two Dimensional Categorization of Learning Styles.
Source: Kolb (1984).

Kolb
Another conception of learning style is a two-dimensional organization offered by Kolb
(1984). One dimension describes a preference for learning from concrete experiences
(such as hands-on activities) versus a preference for learning in the abstract (such as
principles taught by lecture or text). The second dimension expresses a preference for
deriving knowledge through reflection (thinking as a source of generalizations) versus
active experimentation (trying out newly derived principles). As shown in Figure 12.6,
individual learning style is determined by a person’s location in this two-dimensional space
(based on Claxton & Murrell, 1987). However, there is little data to show that these indi-
viduals indeed learn differently. Style is solely based on self-report inventories, and the
theorist’s a priori choice of dimensions.

Sternberg
Another ambitious attempt to match student and instructional style has been reported by
Sternberg (e.g., Grigorenko & Sternberg, 1997). They proposed three cognitive styles:
analytical, creative, or practical thinking. As examples, analytic teaching involves asking
students to compare and contrast, evaluate, or critique an idea. The creative style asks
students to invent or answer “imagine/what if?” types of questions. Practical thinking is
taught by asking students to demonstrate how something could apply in the real world.
To implement this model in the classroom requires three elements to be matched: the
 Learning Styles 349

student’s learning style, the teacher’s instructional style, and a matching testing style.
This suggests an impediment to the use of style theories in educational settings, where
such matching cannot be practically accomplished.

Learning Styles?
Many of us believe we are especially effective learners in one modality or another. Are
those perceptions accurate and do they correspond with actual performance? Kratzig and
Arbuthnott (2006) compared three measures obtained from college students:

Self-labeled learning style (auditory, visual, or kinesthetic)

Learning style inventory results (auditory, visual, or kinesthetic)
Learning actual material that was presented auditorily, visually, or kinesthetically.

Students first categorized themselves as being primarily auditory, visual, or kinesthetic

learners. The students then completed a learning styles inventory to see if it matched the
students’ choices. Finally, the students took actual tests of auditory, visual, and kines-
thetic learning. For example, the students tried to remember words they heard (auditory)
or saw (visual). The results showed no correspondence among the several measures.
Fewer than half of the participants’ self-reported dominant learning style corresponded
with the inventory’s determination of the dominant style; and the inventory’s assignment
of style did not correspond to actual numbers of items remembered from tasks involving
verbal, drawn, and handled objects.
The proper design for research testing a learning style hypothesis needs to assign
some students to instructional format that matches their style, and other students to a
non-matching style. The results then need to show that performance is better among
students matched to their instructional style. The fact that learners state a preference for
a particular learning style does not prove the reality of these styles. Similarly, the fact that
some instructional methods are better than others says nothing about learning style.
Videos and graphics may be more effective teaching tools for all or most learners, not just
visual learners (Pashler, McDaniel, Rohrer, & Bjork, 2009).

Self-Control
If not learning styles, there are other factors that do characterize good academic perform-
ers. These are not so much learning styles as they are traits: positive personal attributes.
These traits include self-control, grit, and growth mindset.
The classic demonstration of self-control in children is the “marshmallow” experi-
ment conducted by Walter Mischel and his students (see Mischel, Shodan, & Rodriguez,
1989). Two- to four-year-old children were given a choice between receiving one marsh-
mallow now or a few marshmallows later. Self-control (S-C) was shown if the child
waited for the larger reward.
Self-control shows a developmental trend. For example, 3-year-olds choose the
immediate small reward, whereas 5-year-olds display more self-control (Logue, Forzano,
& Ackerman, 1996). Adolescents and adults are quite capable of exercising control and
make the choice associated with a delayed large reward, as long as the reinforcers are
points or tokens. When food is the reinforcer, even adults sometimes break down and
choose immediate reinforcement.
350 Individual Differences in Learning and Memory

Years after his initial study, Mischel decided to follow-up on the children who had
participated. He wondered whether early self-control, or lack thereof, related to later
development. It turns out that a number of long-term benefits accrued to the children
who developed self-control early. For instance, those children who had earlier shown
better S-C later had fewer behavior problems in school and higher SAT scores (Mischel,
Shodan, & Peake, 1988).
In the public’s mind, self-control is often referred to as willpower. However, by break-
ing S-C down to component processes, we have a better idea what it is. Self-control
controls attention: directing attention away from temptations and focusing on distrac-
tions. Emotional reactions are restrained, by distancing oneself from the competing
option (Baumeister & Tierney, 2011). Behavior is regulated by using principles of habit
learning and behavior modification.
Research from a number of disciplines suggests that self-control, and related con-
cepts such as self-discipline, conscientiousness, and the ability to delay gratification, are
important predictors of educational attainment, income, and even longevity (Roberts
et al., 2007; Tangney et al., 2004). A major community study followed 1,000 children in
New Zealand from birth to age 32. Measures of self-control that were obtained between
the ages of 3 and 11 had significant correlations with adult measures of health (such as
the absence of various diseases; substance abuse); wealth (income; socioeconomic sta-
tus; financial difficulties); positive social outcomes (higher levels of education); and fewer
negative social outcomes (e.g., criminal convictions) (Moffitt et al., 2011).
The concept of grit refers to the tendency to pursue long-term goals (Duckworth
et al., 2007). Passion and perseverance are two adjectives used to describe grit. Grit is
effort sustained over years, whereas self-control is resisting the hourly temptations
(Duckworth & Gross, 2014). The difference between working towards college graduation
versus making yourself study for tomorrow’s exam.
A growth mindset is an individual’s belief that intelligence and other personal charac-
teristics can be changed for the better. The capacity to learn, remember, solve problems,
etc., can be improved. In the contrasting fixed mindset, a person believes that intellectual
ability is fixed and unchanging (e.g., Dweck & Yeager, 2019). People attribute their suc-
cesses or failures to forces that are controllable or not. After a setback, the student with
growth mindset blames failure on lack of effort; the fixed mindset student blames failure
on lack of ability. Lack of effort can be remedied in the future; lack of intelligence cannot.
The importance of self-control, and the related concepts of grit and growth mindset,
have gained currency among researchers, social scientists, and educators. The answer to
the title of one popular book, How Children Succeed (Tough, 2012), is persistence. Where
the IQ score was once believed to be the best predictor of academic success, we may
now grant at least equal validity to self-control or self-discipline (Roberts et al., 2007).

SOCIAL AND CULTURAL DIFFERENCES

Different cultures emphasize different aspects of remembering. Verbal knowledge is
more important in some societies, whereas spatial or visual memory is more useful to
others. Is memory enhanced in societies that value remembering? To what extent do
schooling and education change the way we learn and use memory? These are questions
 Social and Cultural Differences 351

that cross-cultural studies of learning and memory can address. By studying other cul-
tures, we may discern universal aspects of memory and, conversely, how our own con-
ceptions of memory are shaped by the culture and society we live in.

Epic Memories
Back in high school you probably learned about the Greek poet Homer and his epic tales,
the Iliad and the Odyssey. Maybe you were stunned to hear that these stories were
repeated from memory, without a written version to consult. How do poets and perform-
ers do this?
Earlier in the last century, anthropologists began to study cultures whose literature
was primarily oral rather than written. In the 1930s, Lord studied the performance of
poets who sang folk tales in what was then Yugoslavia, where this oral tradition existed
(e.g., Lord, 1960). Songs sometimes ran to thousands of lines and would take hours to
perform. Experienced singers claimed to know 30 or more of these epics. New tales
were learned by listening to other performers, and singers claimed they could learn a new
song after hearing it once. Such claims reinforced a popular belief that memory is espe-
cially well developed in nonliterate societies. But had these accomplished performers
really memorized these poems?
Lord found that the singers had a different skill than what we are daily exposed to in
Western societies. Our singers do indeed memorize the words to relatively short songs.
The singers in the oral tradition, however, remembered the outline of the story and many
details, and recomposed the story for each performance. The singers knew formulas,
fillers, and different ways of phrasing the same idea. They would then select an appropri-
ate expression to fit the rhyme or meter. The same story might be related, yet each per-
formance would be somewhat different. Although the performers claimed to remember
“word for word,” this phrase has no correspondence in a culture without writing; that is,
what is a word? Instead, the performers were referring to the veracity of their recall. To
suggest that these folk performers do not remember the tales verbatim in no way dimin-
ishes their abilities. Instead, we now perceive that their talent is of a different sort: They
create anew each time that they perform.
This reconstructive aspect of memory is in accord with current conceptions of mem-
ory retrieval. Remembering is not so much the activation of a static trace stored in mem-
ory but instead is the reconstruction of the earlier experience.

Experimental Studies
Our theories assume there are certain features of memory, for example, separate short-
term and long-term memory systems. There are also memory strategies, such as organi-
zation or imagery. Are these universal features of memory, or are they due to schooling,
television, or other constants of Western societies? Cross cultural comparisons may help
answer such questions.
A modern classic in cross-cultural research is D. Wagner’s “Memories of Morocco”
(1978). One of his research goals was to see whether certain memory capacities were
universal, such as short-term memory. If so, then children with widely differing life expe-
riences should still be similar. A second goal was to determine whether certain memory
352 Individual Differences in Learning and Memory

capacities were culturally influenced. If so, the use of strategies such as rehearsal might
differ among same-aged children as a function of schooling or urbanization.
Wagner tested short-term memory for a sequence of seven cards, each having an
animal drawn on it. The cards were shown one at a time before being turned upside down
and placed in a row before the participant. A test card was then shown, and the child had
to indicate where that picture was in the list. This is a version of the memory games we
played as children. Children ranging in age from 6 to 19 were tested: children from an
urban and a rural area of Morocco, and children who attended school or were unschooled.
Correct recall of the test card’s location increased with age. This is not surprising, given
what we reviewed earlier about the development of memory. The new finding was that
schooled children, particularly adolescents and older, remembered more than did
unschooled children. This was especially so in remembering the first cards in the list.
Schooling was associated with a primacy effect, which most likely reflects the develop-
ment of rehearsal. Wagner’s results suggest that intentional learning strategies like
rehearsal are learned in school.
These results are supported by another series of studies, this time done in the Yucatan
Peninsula of Mexico (Sharp, Cole, & Lave, 1979). In a free-recall experiment, children and
adults were read a list of 20 common nouns from several categories (e.g., animals, cloth-
ing). Separate subject groups differed in age, education, and ethnic background. Adults and
adolescents with schooling recalled more of the words. But more importantly, the educated
adults (but not the uneducated ones) categorized items during recall. They clustered seman-
tically similar items at output (i.e., all the animal names, and then the tools, etc.). This study
shows that this strategy of organization can be a function of schooling and not just of age.
What about a popular notion that other cultures may excel at different forms of mem-
ory? One hypothesis is that people living in hunter-gatherer societies have better-devel-
oped visual or spatial memories, necessary for survival in the environments they occupy,
than people from industrialized societies. Kearins (1981) tested this idea by measuring
object location memory in Australian Aborigines and in Australians of European descent.
The participants were all school children, aged 12 to 16 years. The stimuli were objects
placed within a grid and shown briefly to the subjects. Each array contained between 12
and 20 items, either all man-made (e.g., thimble, eraser) or all natural objects (rock, leaf).
After 30 seconds of viewing the array, the experimenter heaped all the objects in a pile,
and the child then had to replace each item back in its original location. The Aboriginal
children excelled in this test of object location. In each variation of this test (e.g., natural
or man-made objects), the Aboriginal children remembered better than the children of
European descent. One striking finding was the number of times that the Aboriginal kids
had perfect scores: 75 percent were perfect on at least one of four trials and 41 percent
on two trials, whereas only 18 percent of the European children were perfect on one trial
and none on two.

Contemporary Cultural Psychology

The initial goals of cross-cultural research were to seek cognitive universals and specific
adaptations, such as the effects of schooling on rehearsal, or object memory in aboriginal
 Social and Cultural Differences 353

populations. We now realize that how we test memory affects how well memory appears
to function. If we ask people to display memory in a way that resembles what they do in
school (e.g., remembering lists of items in the sequence the experimenter presented
them), then people with schooling do better. The current trend in research is to investigate
cognitive skills as they relate to everyday behavior within a culture, rather than as
culture-free. Indeed, this trend is represented by a change from the label cross-cultural
psychology to simply cultural psychology (Rogoff & Chavajay, 1995).

Cultural Studies of Earliest Memories

In her book Battle Hymn of the Tiger Mother, Amy Chua (2011) described her Asian-
American style for educating her children: high expectations for academic performance;
no tolerance and no excuses for lesser performance; severe limits on play; and criticism
of American parenting. Professor Chua (a law professor) was criticized by some readers
for her caricature of Asian parenting, although this was done tongue in cheek and was
meant to describe only Chua’s parenting. Any stereotypes about styles are just that:
stereotypes. Just as one cannot say that all American parents are the same, we cannot
attribute sameness to any other group within a population. Nevertheless, there may be
some lessons in making cross-cultural comparisons.
One active topic of research has been on autobiographical memories in Western and
Eastern cultures. Western cultures (such as American or European) are said to emphasize
the self: self-esteem, personal autonomy, and individual success. Certain East Asian cul-
tures (e.g., China, Japan, or Korea) are said to place more emphasis on the group: the
family, community, or nation. Success is not personal but reflects on the group. Similarly,
failure reflects shame on the group.
These ideas have been applied to research on the earliest childhood memories. When
American students, the typical subjects in psychology experiments, are asked to recall
their earliest memories, they remember more personal events: what they did, what hap-
pened to them, and their feelings at the time. Asian students recall more social, group, or
interpersonal memories. They report family gatherings, ceremonies, or events; and the
reaction of the group (e.g., Ross & Wang, 2010).
The age of self-reported first memories are earlier in American subjects than in Asian-
American or native East Asian subjects. Figure 12.6 presents the age of first memories
from a sampling of studies (sources cited in Wang, March 2008). First memories of
Americans often date to around age 3.5 years, whereas earliest memories among East
Asians are after age 4.
Does this mean that Asian children develop episodic memories later? Not necessarily.
What likely differs is the retrievability of early memories. The wording of the recall request
affects the content of the remembered event. If the instructions are worded so that the
Asian subjects are primed to think of themselves as unique individuals, the personal
prime leads to earlier first memories (Ross & Wang, 2010).
The age of earliest memories also varies across individuals within cultures. Children
who receive more parental attention, such as a first child or an only child, report earlier
memories than later born siblings. As shown in Figure 12.6, this indeed was found in a
comparison of Chinese adults who had been only-children to those who had siblings
354 Individual Differences in Learning and Memory

Age first memory (months)

0 20 40 60

European American 39

Asian American 45

Korean 55

New Zealand European 43

New Zealand Maori 33

New Zealand Asian 58

Asian 50

Asian-primed 43

Israeli 37

Kibbutzim 50

Israeli first-born 40

Israeli later-born 44

Chinese - only child 39

Chinese siblings 48

Figure 12.6
Cross-National Comparisons of Age of First Memory. Adults were asked to report their earliest memories.
Shown are the average ages (in months) of the first memories.
Source: From data reported in Wang (2006).

(Wang, Leichtman, & White, 1998). The content also varied. Chinese only-children more
often mentioned personal early memories; Chinese siblings more often cited family.
In another case, children who had grown up in an Israeli Kibbutzim, where children
from several families are reared together, had later first memories than non-Kibbutzim
children. The Kibbutzim children likely had experienced a different social environment.
One difference among cultures is how children are socialized. For instance, conversations
that parents have with young children shape what eventually gets remembered. Western
parents may emphasize the individual self, with questions such as “What did you think?” or
“How did you feel?” Eastern culture mothers are more likely to explain to the children what
they should feel, or place the day’s events in a group context (Ross & Wang, 2010).
 Summary 355

SUMMARY
The experimental approach to learning focuses on variables that can be manipulated by
researchers, in order to determine the effects of such variables on learning. A correla-
tional approach focuses on variables that cannot be manipulated, such as gender or age,
but are nevertheless important determinants of learning and memory.

The Genetics of Learning Ability

Learning ability is affected by genetic predispositions. Tryon used selective breeding to
produce strains of Maze Bright and Maze Dull rats that differed in learning ability. Genetic
differences in learning have also been demonstrated by inducing genetic mutations in
fruit flies, and transplanting genes between mice. Environment plays an important role
even where there are genetic differences, as in the studies of deprived or enriched rear-
ing conditions.
In human studies, identical twins are more alike for some kinds of memory than are
fraternal twins. Even among genetically identical individuals, a given gene will not make
a difference unless the environment causes the gene to be expressed.
There are metabolic costs associated with increased learning ability. Evolution seeks
an optimal balance among the metabolic-intensive activities necessary for life.

Age Differences in Learning and Memory

Both learning and memory show developmental trends from infancy to old age.
Newborns and human infants learn conditioned responses such as eye-blink, head-turn,
sucking responses, and olfactory conditioning. Classical conditioning declines with
increasing age. Children and adults with intellectual or developmental disabilities also
learn classically conditioned responses, often at about the same rate as nondisabled
participants.
Episodic memory develops with age in children. The phenomenon of childhood amne-
sia may indicate an absence of memories from the first years of life, possibly due to
neural maturation, or simply an inability for adults to recollect those years. However, the
method of elicited imitation shows that infants as young as 11 months old can reproduce
specific sequences of activities.
What is it that develops as children age? Children and adults differ in their capacity of
short-term or working memory. Young children have a smaller knowledge base that limits
new learning. Younger children do not use strategies, such as rehearsal and elaboration,
as much as older children. Knowledge about the functioning of memory develops early in
children’s lives, but metamemory judgments are neither accurate nor sophisticated.
Some aspects of memory decline with aging. This may be due to slowing of mental
processing rather than a reduction in the amount that can be retained. Metamemory
beliefs are unchanged in the elderly. However, negative expectancies and stereotypes
affect the belief that one’s own memory is less competent. Memory performance may
decline because of other factors associated with aging, such as illness, depression, or the
side effects of medications.
356 Individual Differences in Learning and Memory

Exceptional Memory
The history of psychology has notable examples of mnemonists, people with exceptional
memories. Luria’s friend S employed strategies such as imagery, and mnemonics such as
the method of locations. His sensory experience was cross-modal, cutting across sen-
sory modalities, a property called synesthesia. Other mnemonists sometimes use verbal
elaboration, mentally coding new information into meaningful items.
Recent research has explored cases of exceptional memory by individuals who claim
they can remember nearly every day of their lives. People with Highly Superior
Autobiographical Memory do have exceptional personal or autobiographical memory.
Their other forms of memory are often unremarkable when assessed by standard tests
of short-term, episodic, or semantic memory.

Gender and Cognitive Abilities

The truism is that men are better at spatial and women at verbal tasks. Even when per-
formance does differ by gender, this does not necessarily indicate that this is due to
inherent biological differences between the sexes. Gender differences have been shown
to occur when there are differences in knowledge or experience, self-handicapping, or
gender expectancies or stereotypes about how one should perform.

Learning Styles
Several theories of learning styles have been developed, but there is little empirical work
to validate that, in fact, people of different styles learn differently. Well-developed style
theories include Kolb’s two dimensions of concrete versus abstract learning, and reflec-
tive versus active application of knowledge. Style theorists say that students’ learning
style should be matched to corresponding teaching and evaluation styles.
Research from a number of disciplines suggests that self-control, and related traits of
self-discipline, conscientiousness, or persistence, are important predictors of educational
success. Walter Mischel conducted follow-up studies on children from his initial “marsh-
mallow” experiments. He found that better self-control was associated with fewer behav-
ior problems later in school and higher SAT scores.
Other measures of childhood self-control correlate with adult measures of health,
wealth, educational attainment, and (negatively) with criminal convictions.

Social and Cultural Differences

Cross-cultural studies of learning and memory can address questions about universal
features of memory; different demands placed on memory by different environments; or
specific effects of social changes such as schooling. Older descriptions of memory in
nonliterate societies, while impressive, do not often produce verifiable data. Newer
research provides better control and more nuanced findings. Singers in an oral tradition,
for example, remember the outline and general features of an epic story, but otherwise
recompose the story during each performance. Experiments in Morocco and Mexico
show that recall increases with age, but that schooling enhances memory even more so
through the development of strategies such as rehearsal.
 Summary 357

Cross-cultural studies of our earliest memories suggest that adults of Western cul-
tures have earlier childhood recollections that do adults from Eastern cultures. The source
of this difference is likely the emphasis on individualism in Western cultures, and on the
social unit in Eastern cultures. It may also be due to how toddlers are socialized to report
memories during those early parent-child conversations.
GLOSSARY

accessible memory information that can be remembered. Not all memories are acces-
sible, even if they are present (available).
acquisition learning or encoding; acquiring knowledge or new behaviors. Acquisition is
usually measured as a function of study, practice, or learning trials.
amnesia forgetting due to an impairment in the retention or retrieval of memories that
once were available; or in the formation of memories which normally would have
been encoded.
amnesic syndrome a type of amnesia characterized by the inability to form or encode
new episodic and semantic long-­term memories. Short-­term memory, implicit mem-
ory, and procedural learning are usually not affected. Amnesic syndrome is caused
mainly by damage to the hippocampus, and possibly other areas of the brain.
anterograde amnesia the inability to form new memories or acquire new knowledge
since the onset of the amnesia. (Not remembering events that occurred before the
onset of amnesia is retrograde).
applied research the application of principles of learning and memory to solving practi-
cal problems. Applied research is often conducted in field settings using ecologically
realistic tasks, situations, and subject populations.
approach–avoidance conflict behavioral vacillation and uncertainty in a learning situa-
tion in which a response has both positive and negative consequences, such as food
and shock.
arousal a state of heightened mental or physiological attention, readiness, or emotion.
Arousal can refer to a psychological state of alertness; to a physiological state
assessed by measures of heart rate or the skin conductance response (SCR); or to a
trait, such as anxiousness or impulsivity.
avoidance learning a contingency, or rule, stating that performance of an instrumental
response prevents the occurrence of the aversive consequence. Avoidance is also
called negative reinforcement: negative because the response removes or prevents
the aversive outcome; reinforcement because the instrumental response increases
in frequency.
basic research has as its purpose an understanding of the fundamental processes of
learning and memory. In order to demonstrate cause-­and-­effect relationships between
 Glossary 359

key variables, basic research uses simplified tasks and situations, in which extrane-
ous variables are eliminated or controlled.
behavior modification applies the principles of operant learning to changing behaviors
in a variety of applied settings. Also known simply as behavior mod.
behavioral approach describes learning in terms of the relationship among antecedent
stimuli that precede behavior, the behavior itself, and the consequences that follow
behavior. The behavioral approach emphasizes observable stimuli, responses, and
outcomes.
blocking effect conditioning to a new CS is blocked when it is presented in compound
with another CS that already signals the US.
Brown–Peterson distractor task used to test short-­ term memory. The to-­ be-­
remembered item (often three letters or three words) is presented; a distracting task
(such as counting backwards) is performed for up to 30 seconds to prevent rehearsal;
and then recall is attempted.
central executive the working memory component that roughly corresponds with con-
scious attention. The central executive focuses, allocates, and distributes attention
across the other memory stores and across multiple tasks.
classical conditioning a method developed by Pavlov for establishing an association
between two (or more) stimuli. A neutral stimulus (the conditioned stimulus) is paired
with a significant stimulus (the unconditioned stimulus). The development of a condi-
tioned response to the conditioned stimulus indicates that an association has been
learned.
cognitive approach knowledge is encoded, transformed, stored, and retrieved in the
mind and/or brain. The cognitive approach focuses on the internal representations of
knowledge and memory.
cognitive map the mental representation of a spatial environment is analogous to a
map, placing specific objects, places, and routes in context with the surrounding area
(see route learning).
cognitive training explicit practice on cognitive tasks with the intention of improving
overall cognitive (intellectual) functioning.
compensatory response when certain drugs are used as unconditioned stimuli, the
conditioned response may be the opposite of the unconditioned response. The
learned bodily response compensates for, or balances out, the effect of the drug.
compound CS two or more conditioned stimuli are presented simultaneously or in
close succession. For example, a tone and a light are presented together, fol-
lowed by the US (e.g., tone + light → food).
conditioned inhibition a type of classical conditioning which occurs when a CS is pre-
sented without the US, in a context in which the US is otherwise expected. The
inhibitory CS reliably signals that the absence of the US (in contrast to an excitatory
CS which reliably signals the presence of the US).
conditioned response (CR) the response to the conditioned stimulus that is learned
during classical conditioning.
conditioned stimulus (CS) or to-­be-­conditioned stimulus, that is paired with the uncon-
ditioned stimulus. The CS is usually a neutral stimulus to begin with (such as a tone).
360 Glossary

consolidation theory the theory that memory is initially stored in a temporary form in
the brain (a biological short-­term memory), and that over time the memory consoli-
dates into a more permanent form (a biological long-­term memory).
contiguity the principle which says that two events become associated when they
occur close together in time, place, or thought.
contingency one event is dependent on the occurrence of another. In instrumental
learning, the reinforcer is contingent upon the occurrence of the instrumental
response. In classical conditioning, the unconditioned stimulus is contingent on the
occurrence of the conditioned stimulus.
cued recall a cue or prompt is presented to aid recall of an item from memory.
forgetting curve the amount of studied material that can be remembered at different
times after learning. The course of forgetting over time. In Ebbinghaus’s experiments,
most forgetting occurred shortly after learning; forgetting continues at a slower rate
over longer retention intervals.
declarative memory, declarative knowledge memory or knowledge that can be con-
sciously recalled and reported. Declarative includes semantic memory and episodic
memory. Also known as explicit memory, because memory is explicitly questioned.
Declarative knowledge contrasts with procedural memory’s facilitated performance
of skilled behavior.
delay of reinforcement (delayed reward) a reinforcement schedule in which the
instrumental response is not reinforced immediately, but reinforcement is given after
a delay interval.
depth of processing the theory that remembering depends on the amount of cognitive
elaboration, or processing depth. Shallow processing leads to poor retention, whereas
deep processing that involves meaningful analysis and comprehension produces bet-
ter retention.
discrimination (a) in classical conditioning, one CS (the CS+) is paired with the US
whereas another CS (the CS−) is presented without the US. Discrimination
learning is shown when there is more responding to the CS+ than to the CS−.
(b) In instrumental conditioning, the instrumental response is reinforced in the pres-
ence of the discriminative stimulus (S+). This response is not reinforced in the
absence of the discriminative stimulus, or in the presence of S–, a stimulus signaling
that reinforcement is not available. Discrimination learning is shown when there is
more responding during S+ than during S–.
discriminative stimulus (S+ or SD) in instrumental learning, a stimulus that signals
the availability of reinforcement. Responses during the discriminative stimulus are
reinforced. Responses in the absence of the S+ are not reinforced.
dishabituation reinstatement of the orienting to a habituated stimulus by presentation
of a different stimulus, e.g., after habituating the OR to a tone, a light stimulus could
reinstate the response to the tone.
dissociation a research technique used to demonstrate that a treatment variable has
opposite effects on different memory systems, tasks, or measures. Dissociation is
used to differentiate among memory systems.
dopamine one of several neurotransmitters used by neurons to communicate. Dopine
seems to be involved in instrumental conditioning and reward learning.
 Glossary 361

drive a motivational need or desire to obtain a given reinforcer (such as food) or to ter-
minate an aversive reinforcer (such as shock).
drive reduction a theory which states that stimuli are reinforcing because they reduce
primary biological drives, such as hunger or pain.
dual coding an item is encoded into memory in two modalities or forms, e.g., an object
may be remembered both as an image and as the word for the object.
dual-­process theory two processes, habituation, and sensitization occur concurrently.
The magnitude of the orienting response is a combination of these two opposing
tendencies.
elaborative rehearsal type of processing in which there is active cognitive interaction
with, reflection on, or use of the to-­be-­remembered information. Corresponds to
deep processing in a depth-­of-­processing approach.
electroconvulsive therapy (ECT) the use of electroconvulsive shock (ECS) as a psychi-
atric therapy. It often produces amnesic side effects and so is of interest to memory
researchers.
empiricism the philosophical belief that the origin of knowledge is through
experience.
encoding the acquisition of information, or the formation of a memory trace. In the
memory stages approach, encoding is the first stage, followed by storage and
retrieval.
encoding–retrieval paradigm an experimental manipulation that varies the encoding
conditions in combination with the retrieval conditions, e.g., encoding-­and retrieval-­
conditions can match (studying and testing occur in the same room) or differ (testing
occurs in a different room). Examples include mood-­dependent remembering and
drug state-­dependent learning.
encoding specificity information is encoded into memory in the presence of specific
cues, either environmental, physiological, or mental. Retrieval is best when those
cues are present at the time of retrieval.
episodic buffer the component of working memory that integrates information across
the phonological and visual stores, and information entry and retrieval from long-­term
memory.
episodic memory one’s own, individual, personal memories. Episodic memories con-
tain temporal and contextual information about when and where the events occurred.
Some of our autobiographical memories are episodic (but some are semantic).
Episodic can be characterized as “I remember…”
epistemology the field of philosophy that focuses on how we come to have
knowledge.
evaluative conditioning an affectively neutral stimulus, often a word or a picture, is
paired with another word or picture that evokes a positive or negative emotional
reaction. The emotional tone of the neutral stimulus becomes positive or negative as
a function of this conditioning experience.
evolution the biological change in organisms over generations as they better adapt to
the environment to which they are exposed.
explicit memory task a test of memory that contains a direct request to recall some-
thing from episodic or semantic memory.
362 Glossary

exposure therapy psychotherapy treatment that uses repeated or prolonged exposure

to fear-­inducing stimuli as a means of reducing the fear. Stimulus exposure might
reduce fear through, for example, habituation or extinction.
extinction in classical conditioning, the CS is presented alone without the US; in instru-
mental conditioning, the previously reinforced response is no longer followed by rein-
forcement. In either case, the learned response diminishes, or extinguishes, across
extinction trials.
false memory the mistaken recall of an item, stimulus, or event that did not actually
occur. Often refers specifically to the Deese-­Roediger-­McDermott procedure for pro-
ducing false recall, in which lists of highly associated words are presented, and the
subject recalls an associated word that was not on the presented list.
feeling of knowing unable to recall an item or fact now (such as a name or a word), but
subjectively feeling sure the information is known and is present in memory. Related
to the tip-­of-­the-­tongue phenomenon.
flashbulb memory a particularly vivid memory for a surprising, emotional, and conse-
quential event. It is as if a picture of a moment in time is stored in memory.
free recall a method for studying learning and memory for lists of items. The items can
be recalled in any order, meaning recall is free or unconstrained (in contrast with serial
recall).
functional amnesia (also called psychogenic amnesia) a type of amnesia associated
with psychological trauma. Functional amnesia is retrograde: there is forgetting for
the past, or some part of the past, that preceded or coincided with the traumatic
event.
functional approach learning and memory are ways of adapting to the changes and
inconstancies in their environment. Learning is considered in light of survival, adapta-
tion, and evolution.
generalization (a) after classical conditioning to one CS, the conditioned response is
made to stimuli that are similar to the trained CS (i.e., conditioning generalizes to
similar stimuli). (b) In instrumental conditioning, the instrumental response occurs in
the presence of stimuli similar to the initially trained discriminative stimulus (i.e.,
responding generalizes to other discriminative stimuli).
habit a highly trained response or behavior that can be performed without conscious
intention or conscious direction. The response is elicited by conditioned stimuli, often
without conscious awareness.
habit slip the intrusion of a habit when an alternative behavior had been intended; or
the performance of a habitual behavior in the absence of intention to do so. Also
called action slip.
habituation the decrease in orienting (and other) reactions to a stimulus that is repeat-
edly presented in the absence of any consequent stimuli. Habituation is sometimes
used to refer to the procedure (frequent presentation of a stimulus) or to its effect
(decrease in responding).
hypermnesia enhanced remembering. Remembering increases over time, in contrast
to the forgetting that we expect to occur over time. Hypermnesia is usually demon-
strated when more is recalled across successive attempts at reproduction of the
studied material.
 Glossary 363

imagination inflation frequently imagining or thinking about some event or action

leads to an increase in the belief that the action or event actually happened.
implicit learning improvement in the performance of cognitive, motor, or perceptual
skill that develops with training, independent of conscious awareness of specific
details of the tasks.
implicit memory task a test to assesses the effects of prior experience indirectly, by
measuring performance on some task instead of requiring explicit recall. For exam-
ple, responding more quickly to a word that is primed in memory.
incentive motivation a theory that states that stimuli are reinforcing because they
excite a drive, need, or desire for the reinforcer.
incidental learning learning that occurs without intention to remember, but incidental
to other processing of the information.
instrumental conditioning an experimental contingency, or rule, stating that perfor-
mance of a designated instrumental response leads to the occurrence of a particular
consequence. Also known as goal-­directed behavior, operant learning, and trial-­and-­
error learning. Instrumental conditioning includes positive reinforcement (or reward
learning); negative reinforcement (or avoidance learning); and punishment.
instrumental response in instrumental conditioning, the occurrence of positive or neg-
ative reinforcement is made contingent on (is dependent on) performance of the
instrumental response. Change in the response (e.g., its size or frequency) is the goal
of conditioning. Also called operant response.
isolation effect an unusual item or an item presented in a distinctive manner and is
therefore especially remembered. Also known as the von Restorff effect.
knowledge of results feedback concerning the success or accuracy of a response that
is given to the participant during or after practice. This term usually occurs in the
context of motor-­skill learning.
Korsakoff’s syndrome a form of amnesia that is associated with prolonged alcohol
abuse and the thiamine vitamin deficiency that accompanies such use.
Forgetting can be both retrograde (loss of previously recalled memories) and
anterograde (impaired encoding of new memories).
latent learning knowledge or behavior that is not displayed in performance. Learning
occurred but the knowledge is hidden unless prompted by the right circumstances.
learned helplessness produced by an experimental contingency, or rule, in which there
is an explicit lack of contingency between response and an aversive outcome. There
is not an instrumental response that produces punishment, nor is there a response
that avoids it.
learning a relatively permanent change in behavior or behavioral repertoire that occurs
as a result of experience. Learning refers to the acquisition of knowledge or
behavior.
learning curve a graph showing the acquisition of learning, plotting practice or repeti-
tions along the X axis, and size or frequency of a response along the Y axis.
learning style categorizing individual differences in how people learn. This can refer to
the dominant sensory dimensions (visual, verbal, or tactile learning); or cognitive
style (left brain learner, rational and verbal, versus right brain learner, more visual and
intuitive).
364 Glossary

long-­term memory a memory storage system that retains information indefinitely, has
a virtually unlimited capacity, and stores information across various modalities (verbal,
spatial, emotional, etc.).
maintenance rehearsal type of processing characterized by passive repetition or recy-
cling of information in order to keep it available in short-­term memory. Corresponds
to shallow processing in depth-­of-­processing approach.
massed practice (massed repetition) when to-­be-­remembered material is repeated,
the second presentation occurs immediately after the first presentation (or shortly
after), rather than after a longer interval between the repetitions.
meaningfulness to-­be-­remembered information can vary in how well-­known, familiar,
significant, or relevant it is. Meaningfulness is experimentally defined by imagery,
familiarity, number of associations, or frequency of occurrence.
memory (a) the representation of information from past experiences; (b) learning refers
to the acquisition of knowledge or behavior, whereas memory refers to retention and
recall of knowledge or behavior.
memory span the capacity of short-­term memory is defined as the longest string of
items that can be immediately recalled, in correct order. The digit span is often used
to measure the capacity of STM.
mere exposure effect the development of a preference for a stimulus that occurs
through exposure to the stimulus, in the absence of rewards, problems, or other
tasks that engage active processing.
metamemory knowledge about learning and remembering in general, as opposed to
the specific knowledge and memories we have. Metamemory includes beliefs about
how learning and memory generally work, and our own self-­knowledge and expecta-
tions about encoding, storage, and retrieval in specific circumstances.
misinformation effect the presentation of incorrect information is remembered instead
of the previously studied correct information. In a misinformation design, correct
information is first presented, and then the misinformation is presented.
mnemonics techniques or strategies used to aid encoding and retrieval in everyday
memory. Mnemonics can range from simple devices, such as using an acronym to
cue the first letters of the to-­be-­remembered items; to elaborate coding systems
used to memorize large sets of information.
motor-­skill learning acquisition of precisely adjusted movements in which the amount,
direction, and duration of responding corresponds to variations in the regulating stimuli.
n-­back task used to assess working memory and executive functioning. A continuous
sequence of items (e.g., letters, numbers) is presented. As each item occurs, the
subject must decide whether that item matches the one presented n-­steps earlier,
e.g., if n = 2, does the current letter match the one that was two letters before.
neophobia fear of the new, is often used to label the avoidance of new tastes or foods.
neural network models mathematical and computer simulations of memory as a net-
work of hypothetical neurons. Neural network statistically simulate the changes in learn-
ing and memory that occur over trials or over time. Also called connectionist models.
neuroscience approach seeks to understand the underlying biological basis for learn-
ing and memory; the changes that occur in the nervous system during learning and
remembering.
 Glossary 365

omission an experimental contingency stating that performance of an instrumental

response will prevent a positive reinforcer. Omission should decrease the frequency
of the response.
operant learning B. F. Skinner’s version of instrumental learning. The organism’s
response (such as a rat’s bar press) operates on the environment to produce a certain
outcome (such as food).
organization grouping together items that are similar, related, or associated to facilitate
recall.
orienting response (OR) reactions to a novel stimulus. The OR is a composite of sev-
eral physiological and behavioral responses, including startle, arousal, and sense
receptor focusing.
paired-­associate learning a method in which pairs of items are presented for study,
labeled stimulus and response (abbreviated S and R). The task is to learn to recall the
response item that is paired with each stimulus.
partial reinforcement a reinforcement schedule in which the instrumental response is
reinforced some of the time, not every time.
partial reinforcement extinction effect extinction of an instrumental response is
slower if the response was originally conditioned with a partial reinforcement
schedule rather than with continuous reinforcement.
perceptual learning learning to perceive differences among stimuli that develops with
repeated exposure to the stimuli.
performance (a) measures of behavior or memory that are used to indicate whether
learning has occurred. (b) The organism’s actual behavior may not always be a reliable
indicator of what has been learned.
persistence the continued performance of a given instrumental response even though
the outcome is aversive (such as no reward, reduced reward, or punishment).
phobia a strong and excessive fear of a specific object or situation, e.g., fear of snakes
and spiders, heights, or speaking in public.
phonological loop working memory’s verbal short-­term store that is used in rehearsal,
verbal problem solving, and language processing.
positive reinforcement an experimental contingency stating that performance of an
instrumental response will lead to the occurrence of a particular consequence.
Positive means the response produces an outcome, and reinforcement means the
response increases in frequency.
Premack principle a principle of reinforcement that says a higher-­probability activity will
reinforce a lower-­probability activity; or a more-­preferred activity will reinforce a less-­
preferred activity.
prepared learning some types of learning occur easily, rapidly, and early in life, such as
taste aversions, phobias, and language. Evolutionary psychology suggests that organ-
isms have evolved to readily acquire knowledge that is essential to their survival.
primacy effect enhanced recall of the first items in a list relative to the middle items. A
sub-­component of the serial-­position effect.
priming a stimulus that has been recently experienced or has been recently activated
in memory by an associated cue. A priming effect is a facilitated response to the
primed stimulus.
366 Glossary

proactive interference items studied or presented earlier interfere with recall of the
current or more-­recently presented items.
procedural learning (procedural memory) our store of knowledge of how to do
things. This includes perceptual, motor, and cognitive skills. Procedural knowl-
edge can be characterized as “knowing how to.”
propositional learning the relationships among events (stimuli, responses, conse-
quences) can be stated as a series of facts or propositions. Propositional learning is
the acquisition of these facts through exposure to and interaction with events.
prospective memory remembering to do something at the appropriate time in the
future. Remembering to remember.
psychogenic amnesia see functional amnesia.
punishment an experimental contingency, or rule, stating that performance of an
instrumental response is followed by an aversive consequence. This contin-
gency should decrease the frequency of the response. In the sense used here,
punishment is more than just withholding reward (which is called omission)
but is instead the application of an aversive consequence.
recall test a test of learning or memory that requires the recall or reproduction of stud-
ied information or items.
recency effect enhanced recall of the last items in a list relative to the middle items. A
subcomponent of the serial-­position effect.
recognition test a test of learning or memory that presents previously studied items
along with distractors, to see if the studied item can be identified.
reconsolidation the retrieval of a consolidated long-­term memory returns it to a modi-
fiable state. At this time, the memory can be altered, and the changed memory is
reconsolidated back to long-­term memory. See consolidation theory.
reconstructive memory the view that remembering is the reconstruction of an event
on the basis of memory fragments, cues, and present knowledge. Memory recon-
struction contrasts with the view that remembering is simply recalling a static, stored
representation of the actual event.
recovered memory remembering something that was previously forgotten or unre-
called. Often used in the context of unremembered trauma.
rehearsal repeating or recycling to-­be-­remembered items, either vocally or mentally.
reinforcement in instrumental conditioning, the contingency or rule that relates perfor-
mance of the response to an outcome or consequence, which strengthens (rein-
forces) the likelihood of the response occurring in the future. See also positive
reinforcement, avoidance learning.
reinstatement a classical conditioning procedure that restores an extinguished
response. Following extinction of the conditioned response, presentation of the
unconditioned stimulus alone can reinstate the extinguished response.
relearning test a test of learning or memory in which previously studied material is
restudied. The amount of savings or ease of relearning is taken as a measure of
the strength of the original learning.
remembering versus knowing judgments about whether an item is clearly remem-
bered as having been experienced before, versus an item that seems familiar but
produces no specific memory.
 Glossary 367

remote associations in learning a list of items, associations may develop among items
that are not adjacent (i.e., not contiguous).
renewal a classical conditioning procedure that restores an extinguished response.
Following extinction of the conditioned response, exposure to the context in which
conditioning occurred can lead to recovery of the extinguished response.
Rescorla-­Wagner model a theory of classical conditioning. The amount of conditioning
to a CS depends on the degree to which the US is surprising, or unexpected. The
theory describes the changes in conditioning to individual CSs that occur when mul-
tiple CSs are presented.
resistance to extinction a response should diminish during extinction. However, a
given response may persist during extinction even though it is no longer rewarded.
Certain reinforcement conditions during training enhance resistance to extinction.
response–reinforcer learning theory an explanation of instrumental learning that
asserts learning is the formation of a connection between the instrumental
response and the reinforcer.
retrieval recalling, remembering, or reactivating a stored memory. In the stage approach
to describing memory, retrieval is the final stage, preceded by encoding and
storage.
retroactive interference current or recently presented items interfere with recall of
items studied earlier.
retrograde amnesia forgetting of events that occurred before the onset of the disorder.
Memories that were once available, or that otherwise would have been remembered,
are forgotten. Retrograde amnesia is not remembering events that occurred before
the onset of amnesia. Not remembering events that occur after the onset of amnesia
is anterograde.
route learning, route knowledge knowing a sequence of routes, directions, or paths
through a spatial environment. In contrast to a cognitive map, the representation of
route knowledge is sequential, point-­to-­point, and habit-­like (see cognitive map).
schedule of reinforcement the specific rule that relates the delivery of reward to
the timing or frequency of instrumental responding. In general terms, reinforce-
ment can be given after a certain number of instrumental responses (ratio sched-
ule) or a response that occurs after some specified amount of time (interval
schedule).
schema (schemas) outlines of general knowledge that are stored in semantic memory.
Schemas guide the perception, organization, and later recall of familiar events. Also
called scripts.
secondary reinforcer a neutral stimulus that has been paired with a primary reinforcer
acquires the capacity to reinforce on its own, e.g., a clicker that has been followed by
food can become a secondary reinforcer.
second-­order conditioning in first-­order conditioning, a CS is paired with a US. In
second-­order conditioning, a second CS is paired with the first CS. Second-­order
conditioning is demonstrated when the second CS evokes the conditioned
response.
self-control the capacity to inhibit an immediate response for reward, in preference for
a better reward at a later time.
368 Glossary

semantic memory our store of general knowledge; generic knowledge that most of us
share. Semantic memory is like dictionary or encyclopedic knowledge. Semantic can
be characterized as “I know that…”
semantic network the theory that knowledge is stored in semantic memory as a net-
work of interconnected associations, relationships, or pathways. These connections
vary in strength or distance that separate one item from another.
sensitization an increase in responsiveness due to repeated stimulus presentation,
and thus the opposite of habituation. Sensitization is likely to occur with intense stim-
uli and may reflect heightened arousal of the nervous system.
serial learning a method for studying learning and memory in which a list of items is
learned and reproduced in the same sequence in which the items were presented.
serial-­position effect in recalling a list of items, the first items are well remembered
(primacy) and the last items are well remembered (recency), relative to relatively
poorer recall of the middle items.
shaping to condition a behavior that is not currently in the organism’s repertoire, one
can reinforce a series of successive approximations to the desired response. Shaping
is used to train novel behaviors, or sequences of behaviors.
short-­term memory (STM) memory storage that is limited both in its duration and its
capacity. As assessed in laboratory tests, STM retains on the order of five to seven
items, for several seconds to less than a minute in the absence of rehearsal. Working
memory is a later extended version of STM.
skin conductance response (SCR) changes in electrical conductivity in the skin that
are associated with arousal or emotionality. SCRs are measured by electrodes placed
on the skin (the wrist or palm, for example).
Skinner box an operant conditioning chamber. A controlled environment designed for
the presentation of stimuli (such as a tone or light); recording responses, such as bar
pressing; and the administration of reinforcers such as food or shock.
social learning learning through observation or imitation of another’s experiences, or
through verbal communication, rather than through direct personal experience. Also
referred to as vicarious learning.
social reinforcement praise, attention, facial expressions, or physical contact given by
parents, peers, or others that act as reinforcers to enhance or decrease instrumental
behavior.
source memory remembering the true origin of a memory. Source memory is usually
used in reference to a failure to distinguish between what was actually experienced
and what was imagined, thought of, or heard about.
spaced practice (spaced repetition) when to-­be-­remembered material is repeated,
the second presentation occurs after a delay rather than immediately after the first
presentation. Also called distributed repetition.
spacing effect when to-­be-­remembered material is repeated, learning is better when
there is greater separation between the first and second presentations, than when
repetitions are massed.
spatial learning, spatial memory the acquisition, retention, and internal representa-
tion of spatial information. Spatial can vary along a number of dimensions, e.g., being
episodic, semantic, or procedural; short-­term or long-­term; implicit or explicit.
 Glossary 369

spontaneous recovery the return of a response that had diminished or ceased.

Following either habituation or extinction of a stimulus, the response may
spontaneously recover after a period of time without stimulus presentations.
spreading activation activation of a memory spreads to associated or connected items
in memory, allowing related knowledge to become activated. Often used in the con-
text of semantic networks and priming.
statistical learning associations, rules, or algorithms are acquired through extensive
experience, that reflect actual consistent patterns experienced in the world.
stimulus control learned responding is brought under the control of conditioned
stimuli (in classical conditioning) or discriminative stimuli (in instrumental
conditioning).
stimulus–response learning theory an explanation of instrumental learning that
asserts learning is the formation of a connection between the discriminative
stimulus and the instrumental response.
storage in the stage approach to describing memory, storage is the second stage,
preceded by encoding and followed by retrieval. Once something has been
encoded, it needs to be retained or stored in long-­term memory.
targeted memory reactivation a method to manipulate memory processing during sleep.
Sensory stimuli (e.g., music, background odor, etc.) are presented during the learning
session. Presentation of those stimuli during sleep reactivates the recently learned
material, leading to enhanced consolidation and subsequently, better retention.
taste-­aversion learning a taste is used as the conditioned stimulus that is paired with
an unconditioned stimulus that induces illness. The conditioned response is the
development of aversion or avoidance of the taste.
testing effect the benefit to learning of taking a preliminary test on recently studied
material rather than engaging in additional study.
trial-­and-­error learning solving a problem by trying various responses until one pro-
duces the desired outcome. Trial-­ and-­
error was the term used by Edward Lee
Thorndike for what later became known as instrumental conditioning and goal
directed learning.
two-­process theory the theory that avoidance learning is governed by classical and
instrumental conditioning. The first process conditions fear to the warning signal via
classical conditioning. In the second process, escape from the warning signal is rein-
forced with fear reduction via instrumental conditioning. Also known as the Watson–
Mowrer theory.
unconditioned response (UR) the response to the unconditioned stimulus. This
response is sometimes reflexive.
unconditioned stimulus (US) a stimulus that is significant prior to the onset of condi-
tioning (such as food or shock), and which usually elicits a substantial response or
emotion (the unconditioned response).
verbal learning (a) methods used to study variables that affect acquisition, retention,
and recall of verbal items (usually words). These methods include serial learning,
paired associate learning, and free recall. (b) The verbal learning approach that
assumed human learning was analogous to conditioning, and followed similar princi-
ples of associative learning.
370 Glossary

visual-­spatial sketchpad (or visuospatial sketchpad) working memory’s visual and

spatial short-­term memory store. The visual-­ spatial sketchpad is used for the
brief retention or manipulation of visual images and spatial information.
von Restorff effect an unusual item or an item presented in a distinctive manner is
especially remembered. See isolation effect.
working memory (WM) a system of short-­term memory that allows for both retention
and manipulation of information. Alan Baddeley proposed four components of WM.
Temporary storage occurs in either the phonological store or the visual-­spatial mem-
ory. The central executive allocates attention. The episodic memory buffer connects
working memory to long-­term memory.
Yerkes-­Dodson law the effect of arousal on performance follows an inverted U-­shaped
curve. Peak performance occurs at a moderate or intermediate level of arousal; per-
formance is poorer when arousal is too low or too high.
REFERENCES

Aarons, L. (1976). Sleep-assisted instruction. Psychological Bulletin, 83, 1–40.

Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned helplessness in humans: Critique and
reformulation. Journal of Abnormal Psychology, 87, 49–74.
Adams, C. D. (1982). Variations in the sensitivity of instrumental responding to reinforcer devaluation. Quarterly
Journal of Experimental Psychology, 34B, 77–98.
Adams, J. A. (1987). Historical review and appraisal of research on the learning, retention, and transfer of human
motor skills. Psychological Bulletin, 101, 41–74.
Ader, R., & Cohen, N. (1975). Behaviorally conditioned immunosuppression. Psychosomatic Medicine, 37,
333–340.
Ader, R., & Cohen, N. (1993). Psychoneuroimmunology: Conditioning and stress. Annual Review of Psychology,
44, 53–85.
Akhtar, A., Justice, L. V., Morrison, C. M., & Conway, M. A. (2018). Fictional first memories. Psychological
Science, 29, 1612–1619.
Alba, J. W., & Hasher, L. (1983). Is memory schematic? Psychological Bulletin, 93, 203–231.
Alberini, C. M. (2005). Mechanisms of memory stabilization: Are consolidation and reconsolidation similar or
distinct processes? Trends in Neurosciences, 28, 51–56.
Alessandri, S. M., Sullivan, M. W., & Lewis, M. (1990). Violation of expectancy and frustration in early infancy.
Developmental Psychology, 26, 738–744.
Amsel, A. (1962). Frustrative nonreward in partial reinforcement and discrimination learning: Some recent his-
tory and a theoretical extension. Psychological Review, 69, 306–328.
Anderson, J. R. (1983). The architecture of cognition. Cambridge: Harvard University Press.
Anderson, M. C., & Green, C. (2001). Suppressing unwanted memories by executive control. Nature, 410,
366–369.
Anderson, M. C., & Levy, B. J. (2009). Suppressing unwanted memories. Current Directions in Psychological
Science, 18, 189–194.
Anderson, R. C., & Pichert, J. W. (1978). Recall of previously unrecallable information following a shift in perspec-
tive. Journal of Verbal Learning and Verbal Behavior, 17, 1–12.
Aporta, C., & Higgs, E. (2005). Satellite culture: Global positioning systems, Inuit wayfinding, and the need for a
new account of technology. Current Anthropology, 46, 729–753.
Appleby, J. (2021, March 9). Ouch! Needle-phobic people scarred by so many images of Covid shots. Kaiser
Family Foundation website, news article.
Arbuckle, T. Y., Gold, D. P., Andres, D., Schwartzman, A., & Chaikelson, J. (1992). The role of psycho social con-
text, age, and intelligence in memory performance of older men. Psychology and Aging, 7, 25–36.
372 References

Arkes, H. R., & Harkness, A. R. (1980). Effect of making a diagnosis on subsequent recognition of symptoms.
Journal of Experimental Psychology: Human Learning and Memory, 6, 568–575.
Arnold, M. E., Petros, T. V., Beckwith, B. E., Coons, G., & Gorman, N. (1987). The effects of caffeine, impulsivity,
and sex on memory for word lists. Physiology and Behavior, 41, 25–30.
Arnst, C. (2003, August 21). I can’t remember. Businessweek, 42–44.
Arrigo, J. M., & Pezdek, K. (1997). Lessons from the study of psychogenic amnesia. Current Directions in
Psychological Science, 6, 148–152.
Ashcraft, M. H., & Krause, J. A. (2007). Working memory, math performance, and math anxiety. Psycho-nomic
Bulletin & Review, 14, 243–248.
Aspinwall, L. G., & Taylor, S. E. (1992). Modeling cognitive adaptation: A longitudinal investigation of the impact
of individual differences and coping on college adjustment and performance. Journal of Personality and Social
Psychology, 63, 989–1003.
Atkinson, R. C., & Raugh, M. R. (1975). An application of the mnemonic keywords method to the acquisition of
Russian vocabulary. Journal of Experimental Psychology: Human Learning and Memory, 1, 126–133.
Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: A proposed system and its control procedures. In
K. W. Spence & J. T. Spence (Eds.), The psychology of learning and motivation (Vol. 2, pp. 80–195). New York:
Academic Press.
Atkinson, R. C., & Shiffrin, R. M. (1971, August). The control of short-term memory. Scientific American,
152–161.
Au, J. et al. (2016) There is no convincing evidence that working memory training is NOT effective. Psychonomic
Bulletin & Review, 23, 331–337
Averill, J. R., & Rosenn, M. (1972). Vigilant and nonvigilant coping strategies and psychophysiological stress
reactions during anticipation of electric shock. Journal of Personality and Social Psychology, 23, 128–141.
Ayduk, O., & Kross, E. (2008). Enhancing the pace of recovery: Self-distanced analysis of negative experiences
reduces blood pressure reactivity. Psychological Science, 19, 229–231.
Ayllon, T. (1963). Intensive treatment of psychotic behavior by stimulus satiation and food reinforcement.
Behavior Research and Therapy, 1, 53–61.
Ayllon, T., & Azrin, N. H. (1964). Reinforcement and instructions with mental patients. Journal of the Experimental
Analysis of Behavior, 7, 327–331.
Baddeley, A. D. (1966). Short-term memory for word sequences as a function of acoustic, semantic, and formal
similarity. Quarterly Journal of Experimental Psychology, 18, 362–365.
Baddeley, A. D. (1978). The trouble with levels: A reexamination of Craik and Lockhart’s framework for memory
research. Psychological Review, 85, 139–152.
Baddeley, A. D. (1992a). Implicit memory and errorless learning. In L. R. Squire & N. Butters (Eds.),
Neuropsychology of memory (2nd ed., pp. 309–314). New York: Guilford Press.
Baddeley, A. D. (1992b). Working memory. Science, 255, 556–559.
Baddeley, A. D. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive
Sciences, 4, 417–423.
Baddeley, A. D., & Hitch, G. (1974). Working memory. In G. H. Bower (Ed.), The psychology of learning and
motivation (Vol. 8, pp. 47–89). New York: Academic Press.
Baddeley, A. D., & Longman, D. J. A. (1978). The influence of length and frequency of training session on the
rate of learning to type. Ergonomics, 21, 627–635.
Baddeley, A. D., Thomson, N., & Buchanan, M. (1975). Word length and the structure of short-term memory.
Journal of Verbal Learning and Verbal Behavior, 14, 575–589.
Baddeley, A. D., & Wilson, B. (1986). Amnesia, autobiographical memory, and confabulation. In D. C. Rubin (Ed.),
Autobiographical memory (pp. 225–252). New York: Cambridge University Press.
Badia, P. (1990). Memories in sleep: Old and new. In R. R. Bootzin, J. F. Kihlstrom, & D. L. Schacter (Eds.), Sleep
and cognition (pp. 67–76). Washington, DC: American Psychological Association.
 References 373

Baeyens, F., Eelen, P., & Crombez, G. (1995). Pavlovian associations are forever: On classical conditioning and
extinction. Journal of Psychophysiology, 9, 127–141.
Bahrick, H. P. (1979). Maintenance of knowledge: Questions about memory we forgot to ask. Journal of
Experimental Psychology: General, 108, 296–308.
Bahrick, H. P. (1984). Memory for people. In J. E. Harris & P. E. Morris (Eds.), Everyday memory, actions, and
absentmindedness (pp. 19–34). New York: Academic Press.
Bahrick, H. P., Bahrick, P. O., & Wittlinger, R. P. (1975). Fifty years of memory for names and faces: A cross-sec-
tional approach. Journal of Experimental Psychology: General, 104, 54–75.
Bahrick, H. P., & Hall, L. K. (1993). Long intervals between tests can yield hypermnesia: Comments on Wheeler
and Roediger. Psychological Science, 4, 206–208.
Bahrick, L. E., Parker, J. F., Fivush, R., & Levitt, M. (1998). The effects of stress on young children’s memory for
a natural disaster. Journal of Experimental Psychology: Applied, 4, 308–331.
Baird, R. B. (2003). Experts sometimes show more false recall than novices: A cost of knowing too much.
Learning and Individual Differences, 13, 349–355.
Balota, D. A., Duchek, J. M., & Paullin, R. (1989). Age-related differences in the impact of spacing, lag, and
retention interval. Psychology and Aging, 4, 3–9.
Balter, M. (2010). Did working memory spark creative culture? Science, 32, 160–163.
Banaji, M. R., & Crowder, R. G. (1989). The bankruptcy of everyday memory. American Psychologist, 44,
1185–1193.
Banbury, S. P., & Berry, D. C. (1997). Habituation and dishabituation to speech and office noise. Journal of
Experimental Psychology: Applied, 3, 181–195.
Banbury, S. P., Macken, W. J., Tremblay, S., & Jones, D. M. (2001). Auditory distraction and short-term memory:
Phenomena and practical implications. Human Factors, 43, 12–29.
Bandura, A. (1965). Influence of models’ reinforcement contingencies on the acquisition of imitative responses.
Journal of Personality and Social Psychology, 1, 589–595.
Barlow, D. H. (2000). Anxiety and its disorders: The nature and treatment of anxiety and panic. New York:
Guilford Press.
Bartlett, F. C. (1932, reprinted 1995). Remembering: A study in experimental and social psychology. New York:
Cambridge University Press.
Bauer, P. J. (1996). What do infants recall of their lives? Memory for specific events by one- to two-year-olds.
American Psychologist, 51, 29–41.
Bauer, P. J. (2006). Constructing a past in infancy: A neuro-developmental account. Trends in Cognitive Sciences,
10, 175–181.
Bauer, P. J., & Larkina, M. (2014). The onset of childhood amnesia in childhood: A prospective investigation of the
course and determinants of forgetting early-life events. Memory, 22, 907–924.
Baumeister, R. F., & Tierney, J. (2011). Willpower. New York: Penguin Press.
Bechara, A., Damasio, H., Tranel, D., & Damasio, A. R. (1997). Deciding advantageously before knowing the
advantageous strategy. Science, 275, 1293–1295.
Bechara, A., Tranel, D., Damasio, H., Adolphs, R., Rockland, C., & Damasio, A. R. (1995). Double dissociation of
conditioning and declarative knowledge relative to the amygdala and hippocampus in humans. Science, 269,
1115–1118.
Beilock, S. L., Rydell, R. J., & McConnell, A. R. (2007). Stereotype threat and working memory: Mechanisms,
alleviation, and spillover. Journal of Experimental Psychology: General, 136, 256–276.
Bellugi, U., Klima, E. S., & Siple, P. (1974–1975). Remembering in signs. Cognition, 3, 93–125.
Benjamin, M., McKeachie, W. J., Lin, Y.-G., & Holinger, D. P. (1981). Test anxiety: Deficits in information process-
ing. Journal of Educational Psychology, 73, 816–824.
Berns, G. S., Brooks, A. M., & Spivak, M. (2012). Functional MRI in awake unrestrained dogs. PLoS ONE, 7(5),
e38027.
374 References

Bernstein, I. L. (1991). Aversion conditioning in response to cancer and cancer treatment. Clinical Psychology
Review, 11, 185–191.
Berntsen, D. (1998). Voluntary and involuntary access to autobiographical memory. Memory, 6, 113–141.
Berntsen, D. (2010). The unbidden past: Involuntary autobiographical memories as a basic more of remember-
ing. Current Directions in Psychological Science, 19, 138–143.
Berrios, G. E., Markova, I. S., & Girala, N. (2000). Functional memory complaints: Hypochondria and disorgani-
zation. In G. E. Berrios & J. R. Hodges (Eds.), Memory disorders in psychiatric practice (pp. 384–399). Cambridge,
UK: Cambridge University Press.
Bevan, W., & Steger, J. A. (1971). Free recall and abstractness of stimuli. Science, 172, 597–599.
Biegler, P., & Vargas, P. (2016). Feeling is believing: Evaluative conditioning and the ethics of pharmaceutical
advertising. Journal of Bioethical Inquiry, 13, 271–279.
Birch, L. L., Gunder, L., Grimm-Thomas, K., & Laing, D. G. (1998). Infants’ consumption of a new food enhances
acceptance of similar foods. Appetite, 30, 283–295.
Birch, L. L., & Marlin, D. W. (1982). I don’t like it, I never tried it: Effects of exposure on two-year-old children’s
food preferences. Appetite, 3, 353–360.
Birnbaum, I. M., Parker, E. S., Hartley, J. T., & Noble, E. P. (1978). Alcohol and memory: Retrieval processes.
Journal of Verbal Learning and Verbal Behavior, 17, 325–335.
Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe &
A. P. Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 185–205). Cambridge, MA: The MIT Press.
Bjork, R. A., & Kroll, J. F. (2015). Desirable difficulties in vocabulary learning. American Journal of Psychology,
128, 241–252.
Blake, M. J. (1967). Time of day effects on performance in a range of tasks. Psychonomic Science, 9, 349–350.
Bloom, L. C., & Mudd, S. A. (1991). Depth of processing approach to face recognition: A test of two theories.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 556–565.
Bolles, R. C. (1970). Species-specific defense reactions and avoidance learning. Psychological Review, 77, 32–48.
Bolles, R. C. (1972). Reinforcement, expectancy, and learning. Psychological Review, 79, 394–409.
Boorstin, D. J. (1983). The discoverers. New York: Random House.
Booth, D. (1990). Learned role of tastes in eating motivation. In E. D. Capaldi & T. L. Powley (Eds.), Taste, expe-
rience, and feeding (pp. 179–194). Washington, DC: American Psychological Association.
Bornstein, M. H., Kessen, W., & Weiskopf, S. (1976). Color vision and hue categorization in young human infants.
Journal of Experimental Psychology: Human Perception and Performance, 2, 115–129.
Borresen, C. R., & Klingsporn, M. J. (1992). Some perceptual and cognitive factors in mirror tracing: Their limits.
Journal of General Psychology, 119, 365–384.
Bourne, L. E., & Archer, E. J. (1956). Time continuously on target as a function of distribution of practice. Journal
of Experimental Psychology, 51, 25–33.
Bouton, M. E., Westbrook, R. F., Corcoran, K. A., & Maren, S. (2006). Contextual and temporal modulation of
extinction: Behavioral and biological mechanisms. Biological Psychiatry, 60, 352–360.
Bovjberg, D. H., Redd, W. H., Maier, L. A., Holland, J. C., Lesko, L., Niedzwiecki, D.,… Hawkes, T. (1990).
Anticipatory immune suppression in women receiving cyclic chemotherapy for ovarian cancer. Journal of
Consulting and Clinical Psychology, 58, 153–157.
Bower, G. H. (1970). Analysis of a mnemonic device. American Scientist, 58, 496–510.
Bower, G. H. (1981). Mood and memory. American Psychologist, 36, 129–148.
Bower, G. H., & Clark, M. C. (1969). Narrative stories as mediators for serial learning. Psychonomic Science, 14,
181–182.
Bower, G. H., Clark, M. C., Lesgold, A. M., & Winzenz, D. (1969). Hierarchical retrieval schemes in recall of cat-
egorized word lists. Journal of Verbal Learning and Verbal Behavior, 8, 323–343.
Bower, G. H., & Hilgard, E. R. (1981). Theories of learning (5th ed.). Englewood Cliffs, NJ: Prentice Hall.
 References 375

Bower, G. H., & Karlin, M. B. (1974). Depth of processing pictures of faces and recognition memory. Journal of
Experimental Psychology, 103, 751–757.
Bower, G. H., Monteiro, K. P., & Gilligan, S. G. (1978). Emotional mood as a context of learning and recall. Journal
of Verbal Learning and Verbal Behavior, 17, 573–585.
Bowers, J. S., & Schacter, D. L. (1990). Implicit memory and test awareness. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 16, 404–416.
Brainerd, C. J., Forrest, T. J., Karibian, D., & Reyna, V. F. (2006). Development of the false-memory illusion.
Developmental Psychology, 42, 962–979.
Bransford, J. D., Nitsch, K. E., & Franks, J. J. (1977). Schooling and the facilitation of knowing. In R. C. Anderson,
R. J. Spiro, & W. E. Montague (Eds.), Schooling and the acquisition of knowledge (pp. 31–55). Hillsdale, NJ:
Erlbaum.
Breen, N., Caine, D., Colthart, M., Hendy, J., & Roberts, C. (2000). Towards an understanding of delusions of
misidentification: Four case studies. Mind & Language, 15, 74–100.
Breland, K., & Breland, M. (1961). The misbehavior of organisms. American Psychologist, 61, 681–684.
Brewer, J. B., Zhao, Z., Desmond, J. E., Glover, G. H., & Gabrieli, J. D. E. (1998). Making memories: Brain activity
that predicts how well visual experience will be remembered. Science, 281, 1185–1187.
Brewin, C. R., & Andrews, B. (1998). Recovered memories of trauma: Phenomenology and cognitive mecha-
nisms. Clinical Psychology Review, 18, 949–970.
Broadbent, D. E., Cooper, P. F., FitzGerald, P., & Parkes (1982). The Cognitive Failures Questionnaire (CFQ) and
its correlates. British Journal of Clinical Psychology, 21, 1–16.
Broadhurst, P. L. (1963). The science of animal behavior. Baltimore: Penguin.
Brooks, J. O., & Watkins, M. J. (1990). Further evidence of the intracity of the memory span. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 16, 1134–1141.
Brown, A. S. (2003). A review of the déjà vu experience. Psychological Bulletin, 129, 394–413.
Brown, A. S., & Murphy, D. R. (1989). Cryptoamnesia: Delineating inadvertent plagiarism. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 15, 432–442.
Brown, J. S. (1958). Some tests of the decay theory of immediate memory. Quarterly Journal of Experimental
Psychology, 10, 12–21.
Brown, J. S. (1969). Factors affecting self-punitive locomotor behavior. In B. A. Campbell & R. M. Church (Eds.),
Punishment and aversive behavior (pp. 467–514). New York: Appleton, Century, Crofts.
Brown, M. F., & Moore, J. A. (1997). In the dark II: Spatial choice when access to extrinsic spatial cues is elimi-
nated. Animal Learning & Behavior, 25, 335–346.
Brown, R. T., & Kulick, J. (1977). Flashbulb memories. Cognition, 5, 73–99.
Brown, R. T., & McNeil, D. (1966). The “tip of the tongue” phenomenon. Journal of Verbal Learning and Verbal
Behavior, 5, 325–337.
Bruce, V., & Young, A. (1986). Understanding face recognition. British Journal of Psychology, 77, 305–327.
Bruer, J. T. (1997). Education and the brain: A bridge too far. Educational Researcher, 26(8), 4–16.
Burger, J. M., Kolss, M., Pont, J., & Kawecki, T. J. (2008). Learning ability and longevity: A symmetrical evolution-
ary trade-off in Drosophila. Evolution, 62, 1294–1304.
Burgess, P. W. (2000). Strategy application disorder: The role of the frontal lobes in human multitasking.
Psychological Research, 63, 279–288.
Burgoyne, A. P., & Engle, R. W. (2020). Attention Control: A Cornerstone of Higher-Order Cognition Current
Directions in Psychological Science, 29, 624–630.
Burns, D. J. (1990). The generation effect: A test between single- and multifactor theories. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 16, 1060–1067.
Bushman, B. J., & Baumeister, R. F. (1998). Threatened egotism, narcissism, self-esteem, and direct and dis-
placed aggression: Does self-love or self-hate lead to violence? Journal of Personality and Social Psychology, 75,
219–239.
376 References

Butters, N., & Cermak, L. S. (1986). A case study of the forgetting of autobiographical knowledge: Implications
for the study of retrograde amnesia. In D. C. Rubin (Ed.), Autobiographical memory (pp. 253–272). New York:
Cambridge University Press.
Byrne, R. (1979). Memory for urban geography. Quarterly Journal of Experimental Psychology, 31, 147–154.
Cabeza, R., & Lennartson, E. R. (2005). False memory across languages: Implicit associative response vs. fuzzy
trace views. Memory, 13, 1–5.
Cameron, C. L., Cella, D., Herndon, J. E. II, Kornblith, A. B., Zuckerman, E., Henderson, E., … Holland, J. C.
(2001). Persistent symptoms among survivors of Hodgkin’s disease: An explanatory model based on classical
conditioning. Health Psychology, 20, 71–75.
Campitelli, G., & Gobet, F. (2011). Deliberate practice: Necessary but not sufficient. Current Directions in
Psychological Science, 20, 280–285.
Candland, D. K. (1993). Feral children and clever animals. New York: Oxford University Press.
Cann, A., & Ross, D. A. (1989). Olfactory stimuli as context cues in human memory. American Journal of
Psychology, 102, 91–102.
Capaldi, E. J. (1971). Memory and learning: A sequential viewpoint. In W. K. Honig & P. H. R. James (Eds.),
Animal memory (pp. 112–154). New York: Academic Press.
Capretta, P. J., & Berkun, M. M. (1962). Validity and reliability of certain measures of psychological stress.
Psychological Reports, 10, 875–878.
Caramazza, A., & Hillis, A. E. (1991). Lexical organization of nouns and verbs in the brain. Nature, 349,
788–790.
Ceci, J. S., & Bronfenbrenner, U. (1985). “Don’t forget to take the cupcakes out of the oven”: Prospective mem-
ory, strategic time-monitoring, and context. Child Development, 56, 152–164.
Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: A
review and quantitative synthesis. Psychological Bulletin, 132, 354–380.
Chan, J. C. K., McDermott, K. B., & Roediger, H. L., III. (2006). Retrieval-induced facilitation: Initially nontested
material can benefit from prior testing of related material. Journal of Experimental Psychology: General, 135,
553–571.
Chan, M. S., Jones, C. R., Jamieson, K. H., & Albarracin, D. (2017). Debunking: A meta-analysis of the psycho-
logical efficacy of messages countering misinformation. Psychological Science, 28, 1531–1546.
Chance, P. (1992, November). The rewards of learning. Phi Delta Kappan, 200–207.
Chapman, G. B. (1991). Trial order affects cue interaction in contingency judgement. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 17, 837–854.
Charness, N. (1987). Component processes in bridge bidding and novel problem-solving tasks. Canadian Journal
of Psychology, 41, 223–243.
Chase, W. G., & Ericsson, K. A. (1981). Skilled memory. In J. R. Anderson (Ed.), Cognitive skills and their acqui-
sition (pp. 141–180). Hillsdale, NJ: Erlbaum.
Chen, M. K., Lakshminarayanan, V., & Santos, L. R. (2006). How basic are behavioral biases? Evidence from
capuchin monkey trading behavior. Journal of Political Economy, 114, 517–537.
Cheyne, J. A., Carriere, J. S., & Smilek, D. (2006). Absent-mindedness: Lapses of conscious awareness and
everyday cognitive failures. Consciousness and Cognition, 15, 578–592.
Cheyne, J. A., Goyeche, J. R. M., & Walters, R. H. (1969). Attention, anxiety, and rules in resistance to extinction
in children. Journal of Experimental Child Psychology, 7, 231–244.
Chi, M. T. H. (1978). Knowledge structures and memory development. In R. Siegler (Ed.), Children’s thinking:
What develops? (pp. 73–96). Hillsdale, NJ: Erlbaum.
Chorover, S. L., & Schiller, P. H. (1965). Short-term retrograde amnesia in rats. Journal of Comparative and
Physiological Psychology, 59, 73–78.
Choy, Y., Fyer, A. J., & Lipsitz, J. D. (2007). Treatment of specific phobia in adults. Clinical Psychology Review, 27,
266–286.
 References 377

Christianson, S.-A., Loftus, E. F., Hoffman, H., & Loftus, G. R. (1991). Eye fixations and memory for emotional
events. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 693–701.
Christianson, S.-A., & Nilsson, L.-G. (1984). Functional amnesia as induced by a psychological trauma. Memory
& Cognition, 12, 142–155.
Chua, A. (2011). Battle hymn of the tiger mother. New York; Penguin.
Chung, P. H., & Byrne, M. D. (2008). Cue effectiveness in mitigating postcompletion errors in a routine proce-
dural task. International Journal of Human-Computer Studies, 6, 217–232.
Clark, D. M., & Teasdale, J. D. (1982). Diurnal variation in clinical depression and accessibility of positive and
negative experiences. Journal of Abnormal Psychology, 91, 87–95.
Claxton, C. S., & Murrell, P. H. (1987). Learning styles: Implications for improving education. Washington, DC:
Association for the Study of Higher Education.
Cohen, G. (1989). Memory in the real world. Hillsdale, NJ: Erlbaum.
Cohen, G., & Faulkner, D. (1986). Memory for proper names: Age differences in retrieval. British Journal of
Developmental Psychology, 4, 187–197.
Collins, A. M., & Loftus, E. F. (1975). A spreading activation theory of semantic processing. Psychological Review,
82, 407–428.
Collins, A. M., & Quillian, M. R. (1969). Retrieval time from semantic memory. Journal of Verbal Learning and
Verbal Behavior, 8, 240–247.
Conrad, R. (1958). Accuracy of recall using key set and telephone dial, and the effect of a prefix digit. Journal of
Applied Psychology, 42, 285–288.
Conrad, R. (1964). Acoustic confusions in immediate memory. British Journal of Psychology, 55, 75–83.
Conrad, R. (1970). Short-term memory processes in the deaf. British Journal of Psychology, 61, 179–195.
Conway, M. A., Cohen, G., & Stanhope, N. (1991). On the very long-term retention of knowledge acquired
through formal education: Twelve years of cognitive psychology. Journal of Experimental Psychology: General,
120, 395–409.
Conway, M. A., Gardiner, J. M., Perfect, T. J., Anderson, S. J., & Cohen, G. (1997). Changes in memory aware-
ness during learning: The acquisition of knowledge by psychology undergraduates. Journal of Experimental
Psychology: General, 126, 393–413.
Cook, M., & Mineka, S. (1990). Selective associations in the observational conditioning of fear in rhesus mon-
keys. Journal of Experimental Psychology: Animal Behavior Processes, 16, 372–389.
Coolidge, F. L., & Wynn, T. (2009). The rise of homo sapiens: The evolution of modern thinking. Chichester, UK:
Wiley-Blackwell.
Cools, R., & Robbins, T. W. (2004). Chemistry of the adaptive mind. Philosophical Transactions of the Royal
Society London A, 362, 2871–2888.
Cooney, J. B., & Swanson, H. L. (1990). Individual differences in memory for mathematical story problems:
Memory span and problem perception. Journal of Educational Psychology, 82, 570–577.
Cooper, C. J. (1964). Some relationships between paired-associates learning and foreign-language aptitude.
Journal of Educational Psychology, 55, 132–138.
Cooper, R. M., & Zubek, J. P. (1958). Effects of enriched and restricted early environments on the learning ability
of bright and dull rats. Canadian Journal of Psychology, 12, 159–164.
Corkin, S. (2002). What’s new with the patient H. M.? Nature Reviews: Neuroscience, 3, 153–160.
Corkin, S. (2013). Permanent present tense: The unforgettable life of the amnesic patient, H. M. New York: Basic
Books.
Cornell, E. H., & Heth, C. D. (1979). Response versus place learning by human infants. Journal of Experimental
Psychology: Human Learning and Memory, 5, 188–196.
Cornell, E. H., Heth, C. D., Kneubuhler, Y., & Sehgal, S. (1996). Serial position effects in children’s route reversal
errors: Implications for police search operations. Applied Cognitive Psychology, 10, 301–326.
378 References

Cowan, N. (2010). The magical mystery four: How is working memory capacity limited, and why? Current
Directions in Psychological Science, 19, 51–57.
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of
Verbal Learning and Verbal Behavior, 11, 671–684.
Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal
of Experimental Psychology: General, 104, 268–294.
Cronbach, L. J. (1957). The two disciplines of scientific psychology. American Psychologist, 12, 671–684.
Crowder, R. G. (1972). Visual and auditory memory. In J. F. Kavanagh & I. G. Mattingly (Eds.), Language by ear
and by eye (pp. 251–275). Cambridge, MA: The MIT Press.
Crowder, R. G. (1993). Short-term memory: Where do we stand? Memory & Cognition, 21, 142–145.
Crutcher, R. J., & Healy, A. F. (1989). Cognitive operations and the generation effect. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 15, 669–675.
Cuddy, E., & Reeves, R. V. (2014). Hitting kids: American parenting and physical punishment. Brookings Institution
Report. https://www.brookings.edu/research/hitting-kids-american-parenting-and-physical-punishment/
Cuddy, L. J., & Jacoby, L. L. (1982). When forgetting helps memory: An analysis of repetition effects. Journal of
Verbal Learning and Verbal Behavior, 21, 451–467.
Cutler, S. J., & Grams, A. E. (1988). Correlates of self-reported everyday memory problems. Journal of
Gerontology: Social Sciences, 43, S82–S90.
Cytowic, R. E., & Wood, F. B. (1982). Synesthesia 1: A review of major theories and their brain basis. Brain and
Cognition, 1, 23–35.
Dadds, M. R., Bovbjerg, D. H., Redd, W. H., & Cutmore, T. R. H. (1997). Imagery in human classical conditioning.
Psychological Bulletin, 122, 89–103.
Dallal, N. L., & Meck, W. H. (1990). Hierarchical structures: Chunking by food type facilitates spatial memory.
Journal of Experimental Psychology: Animal Behavior Processes, 16, 69–84.
Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal
Learning and Verbal Behavior, 19, 450–466.
Daneman, M., & Merikle, P. M. (1996). Working memory and language comprehension: A meta-analysis.
Psychonomic Bulletin & Review, 3, 422–433.
Dark, V. J., & Benbow, C. P. (1990). Enhanced problem solving translation and short-term memory: Components
of mathematical skill. Journal of Educational Psychology, 82, 420–429.
Dark, V. J., & Benbow, C. P. (1991). Differential enhancement of working memory with mathematical versus
verbal precocity. Journal of Educational Psychology, 83, 48–60.
Darwin, C. (1859). On the origin of species by means of natural selection, or the preservation of favoured races
in the struggle for life. London: John Murray (New York: Modern Library, 1967).
Daum, I., Channon, S., & Canavan, A. G. M. (1989). Classical conditioning in patients with severe memory
problems. Journal of Neurology, Neurosurgery, and Psychiatry, 52, 47–51.
Daum, I., & Schugens, M. M. (1996). On the cerebellum and classical conditioning. Current Directions in
Psychological Science, 2, 58–61.
Daum, I., Schugens, M. M., Ackerman, H., Lutzenberger, W., Dichgans, J., & Birbaumer, N. (1993). Classical
conditioning after cerebellar lesions in humans. Behavioral Neuroscience, 107, 748–756.
Davey, G. C. L. (1995). Preparedness and phobias: Specific evolved associations or a generalized expectancy
bias? Behavioral and Brain Sciences, 18, 289–325.
Davey, G. C. L., & Matchett, G. (1994). Unconditioned stimulus rehearsal and the retention and enhancement of
differential “fear” conditioning: Effects of trait and state anxiety. Journal of Abnormal Psychology, 103,
708–718.
Davies, G., & Robertson, N. (1993). Recognition memory for automobiles: A developmental study. Bulletin of
the Psychonomic Society, 31, 103–106.
Davis, H. P., & Bernstein, P. A. (1992). Age related changes in explicit and implicit memory. In L. Squire & N.
Butters (Eds.), The neuropsychology of memory (2nd ed., pp. 249–261). New York: Guilford Press.
 References 379

Davis, M. (1974). Sensitization of the rat startle response by noise. Journal of Comparative and Physiological
Psychology, 87, 571–581.
Davis, M. (2006). Neural systems involved in fear and anxiety measured with fear-potentiated startle. American
Psychologist, 61, 741–756.
Dawson, M. E., & Schell, A. M. (1987). Human autonomic and skeletal classical conditioning: The role of con-
scious cognitive factors. In G. C. L. Davey (Ed.), Cognitive processes and Pavlovian conditioning in humans (pp.
27–55). Chichester, UK: Wiley.
Dawson, M. E., Schell, A. M., & Banis, H. T. (1986). Greater resistance to extinction of electrodermal responses
conditioned to potentially phobic CSs: A noncognitive process? Psychophysiology, 23, 552–561.
Deese, J. (1959). On the prediction of occurrence of particular verbal intrusions in immediate memory. Journal
of Experimental Psychology, 58, 17–22.
De Houwer, J., Baeyens, F., & Field, A. P. (2005). Associative learning of likes and dislikes: Some current contro-
versies and possible ways forward. Cognition and Emotion, 19, 161–174.
Dekker, E., Pelser, H., & Groen, J. (1957). Conditioning as a cause of asthmatic attacks: A laboratory study.
Journal of Psychosomatic Research, 2, 97–108.
Delgado, M. R., Jou, R. L., LeDoux, J. E., & Phelps, E. A. (2009). Avoiding negative outcomes: Tracking the mech-
anisms of avoidance learning in humans during fear conditioning. Frontiers in Behavioral Neuroscience, 3, 1–9.
Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). California verbal learning test (manual). San Antonio,
TX: Psychological Corporation.
Dempster, F. N. (1988). The spacing effect: A case study in the failure to apply the results of psychological
research. American Psychologist, 43, 627–634.
Dempster, F. N. (1996). Distributing and managing the conditions of encoding and practice. In E. L. Bjork & R. A.
Bjork (Eds.), Memory (pp. 317–344). New York: Academic Press.
deQuervain, D. J., Roozendaal, B., & McGaugh, J. L. (1998). Stress and glucocorticoids impair retrieval of long-
term spatial memory. Nature, 394, 787–790.
Dess, N. K., & Chapman, C. D. (1998). “Humans and animals”? On saying what we mean. Psychological
Science, 9, 156–157.
Detterman, D. K. (1975). The von Restorff effect and induced amnesia: Production by manipulation of sound
intensity. Journal of Experimental Psychology: Human Learning and Memory, 1, 614–628.
de Wijk, R. A., Schab, F. R., & Cain, W. S. (1995). Odor identification. In F. R. Schab & R. G. Crowder (Eds.),
Memory for odors (pp. 21–37). Hillsdale, NJ: Erlbaum.
Diamond, A. (2013). Executive functions. Annual Review of Psychology, 64, 135–168.
Diamond, A., & Lee, K. (2011). Interventions shown to aid executive function development in children 4 to 12
years old. Science, 333, 959–964.
Dickinson, A. (1980). Contemporary animal learning theory. Cambridge: Cambridge University Press.
DiClemente, D. F., & Hantula, D. A. (2003). Applied behavioral economics and consumer choice. Journal of
Economic Psychology, 24, 589–602.
Diekelmann, S., & Born, J. (2010). The memory function of sleep. Nature Reviews Neuroscience, 11, 114–126.
Dienes, Z., & Berry, D. (1997). Implicit learning: Below the subjective threshold. Psychonomic Bulletin & Review,
4, 3–23.
Dog roams 840 miles, finds family after move. (1983, March 27). Charlotte Observer, p. 12A.
Dollard, J., & Miller, N. E. (1950). Personality and psychotherapy. New York: McGraw-Hill.
Domjan, M. (1976). Determinants of the enhancement of flavored-water intake by prior exposure. Journal of
Experimental Psychology: Animal Behavior Processes, 2, 17–27.
Domjan, M., & Purdy, J. E. (1995). Animal research in psychology: More than meets the eye of the general
psychology student. American Psychologist, 50, 496–503.
Duckworth, A. L., & Gross, J. J. (2014). Self-control and grit: Related but separable determinants of success.
Current Directions in Psychological Science, 23, 319–325.
380 References

Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007). Grit: Perseverance and passion for long-
term goals. Journal of Personality and Social Psychology, 92, 1087–1101.
Dulany, D. E. (1968). Awareness, rules, and propositional control: A confrontation with S—R behavior theory. In
T. R. Dixon & D. H. Horton (Eds.), Verbal behavior and general behavior theory (pp. 340–385). Englewood Cliffs,
NJ: Prentice Hall.
Dunlosky, J., Rawson, K. A., March, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students’ learning
with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological
Science in the Public Interest, 14, 4–58.
Dunning, D., Johnson, K., Ehrlinger, J., & Kruger, J. (2003). Why people fail to recognize their own incompe-
tence. Current Directions in Psychological Science, 12, 83–87.
Dweck, C. S., & Wagner, A. R. (1970). Situational cues and correlation between CS and US as determinants of
the conditioned emotional response. Psychonomic Science, 18, 145–147.
Dweck, C. S., & Yeager, D. S. (2019). Mindsets: A view from two eras. Perspectives on Psychological Science,
14, 481–496.
Dwyer, D. M., Mundy, M. E., & Honey, R. C. (2011). The role of stimulus comparison in human perceptual learn-
ing: Effects of distractor placement. Journal of Experimental Psychology: Animal Behavior Processes, 37,
300–307.
Eacott, M. J., & Crawley, R. A. (1998). The offset of childhood amnesia: Memory for events that occurred before
age 3. Journal of Experimental Psychology: General, 127, 22–33.
Ebbinghaus, H. (1902). Grundzüge der Psychologie (Figure 51, p. 646). Leipzig: Verlag Von Veit.
Ebbinghaus, H. (1964). Memory: A contribution to experimental psychology (H. A. Ruger & C. E. Bussenius,
Trans.). New York: Dover. (Original work published 1885).
Eich, E. (1985). Context, memory, and integrated item/context imagery. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 11, 764–770.
Eich, E. (1995). Mood as a mediator of place dependent memory. Journal of Experimental Psychology: General,
124, 293–308.
Eich, J. W., Weingartner, H., Stillman, R. C., & Gillin, J. C. (1975). State dependent accessibility of retrieval cues
in the retention of a categorized list. Journal of Verbal Learning and Verbal Behavior, 14, 408–417.
Eikelboom, R., & Stewart, J. (1982). Conditioning of drug-induced physiological responses. Psychological
Review, 89, 507–528.
Einstein, G. O., McDaniel, M. A., & Lackey, S. (1989). Bizarre imagery, interference, and distinctiveness. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 15, 137–146.
Eisenberger, R., & Cameron, J. (1996). Detrimental effects of reward: Reality or myth. American Psychologist,
51, 1153–1166.
Eisenberger, R., Heerdt, W. A., Hamdin, M., Zimet, S., & Bruckmeir, G. (1979). Transfer of persistence across
behaviors. Journal of Experimental Psychology: Human Learning and Memory, 5, 522–530.
Ellard, C. (2010). You are here: Why we can find our way to the moon, but get lost in the mall. New York:
Doubleday.
Ellis, N. C., & Hennelly, R. A. (1980). A bilingual word-length effect: Implications for intelligence testing and the
relative ease of mental calculation in Welsh and English. British Journal of Psychology, 71, 43–52.
Emmons, W. H., & Simon, C. W. (1956). The non-recall of material presented during sleep. American Journal of
Psychology, 69, 76–81.
Engen, T., Kuisma, J. E., & Eimas, P. D. (1973). Short-term memory of odors. Journal of Experimental Psychology,
99, 222–225.
Engle, R. W., & Marshall, K. (1983). Do developmental changes in digit span result from acquisition strategies?
Journal of Experimental Child Psychology, 36, 429–436.
Engle, R. W., Tuholski, S. W., Laughlin, J. E., & Conway, A. R. A. (1999). Working memory, short-term memory,
and general fluid intelligence. Journal of Experimental Psychology: General, 128, 309–331.
 References 381

Erber, J. T., & Rothberg, S. T. (1991). Here’s looking at you: The relative effects of age and attractiveness on
judgements about memory failure. Journal of Gerontology, 46, P116–P123.
Erdelyi, M. H., & Kleinbard, J. (1978). Has Ebbinghaus decayed with time? The growth of recall (hypermnesia)
over days. Journal of Experimental Psychology: Human Learning and Memory, 4, 275–289.
Ericsson, K. A., & Charness, N. (1994). Expert performance: Its structure and acquisition. American Psychologist,
49, 725–747.
Ericsson, K. A., Chase, W. G., & Faloon, S. (1980). Acquisition of a memory skill. Science, 208, 1181–1182.
Ericsson, K. A., Krampe, R. T., & Tesch-Romer, C. (1993). The role of deliberate practice in the acquisition of
expert performance. Psychological Review, 100, 363–406.
Erikson, G. C., Hager, L. B., Houseworth, C., Dungan, J., Petros, T., & Beckwith, B. E. (1985). The effects of caf-
feine on memory for word lists. Physiology and Behavior, 35, 47–51.
Estes, W. K. (1944). An experimental study of punishment. Psychological Monographs, 57 (3, No. 263), 1–40.
Estes, W. K. (1972). An associative basis for coding and organization in memory. In A. W. Melton & E. Martin
(Eds.), Coding processes in human memory (pp. 161–190). Washington, DC: V. H. Winston.
Evans, G. W., Fellows, J., Zorn, M., & Doty, K. (1980). Cognitive mapping and architecture. Journal of Applied
Psychology, 65, 474–478.
Evans, G. W., & Johnson, D. (2000). Stress and open-office noise. Journal of Applied Psychology, 85, 779–783.
Evans, H. (2004). They made America. New York: Back Bay Books.
Eysenck, H. J. (1981). A model for personality. Berlin: Springer.
Eysenck, M. W. (1979). Anxiety, learning, and memory: A reconceptualization. Journal of Research in Personality,
13, 363–385.
Eysenck, M. W., & Eysenck, M. C. (1979). Processing depth, elaboration of encoding, memory stores, and
expended processing capacity. Journal of Experimental Psychology: Human Learning and Memory, 5,
472–484.
Fabiani, M., & Donchin, E. (1995). Encoding processes and memory organization: A model of the von Restorff
effect. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 224–240.
Fagan, J. W., Ohr, P. S., Flechenstein, L. K., & Ribner, D. R. (1985). The effect of crying on long-term memory in
infancy. Child Development, 56, 1584–1592.
Fein, G., Davenport, L., Yingling, C. D., & Galin, D. (1988). Verbal and nonverbal memory deficits in pure dyslexia.
Developmental Neuropsychology, 4, 181–197.
Feingold, A. (1988). Cognitive gender differences are disappearing. American Psychologist, 43, 95–103.
Feld, G. B., & Diekelmann, S. (2020). Building the bridge: Outlining steps toward an applied sleep-and-memory
research program. Current Directions in Psychological Research, 29(6), 554–562
Fendrich, D. W., Healy, A. F., & Bourne, L. E. (1991). Long-term repetition effects for motoric and perceptual
procedures. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 137–151.
Finkel, D., & McGue, M. (1993). The origins of individual differences in memory among the elderly: A behavior
genetic analysis. Psychology and Aging, 8, 527–537.
Finn, B., & Roediger, H. L. III. (2011). Enhancing retention through reconsolidation: Negative emotional arousal
following retrieval enhances later recall. Psychological Science, 22, 781–786.
Fitzgerald, H. E., & Brackbill, Y. (1976). Classical conditioning in infancy: Development and constraints.
Psychological Bulletin, 83, 353–376.
Fivush, R., & Nelson, K. (2004). Culture and language in the emergence of autobiographical memory.
Psychological Science, 15, 573–577.
Flavel, J. H. (1971). First discussant’s comments: What is memory development the development of? Human
Development, 14, 272–278.
Flavell, J. H., Beach, D. R., & Chinsky, J. M. (1966). Spontaneous verbal rehearsal in a memory task as a function
of age. Advances in Methods and Practices in Psychological Science, 4(2), 1–20.
382 References

Foa, E. B., Gillihan, S. J., & Bryant, R. A. (2013). Challenges and successes in dissemination of evidence-based
treatments for posttraumatic stress: Lessons learned from prolonged exposure therapy for PTSD. Psychological
Science in the Public Interest, 14, 65–111.
Foa, E. B., Steketee, G., Grayson, J. B., Turner, R. M., & Latimer, P. R. (1984). Deliberate exposure and blocking
of obsessive-compulsive rituals: Immediate and long-term effects. Behavior Therapy, 15, 450–472.
Foer, J. (2011). Moonwalking with Einstein. New York: Penguin Books.
Fogarty, S. J., & Helmsley, D. R. (1983). Depression and the accessibility of memories—a longitudinal study.
British Journal of Psychiatry, 142, 232–237.
Folkard, S., & Monk, T. H. (1980). Circadian rhythms in human memory. British Journal of Psychology, 71,
295–307.
Folkman, S., & Lazarus, R. S. (1985). If it changes it must be a process: Study of emotion and coping during three
stages of a college examination. Journal of Personality and Social Psychology, 48, 150–170.
Freedman, D. G. (1958). Constitutional and environmental interactions in rearing of four breeds of dogs. Science,
127, 585–586.
Frensch, P. A., & Runger, D. (2003). Implicit learning. Current Directions in Psychological Science, 12, 13–18.
Freud, S. (1901/1960). The psychopathology of everyday life. New York: Norton.
Friedman, N. P., & Miyake, A. (2017). Unity and diversity of executive functions: Individual differences as a win-
dow on cognitive structure. Cortex, 86, 186–204.
Frings, M., et al. (2006). Involvement of the human cerebellum in short-term and long-term habituation of the
acoustic startle response: A serial PET study. Clinical Neurophysiology, 117, 1290–1300.
Fujiwara, E., et al. (2008). Functional retrograde amnesia: A multiple case study. Cortex, 44, 29–435.
Gabrieli, J. D. E., Corkin, S., Mickel, S. F., & Growdon, J. H. (1993). Intact acquisition and long-term retention of
mirror-tracing skill in Alzheimer’s disease and global amnesia. Behavioral Neuroscience, 107, 899–910.
Gallistel, C. R., Fairhurst, S., & Balsam, P. (2004). The learning curve: Implications of a quantitative analysis.
Proceedings of the National Academy of Science, 101, 13124–13131.
Garcia, J., & Koelling, R. A. (1966). Relation of cue to consequence in avoidance learning. Psychonomic Science,
4, 123–124.
Gartman, L., & Johnson, N. F. (1972). Massed versus distributed repetition of homographs: A test of the differ-
ential-encoding hypothesis. Journal of Verbal Learning and Verbal Behavior, 11, 801–808.
Gatchel, R. J. (1975). Effects of interstimulus interval length on short-term and long-term habituation of auto-
nomic components of the orienting response. Physiological Psychology, 3, 133–136.
Gathercole, S. E. (1994). The nature and uses of working memory. In P. E. Morris & M. Gruneberg (Eds.),
Theoretical aspects of memory (pp. 50–78). London: Routledge.
Geller, J., Still, M. L., Dark, V. J., & Carpenter, S. K. (2018). Would disfluency by any other name still be disfluent?
Examining the disfluency effect with cursive handwriting. Memory & Cognition, 46, 1109–1126.
Gershman, S. J., Jones, C. E., Norman, K. A., Monfils, M.-H., & Niv, Y. (2013). Gradual extinction prevents the
return of fear: Implications for the discovery of state. Frontiers in Behavioral Neuroscience, 7, 164.
Gershoff, E. T., & Grogan-Kaylor, A. (2016). Spanking and child outcomes: Old controversies and new meta-anal-
yses. Journal of Family Psychology, 30, 453–469.
Gesell, A., & Thompson, H. (1929). Learning and growth in identical twins: An experimental study by the method
of co-twin control. Genetic Psychology Monographs, 6, 1–23.
Gibson, E. J. (1969). Principles of perceptual learning and development. New York: Appleton, Century, Crofts.
Gibson, E. J., & Walk, R. D. (1956). The effect of prolonged exposure to visually presented patterns on learning
to discriminate them. Journal of Comparative and Physiological Psychology, 49, 239–242.
Gibson, J. J., & Gibson, E. J. (1955). Perceptual learning: Differentiation or enrichment? Psychological Review,
62, 32–41.
Gilliland, K., & Andress, D. (1981). Ad lib caffeine consumption, symptoms of caffeinism, and academic perfor-
mance. American Journal of Psychiatry, 138, 512–514.
 References 383

Gladwell, M. (2005). Blink. New York: Little, Brown and Company.

Glanzer, M., & Cunitz, A. R. (1966). Two storage mechanisms in free recall. Journal of Verbal Learning and Verbal
Behavior, 5, 351–360.
Glass, D. C., & Singer, J. E. (1972). Urban stress: Experiments in noise and social stressors. New York: Academic
Press.
Glisky, E. L., & Schacter, D. A. (1989). Extending the limits of complex learning in organic amnesia: Computer
training in a vocational domain. Neuropsychologia, 27, 107–120.
Gobet, F., & Campitelli, G. (2007). The role of domain-specific practice, handedness, and starting age in chess.
Developmental Psychology, 43, 159–172.
Gobet, F., & Simon, H. A. (1996). Recall of rapidly presented chess positions is a function of skill. Psycho-nomic
Bulletin & Review, 3, 159–163.
Godden, D. R., & Baddeley, A. D. (1975). Context dependent memory in two natural environments: On land and
underwater. British Journal of Psychology, 66, 325–331.
Goettl, B. P., Yadrick, R. M., Connolly-Gomez, C., Regian, W., & Shebilske, W. L. (1996). Alternating task modules
in isochronal distributed training of complex tasks. Human Factors, 38, 330–346.
Goff, L. M., & Roediger, H. L. (1998). Imagination inflation for action events: Repeated imaginings lead to illusory
recollections. Memory & Cognition, 26, 20–33.
Gold, P. E., & Korol, D. L. (2012, December). Making memories matter. Frontiers in Integrative Neuroscience, 6,
article 116.
Goldfarb, E. V., Tompary, A., Davachi, L., & Phelps, E. A. (2019). Acute stress throughout the memory cycle:
Diverging effects on associative and item memory. Journal of Experimental Psychology: General, 148,
13–29.
Goldman, M. S. (1983). Cognitive impairment in chronic alcoholics. American Psychologist, 38, 1045–1054.
Goldstone, R. L. (1998). Perceptual learning. Annual Review of Psychology, 49, 585–612.
Golkar, A., Selbing, I., Flygare, O., Ohman, A., & Olsson, A. (2013). Other people as means to a safe end:
Vicarious extinction blocks the return of learned fear. Psychological Science, 24, 2182–2190.
Goodwin, K. A., Meissner, C. A., & Ericsson, K. A. (2001). Toward a model of false recall: Experimental manipu-
lation of encoding context and the collection of verbal reports. Memory & Cognition, 29, 806–819.
Gormezano, I. (1966). Classical conditioning. In J. B. Sidowski (Ed.), Experimental methods and instrumentation
in psychology (pp. 385–420). New York: McGraw-Hill.
Graf, P., & Schacter, D. L. (1989). Unitization and grouping mediate dissociations in memory for new associa-
tions. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 930–940.
Graf, P., Squire, L., & Mandler, G. (1984). The information that amnesic patients do not forget. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 10, 164–178.
Greene, R. L. (1986). Sources of recency effects in free recall. Psychological Bulletin, 99, 221–228.
Greenough, W. T. (1985). The possible role of experience dependent synaptogenesis, or synapses on demand,
in the memory process. In N. M. Weinberger, J. L. McGaugh, & G. Lynch (Eds.), Memory systems of the brain
(pp. 77–103). New York: Guilford.
Greenspoon, J. (1955). The reinforcing effect of two spoken sounds on the frequency of two responses.
American Journal of Psychology, 68, 409–416.
Grigorenko, E. L., & Sternberg, R. J. (1997). Styles of thinking, abilities, and academic performance. Exceptional
Children, 63, 295–312.
Grillon, C. (2002). Startle reactivity and anxiety disorders: Aversive conditioning, context, and neurobiology.
Biological Psychiatry, 52, 958–975.
Groninger, L. D., Groninger, D. H., & Stiens, J. (1995). Learning the names of people: The role of image media-
tors. Memory, 3, 147–167.
Groves, P. M., & Thompson, R. F. (1970). Habituation: A dual-process theory. Psychological Review, 77,
419–450.
384 References

Gulya, M., Galluccio, L., Wilk, A., & Rovee-Collier, C. (2001). Infants’ long-term memory for a serial list:
Recognition and reactivation. Developmental Psychobiology, 38, 174–185.
Gunter, B., Berry, C., & Clifford, B. R. (1981). Proactive interference effects with television news items: Further
evidence. Journal of Experimental Psychology: Human Learning and Memory, 7, 480–487.
Gunther, L. M., Miller, R. R., & Matute, H. (1997). CSs and USs: What’s the difference? Journal of Experimental
Psychology: Animal Behavior Processes, 23, 15–30.
Gustavson, C. R. (1977). Comparative and field aspects of learned food aversions. In L. M. Barker, M. R. Best,
& M. Domjan (Eds.), Learning mechanisms in food selection (pp. 23–44). Waco, TX: Baylor University Press.
Gustavson, C. R., & Garcia, J. (1974, August). Pulling a gag on the wily coyote. Psychology Today, 8(3), 68–72.
Guthrie, J. P., Ash, R. A., & Bendapudi, V. (1995). Additional validity evidence for a measure of morningness.
Journal of Applied Psychology, 80, 186–190.
Hall, G. (1994). Pavlovian conditioning: Laws of association. In N. J. Mackintosh (Ed.), Animal learning and cog-
nition (pp. 15–43). San Diego, CA: Academic Press.
Hamann, S. B., Squire, L. R., & Schacter, D. L. (1995). Perceptual thresholds and priming in amnesia.
Neuropsychology, 9, 3–15.
Hambrick, D. Z., & Meinz, E. J. (2011). Limits on the predictive power of domain-specific experience and knowl-
edge in skilled performance. Current Directions in Psychological Science, 20, 275–279.
Hammerl, M., Bloch, M., & Silverthorne, C. P. (1997). Effects of US-alone presentations on human evaluative
conditioning. Learning and Motivation, 28, 491–509.
Hammond, D., Fong, G. T., Borland, R., Cummings, K. M., McNeill, A., & Driezen, P. (2007). Text and graphic
warnings on cigarette packages: Findings from International tobacco control four country study. American
Journal of Preventive Medicine, 32, 210–217.
Hampson, E., & Kimura, D. (1988). Reciprocal effects of hormonal fluctuations of human motor and perceptu-
al-spatial skills. Behavioral Neuroscience, 102, 456–459.
Hanley-Dunn, P., & McIntosh, J. L. (1984). Meaningfulness and recall of names by young and old adults. Journal
of Gerontology, 39, 583–585.
Harlow, H. F. (1959). Love in infant monkeys. Scientific American, 200(6), 68–74.
Harp, S. F., & Mayer, R. E. (1998). How seductive details do their damage: A theory of cognitive interest in sci-
ence learning. Journal of Educational Psychology, 90, 414–434.
Harrington, D. L., Haaland, K. Y., Yeo, R. A., & Marder, E. (1990). Procedural memory in Parkinson’s disease:
Impaired motor but not visuoperceptual learning. Journal of Clinical and Experimental Neuro-psychology, 12,
323–339.
Harris, B. (1979). Whatever happened to Little Albert? American Psychologist, 34, 151–160.
Harris, J. E. (1978). External memory aids. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical
aspects of memory (pp. 172–179). London: Academic Press.
Harris, J. E. (1980). Memory aids people use: Two interview studies. Memory & Cognition, 8, 31–38.
Hart, J. T. (1965). Memory and the feeling-of-knowing experience. Journal of Educational Psychology, 56, 208–216.
Hasher, L., & Zacks, R. T. (1988). Working memory, comprehension, and aging: A review and a new view. The
Psychology of Learning and Motivation, 22, 193–225.
Hastie, R., & Kumar, P. A. (1979). Person memory: Personality traits as organizing principles in memory for
behaviors. Journal of Personality and Social Psychology, 37, 25–38.
Haubrich, J., Bernabo, M., Baker, A. G., & Nader, K. (2020). Impairments to consolidation, reconsolidation, and
long-term memory maintenance lead to memory erasure. Annual Review of Neuroscience, 43, 297–314.
Hawkins, R. D., Cohen, T. E., Greene, W., & Kandel, E. R. (1998). Relationships between dishabituation, sensiti-
zation, and inhibition of the gill- and siphon-withdrawal reflex in Aplysia californica: Effects of response measure,
test time, and training stimulus. Behavioral Neuroscience, 112, 24–38.
Healy, A. F., Clawson, D. M., McNamara, D. S., Marmie, W. R., Schneider, V. I., Rickard, T. C., et al. (1993). The
long-term retention of knowledge and skills. The Psychology of Learning and Motivation, 30, 135–164.
 References 385

Healy, A. F., & McNamara, D. S. (1996). Verbal learning and memory: Does the modal model still work? Annual
Review of Psychology, 47, 143–172.
Hearst, E., & Jenkins, H. M. (1974). Sign tracking: The stimulus-reinforcer relation and directed action. Austin, TX:
Psychonomic Society.
Hebb, D. O. (1949). Organization of behavior. New York: Wiley.
Hebb, D. O. (1955). Drives and the C. N. S. (conceptual nervous system). Psychological Review, 62, 243–254.
Heindel, W. C., Butters, N., & Salmon, D. P. (1988). Impaired learning of a motor skill in patients with Huntington’s
disease. Behavioral Neuroscience, 102, 141–147.
Hepper, P. G. (1989). Foetal learning: Implications for psychiatry? British Journal of Psychiatry, 155, 289–293.
Herlitz, A., Nilsson, L. G., & Backman, L. (1997). Gender differences in episodic memory. Memory & Cognition,
25, 801–811.
Hermans, D., Craske, M. G., Mineka, S., & Lovibond, P. F. (2006). Extinction in human fear conditioning. Biological
Psychiatry, 60, 361–368.
Herrmann, D. J. (1987). Task appropriateness of mnemonic techniques. Perceptual and Motor Skills, 64,
171–178.
Herrmann, D. J., Crawford, M., & Holdsworth, M. (1992). Gender-linked differences in everyday memory perfor-
mance. British Journal of Psychology, 83, 221–231.
Heuer, F., & Reisberg, D. (1992). Emotion, arousal, and memory for detail. In S.-A. Christianson (Ed.), The hand-
book of emotion and memory (pp. 151–180). Hillsdale, NJ: Erlbaum.
Hiby, E. F., Rooney, N. J., & Bradshaw, J. W. (2004). Dog training methods: Their use, effectiveness, and interac-
tion with behavior and welfare. Animal Welfare, 13, 63–69.
Hicks-Pass, S. (2009). Corporal punishment in America today: Spare the rod, spoil the child? A systematic review
of the literature. Best Practices in Mental Health: An International Journal, 5, 71–88.
Hilgard, E. R., & Marquis, D. G. (1940). Conditioning and learning. NY: Appleton-Century.
Hill, W. F. (1978). Effects of mere exposure on preferences in nonhuman animals. Psychological Bulletin, 85,
1177–1198.
Hills, T., & Hertwig, R. (2011). Why aren’t we smarter already: Evolutionary trade-offs and cognitive enhance-
ments. Current Directions in Psychological Science20, 373–377
Hintzman, D. L. (1990). Human learning and memory: Connections and dissociations. Annual Review of
Psychology, 41, 109–139.
Hintzman, D. L. (2011). Research strategy in the study of memory: Fads, fallacies, and the search for the “coor-
dinates of truth”. Perspectives on Psychological Science, 6, 253–271.
Hintzman, D. L., Block, R. A., & Summers, J. J. (1973). Modality tags and memory for repetitions: Locus of the
spacing effect. Journal of Verbal Learning and Verbal Behavior, 12, 229–238.
Hirnstein, M., Steubs, J., Moèand, A., & Hausmann, M. (2022). Sex/gender differences in verbal fluency and
verbal-episodic memory: A meta-analysis. Perspectives on Psychological Science, 1–24
Hiroto, D. S. (1974). Locus of control and learned helplessness. Journal of Experimental Psychology, 102,
187–193.
Hirst, W., et al. (2009). Long-term memory for the terrorist attack of September 11: Flashbulb memories, event
memories, and the factors that influence their retention. Journal of Experimental Psychology: General, 138,
161–176.
Hirst, W., et al. (2015). A ten-year follow-up of a study of memory for the attack of Septeber 11, 2001: Flashbulb
memories and memories for flashbulb events. Journal of Experimental Psychology: General, 144, 604–623.
Hirst, E. S., Dozier, C. L., & Payne, S. W. (2016). Efficacy and preference for reinforcement and response cost in
token economies. Journal of Applied Behavior Analysis, 49, 329–345.
Hirst, W., Spelke, E., Reaves, C. C., Caharack, G., & Neisser, U. (1980). Dividing attention without alternation or
automaticity. Journal of Experimental Psychology: General, 109, 98–117.
386 References

Hoehl, S., Hellmer, K., Johansson, M., & Gredebäck, G. (2017). Itsy bitsy spider…: Infants react with increased
arousal to spiders and snakes. Frontiers in Psychology, 8, 1710.
Hogervorst, E., DeJager, C., Budge, M., & Smith, A. D. (2004). Serum levels of estradiol and testosterone and
performance in different cognitive domains in healthy elderly men and women. Psychoneuroendocrinology, 29,
405–421.
Hollis, K. L. (1997). Contemporary research on Pavlovian conditioning: A new “functional” analysis. American
Psychologist, 52, 956–965.
Holloway, F. A. (1978). State dependent retrieval based on time of day. In B. Ho, D. Richards, & D. Chute (Eds.),
Drug discrimination and state dependent learning (pp. 319–344). New York: Academic Press.
Horton, D. L., & Mills, C. B. (1984). Human learning and memory. Annual Review of Psychology, 35, 361–394.
Hostetler, A. J. (1988, January). If you said it once, you’ll say it again. APA Monitor, 9.
Houston, J. P. (1983). Psychology: A closed system of self-evident information? Psychological Reports, 52,
203–208.
Hovland, C. I. (1951). Human learning and retention. In S. S. Stevens (Ed.), Handbook of experimental psychol-
ogy (pp. 613–688). New York: Wiley.
Howe, M. L. (2000). The fate of early memories: Developmental science and the retention of early childhood
experiences. Washington, DC: American Psychological Association.
Hudson, J. A. (1990). The emergence of autobiographical memory in mother-child conversation. In R. Fivush & J. A.
Hudson (Eds.), Knowing and remembering in young children (pp. 166–196). New York: Cambridge University Press.
Hull, C. L. (1943). Principles of behavior. New York: Appleton, Century, Crofts.
Hull, C. L. (1949). Behavior postulates and corollaries—1949. Psychological Review, 57, 173–180.
Hulme, C., & Mackenzie, S. (1992). Working memory and severe learning difficulties. Mahwah, NJ: Erlbaum.
Hunt, R. R. (1995). The subtlety of distinctiveness: What von Restorff really did. Psychonomic Bulletin & Review,
2, 105–112.
Hunt, R. R., & Elliott, J. M. (1980). The role of nonsemantic information in memory: Orthographic distinctiveness
effects on retention. Journal of Experimental Psychology: General, 109, 49–74.
Hurley, D. (2013). Smarter: The new science of building brain power. New York: Penguin.
Husain, M., & Mehta, M. A. (2011). Cognitive enhancement by drugs in health and disease. Trends in Cognitive
Sciences, 15, 28–36.
Huston, J. P., Mondadori, C., & Waser, P. G. (1974). Facilitation of learning by reward of post-trial memory pro-
cesses. Experientia, 30, 1038–1040.
Hyde, T. S., & Jenkins, J. J. (1969). Differential effects of incidental tasks on the organization of recall of a list of
highly associated words. Journal of Experimental Psychology, 82, 472–481.
Ilieva, I., Boland, J., & Farah, M. J. (2013). Objective and subjective cognitive enhancing effects of mixed amphet-
amine salts in healthy people. Neuropharmacology, 64, 496–505.
Ince, L. P., Brucker, B. S., & Alba, A. (1978). Reflex conditioning of a spinal man. Journal of Comparative and
Physiological Psychology, 92, 796–802.
Inhoff, A. W., & Gordon, A. M. (1997). Eye movements and eye-hand coordination during typing. Current
Directions in Psychological Science, 6, 153–157.
Inoue, S., & Matsuzawa, T. (2007). Working memory of numerals in chimpanzees. Current Biology, 17,
1004–1005.
Ishikawa, T., Fujiwara, H., Imai, O., & Okabe, A. (2008). Wayfinding with a GPS-based mobile navigation system:
A comparison with maps and direct experience. Journal of Environmental Psychology, 28, 74–82.
Israel, L., & Schacter, D. L. (1997). Pictorial encoding reduces false recognition of semantic associates.
Psychonomic Bulletin & Review, 4, 577–581.
Jackson, J., Donaldson, D. I., & Dering, B. (2021). The morning after the night before: Alcohol-induced blackouts
impair next day recall in sober young adults. Plos One, 16(5), e0250827.
Jacobs, A. (2005, July 31). The Adderall advantage. New York Times, p. ED16.
 References 387

Jacoby, L. L., Woloshyn, V., & Kelley, C. M. (1989). Becoming famous without being recognized: Unconscious
influences of memory produced by divided attention. Journal of Experimental Psychology: General, 118, 115–125.
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on
working memory. Proceedings of the National Academy of Sciences, USA, 105, 6829–6833.
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Priti Shah, P. (2011). Short- and long-term benefits of cognitive
training. Proceedings of the National Academy of Sciences, USA, 108, 10081–10086.
James, W. (1890). The principles of psychology (2 vols.). New York: Dover reprint edition, (1950).
Janet, P. (1907). The major symptoms of hysteria. New York: Macmillan.
Janis, I. L., & Astrachan, M. (1951). The effects of electroconvulsive treatments on memory efficiency. Journal
of Abnormal and Social Psychology, 46, 501–511.
Janowsky, J. S., Oviatt, S. K., & Orwoll, E. S. (1994). Testosterone influences spatial cognition in older men.
Behavioral Neuroscience, 108, 325–332.
Jauhar, S. (2002, May 5). Jolts of anxiety. New York Times Magazine, pp. 16, 18, 22, 24.
Jenkins, H. M., & Moore, B. R. (1973). The form of the autoshaped response with food or water reinforcers.
Journal of the Experimental Analysis of Behavior, 20, 163–181.
Jenkins, J. G., & Dallenbach, K. M. (1924). Obliviscence during sleep and waking. American Journal of
Psychology, 35, 605–612.
Jenkins, J. J., & Russell, W. A. (1952). Associative clustering during recall. Journal of Abnormal and Social
Psychology, 47, 818–821.
Jensen, A. R. (1962). Temporal and spatial effects of serial position. American Journal of Psychology, 75,
390–400.
Johnson, M. K., Nolde, S. F., Mather, M., Kounios, J., Schacter, D. L., & Curran, T. (1997). The similarity of brain
activity associated with true and false recognition memory depends on test format. Psychological Science, 8,
250–257.
Johnson, M. K., & Raye, C. L. (1981). Reality monitoring. Psychological Review, 88, 67–85.
Jones, B. F., & Hall, J. W. (1982). School applications of the mnemonic keyword method as a study strategy by
eighth graders. Journal of Educational Psychology, 74, 230–237.
Jones, W. T. (1952). A history of Western philosophy. New York: Harcourt, Brace & World.
Josselyn, S. A., & Tonegawa, S. (2020). Memory engrams: Recalling the past and imagining the future. Science,
367, eaaw4325.
Kail, R. (1990). The development of memory in children (3rd ed.). New York: Freeman.
Kalish, H. I. (1958). The relationship between discriminability and generalization: A re-evaluation. Journal of
Experimental Psychology, 55, 637–644.
Kamin, L. J. (1956). The effects of termination of the CS and avoidance of the US on avoidance learning. Journal
of Comparative and Physiological Psychology, 49, 420–424.
Kamin, L. J. (1969). Predictability, surprise, attention, and conditioning. In B. A. Campbell & R. M. Church (Eds.),
Punishment (pp. 279–296). New York: Appleton, Century, Crofts.
Kaminski, J., Call, J., & Fischer, J. (2004). Word learning in the domestic dog: Evidence for “fast mapping”.
Science, 304, 1682–1683.
Kandel, E. R. (2001). The molecular biology of memory storage: A dialogue between genes and synapses.
Science, 294, 1030–1038.
Kandel, E. R. (2006). In search of memory. New York: Norton.
Kang, S. H. K., McDermott, K. B., & Roediger, H. L., III. (2007). Test format and corrective feedback modify the
effect of testing on long-term retention. European Journal of Cognitive Psychology, 19, 528–558.
Karni, A., Tanne, D., Rubenstein, B. S., Askenasy, J. M., & Sagi, D. (1994). Dependence on REM sleep of over-
night improvement of a perceptual skill. Science, 265, 679–682.
Kassam, K. S., Gilbert, D. T., Swencionis, J. K., & Wilson, T. D. (2009). Misconceptions of memory: The Scooter
Libby effect. Psychological Science, 20, 551–553.
388 References

Kausler, D. H., Wiley, J. G., & Lieberwitz, K. J. (1992). Adult age differences in short-term memory and subse-
quent long-term memory for actions. Psychology and Aging, 7, 309–316.
Kawecki, T. J. (2010). Evolutionary ecology of learning: Insights from fruit lies. Population Ecology, 52, 15–25.
Kazdin, A. E. (1972). Response cost: The removal of conditioned reinforcers for therapeutic change. Behavior
Therapy, 3, 533–546.
Kazdin, A. E. (1994). Behavior modification in applied settings (5th ed.). Pacific Grove, CA: Brooks/Cole.
Kearins, J. M. (1981). Visual spatial memory in Australian aboriginal children of desert regions. Cognitive
Psychology, 13, 434–460.
Keenan, J., McWhinney, B., & Mayhew, D. (1977). Pragmatics in memory: A study of natural conversation.
Journal of Verbal Learning and Verbal Behavior, 17, 549–560.
Keeney, T. J., Cannizzo, S. R., & Flavell, J. H. (1967). Spontaneous and induced rehearsal in a recall task. Child
Development, 38, 953–966.
Kempermann, G., Kuhn, H. G., & Gage, F. H. (1997). More hippocampal neurons in adult mice living in an
enriched environment. Nature, 386, 493–495.
Kenny, L. M., Bryant, R. A., Silove, D., Creamer, M., O’Donnell, M., & McFarlane, A. C. (2009). Distant
memories: A prospective study of vantage point of trauma memories. Psychological Science, 20,
1049–1052.
Keppel, G. (1964). Facilitation in short- and long-term retention of paired associates following distributed prac-
tice in learning. Journal of Verbal Learning and Verbal Behavior, 3, 91–111.
Kesner, R. P., Hopkins, R. O., & Chiba, A. A. (1992). Learning and memory in humans, with an emphasis on the
role of the hippocampus. In L. R. Squire & N. Butters (Eds.), Neuropsychology of memory (pp. 106–121).
New York: Guilford Press.
Kientzle, M. J. (1949). Ability patterns under distributed practice. Journal of Experimental Psychology, 39,
532–537.
Kiewra, K. A., DuBois, N. F., Christian, D., McShane, A., Meyerhoffer, M., & Roskelley, D. (1991). Notetaking
functions and techniques. Journal of Educational Psychology, 83, 240–245.
Kihlstrom, J. F. (1987). The cognitive unconscious. Science, 237, 1445–1452.
Kim, J., Allen, C. T., & Kardes, F. R. (1996). An investigation of the mediational mechanisms underlying attitudinal
conditioning. Journal of Marketing Research, 33, 318.
Kimble, G. A. (1967). The definition of learning and some useful distinctions. In G. A. Kimble (Ed.), Foundations
of conditioning and learning (pp. 82–99). New York: Appleton, Century, Crofts.
Kindt, M., Soeter, M., & Vervliet, B. (2009). Beyond extinction: Erasing fear responses and preventing the return
of fear. Nature Neuroscience, 12, 256–258.
Kirby, F. D., & Shields, F. (1972). Modification of arithmetic response rate and attending
Klatzky, R. L. (1984). Memory and awareness: An information-processing perspective. New York: Freeman.
Klein, D. C., & Seligman, M. E. P. (1976). Reversal of performance deficits and perceptual deficits in learned
helplessness and depression. Journal of Abnormal Psychology, 85, 11–26.
Klingberg, T. (2009). The overflowing brain. Oxford, UK: Oxford University Press.
Knowlton, B., McGowan, M., Olton, D. S., & Gamzu, E. (1985). Hippocampal stimulation disrupts spatial working
memory even 8h after acquisition. Behavioral & Neural Biology, 44, 325–337.
Kolb, B., & Whishaw, I. Q. (1985). Fundamentals of human neuropsychology (2nd ed.). New York: Freeman.
Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development. Englewood
Cliffs, NJ: Prentice Hall.
Kopelman, M. D. (1987). Amnesia: Organic and psychogenic. British Journal of Psychiatry, 150, 428–442.
Kopelman, M. D. (1994). Working memory in the amnesic syndrome and degenerative dementia.
Neuropsychology, 8, 555–562.
Koriat, A. (1997). Monitoring one’s own knowledge during study: A cue-utilization approach to judgments of
learning. Journal of Experimental Psychology: General, 126, 349–370.
 References 389

Koulack, D., & Goodenough, D. R. (1976). Dream recall and dream recall failure: An arousal-retrieval model.
Psychological Bulletin, 83, 975–984.
Kozlowski, L. T., & Bryant, K. J. (1977). Sense of direction, spatial orientation, and cognitive maps. Journal of
Experimental Psychology: Human Perception and Performance, 3, 590–598.
Kraiger, K., Ford, J. K., & Salas, E. (1993). Application of cognitive, skill-based, and affective theories of learning
outcomes to new methods of training evaluation. Journal of Applied Psychology, 78, 311–328.
Kratzig, G., Arbuthnott, K. (2006). Perceptual learning style and proficiency: A test of the hypothesis. Journal of
Educational Psychology, 98, 238–246.
Kreutzer, M. A., Leonard, C., & Flavell, J. H. (1975). An interview study about children’s knowledge about mem-
ory. Monographs of the Society for Research in Child Development, 40 (1, Serial No. 159).
Kritchevsky, M., Chang, J., & Squire, L. R. (2004). Functional amnesia: Clinical description and neuropsycholog-
ical profile of 10 cases. Learning & Memory, 11, 213–226.
Krug, D., Davis, T. B., & Glover, J. A. (1990). Massed versus distributed repeated reading: A case of forgetting
helping recall? Journal of Educational Psychology, 82, 366–371.
Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it: How difficulties in recognizing one’s own incom-
petence lead to inflated self-assessments. Journal of Personality and Social Psychology, 77, 1121–1134.
Kuhara-Kojima, K., & Hatano, G. (1991). Contribution of content knowledge and learning ability to the learning of
facts. Journal of Educational Psychology, 83, 253–263.
Kulhavy, R. W., Dyer, H. W., & Silver, L. (1975). The effects of notetaking and test expectancy on the learning of
text material. Journal of Educational Research, 68, 363–365.
Kulhavy, R. W., Schwartz, N. H., & Peterson, S. (1986). Working memory: The encoding process. In G. D. Phye &
T. Andre (Eds.), Cognitive classroom learning (pp. 115–140). New York: Academic Press.
Kunst-Wilson, W. R., & Zajonc, R. B. (1980). Affective discrimination of stimuli that cannot be recognized.
Science, 207, 557–558.
Kurbat, M. A., Shevell, S. K., & Rips, L. J. (1998). A year’s memories: The calendar effect in autobiographical
memory. Memory & Cognition, 26, 532–552.
Kvavilashvili, L. (1992). Remembering intentions: A critical review of existing experimental paradigms. Applied
Cognitive Psychology, 6, 507–524.
Kvavilashvili, L., & Mandler, G. (2004). Out of one’s mind: A study of involuntary semantic memories. Cognitive
Psychology, 48, 47–94.
Kyllonen, P. C., Tirre, W. C., & Christal, R. E. (1991). Knowledge and processing speed as determinants of asso-
ciative learning. Journal of Experimental Psychology: General, 120, 57–79.
Landauer, T. K. (1969). Reinforcement as consolidation. Psychological Review, 76, 82–96.
Landauer, T. K., & Bjork, B. J. (1978). Optimum rehearsal patterns and name learning. In M. M. Gruneberg, P. E.
Morris, & R. N. Sykes (Eds.), Practical aspects of memory (pp. 625–632). New York: Academic Press.
Lally, P., van Jaarsveld, C. H., Potts, H. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation
in the real world. European Journal of Social Psychology, 40, 998–1009.
Lashley, K. S. (1917). The effect of strychnine and caffeine upon rate of learning. Psychobiology, 1, 141–170.
Lashley, K. S. (1929). Brain mechanisms and intelligence. Chicago: University of Chicago Press (Dover reprint, 1963).
Lashley, K. S. (1951). The problem of serial order in behavior. In L. A. Jeffress (Ed.), Cerebral mechanisms in
behavior (pp. 112–136). New York: Wiley.
Law, C., & Boisseau, C. L. (2019). Exposure and response prevention in the treatment of obsessive-compulsive
disorder: Current perspectives. Psychology Research and Behavior Management, 12, 1167–1174.
Lawton, C. A. (1994). Gender differences in way-finding strategies: Relationship to spatial ability and spatial
anxiety. Sex Roles, 30, 765–779.
Lechner, H. A., Squire, L. R., & Byrne, J. H. (1999). 100 years of consolidation: Remembering Muller and
Pilzecker. Learning & Memory, 6, 77–87.
LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–184.
390 References

LeDoux, J. E., Moscarello, J., Sears, R., & Campese, V. (2017). The birth, death and resurrection of avoidance: A
reconceptualization of a troubled paradigm. Molecular Biology, 22, 24–36.
Lee, C. L., & Estes, W. K. (1977). Order and position in primary memory for letter strings. Journal of Verbal
Learning and Verbal Behavior, 16, 395–416.
Lemonick, M. D. (1995, July 24). Einstein strikes again. Time, p. 55.
Lenehan, M. (1986, April). Four ways to walk a dog. The Atlantic Monthly, 35–48, 89–99.
Lenneberg, E. H. (1967). Biological foundations of language. New York: Wiley.
LePort, A. K. R., Matfield, A. T., Dickinson-Anson, H., Fallon, J. H., Stark, C. E. L., Kruggel, F., … McGaugh, J. L.
(2012). Behavioral and neuroanatomical investigation of highly superior autobiographical memory (HSAM).
Neurobiology of Learning and Memory, 98, 78–92.
Lepper, M. R., Greene, D., & Nisbett, R. E. (1973). Undermining children’s intrinsic interest with extrinsic
rewards: A test of the overjustification hypothesis. Journal of Personality and Social Psychology, 28, 129–137.
Lettvin, J. Y., Maturana, H. R., McCulloch, W. S., & Pitts, W. H. (1959). What the frog’s eye tells the frog’s brain.
Proceedings of the IRE, 1940–1951.
Levin, I. P., & Hinrichs, J. V. (1995). Experimental psychology: Contemporary methods and applications. Madison,
WI: WCB Brown & Benchmark.
Levine, M., Marchon, I., & Hanley, G. L. (1984). The placement and misplacement of you-are-here maps.
Environment and Behavior, 16, 139–157.
Lewandowsky, S., Ecker, U., Seifert, C., Schwarz, N., & Cook, J. (2012). Misinformation and its correction:
Continued influence and successful debiasing. Psychological Science in the Public Interest, 13, 106–131.
Lewis, D. J. (1979). Psychobiology of active and inactive memory. Psychological Bulletin, 86, 1054–1083.
Lewis, D. J., & Duncan, C. P. (1958). Expectation and resistance to extinction of a lever-pulling response as a function
of percentage of reinforcement and number of acquisition trials. Journal of Experimental Psychology, 55, 121–128.
Lieberman, D. A., McIntosh, D. C., & Thomas, G. V. (1979). Learning when reward is delayed: A marking hypoth-
esis. Journal of Experimental Psychology: Animal Behavior Processes, 5, 224–242.
Linton, M. (1982). Transformations of memory in everyday life. In U. Neisser (Ed.), Memory Observed (pp. 77–
91). San Francisco: Freeman.
Lobb, H., & Hardwick, C. (1976). Eyelid conditioning and intellectual level: Effects of repeated acquisition and
extinction. American Journal of Mental Deficiency, 80, 423–430.
LoBue, V., Rakison, D. H., & DeLoache, J. S. (2010). Threat perception across the life span: Evidence of multiple
converging pathways. Current Directions in Psychological Science, 19, 375–379.
Locke, J. L., & Fehr, F. S. (1971). Young children’s use of the speech code in a recall task. Journal of Experimental
Child Psychology, 10, 367–373.
Lockhead, G. R., & Crist, W. B. (1980). Making letters distinctive. Journal of Educational Psychology, 72,
483–493.
Loftus, E. F. (1979). Eyewitness testimony. Cambridge: Harvard University Press.
Loftus, E. F., & Burns, T. E. (1982). Mental shock can produce retrograde amnesia. Memory & Cognition, 10,
318–323.
Loftus, E. F., & Ketcham, K. (1994). The myth of repressed memory. New York: St. Martin’s Press.
Loftus, E. F., & Loftus, G. R. (1980). On the permanence of stored information in the brain. American Psychologist,
35, 409–420.
Loftus, E. F., Loftus, G. R., & Messo, J. (1987). Some facts about “weapon focus”. Law and Human Behavior, 11,
55–62.
Loftus, E. F., Miller, D. G., & Burns, H. J. (1978). Semantic integration of verbal information into a visual memory.
Journal of Experimental Psychology: Human Learning and Memory, 4, 19–31.
Loftus, E. F., & Pickrell, J. E. (1995). The formation of false memories. Psychiatric Annals, 25, 720–725.
Logan, F. A., & Wagner, A. R. (1962). Direction of change in CS in eyelid conditioning. Journal of Experimental
Psychology, 64, 325–326.
 References 391

Logue, A. W., Forzano, L. B., & Ackerman, K. T. (1996). Self-control in children: Age, preference for reinforcer
amount and delay, and language ability. Learning and Motivation, 27, 269–277.
Lord, A. B. (1960). The singer of tales. Cambridge: Harvard University Press.
Lukowiak, K., & Jacklet, J. W. (1972). Habituation and dishabituation: Interactions between peripheral and cen-
tral nervous systems in Aplysia. Science, 178, 1306–1308.
Luria, A. R. (1968). The mind of a mnemonist. New York: Basic Books.
Lynch, S., & Yarnell, P. R. (1973). Retrograde amnesia: Delayed forgetting after concussion. American Journal of
Psychology, 86, 643–645.
Lyubomirsky, S., Kasri, F., & Zehm, K. (2003). Dysphoric rumination impairs concentration on academic tasks.
Cognitive Therapy and Research, 27, 309–330.
MacFadyen, J. T. (1986, October). Educated monkeys help the disabled to help themselves. Smithsonian, 17,
125–132.
MacKay, D. G., Shafto, M., Taylor, J. K., Marian, D. E., Abrams, L., & Dyer, J. R. (2004). Relations between emo-
tion, memory, and attention: Evidence from taboo Stroop, lexical decision, and immediate memory tasks.
Memory & Cognition 2004, 32, 474–488.
MacKintosh, N. J. (1974). The psychology of animal learning. New York: Academic Press.
MacKintosh, N. J. (1983). Conditioning and associative learning. New York: Oxford University Press.
Madigan, S., & O’Hara, R. (1992). Short-term memory at the turn of the century: Mary Whiton Calkins’s memory
research. American Psychologist, 47, 170–174.
Magnussen, S., et al. (2006). What people believe about memory. Memory, 14, 595–613.
Maguire, E. A., Gadian, D. G., Johnsrude, I. S., Good, C. D., Ashburner, J., Frackowiak, R. S., & Frith, C. D. (2000).
Navigation-related structural change in the hippocampi of taxi drivers. Proceedings of the National Academy of
Science, 97(8), 4398–4403.
Maguire, E. A., Valentine, E. R., Wilding, J. M., & Kapur, N. (2003). Routes to remembering: The brains behind
superior memory. Nature Neuroscience, 6, 90–95.
Maia, T. V., & Cleeremans, A. (2005). Consciousness: Converging insights from connectionist modeling and
neuroscience. Trends in Cognitive Sciences, 9, 397–404.
Maia, T. V., & McClelland, J. L. (2004). A reexamination of the evidence for the somatic marker hypothesis: What
participants really know in the Iowa gambling task. Proceedings of the National Academy of Sciences, 101,
16075–16080.
Maier, N. R. F., & Schneirla, T. C. (1935). Principles of animal psychology. New York: McGraw-Hill.
Maier, S. F., Seligman, M. E. P., & Solomon, R. L. (1969). Pavlovian fear conditioning and learned helplessness:
Effects on escape and avoidance behavior of (a) the CS-US contingency and (b) the independence of the US and
voluntary responding. In B. A. Campbell & R. M. Church (Eds.), Punishment and aversive behavior (pp. 299–342).
New York: Appleton, Century, Crofts.
Mandler, G. (1967). Organization and memory. In G. H. Bower (Ed.), The psychology of learning and motivation
(Vol. 1, pp. 327–372). New York: Academic Press.
Mangiulli, I., Otgaar, H., Jelicic, M., & Merckelbach, H. (2021). A critical review of case studies on dissociative
amnesia. Clinical Psychological Science, 1–21.
Mantyla, T. (1986). Optimizing cue effectiveness: Recall of 500 and 600 incidentally learned words. Journal of
Experimental Psychology, Learning, Memory, and Cognition, 12, 66–71.
Mantyla, T. (2003). Assessing absentmindedness: Prospective memory complaint and impairment in mid-
dle-aged adults. Memory and Cognition, 31, 15–25.
Marks, I. (1977). Phobias and obsessions: Clinical phenomena in search of a laboratory model. In J. D. Maser &
M. E. P. Seligman (Eds.), Psychopathology: Experimental models (pp. 174–213). San Francisco: Freeman.
Marler, P. (1970). A comparative approach to vocal learning: Song development in white-crowned sparrows.
Journal of Comparative and Physiological Psychology, 71, 1–25.
Marslen-Wilson, W. D., & Teuber, H. L. (1975). Memory for remote events in anterograde amnesia: Recognition
of public figures in news photographs. Neuropsychologia, 13, 347–352.
392 References

Martin, E. (1975). Generation-recognition theory and the encoding specificity principle. Psychological Review,
82, 150–153.
Martin, I., & Levey, A. B. (1987). Learning what will happen next: Conditioning, evaluation, and cognitive pro-
cesses. In G. Davey (Ed.), Cognitive processes and Pavlovian conditioning in humans (pp. 57–81). Chichester,
UK: Wiley.
Martin, M., & Jones, G. V. (1984). Cognitive failures in everyday life. In J. E. Harris & P. E. Morris (Eds.), Everyday
memory, actions and absentmindedness (pp. 173–190). New York: Academic Press.
Martin, R. C. (2021). The critical role of semantic working memory in language comprehension and production.
Current Directions in Psychological Science 30, 283–291.
Maslow, A. H. (1937). The influence of familiarization on preferences. Journal of Experimental Psychology, 21,
162–180.
Masserman, J. (1943). Behavior and neurosis: An experimental psychoanalytic approach to psychobiologic prin-
ciples. Chicago: University of Chicago Press.
Mayer, R. E., & Massa, L. J. (2003). Three facets of visual and verbal learners: Cognitive ability, cognitive style,
and learning preference. Journal of Educational Psychology, 95, 833–846.
Maylor, E. A., & Rabbitt, P. M. A. (1993). Alcohol, reaction time, and memory: A meta-analysis. British Journal of
Psychology, 84, 301–317.
McCarthy, R. A., & Warrington, E. K. (1988). Evidence for modality-specific meaning systems in the brain.
Nature, 334, 428–430.
McClosky, M., Wible, C. G., & Cohen, N. J. (1988). Is there a special flashbulb memory mechanism? Journal of
Experimental Psychology: General, 117, 171–181.
McCrae, R. R., & Costa, P. T. (1986). Clinical assessment can benefit from recent advances in personality psy-
chology. American Psychologist, 41, 1001–1003.
McDonald, R. J., & White, N. M. (1993). A triple dissociation of memory systems: Hippocampus, amygdala, and
dorsal striatum. Behavioral Neuroscience, 107, 3–22.
McGaugh, J. L. (1991). Neuromodulation and the storage of information: Involvement of the amygdaloid com-
plex. In R. G. Wister & H. J. Weingartner (Eds.), Perspectives on cognitive neuroscience (pp. 279–299). New
York: Oxford University Press.
McGaugh, J. L. (2000). Memory—a century of consolidation. Science, 287, 248–251.
McGeoch, J. A. (1932). Forgetting and the law of disuse. Psychological Review, 39, 352–370.
McGuire, W. J. (1961). A multiprocess model for paired associate learning. Journal of Experimental Psychology,
62, 335–347.
McIsaac, H. K., & Eich, E. (2004). Vantage point in traumatic memory. Psychological Science, 15, 248–253.
McKeithan, K. B., Reitman, J. S., Rueter, H. H., & Hirtle, S. C. (1981). Knowledge organization and skill differ-
ences in computer programmers. Cognitive Psychology, 13, 307–325.
McLenon, J., & Rogers, M. A. M. (2019). The fear of needles: A systematic review and meta-analysis. Journal of
Advanced Nursing, 75, 30–42.
McMurtry, C. M., et al. (2016). Exposure-based interventions for the management of individuals with high levels
of needle fear across the lifespan: A clinical practice guideline and call for further research. Cognitive Behavior
Therapy, 45, 217–235.
McNally, R. J. (1987). Preparedness and phobias: A review. Psychological Bulletin, 101, 283–303.
Mehus, C. J., & Patrick, M. E. (2021). Prevalence of spanking in US national samples of 35-year-old parents from
1993 to 2017. JAMA Pediatrics, 175(1), 92–94.
Melby-Verlag, M., & Hulme, C. (2013). Is working memory training effective: A meta-analytic review.
Developmental Psychology, 49, 270–291.
Melby-Lervåg, M., & Hulme, C. (2016) There is no convincing evidence that working memory training is effec-
tive: a reply. Psychonomic Bulletin & Review, 23, 324–330.
Melcher, J. M., & Schooler, J. W. (1996). The misremembrance of wines past: Verbal and perceptual expertise
differentially mediate verbal overshadowing of taste memory. Journal of Memory and Language, 35, 231–245.
 References 393

Melton, A. W. (1963). Implications of short-term memory for a general theory of memory. Journal of Verbal
Learning and Verbal Behavior, 2, 1–21.
Menzel, E. W. (1973). Chimpanzee spatial memory organization. Science, 182, 943–945.
Mery, F., & Kawecki, T. J. (2004). An operating cost of learning in Drosophila melanogaster. Animal Behavior, 68,
589–598.
Merzenich, M. M., Jenkins, W. M., William, M., Johnston, P., Schreiner, C., Miller, S. L., & Tallal, P. (1996).
Temporal processing deficits of language-learning impaired children ameliorated by training. Science, 271,
77–81.
Midkiff, E. E., & Bernstein, I. L. (1985). Targets of learned food aversions in humans. Physiology and Behavior,
34, 839–841.
Miller, G. A. (1956). The magical number seven plus-or-minus two: Some limits of our capacity for processing
information. Psychological Review, 63, 81–97.
Miller, L. C., Barrett, C. L., & Hampe, E. (1974). Phobias of childhood in a prescientific era. In A. Davis (Ed.), Child
personality and psychopathology: Current topics (pp. 89–134). New York: Wiley.
Miller, M. E., Adesso, V. J., Fleming, J. P., Gino, A., & Lauerman, R. (1978). Effects of alcohol on the storage and
retrieval processes of heavy social drinkers. Journal of Experimental Psychology: Human Learning and Memory,
4, 246–255.
Miller, N. E. (1948). Theory and experiment relating psychoanalytic displacement to stimulus-response general-
ization. Journal of Abnormal and Social Psychology, 43, 155–178.
Miller, N. E. (1959). Liberalization of basic S-R concepts: Extensions to conflict behavior, motivation, and social
learning. In S. Koch (Ed.), Psychology: A study of a science (Vol. 2, pp. 196–292). New York: McGraw-Hill.
Miller, N. E. (1985). The value of behavioral research on animals. American Psychologist, 40, 423–440.
Miller, N. E., & Dollard, J. (1941). Social learning and imitation. New Haven, CT: Yale University Press.
Miller, R. R., & Springer, A. D. (1973). Amnesia, consolidation, and retrieval. Psychological Review, 80, 69–79.
Milton, F., Muhlert, N., Pindus, D. M., Butler, C. R., Kapur, N., Graham, K. S., & Zeman, A. (2010). Remote mem-
ory deficits in transient epileptic amnesia. Brain, 133, 1368–1379.
Mineka, S. (1979). The role of fear in theories of avoidance learning, flooding, and extinction. Psychological
Bulletin, 86, 985–1010.
Mineka, S., & Zinbarg, R. (2006). A contemporary learning theory perspective on the etiology of anxiety disor-
ders. American Psychologist, 61, 10–26.
Misanin, J. R., Miller, R. R., & Lewis, D. J. (1968). Retrograde amnesia produced by reactivation of a consoli-
dated memory trace. Science, 160, 203–204.
Mischel, W., Shoda, Y., & Peake, P. (1988). The nature of adolescent competencies predicted by preschool delay
of gratification. Journal of Personality and Social Psychology, 54, 687–696.
Mischel, W., Shoda, Y., & Rodriguez, M. L. (1989). Delay of gratification in children. Science, 244, 933–938.
Mishkin, M., & Appenzeller, T. (1987, June). The anatomy of memory. Scientific American, 256(6), 80–89.
Miyake, A., & Friedman, N. P. (2012). The nature and organization of individual differences in executive functions:
Four general conclusions. Current Directions in Psychological Science, 21, 8–14.
Miyatsu, T., Nguyen, K., & McDaniel, M. A. (2018). Five popular study strategies: Their pitfalls and optimal imple-
mentations. Perspectives on Psychological Science, 13, 390–407.
Moffitt, T. E., Arseneault, L., Belsky, D., Dickson, N., Hancox, R. J., Harrington, H. …Caspi, A. (2011). A gradient
of childhood self-control predicts health, wealth, and public safety. Proceedings of the National Academy of
Science, 108, 2693–2698.
Montello, D. R. (2010). You are where? The function and frustration of you-are-here (YAH) maps. Spatial Cognition
& Computation, 10, 94–104.
Moore, T. E. (1982). Subliminal advertising: What you see is what you get. Journal of Marketing, 46, 38–47.
Morris, R. G. M. (1981). Spatial localization does not require the presence of local cues. Learning and Motivation,
12, 239–260.
394 References

Morris, P. E., & Fritz, C. O. (2000). The name game: Using retrieval practice to improve the learning of names.
Journal of Experimental Psychology: Applied, 6, 124–129.
Morrison, A. B., & Chein, J. M. (2011). Does working memory training work? The promise and challenges of
enhancing cognition by training working memory. Psychonomic Bulletin and Review, 18, 46–60.
Moscovith, M., Winocur, G., & McLachlan, D. (1986). Memory as assessed by recognition and reading time in
normal and memory-impaired people with Alzheimer’s Disease and other neurological disorders. Journal of
Experimental Psychology: General, 115, 331–347.
Mowrer, H. O. (1947). On the dual nature of learning—a reinterpretation of “conditioning” and “problem solv-
ing”. Harvard Educational Review, 17, 102–148.
Muller, G. E., & Pilzecker, A. (1900). Experimentalle Beitrage zur Lehre vom Gedachtnis. Zeitschrift fur
Psychologie, 1, 1–300.
Muller, K. D., et al. (2015). Verbal fluency and early memory decline: Results from the Wisconsin registry for
Alzheimer’s prevention. Archives of Clinical Neuropsychology, 30, 448–457.
Mynatt, C. R., & Doherty, M. E. (1999). Understanding human behavior. Boston: Allyn & Bacon.
Nadar, K., Schafe, G. E., & LeDoux, J. E. (2000). Fear memories require protein synthesis in the amygdala for
reconsolidation after retrieval. Nature, 406, 722–726.
Nairne, J. S., Pandeirada, J. N. S., & Thompson, S. R. (2008). Adaptive memory: The comparative value of sur-
vival processing. Psychological Science, 19, 176–180.
Naveh-Benjamin, M. (1991). A comparison of training programs intended for different types of test-anxious
students: Further support for an information processing model. Journal of Educational Psychology, 83,
134–139.
Naveh-Benjamin, M., & Ayres, T. J. (1986). Digit span, reading rate, and linguistic relativity. Quarterly Journal of
Experimental Psychology: Human Experimental Psychology, 38A, 739–751.
Neisser, U. (1967). Cognitive psychology. New York: Appleton, Century, Crofts.
Neisser, U. (1978). Memory: What are the important questions. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes
(Eds.), Practical aspects of memory (pp. 3–24). New York: Academic Press. (Reprinted in Neisser, 1982).
Neisser, U. (1982). Snapshots or benchmarks. In U. Neisser (Ed.), Memory observed (pp. 43–48). San Francisco:
Freeman.
Nelson, C. A. (1995). The ontogeny of human memory: A cognitive neuroscience approach. Developmental
Psychology, 31, 723–738.
Nelson, D. L., & Schreiber, T. A. (1992). Word concreteness and word structure as independent determinants of
recall. Journal of Memory and Language, 31, 237–260.
Nelson, K., & Fivush, R. (2004). The emergence of autobiographical memory: A social cultural developmental
theory. Psychological Review, 111, 486–511.
Nelson, T. O. (1976). Reinforcement and human memory. In W. K. Estes (Ed.), Handbook of learning and cogni-
tive processes (Vol. 3, pp. 207–246). Hillsdale, NJ: Erlbaum.
Nelson, T. O. (1978). Detecting small amounts of information in memory: Savings for nonrecognized items.
Journal of Experimental Psychology: Human Learning and Memory, 4, 453–468.
Nelson, T. O., & Gerler, D. (1984). Accuracy of feeling-of-knowing judgments for predicting perceptual identifica-
tion and relearning. Journal of Experimental Psychology: General, 113, 282–300.
Nelson, T. O., & Leonesio, R. J. (1988). Allocation of self-paced study time and the “labor-in-vain effect”. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 14, 676–686.
Nemiah, J. C. (1979). Dissociative amnesia. In J. H. Kihlstom & F. J. Evans (Eds.), Functional disorders of mem-
ory (pp. 303–323). Hillsdale, NJ: Erlbaum.
Neumann, D. L., Lipp, O. V., & Siddle, D. A. T. (1997). Conditioned inhibition of autonomic Pavlovian conditioning
in humans. Biological Psychology, 46, 223–233.
Neuringer, A., Kornell, N., & Olaf, M. (2001). Stability and variability in extinction. Journal of Experimental
Psychology: Animal Behavior Processes, 27, 79–94.
 References 395

Newell, A., & Rosenbloom, P. S. (1981). Mechanisms of skill acquisition and the law of practice. In J. R. Anderson
(Ed.), Cognitive skills and their acquisition (pp. 1–55). Hillsdale, NJ: Erlbaum.
Nickerson, R. S., & Adams, M. J. (1979). Long-term memory for a common object. Cognitive Psychology, 11,
287–307.
Nilsson, L. (1987). Motivated forgetting: Dissociation between performance data and subjective reports.
Psychological Research, 49, 183–188.
Nissen, M. J., & Bullemer, P. (1987). Attentional requirements of learning: Evidence from performance meas-
ures. Cognitive Psychology, 19, 1–32.
Noice, H., & Noice, T. (1997). The nature of expertise in professional acting: A cognitive view. Mahwah, NJ:
Erlbaum.
Noice, H., & Noice, T. (2006). What studies of actors and acting can tell us about memory and cognitive function-
ing. Current Directions in Psychological Science, 15, 14–18.
Norman, D. A. (1981). The categorization of action slips. Psychological Review, 88, 1–15.
Norman, G. R., Brooks, L. R., & Allen, S. W. (1989). Recall by expert medical practitioners and novices as a
record of processing attention. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15,
1166–1174.
Nyberg, L., Cabeza, R., & Tulving, E. (1996). PET studies of encoding and retrieval: The HERA model. Psychonomic
Bulletin & Review, 3, 135–148.
Oerter, R. (2003). Biological and psychological correlates of exceptional performance and development. Annals
of the New York Academy of Sciences, 999, 451–460.
Ogden, J. A., & Corkin, S. (1991). Memories of H. M. In W. C. Abraham, M. C. Corballis, & K. G. White (Eds.),
Memory mechanisms (pp. 195–215). Hillsdale, NJ: Erlbaum.
Ohman, A., Fredrikson, M., Hugdahl, K., & Rimmo, P. (1976). The premise of equipotentiality in human classical
conditioning: Conditioned electrodermal responses to potentially phobic stimuli. Journal of Experimental
Psychology: General, 105, 313–337.
Ohman, A., & Mineka, S. (2001). Fears, phobias, and preparedness: Toward an evolved module of fear and fear
learning. Psychological Review, 108, 483–522.
Olds, J., & Milner, P. (1954). Positive reinforcement produced by electrical stimulation of septal area and other
regions of rat brain. Journal of Comparative and Physiological Psychology, 47, 419–427.
Olson, G. M. (1976). An information-processing analysis of visual memory and habituation in infants. In T. J. Tighe
& R. N. Leaton (Eds.), Habituation: Perspectives from child development, animal behavior, and neurophysiology
(pp. 239–277). Hillsdale, NJ: Erlbaum.
Olsson, A., & Phelps, E. A. (2004). Learned fear of unseen faces after Pavlovian, observational, and instructed
fear. Psychological Science, 15, 822–828.
Olsson, A., & Phelps, E. A. (2007). Social learning of fear. Nature Neuroscience, 10, 1095–1102.
Olton, D. S. (1977, June). Spatial memory. Scientific American, 236, 82–98.
Olton, D. S. (1979). Mazes, maps, and memory. American Psychologist, 34, 583–596.
Olton, D. S., & Samuelson, R. J. (1976). Remembrance of places passed: Spatial memory in rats. Journal of
Experimental Psychology: Animal Behavior Processes, 2, 97–116.
Ornstein, P. A., Naus, M. J., & Liberty, C. (1975). Rehearsal and organizational processes in children’s memory.
Child Development, 46, 818–830.
Orr, S. P., Metzger, L. J., Lasko, N. B., Macklin, M. L., Peri, T., & Pitman, R. K. (2000). De novo conditioning in
trauma-exposed individuals with and without posttraumatic stress disorder. Journal of Abnormal Psychology,
109, 290–298.
Osgood, C. E. (1953). Method and theory in experimental psychology. New York: Oxford University Press.
Overman, W. H., Pate, B. J., Moore, K., & Peuster, A. (1996). Ontogeny of place learning in children as measured
in the radial arm maze, Morris search task, and the open field task. Behavioral Neuroscience, 110, 1205–1228.
Paivio, A. (1969). Mental imagery in associative learning and memory. Psychological Review, 76, 241–262.
396 References

Paller, K. A. (2017) Sleeping in a Brave New World: Opportunities for improving learning and clinical outcomes
through targeted memory reactivation. Current Directions in Psychological Science, 26, 532–537.
Palombo, D. J., Alain, C., Söderlund, H., Khuu, W., & Levine, B. (2015). Severely deficient autobiographical
memory (SDAM) in healthy adults: A new mnemonic syndrome. Neuropsychologia, 72, 105–118.
Papka, M., Ivy, R. B., & Woodruff-Pak, D. S. (1997). Eyeblink classical conditioning and awareness revisited.
Psychological Science, 8, 404–408.
Park, D. C., Smith, A. D., & Cavanaugh, J. C. (1990). Metamemories of memory researchers. Memory and
Cognition, 18, 321–327.
Parker, E. S., Cahill, L., & McGaugh, J. L. (2006). A case of unusual autobiographical memory. Neurocase, 12,
35–49.
Parkin, A. J., & Streete, S. (1988). Implicit and explicit memory in young children and adults. British Journal of
Psychology, 79, 361–369.
Pashler, H., McDaniel, M., Rohrer, D., & Bjork, R. (2009). Learning styles: Concepts and evidence. Psychological
Science in the Public Interest, 9, 105–119.
Pashler, H., Rohrer, D., Cepeda, N. J., & Carpenter, S. K. (2007). Enhancing learning and retarding forgetting:
Choices and consequences. Psychonomic Bulletin & Review, 14, 187–193.
Pavlov, I. P. (1927/1960). Conditioned reflexes (G. V. Anrep, Trans.) (Dover reprint, 1960). New York: Dover.
Pearce, J. M. (1997). Animal learning and cognition: An introduction (2nd ed.). East Sussex, UK: Psychology
Press.
Pearce, S. A., Isherwood, S., Hrouda, D., Richardson, P. H., Erskine, A., & Skinner, J. (1990). Memory and pain:
Tests of mood congruity and state dependent learning in experimentally induced and clinical pain. Pain, 43,
187–193.
Penfield, W. (1955). The permanent record of the stream of consciousness: Proceedings of the 14th international
congress of psychologists. Acta Psychologica, 11, 46–69.
Penfield, W., & Rasmussen, T. (1950). The cerebral cortex of man. New York: Macmillan.
Perfetti, C. A., & Lesgold, A. M. (1979). Coding and comprehension in skilled reading and implications for reading
instruction. In L. B. Resnick & P. A. Weaver (Eds.), Theory and practice of early reading (pp. 57–84). Hillsdale, NJ:
Erlbaum.
Perkins, C. C. (1968). An analysis of the concept of reinforcement. Psychological Review, 75, 155–172.
Perlmutter, M. (1978). What is memory aging the aging of? Developmental Psychology, 14, 330–345.
Peters, F., Villeneuve, S., & Belleville, S. (2014). Predicting progression to dementia in elderly subjects with mild
cognitive impairment using both cognitive and neuroimaging predictors. Journal of Alzheimer’s Disease, 38,
307–318.
Peterson, L. R., & Peterson, M. J. (1959). Short-term retention of individual items. Journal of Experimental
Psychology, 58, 193–198.
Petri, H. L., & Mishkin, M. (1994). Behaviorism, cognitivism and the neuropsychology of memory. American
Scientist, 82, 30–37.
Petros, T. V., Beckwith, B. E., & Anderson, M. (1990). Individual differences in the effects of time of day and
passage difficulty on prose memory in adults. British Journal of Psychology, 81, 63–72.
Pezdek, K., Whetstone, T., Reynolds, K., Askari, N., & Dougherty, T. (1989). Memory for real-world scenes: The
role of consistency with schema expectation. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 15, 587–595.
Pilley, J. W., & Reid, A. K. (2011). Border collie comprehends object names as verbal referents. Behavioural
Processes, 86, 184–195.
Pillsbury, W. B., & Sylvester, A. (1940). Retroactive and proactive inhibition in immediate memory. Journal of
Experimental Psychology, 27, 532–545.
Pinker, S. (1994). The language instinct. New York: Morrow.
Pintrich, P. R., & DeGroot, E. V. (1990). Motivational and self-regulated learning components of classroom aca-
demic performance. Journal of Educational Psychology, 82, 33–40.
 References 397

Pitman, R. K., Sanders, K. M., Zusman, R. M., Healy, A. R., Lasko, N. B., Cahill, L., & Orr, S. P. (2002). Pilot study
of secondary prevention of posttraumatic stress disorder with propranolol. Biological Psychiatry, 51, 189–192.
Pliner, O. (1982). The effects of mere exposure on liking for edible substances. Appetite, 3, 283–290.
Plous, S. (1996). Attitudes toward the use of animals in psychological research and education: Results from a
national survey of psychology majors. Psychological Science, 7, 352–358.
Polsky, R. (2000). Can aggression in dogs be elicited through the use of electronic pet containment systems?
Journal of Applied Animal Welfare Science, 3, 345–357.
Powers, R. B. (2006). The echo maker. New York: Farrar, Straus & Giroux.
Powers, R. B., & Osborne, J. G. (1976). Fundamentals of behavior. St. Paul, MN: West Publishing.
Premack, D. (1962). Reversibility of the reinforcement relation. Science, 136, 235–237.
Pressley, M., & Dennis-Rounds, J. (1980). Transfer of a mnemonic keyword strategy at two age levels. Journal
of Educational Psychology, 72, 575–582.
Pressley, M., & Ghatala, E. S. (1989). Metacognitive benefits of taking a test for children and young adolescents.
Journal of Experimental Child Psychology, 47, 430–450.
Pressley, M., Levin, J. R., & Ghatala, E. S. (1988). Strategy comparison opportunities promote long-term strategy
use. Contemporary Educational Psychology, 13, 157–168.
Pressley, M., Levin, J. R., Hall, J. W., Miller, G. E., & Berry, J. K. (1980). The keyword method and foreign word
acquisition. Journal of Experimental Psychology: Human Learning and Memory, 6, 163–173.
Pressley, M., McDaniel, M. A., Turnure, J. E., Wood, E., & Ahmad, M. (1987). Generation and precision of elab-
oration: Effects on intentional and incidental learning. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 13, 291–300.
Price, J., & Davis, B. (2008). The woman who can’t forget: The extraordinary story of living with the most
remarkable memory known to science—A memoir. New York: Free Press.
Rachman, S., & Lopatka, C. (1988). Return of fear: Underlearning and overlearning. Behaviour Research and
Therapy, 26, 99–104.
Rahhal, T. A., Hasher, L., & Colcombe, S. J. (2001). Instructional manipulations and age differences in memory:
Now you see them and now you don’t. Psychology and Aging, 16, 697–706.
Raichle, M. E. (1994). Images of the mind: Studies with modern imaging techniques. Annual Review of
Psychology, 45, 333–356.
Rajaram, S., & Roediger, H. L. (1993). Direct comparison of four implicit memory tests. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 19, 765–776.
Ramaswami, M. (2014). Network plasticity in adaptive filtering and behavioral habituation. Neuron 82,
1216–1229.
Ramond, C. K. (1954). Selective learning as a function of hunger. Journal of Experimental Psychology, 48,
265–270.
Rampon, C., Jiang, C. H., Dong, H., Tang, Y-P., Lockhart, D., Schultz, P., … & Hu, Y. (2000a). Effects of environ-
mental enrichment on gene expression in the brain. Proceedings of the National Academies of Science, 97,
12880–12884.
Rankin, C. H., et al. (2009). Habituation revisited: An updated and revised description of the behavioral charac-
teristics of habituation. Neurobiology of Learning and Memory, 92, 135–138.
Rasch, B., Buchel, C., Gais, S., & Born, J. (2007). Odor cues during slow-wave sleep prompt declarative memory
consolidation. Science, 315, 1426–1429.
Rauscher, F. H., Shaw, G. L., & Ky, K. N. (1993). Music and spatial task performance. Nature, 365, 611.
Reason, J. T. (1990). Human error. Cambridge: Cambridge University Press.
Reber, A. S. (1967). Implicit learning of artificial grammar. Journal of Verbal Learning and Verbal Behavior, 6,
855–863.
Reber, A. S. (1976). Implicit learning of synthetic languages: The role of instructional set. Journal of Experimental
Psychology: Human Learning and Memory, 2, 88–94.
398 References

Reber, A. S. (1993). Implicit learning and tacit knowledge: An essay on the cognitive unconscious. New York:
Oxford University Press.
Reber, R., Winkielmanm, P., & Schwartz, N. (1998). Effects of perceptual fluency on affective judgements.
Psychological Science, 9, 45–48.
Reed, G. (1979). Everyday anomalies of recall and recognition. In J. F. Kihlstrom & F. J. Evans (Eds.), Functional
disorders of memory (pp. 1–28). Hillsdale, NJ: Erlbaum.
Reed, P. (2001). Human response rates and causality judgments on schedules of reinforcement. Learning and
Motivation, 32, 332–348.
Reiser, B. J., Black, J. B., & Kalamarides, P. (1986). Strategic memory search processes. In D. C. Rubin (Ed.),
Autobiographical memory (pp. 100–121). Cambridge: Cambridge University Press.
Rescorla, R. A. (1967). Pavlovian conditioning and its proper control procedures. Psychological Review, 74,
71–80.
Rescorla, R. A. (1987). A Pavlovian analysis of goal directed behavior. American Psychologist, 42, 119–129.
Rescorla, R. A. (1988). Pavlovian conditioning: It’s not what you think it is. American Psychologist, 43, 151–161.
Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of
reinforcement and nonreinforcement. In A. H. Black & W. F. Prokasy (Eds.), Classical conditioning II (pp. 64–99).
New York: Appleton, Century, Crofts.
Restle, F. (1957). Discrimination of cues in mazes: A resolution of the “place-vs.-response” question.
Psychological Review, 64, 217–228.
Revelle, W., Humphreys, M. S., Simon, L., & Gilliland, K. (1980). The interactive effect of personality, time of day,
and caffeine: A test of the arousal model. Journal of Experimental Psychology: General, 109, 1–31.
Revusky, S., Coombes, S., & Pohl, R. W. (1982). US pre-exposure: Effects on flavor aversions produced by pair-
ing a poisoned partner with ingestion. Animal Learning & Behavior, 10, 83–90.
Richardson, R., Guanowsky, V., Ahlers, S. T., & Riccio, D. D. (1984). Role of body temperature in the onset of, and
recovery from, hypothermia-induced anterograde amnesia. Physiological Psychology, 12, 125–132.
Rivlin, G. (2004, May 9). Bet on it. New York Times Magazine, pp. 42–47, 74, 80, 81.
Roberts, B. W., Kuncel, N. R., Shiner, R., Caspi, A., & Goldberg, L. R. (2007). The power of personality: The
comparative validity of personality traits, socioeconomic status, and cognitive ability for predicting important life
outcomes. Perspectives on Psychological Science, 2, 313–345.
Robinson, E. S., & Brown, M. A. (1926) Effect of serial position upon memorization. American Journal of
Psychology, 37, 538–552.
Robinson, K. J., & Roediger, H. L. III. (1997). Associative processes in false recall and false recognition.
Psychological Science, 8, 231–237.
Roediger, H. L. III. (1980). The effectiveness of four mnemonics in ordering recall. Journal of Experimental
Psychology: Human Learning and Memory, 6, 558–568.
Roediger, H. L. III. (1990). Implicit memory: Retention without remembering. American Psychologist, 45,
1043–1056.
Roediger, H. L., III, & Karpicke, J. D. (2006). Test-enhanced learning. Psychological Science, 17, 249–255.
Roediger, H. L., III, & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in
lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 803–814.
Roediger, H. L., III, & McDermott, K. B. (2000). Tricks of memory. Current Directions in Psychological Science,
9, 123–127.
Roediger, H. L. III, & Smith, M. A. (2012). The “pure-study” learning curve: The learning curve without cumulative
testing. Memory & Cognition, 40, 989–1002.
Roediger, H. L., III, & Thorpe, L. A. (1978). The role of recall time in producing hypermnesia. Memory & Cognition,
6, 296–305.
Roediger, H. L. III, Watson, J. M., McDermott, K. B., & Gallo, D. A. (2001). Factors that determine false recall: A
multiple regression analysis. Psychonomic Bulletin & Review, 8, 385–407.
Rogers, L. (1989, September). Home, sweet-smelling home. Natural History, 61–66.
 References 399

Rogoff, B., & Chavajay, P. (1995). What’s become of research on the cultural basis of cognitive development.
American Psychologist, 50, 859–877.
Roitblat, H. L., Penner, R. H., & Nachtigall, P. E. (1990). Matching-to-sample by an echolocating dolphin. Journal
of Experimental Psychology: Animal Behavior Processes, 16, 85–95.
Rolls, B. J. (1990). The role of sensory-specific satiety in food intake and selection. In E. D. Capaldi & T. L. Powley
(Eds.), Taste, experience, and feeding (pp. 197–209). Washington, DC: American Psychological Association.
Rosenbaum, R. S., Kohler, S., Schacter, D. L., Moscovitvh, M., Westmacott, R., Black, S. E., … & Tulving, E.
(2005). The case of K. C.: Contributions of a memory-impaired person to memory theory. Neuropsychologia, 43,
989–1021.
Ross, M., & Wang, Q. (2010). Why we remember and what we remember: Culture and autobiographical mem-
ory. Perspectives on Psychological Science, 5, 401–409.
Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and coping with stress. American Psychologist, 41,
813–819.
Rothbaum, B. O., Anderson, P., Zimand, E., Hodges, L., Lang, D., & Wilson, J. (2006). Virtual reality exposure
therapy and standard (in vivo) exposure therapy in the treatment of fear of flying. Behavior Therapy, 37, 80–90.
Rotter, J. B. (1990). Internal versus external control of reinforcement: A case history of a variable. American
Psychologist, 45(4), 489–493.
Rovee, C. K., & Rovee, D. T. (1969). Conjugate reinforcement of infant exploratory behavior. Journal of
Experimental Child Psychology, 8, 33–39.
Rovee-Collier, C. K., Sullivan, M. W., Enright, M., Lucas, D., & Fagan, J. W. (1980). Reactivation of infant memory.
Science, 208, 1159–1162.
Royer, J. M. (1986). Designing instruction to produce understanding. In G. D. Phye & T. Andre (Eds.), Cognitive
classroom learning (pp. 83–113). New York: Academic Press.
Rozin, P., & Kalat, J. W. (1971). Specific hungers and poison avoidance as adaptive specializations of learning.
Psychological Review, 78, 459–486.
Rubin, D. C. (1977). Very long-term memory for prose and verse. Journal of Verbal Learning and Verbal Behavior,
16, 611–621.
Rubin, D. G., & Friendly, M. (1986). Predicting which words get recalled: Measures of free recall, availability,
goodness, emotionality, and pronounceability for 925 nouns. Memory & Cognition, 14, 79–94.
Rundus, D. (1971). Analysis of rehearsal processes in free recall. Journal of Experimental Psychology, 89,
63–77.
Ryan, E. B. (1992). Beliefs about memory changes across the adult life span. Journal of Gerontology:
Psychological Sciences, 47, 41–46.
Ryan, J. D., Althoff, R. R., Whitlow, S., & Cohen, N. J. (2000). Amnesia is a deficit in relational memory.
Psychological Science, 11, 454–461.
Sachs, J. S. (1967). Recognition memory for syntactic and semantic aspects of connected discourse. Perception
& Psychophysics, 2, 437–442.
Sacks, O. (1985). The man who mistook his wife for a hat. New York: Summit Books.
Saffran, J. R. (2009). What can statistical learning tell about infant learning? In A. Woodward (Ed.), Learning and
the infant mind (pp. 29–46). New York: Oxford University Press.
Saffran, J. R., Aslin, R. N., & Newport, E. L. (1996). Statistical learning by 8-month-old infants. Science, 274,
1926–1928.
Sahakian, B., & Morein-Zamir, S. (2007). Professor’s little helper. Nature, 450, 1157–1159.
Sala, G., & Gobet, G. (2019). Cognitive training does not enhance general cognition. Trends in Cognitive Sciences,
23, 9–20.
Salmon, K., Price, M., & Pereira, J. K. (2002). Factors associated with young children’s long-term recall of an
invasive medical procedure: A preliminary investigation. Journal of Developmental and Behavioral Pediatrics, 23,
347–352.
Salthouse, T. A. (1994). The aging of working memory. Neuropsychology, 8, 535–543.
400 References

Sampaio, C., & Wang, R. F. (2009). Category-based errors and the accessibility of unbiased spatial memories: A
retrieval model. Journal of Experimental Psychology: Learning, Memory & Cognition, 5, 1331–1337.
Saufley, W. H., Otaka, S. R., & Bavaresco, J. L. (1985). Context effects: Classroom tests and context independ-
ence. Memory & Cognition, 13, 522–528.
Schacter, D. L. (1983). Amnesia observed: Remembering and forgetting in a natural environment. Journal of
Abnormal Psychology, 92, 236–242.
Schacter, D. L., Verfaellie, M., & Pradere, D. (1996). The neuropsychology of memory illusions: False recall and
recognition in amnesic patients. Journal of Memory and Language, 35, 319–334.
Schacter, D. L., Wang, P. L., Tulving, E., & Freedman, M. (1982). Functional retrograde amnesia: A quantitative
case study. Neuropsychologia, 20, 523–532.
Schafe, G. E., & LeDoux, J. E. (2000). Memory consolidation of auditory Pavlovian fear conditioning requires
protein synthesis and protein kinase a in the amygdala. The Journal of Neuroscience, 20, RC96, 1–5.
Scheier, M. F., & Carver, C. S. (1993). On the power of positive thinking: The benefits of being optimistic. Current
Directions in Psychological Science, 2, 26–30.
Schendel, J. D., & Hagman, J. D. (1982). On sustaining procedural skills over a prolonged retention interval.
Journal of Applied Psychology, 67, 605–610.
Schiller, D., Monfils, M.-H., Raio, C. M., Johnson, D. C., LeDoux, J. E., & Phelps, E. A. (2010). Preventing the
return of fear in humans using reconsolidation update mechanisms. Nature, 463, 49–53.
Schmidt, R. A., & Bjork, R. A. (1992). New conceptualizations of practice: Common principles in three paradigms
suggest new concepts for training. Psychological Science, 3, 207–217.
Schmidt, R. A., Young, D. E., Swinnen, S., & Shapiro, D. C. (1989). Summary knowledge of results for skill acqui-
sition: Support for the guidance hypothesis. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 15, 352–359.
Schneider, V. I., Healy, A. F., & Bourne, L. E., Jr. (2002). What is learned under difficult conditions is hard to forget:
Contextual interference effects in foreign vocabulary acquisition, retention, and transfer. Journal of Memory and
Language, 46, 419–440.
Schooler, J. W., Ryan, R. S., & Reder, L. (1996). The costs and benefits of verbally rehearsing memory for faces.
In D. Herrmann, C. McEvoy, C. Hertzog, P. Hertel, & M. K. Johnson (Eds.), Basic and applied memory research
(Vol. 2., pp. 51–65). Mahwah, NJ: Erlbaum.
Schultz, D. P., & Schultz, S. E. (1996). A history of modern psychology (6th ed.). Fort Worth, TX: Harcourt Brace
& Company.
Schwartz, B. L. (1999). Sparkling at the end of the tongue: The etiology of tip-of-the-tongue phenomenology.
Psychonomic Bulletin & Review, 6, 379–393.
Schwartz, B. N. (1978). Psychology of learning and behavior. New York: Norton.
Scoville, W. B., & Milner, B. (1957). Loss of recent memory after bilateral hippocampal lesions. Journal of
Neurology, Neurosurgery and Psychiatry, 20, 11–19.
Scrivener, E., & Safer, M. A. (1988). Eyewitnesses show hyperamnesia for details about a violent event. Journal
of Applied Psychology, 73, 371–377.
Seamon, J. G., Luo, C. R., Kopecky, J. J., Price, C. A., Rothschild, L., Fung, N. S., & Schwartz, M. A. (2002). Are
false memories more difficult to forget than accurate memories? The effect of retention interval on recall and
recognition. Memory and Cognition, 30, 1054–1064.
Sears, S. F., Jr., Kovacs, A. H., Azzarello, L., Larsen, K., & Conti, J. B. (2004). Innovations in health psychology:
The psychosocial care of adults with implantable cardioverter defibrillators. Professional Psychology: Research
and Practice, 35, 520–526.
Sehulster, J. R. (1989). Content and temporal structure of autobiographical knowledge: Remembering twen-
ty-five seasons of the Metropolitan Opera. Memory & Cognition, 17, 590–606.
Seligman, M. E. P. (1970). On the generality of the laws of learning. Psychological Review, 77, 406–418.
Seligman, M. E. P. (1972). Phobias and preparedness. In M. E. P. Seligman & J. L. Hager (Eds.), Biological bound-
aries of learning (pp. 451–462). New York: Appleton, Century, Crofts.
Seligman, M. E. P. (1975). Helplessness: On depression, development, and death. San Francisco: Freeman.
 References 401

Seligman, M. E. P., & Johnson, J. C. (1973). A cognitive theory of avoidance learning. In F. J. McGuigan & D. B.
Lumsden (Eds.), Contemporary approaches to conditioning and learning (pp. 69–110). Washington, DC: Winston.
Seligman, M. E. P., Maier, S. F., & Geer, J. H. (1968). Alleviation of learned helplessness in the dog. Journal of
Abnormal Psychology, 73, 256–262.
Semb, G. B., Ellis, J. A., & Araujo, J. (1993). Long-term memory for knowledge learned in school. Journal of
Educational Psychology, 85, 305–316.
Shallice, T. (1988). From neuropsychology to mental structure. Cambridge: Cambridge University Press.
Shallice, T., & Burgess, P. (1991). Deficits in strategy application following frontal lobe damage in man. Brain, 114,
727–741.
Shallice, T., & Warrington, E. K. (1970). Independent functioning of verbal memory stores: A neuropsychological
study. Quarterly Journal of Experimental Psychology, 22, 261–273.
Shand, M. A., & Klima, E. S. (1981). Nonauditory suffix effects in congenitally deaf signers of American Sign
Language. Journal of Experimental Psychology: Human Learning and Memory, 6, 464–474.
Shanks, D. R. (2010). Learning: From association to cognition. Annual Review of Psychology, 61, 273–301.
Shanks, D. R., & Dickinson, A. (1991). Instrumental judgment and performance under variations in action-out-
come contingency and contiguity. Memory & Cognition, 19, 353, 360.
Sharp, D., Cole, M., & Lave, C. (1979). Education and cognitive development: The evidence from experimental
research. Monographs of the Society for Research in Child Development, 44 (1–2, Serial No. 178).
Sharps, M. J., Welton, A. L., & Price, J. L. (1993). Gender and task in the determination of spatial cognitive
performance. Psychology of Women Quarterly, 17, 71–83.
Shaughnessy, J. J., Zimmerman, J., & Underwood, B. J. (1972). Further evidence on the MP-DP effect in free-re-
call learning. Journal of Verbal Learning and Verbal Behavior, 11, 1–12.
Sheen, M., Kemp, S., & Rubin, D. (2001). Twins dispute memory ownership: A new false memory phenomenon.
Memory and Cognition, 29, 779–788.
Sheffield, F. D., & Roby, T. B. (1950). Reward value of a non-nutritive sweet taste. Journal of Comparative and
Physiological Psychology, 43, 471–481.
Sherry, D. F., & Schacter, D. L. (1987). The evolution of multiple memory systems. Psychological Review, 94,
439–454.
Sherwin, B. B. (1994). Estrogenic effects on memory in women. Annals of the New York Academy of Science,
743, 213–231.
Shettleworth, S. (1975). Reinforcement and the organization of behavior in golden hamsters: Hunger, environ-
ment, and food reinforcement. Journal of Experimental Psychology: Animal Behavior Processes, 1, 56–87.
Shimamura, A. P., Salmon, D. P., Squire, L. R., & Butters, N. (1987). Memory dysfunction and word priming in
amnesia. Behavioral Neuroscience, 101, 347–351.
Shimp, T., Stuart, E. W., & Engle, R. W. (1991). A program of classical conditioning experiments testing variations
in the CS and context. Journal of Consumer Research, 18, 1–12.
Sholl, M. J. (1987). Cognitive maps as orienting schemata. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 13, 615–628.
Shore, D. I., Stanford, L., MacInnes, J. W., Klein, R. M., & Brown, R. E. (2001). Of mice and men: Virtual Hebb-
Williams mazes permit comparisons across species. Cognitive, Affective, & Behavioral Neuroscience, 1,
83–89.
Siddle, D. A. T. (1985). Effects of stimulus omission and stimulus change on dishabituation of the skin conduct-
ance response. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 206–216.
Siddle, D. A. T., Kuiack, M., & Kroese, B. S. (1983). The orienting reflex. In A. Gale & J. A. Edwards (Eds.),
Physiological correlates of human behavior (pp. 149–170). London: Academic Press.
Siddle, D. A. T., & Lipp, O. V. (1997). Orienting, habituation, and information processing: The effects of omission,
the role of expectancy, and the problem of dishabituation. In P. J. Lang, R. F. Simons, & M. T. Balaban (Eds.),
Attention and orienting: Sensory and motivational processes (pp. 23–40). Mahwah, NJ: Erlbaum.
Siegel, S. (1974). Flavor preexposure and “learned safety”. Journal of Comparative and Physiological Psychology,
87, 1073–1082.
402 References

Siegel, S. (1982). Pharmacological habituation and learning. In M. L. Commons, R. J. Herrnstein, & A. R. Wagner
(Eds.), Quantitative analyses of behavior (Vol. 3, pp. 195–217). Cambridge, MA: Ballinger.
Siegel, S. (1984). Pavlovian conditioning and heroin overdose: Reports by overdose victims. Bulletin of the
Psychonomic Society, 22, 428–430.
Siegel, S. (1991). Feedforward processes in drug tolerance and dependence. In R. G. Lister & H. J. Weingartner
(Eds.), Perspectives on cognitive neuroscience (pp. 405–416). New York: Oxford University Press.
Siegel, S., Hinson, R. E., Krank, M. D., & McCully, J. (1982). Heroin “overdose” death: Contribution of drug-as-
sociated environmental cues. Science, 216, 436–437.
Silverman, I., Choi, J., Mackewn, A., Fisher, M., Moro, J., & Olshansky, E. (2000). Evolved mechanisms under-
lying wayfinding: Further studies on the hunter-gatherer theory of spatial sex differences. Evolution and Human
Behavior, 21, 210–213.
Silverman, I., & Eals, M. (1992). Sex differences in spatial abilities: Evolutionary theory and data. In J. H. Barkow,
L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp.
533–549). New York: Oxford University Press.
Simcock, G., & Hayne, H. (2002). Breaking the barrier? Children fail to translate their preverbal memories into
language. Psychological Science, 13, 225–231.
Simon, D. A., & Bjork, R. A. (2001). Metacognition in motor learning. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 27, 907–912.
Simons, D. J., Boot, W. R., Charness, N., Gathercole, S. E., Chabris, C. F., Hambrick, D. Z., & Stine-Morrow, E.
A. (2016). Do “brain training” programs work? Psychological Science in the Public Interest, 17(3), 103–186.
Simons, D. J., & Chabris, C. F. (2011). What people believe about how memory works: A representative survey
of the U.S. population. Plos One, 6(8), e22757.
Singh, D. (1970). Preference for bar pressing to obtain reward over freeloading in rats and children. Journal of
Comparative and Physiological Psychology, 73, 320–327.
Skinner, B. F. (1938). The behavior of organisms. New York: Appleton, Century, Crofts.
Skinner, B. F. (1953). Science and human behavior. New York: Macmillan.
Skinner, B. F. (1956). A case history in scientific method. American Psychologist, 11, 221–233.
Skinner, B. F. (1971). Beyond freedom and dignity. New York: Knopf.
Skinner, N. F. (1985). University grades and time of day of instruction. Bulletin of the Psychonomic Society, 23, 67.
Skurnik, I., Yoon, C., Park, D. C., & Schwarz, N. (2005). How warnings about false claims become recommenda-
tions. Journal of Consumer Research, 31, 713–724.
Slamecka, N. J., & Graf, P. (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental
Psychology: Human Learning and Memory, 4, 592–604.
Slater, L. (2003, November 2). The cruelest cure. New York Times Magazine, pp. 34–37.
Smeets, T., Otgaar, H., Candel, I., & Wolf, O. T. (2008). True or false? Memory is differentially affected by stress-in-
duced cortisol elevations and sympathetic activity at consolidation and retrieval. Psychoneuroendocrinology, 33,
1378–1386.
Smith, M. E., & Farah, M. J. (2011). Are prescription stimulants “smart pills”? The epidemiology and cognitive
neuroscience of prescription stimulant use by normal healthy individuals. Psychological Bulletin, 137, 717–741.
Smith, S. M. (1979). Remembering in and out of context. Journal of Experimental Psychology: Human Learning
and Memory, 5, 460–471.
Smith, S. M. (1985). Background music and context-dependent memory. American Journal of Psychology, 98,
591–603.
Smith, S. M., & Vela, E. (2001). Environmental context-dependent memory: A review and meta-analysis.
Psychonomic Bulletin & Review, 8, 203–220.
Snowman, J. (1986). Learning tactics and strategies. In G. D. Phye & T. Andre (Eds.), Cognitive classroom learn-
ing (pp. 243–275). New York: Academic Press.
Snyder, J., Schrepferman, L., & St. Peter, C. (1997). Origins of antisocial behavior. Behavior Modification, 21,
187–215.
 References 403

Sokolov, Y. N. (1963). Perception and the conditioned reflex. London: Pergamon Press.
Solomon, P. R., & Pendlebury, W. W. (1992). Aging and memory: A model systems approach. In L. R. Squire &
N. Butters (Eds.), Neuropsychology of memory (2nd ed., pp. 262–276). New York: Guilford Press.
Solomon, R. L., Turner, L. H., & Lessac, M. S. (1968). Some effects of delay of punishment on resistance to
temptation in dogs. Journal of Personality and Social Psychology, 8, 233–238.
Spear, N. E., & Riccio, D. C. (1994). Memory: Phenomenon and principles. Boston: Allyn & Bacon.
Spence, K. W. (1956). Behavior theory and conditioning. New Haven, CT: Yale University Press.
Sporer, S. L. (1991). Deep-deeper-deepest? Encoding strategies and the recognition of human faces. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 17, 323–333.
Squire, L. R. (1987). Memory and brain. New York: Oxford University Press.
Squire, L. R. (1992). Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans.
Psychological Review, 99, 195–231.
Squire, L. R., & Cohen, N. J. (1984). Human memory and amnesia. In J. L. McGaugh, G. Lynch, & N. Weinberger
(Eds.), Neurobiology of learning and memory (pp. 3–64). New York: Guilford Press.
Squire, L. R., Knowlton, B., & Musen, G. (1993). The structure and organization of memory. Annual Review of
Psychology, 44, 453–495.
Squire, L. R., & Slater, P. C. (1975). Forgetting in very long-term memory as assessed by an improved question-
naire technique. Journal of Experimental Psychology: Human Learning and Memory, 1, 50–54.
Squire, L. R., Slater, P. C., & Miller, P. L. (1981). Retrograde amnesia and bilateral electroconvulsive therapy.
Archives of General Psychiatry, 38, 89–95.
Staddon, J. (1995, February). On responsibility and punishment. The Atlantic Monthly, 88–94.
Standing, L., Conezio, J., & Haber, R. N. (1970). Perception and memory for pictures: Single trial learning of 2560
visual stimuli. Psychonomic Science, 19, 73–74.
Steele, K. M., Bass, K. E., & Crook, M. D. (1999). The mystery of the Mozart effect: Failure to replicate.
Psychological Science, 10, 366–369.
Stevens, A., & Coupe, P. (1978). Distortions in judged spatial relations. Cognitive Psychology, 13, 422–437.
Stevenson, H. W., & Zigler, E. F. (1958). Probability learning in children. Journal of Experimental Psychology, 56,
185–192.
Stickgold, R., Hobson, J. A., & Fosse, M. (2001). Sleep, learning, and dreams: Off-line memory reprocessing.
Science, 294, 1052–1057.
Stigler, J. W., Lee, S. Y., & Stevenson, H. W. (1986). Digit memory in Chinese and English: Evidence for a tem-
porally limited store. Cognition, 23, 1–20.
Stolz, S. B., & Lott, D. F. (1964). Establishment in rats of a persistent response producing a net loss of reinforce-
ment. Journal of Comparative and Physiological Psychology, 57, 147–149.
Strayer, D., & Drews, F. (2007). Cell-phone-induced driver distraction. Current Directions in Psychological
Science, 16, 128–131.
Sullivan, R. M., Taborsky-Barbar, S., Mendoza, R., Ition, A., & Leon, M. (1991). Olfactory classical conditioning in
neonates. Pediatrics, 87, 511–518.
Sutherland, A. (2006). Kicked, bitten and scratched. New York: Viking/Penguin.
Swanson, H. L. (1993). Working memory in learning disability subgroups. Journal of Experimental Child
Psychology, 56, 87–114.
Swinnen, S. P. (1990). Interpolated activities during the knowledge-of-results delay and post-knowledge-of-
results interval: Effects on performance and learning. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 16, 692–705.
Swinnen, S. P., Schmidt, R. A., Nicholson, D. E., & Shapiro, D. C. (1990). Information feedback for skill acquisition:
Instantaneous knowledge of results degrades learning. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 16, 706–716.
Talarico, J. M., & Rubin, D. C. (2003). Confidence, not consistency, characterizes flashbulb memories.
Psychological Science, 14, 455–461.
404 References

Talland, G. A. (1968). Disorders of memory and learning. Middlesex, UK: Penguin.

Talwar, S. K., Xu, S., Hawley, E. S., Weiss, S. A., Moxon, K. A., & Chapin, J. K. (2002, May 2). Rat navigation by
remote control. Nature, 417, 37–38.
Tang, Y.-P., Shimizu, E., Dube, G., Rampon, C., Kerchner, G., Zhuo, M., … & Tsien, J. (1999). Genetic enhance-
ment of learning and memory in mice. Nature, 401, 63–69.
Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). Higher self-control predicts good adjustment, less
pathology, better grades, and interpersonal success. Journal of Personality, 72, 271–324.
Tenenbaum, G., Tehan, G., Stewart, G., & Christensen, S. (1999). Recalling a floor routine: The effects of skill and
age on memory for order. Applied Cognitive Psychology, 13, 101–123.
Terry, W. S. (1983). Effects of food primes on instrumental acquisition and performance. Learning and Motivation,
14, 107–122.
Terry, W. S. (1987). Everyday forgetting: Data from a diary study. Psychological Reports, 62, 299–303.
Terry, W. S. (1995). When proper names are not forgotten: Recall of eponymous medical disorders. Perceptual
and Motor Skills, 81, 923–928.
Terry, W. S., & Anthony, S. G. (1980). Arousal and short-term memory: Effects of caffeine and trial spacing on
delayed alternation performance. Animal Learning and Behavior, 8, 368–374.
Thapar, A., Petrill, S. A., & Thompson, L. A. (1994). The heritability of memory in the Western Reserve Twin
Project. Behavior Genetics, 24, 155–160.
Thompson, L. A., Detterman, D. K., & Plomin, R. (1991). Associations between cognitive abilities and scholastic
achievement: Genetic overlap but environmental differences. Psychological Science, 2, 158–165.
Thompson, R. F. (2005). In search of memory traces. Annual Review of Psychology, 56, 1–23.
Thompson, R. F., & Gluck, M. A. (1991). Brain substrates of basic associative learning and memory. In R. G.
Wister & H. J. Weingartner (Eds.), Perspectives on cognitive neuroscience (pp. 24–45). New York: Oxford
University Press.
Thompson, R. F., & Spencer, W. A. (1966). Habituation: A model phenomenon for the study of neuronal sub-
strates of behavior. Psychological Review, 73, 16–43.
Thorndike, E. L. (1898). Animal intelligence: An experimental study of the associative processes in animals.
Psychological Review Monograph Supplements, 2 (Whole No. 8, 1–109).
Thorndike, E. L. (1911). Animal intelligence: Experimental studies. New York: Macmillan.
Thorndyke, P. W., & Stasz, C. (1980). Individual differences in procedures for knowledge acquisition from maps.
Cognitive Psychology, 12, 137–175.
Timberlake, W., & Allison, J. (1974). Response deprivation: An empirical approach to instrumental performance.
Psychological Review, 81, 146–164.
Tinklepaugh, O. L. (1928). An experimental study of representative factors in monkeys. Journal of Comparative
Psychology, 8, 197–236.
Tolman, E. C. (1933). Sign-gestalt or conditioned reflex? Psychological Review, 40, 391–411.
Tolman, E. C. (1948). Cognitive maps in rats and men. Psychological Review, 55, 189–208.
Tolman, E. C., & Honzik, C. H. (1930). Introduction and removal of reward, and maze performance in rats.
University of California Publications in Psychology, 4, 257–275.
Tolman, E. C., Ritchie, B. F., & Kalish, D. (1946). Studies in spatial learning: II. Place learning versus response
learning. Journal of Experimental Psychology, 36, 221–229.
Tolman, E. C., Ritchie, B. F., & Kalish, D. (1947). Studies in spatial learning: IV. The transfer of place learning to
other starting points. Journal of Experimental Psychology, 37, 39–47.
Tonegawa, S., Pignatelli, M., Roy, D. S., & Ryan, T. J. (2015). Memory engram storage and retrieval. Current
Opinion in Neurobiology, 35, 101–109.
Tough, P. (2012). How children succeed. Boston: Houghton Mifflin Harcourt.
Touron, D. R. (2015). Memory avoidance by older adults: When “old dogs” won’t perform their “new tricks”.
Current Directions in Psychological Science, 24, 170–176.
 References 405

Tryon, R. C. (1940). Genetic differences in maze learning in rats. Yearbook of the National Society for the Study
of Education, 39, 111–119.
Tuber, D. S., Hothersall, D., & Voith, V. L. (1974). Animal clinical psychology: A modest proposal. American
Psychologist, 29, 762–766.
Tuber, D. S., Miller, D. D., Caris, K. A., Halter, R., Linden, F., & Hennessy, M. B. (1999). Dogs in animal shelters:
Problems, suggestions, and needed expertise. Psychological Science, 10, 379–386.
Tully, T. (1996). Discovery of genes involved with learning and memory: An experimental synthesis of Hirschian
and Benzerian perspectives. Proceedings of the National Academy of Sciences, 93, 13460–13467.
Tulving, E. (1962). Subjective organization in free recall of “unrelated” words. Psychological Review, 69,
344–354.
Tulving, E. (1983). Elements of episodic memory. Oxford, UK: Clarendon Press.
Tulving, E. (1985). How many memory systems are there? American Psychologist, 40, 385–398.
Tulving, E. (1989). Remembering and knowing the past. American Scientist, 77, 361–367.
Tulving, E., & Pearlstone, Z. (1966). Availability versus accessibility of information in memory for words. Journal
of Verbal Learning and Verbal Behavior, 5, 381–391.
Tulving, E., & Schacter, D. L. (1990). Priming and human memory systems. Science, 247, 301–306.
Tulving, E., Schacter, D. L., & Stark, H. A. (1982). Priming effects in word-fragment completion are independent
of recognition memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 8, 336–342.
Tulving, E., & Thompson, D. M. (1973). Encoding specificity and retrieval processes in episodic memory.
Psychological Review, 80, 352–373.
Tulving, E., & Watkins, M. J. (1975). Structure of memory traces. Psychological Review, 82, 261–275.
Umanath, S., & Coane, J. H. (2020). Face validity of remembering and knowing: Empirical consensus and disa-
greement between participants and researchers. Perspectives on Psychological Science, 15, 1400–1422.
Underwood, B. J. (1964). The representativeness of rote verbal learning. In A. W. Melton (Ed.), Categories of
human learning (pp. 47–56). New York: Academic Press.
Underwood, B. J. (1983). Attributes of memory. Glenview, IL: Scott, Foresman.
Underwood, B. J., Boruch, R. F., & Malmi, R. A. (1978). Composition of episodic memory. Journal of Experimental
Psychology: General, 107, 393–419.
Unsworth, N., & Engle, R. W. (2007). On the division of short-term and working memory: An examination of
simple and complex span and their relation to higher order abilities. Psychological Bulletin, 133,
1038–1066.
Urcelay, G. P., Wheeler, D. S., & Miller, R. R. (2009). Spacing extinction trials alleviates renewal and spontaneous
recovery. Learning & Behavior, 37, 60–73.
van der Kolk, B. A. (1994). The body keeps the score: Memory and the evolving psychobiology of post-traumatic
stress. Harvard Review of Psychiatry, 1, 253–265.
Vance, A., et al. (2018). Tuning out security warnings: Examination of habituation through tracking and field
experiments. MIS Quarterly, 42, 1–26.
Vaughn, W., & Greene, S. L. (1984). Pigeon visual memory capacity. Journal of Experimental Psychology: Animal
Behavior Processes, 10, 256–271.
Verplanken, B., & Wood, W. (2006). Interventions to break and create consumer habits. Journal of Public Policy
& Marketing, 25, 90–103.
Volk, H. E., McDermott, K. B., Roediger, H. L., & Todd, R. D. (2006). Genetic influences on free and cued recall
in long-term memory tasks. Twins Research and Human Genetics, 9, 623–631.
Waddill, P. J., & McDaniel, M. A. (1998). Distinctiveness effects in recall: Differential processing or privileged
retrieval? Memory & Cognition, 26, 108–120.
Wagner, A. D., Schacter, D. L., Rotte, M., Koutstaal, W., Maril, A., Dale, A. M. … & Buckner, R. L. (1998). Building
memories: Remembering and forgetting of verbal experiences as predicted by brain activity. Science, 281,
1188–1191.
406 References

Wagner, A. R. (1969). Frustrative nonreward: A variety of punishment? In B. A. Campbell & R. M. Church (Eds.),
Punishment and aversive behavior (pp. 157–181). New York: Appleton, Century, Crofts.
Wagner, A. R. (1976). Priming in STM: An information processing mechanism for self-generated or retrieval-gen-
erated depression in performance. In T. J. Tighe & R. N. Leaton (Eds.), Habituation: Perspectives from child
development, animal behavior, and neurophysiology (pp. 95–128). Hillsdale, NJ: Erlbaum.
Wagner, A. R., & Brandon, S. E. (1989). Evolution of a structured connectionist model of Pavlovian conditioning
(AESOP). In S. B. Klein & R. B. Mowrer (Eds.), Contemporary learning theories: Pavlovian conditioning and the
status of traditional learning theory (pp. 149–189). Hillside, NJ: Erlbaum.
Wagner, A. R., & Rescorla, R. A. (1972). Inhibition in Pavlovian conditioning: Application of a theory. In R. A.
Boakes & M. S. Halliday (Eds.), Inhibition and learning (pp. 301–336). London: Academic Press.
Wagner, A. R., Siegel, S., Thomas, E., & Ellison, G. D. (1964). Reinforcement history and the extinction of a
conditioned salivary response. Journal of Comparative and Physiological Psychology, 58, 354–358.
Wagner, A. R., & Terry, W. S. (1975). Backward conditioning to a CS following an expected vs. a surprising UCS.
Animal Learning & Behavior, 3, 370–374.
Wagner, D. A. (1978). Memories of Morocco: The influence of age, schooling, and environment on memory.
Cognitive Psychology, 10, 1–28.
Wagner, I. C., et al. (2021, March 3). Durable memories and efficient neural coding through mnemonic training
using the method of loci. Science Advances, 7: eabc7606.
Walcott, C. (1989, November). Show me the way to go home. Natural History, pp. 40–46.
Walker, M. P., Brakefield, T., Seidman, J., Morgan, A., Hobson, J. A., & Stickgold, R. (2003). Sleep and the time
course of motor skill learning. Learning and Memory, 10, 275–284.
Walker, S. (1987). Animal learning: An introduction. London: Routledge & Kegan Paul.
Wang, Q. (March, 2008). Where does our past begin? A sociocultural perspective on the phenomenon of child-
hood amnesia. American Psychological Association, Psychological Science Agenda, online (/science/about/
psa/2008/03/).
Wang, Q., Leichtman, M. D., & White, S. H. (1998). Childhood memory and self-description in young Chinese
adults: The impact of growing up an only child. Cognition, 69, 1, 73–103.
Wammes, J. D., Meade, M. E., & Fernandes, M. A. (2016). The drawing effect: Evidence for reliable and robust
memory benefits in free recall. Quarterly Journal of Experimental Psychology, 69, 1752–1776.
Waring, J. D., Payne, J. D., Schacter, D. L., & Kensinger, E. A. (2010). Impact of individual differences upon
emotion-induced memory trade-offs. Cognition and Emotion, 24, 150–167.
Warrington, E. K., & Weiskrantz, L. (1968a). A new method of testing long-term retention with special reference
to amnesic patients. Nature, 217, 972–974.
Warrington, E. K., & Weiskrantz, L. (1968b). A study of learning and retention in amnesic patients.
Neuropsychologia, 6, 283–291.
Wasserman, E. A., Dorner, W. W., & Kao, S. F. (1990). Contributions of specific cell information to judgments of
interevent contingency. Journal of Experimental Psychology: Learning, Memory, and Cognition, 16, 509–521.
Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3,
1–20.
Watson, J. M., & Strayer, D. L. (2010). Supertaskers: Profiles in extraordinary multitasking ability. Psycho-nomic
Bulletin and Review, 17, 479–485.
Waugh, N. C., & Norman, D. A. (1965). Primary memory. Psychological Review, 72, 89–104.
Weingartner, H., Adefris, W., Eich, J. E., & Murphy, D. L. (1976). Encoding-imagery specificity in alcohol state-de-
pendent learning. Journal of Experimental Psychology: Human Learning and Memory, 2, 83–87.
Weingartner, H., Miller, H., & Murphy, D. L. (1977). Mood-dependent retrieval of verbal associations. Journal of
Abnormal Psychology, 86, 276–284.
Weinstock, S. (1954). Resistance to extinction of a running response following partial reinforcement under
widely spaced trials. Journal of Comparative and Physiological Psychology, 48, 318–322.
 References 407

Weiss, J. M. (1977). Psychosomatic disorders. In J. D. Maser & M. E. P. Seligman (Eds.), Psychopathology:

Experimental models (pp. 232–269). San Francisco: Freeman.
Weldon, M. S., & Roediger, H. L. (1987). Altering retrieval demands reverses the picture superiority effect.
Memory & Cognition, 15, 269–280.
Werker, J. F. (1989, January–February). Becoming a native listener. American Scientist, 77, 54–59.
Wessel, G., Ziemkiewicz, C., Chang, R., & Sauda, E. (2010). The case for a spatial framework for semantic
information. Advanced Visual Interfaces (AVI) Conference in Rome, Conference Proceedings. Charlotte
Visualization Center Technical Report CVC-UNCC-10-07.
West, R. L., Crook, T. H., & Barron, K. L. (1992). Everyday memory performance across the life span: Effects of
age and noncognitive individual differences. Psychology and Aging, 7, 72–82.
Westbrook, J. I., Woods, A., Rob, I., Dunsmuir, W. T. M., & Day, R. O. (2010). Association of interruptions with an
increased risk and severity of medication administration errors. Archives of Internal Medicine, 170, 683–690.
Wetherington, C. L. (1982). Is adjunctive behavior a third class of behavior? Neuroscience & Biobehavioral
Reviews, 6, 329–350.
Wheeler, M. A., & Roediger, H. L. (1992). Disparate effects of repeated testing: Reconciling Ballard’s (1914) and
Bartlett’s (1932) results. Psychological Science, 3, 240–245.
White, N. H., & Milner, P. M. (1992). The psychobiology of reinforcers. Annual Review of Psychology, 43, 443–471.
Whitlow, J. W., & Wagner, A. R. (1984). Memory and habituation. In H. V. S. Peeke & L. Petrinovich (Eds.),
Habituation, sensitization, and behavior (pp. 103–153). New York: Academic Press.
Wickelgren, I. (1997). Getting a grasp on working memory. Science, 275, 1580–1582.
Wickelgren, W. A. (1970). Multitrace strength theory. In D. A. Norman (Ed.), Models of human memory (pp.
65–100). New York: Academic Press.
Wickens, D. D., Dalezman, R. E., & Eggemeier, F. T. (1976). Multiple encoding of word attributes in memory.
Memory & Cognition, 4, 307–310.
Wiggs, C. L. (1993). Aging and memory for frequency of occurrence of novel, visual stimuli: Direct and indirect
measures. Psychology and Aging, 8, 400–410.
Wightman, D. C., & Sistrunk, F. (1987). Part-task training strategies in simulated carrier landing final-approach
training. Human Factors, 29, 245–254.
Wilensky, A. E., Schafe, G. E., Kristensen, M. P., & LeDoux, J. E. (2006). Rethinking the fear circuit: The central
nucleus of the amygdala is required for the acquisition, consolidation, and expression of Pavlovian fear condition-
ing. Journal of Neuroscience, 26, 12387–12396.
Wilkins, S. M., Jones, M. G., Koral, D. L., Gold, P. E., & Manning, C. A. (1997). Age-related differences in an
ecologically based study of route learning. Psychology and Aging, 12, 372–375.
Williams, J. M. G., & Markar, H. R. (1991). Money hidden and rediscovered in subsequent manic phases: A case
of action dependent on mood state. British Journal of Psychiatry, 159, 579–581.
Williams, L. M. (1994). Recall of childhood trauma: A prospective study of women’s memories of childhood
sexual abuse. Journal of Consulting and Clinical Psychology, 62, 1167–1176.
Williamson, V. J., Jilka, S. R., Fry, J., Finkel, S., Mullensiefen, D., & Stewart, L. (2011). How do “ear-worms” start?
Classifying the everyday circumstances of involuntary musical imagery. Psychology of Music, 40, 259–284.
Wilson, M. A., & McNaughton, B. L. (1994). Reactivation of hippocampal ensemble memories during sleep.
Science, 265, 676–679.
Winograd, E. (1981). Elaboration and distinctiveness in memory for faces. Journal of Experimental Psychology:
Human Learning and Memory, 7, 181–190.
Wirtz, P. W., & Harrell, A. V. (1987). Effects of post-assault exposure to attack-similar stimuli on long-term recov-
ery of victims. Journal of Consulting and Clinical Psychology, 55, 10–16.
Wittman, W. T., & Healy, A. F. (1995). A long-term retention advantage for spatial information learned naturally
and in the laboratory. In A. F. Healy & L. E. Bourne (Eds.), Learning and memory of knowledge and skills (pp.
170–205). Thousand Oaks, CA: Sage.
408 References

Wixted, J. T. (1991). The conditions and consequences of maintenance rehearsal. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 17, 963–973.
Wolf, A. S. (1980). Homicide and blackout in Alaskan natives. Journal of Studies on Alcohol, 41, 456–462.
Wolf, M. M., Giles, D. K., & Hall, R. V. (1968). Experiments with token reinforcement in a remedial classroom.
Behavior Research and Therapy, 6, 51–54.
Woloshyn, V. E., Willoughby, T., Wood, E., & Pressley, M. (1990). Elaborative interrogation facilitates adult learn-
ing of factual paragraphs. Journal of Educational Psychology, 82, 513–524.
Wood, W., Tam, L., & Witt, M. G. (2005). Changing circumstances, disrupting habits. Journal of Personality and
Social Psychology, 88, 918–933.
Woodruff-Pak, D. S., Papka, M., & Simon, E. W. (1994). Eyeblink classical conditioning in Down’s syndrome,
fragile X syndrome, and normal adults over and under age 35. Neuropsychology, 8, 14–24.
Woollett, K., & Maguire, E. A. (2011). Acquiring “the Knowledge” of London’s layout drives structural brain
changes. Current Biology, 21, 2109–2114.
Wright, A. A., Cook, R. G., Rivera, J. J., Shyan, M. R., Neiworth, J. J., & Jitsumori, M. (1990). Naming, rehearsal,
and interstimulus interval effects in memory processing. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 6, 1043–1059.
Wright, A. A., Santiago, H. C., Sands, S. F., Kendrick, D. F., & Cook, R. G. (1985). Memory processing of serial
lists by pigeons, monkeys and people. Science, 229, 287–289.
Wulf, G., & Schmidt, R. A. (1989). The learning of generalized motor programs: Reducing the relative frequency
of knowledge of results enhances memory. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 15, 748–757.
Wynn, K. (1992). Addition and subtraction by human infants. Nature, 358, 749–750.
Wynn, K. (1995). Infants possess a system of numerical knowledge. Current Directions in Psychological Science,
4, 172–177.
Yates, F. A. (1966). The art of memory. Chicago: University of Chicago Press.
Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Journal
of Comparative Neurology and Psychology, 18, 459–482.
Young, A. W., Hay, D. C., & Ellis, A. W. (1985). The faces that launched a thousand slips: Everyday difficulties and
errors in recognising people. British Journal of Psychology, 76, 495–523.
Young, G. C. D., & Martin, M. (1981). Processing of information about self by neurotics. British Journal of Clinical
Psychology, 20, 205–212.
Young, M. E. (1995). On the origin of personal causality theories. Psychonomic Bulletin & Review, 2, 83–104.
Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology
Monograph, 9 (Part 2), 1–28.
Zangwill, O. L. (1937). An investigation of the relation between the processes of reproducing and recognizing
simple figures, with special reference to Koffka’s trace theory. British Journal of Psychology, 27, 250–276.
Zoladek, L., & Roberts, W. A. (1978). The sensory bases of spatial memory in the rat. Animal Learning & Behavior,
6, 77–81.
Zola-Morgan, S., & Squire, L. R. (1993). Neuroanatomy of memory. Annual Review of Neuroscience, 16,
547–563.
AUTHOR INDEX

Note: Page numbers in italics refer to figures; numbers in bold refer to tables.

Aarons, L. 251 Aspinwall, L. G. 137 Bavaresco, J. L. 295

Abramson 136 Astrachan, M. 195, 196 Bechara, A. 138, 320
Ackerman, K. T. 350 Atkinson, R. C. 157, 177, 204, Beckwith, B. E. 252, 253
Adams, C. D. 106 205, 210, 230 Beechel, Allen 205
Adams, J. A. 309, 314 Au, J. 225 Beilock, S. L. 222
Adams, M. J. 39 Averill, J. R. 131 Belleville, S 167
Adefris, W. 294 Ayduk, O. 132 Bellugi, U. 214
Akhtar, A. 338 Ayllon, T. 98, 109, 109, 114 Benbow, C. P. 212, 213, 228
Alba, J. W. 70, 257, 293 Ayres, T. J. 208 Benjamin, M. 284
Albarracin, D. 291 Azrin, N. H. 98 Berkun, M. M. 8, 283
Alberini, C. M. 275 Berns, Gregory 100
Alessandri, S. M. 117, 118 Backman, L. 345 Bernstein, I. L. 13, 54
Allen, C. T. 68 Baddeley, A. D. 188, 204, 208, Bernstein, P. A. 156
Allen, S. W. 247 215, 216, 217, 221, 229, Berntsen, D. 277
Allison, J. 95 281, 294, 295, 311, 312, Berrios, G. E. 197
Althoff, R. R. 245 313, 324 Berry, D. C. 43, 206, 207,
Amsel, A. 119 Baeyens, F. 71, 76 320
Anderson, John R. 323, 328 Bahrick (& Hall) 279 Bevan, W. 240
Anderson, M. C. 132, 253 Bahrick, Harry P. 7, 162, 163, Biegler, P. 71
Anderson, R. C. 294 247, 250, 268, 269 Birch, L. L. 41
Andress, D. 14 Bahrick, L. E. 250, 250 Birnbaum, I. M. 294
Andrews, B. 292 Bahrick, P. O. 7, 268, 269 Bjork, R. A. 2, 12, 155, 247,
Anthony, S. G. 14 Baird, R. B. 287 314, 315, 350
Aporta, C. 325 Balter, M. 221 Black, J. B. 293
Appenzeller, T. 226 Banaji, M. R. 14–15 Blake, M. J. 253
Appleby, J. 46 Banbury, S. P. 43, 210 Bloch, M. 70
Araujo, J. 268 Bandura, A. 6, 6 Block, R. A. 245
Arbuckle, T. Y. 341 Barlow, David 44 Bloom, L. C. 238
Arbuthnott, K. 349 Barrett, F. C. 76 Boland, J. 252
Archer, E. J. 311, 312, 313 Barron, K. L. 341 Bolles, Robert C. 130
Arkes, H. R. 258 Bartlett, Frederick 292, 293, Boorstin, Daniel J. 167
Arnold, M. E. 252 298 Booth, D. 42
Arrigo, J. M. 291 Bass, K. E. 20 Bornstein, M. H. 28, 29, 251
Ashcraft, M. H. 222 Bauer, P. J. 335, 339 Borresen, C. R. 310
Aslin, R. N. 166 Baumeister, R. F. 124, 350 Boruch, R. F. 164
410 Author Index

Bourne, L. E. 155, 311, 312, Cahill, L. 343 Collins, A. M. 270, 271, 271,
313 Cain, W. S. 213 272, 296
Bovbjerg, D. H. 55 Calkins, Mary Whiton 151 Conezio, J. 239
Bower, G. H. 149, 159, 167, Call, J. 166 Connolly-Gomez, C. 312
168, 238, 281, 282 Campese, V. 138, 139 Conrad, R. 207, 209, 214
Bowers, J. S. 164 Campitelli, G. 322 Conway, M. A. 161, 216, 268,
Brackbill, Y. 333 Canavan, A. G. M. 194 338
Bradshaw, J. W. 126 Candel, I. 185, 186 Cook, J. 291
Brainerd, C. J. 287 Candland, D. K. 3 Cook, M. 75
Brandon, S. E. 68 Cann, A. 283 Coolidge, F. L. 220, 221
Bransford, J. D. 240 Cannizzo, S. R. 336 Coombes, S. 55
Breen, N. 199 Capaldi, E. J. 120 Cooney, J. B. 229
Breland, Keller 100 Capretta, P. J. 8, 283 Coons, G. 252
Breland, Marion 100 Caramazza, A. 272 Cooper, C. J. 154
Brewer, J. B. 234 Carpenter, P. A. 228 Cooper, P. F. 197
Brewin, C. R. 292 Carpenter, S. K. 246, 260 Cooper, R. M. 331, 332
Broadbent, D. E. 197 Carriere, J. S. 197 Corkin, Suzanne 194, 306, 310
Broadhurst, P. L. 129 Carver, C. S. 137 Cornell, E. H. 149, 309
Bronfenbrenner, U. 346 Cavanaugh, J. C. 170 Coupe, P. 308
Brooks, A. M. 100 Ceci, J. S. 346 Cowan, N. 209
Brooks, J. O. 207 Cepeda, N. J. 246 Craik, F. I. M. 187, 235, 236
Brooks, L. R. 247 Cermak, L. S. 294 Crawford, M. 346
Brown, A. S. 199 Chabris, C. F. 15 Crawley, R. A. 339
Brown, J. S. 122 Chan, J. C. K. 242 Crist, W. B. 153, 153
Brown, John 205 Chan, M. S. 291 Crombez, G. 76
Brown, M. A. 148, 149 Chance, P. 99 Cronbach, L. J. 329, 330
Brown, M. F. 303 Chang, J. 197 Crook, M. D. 20, 341
Brown, R. E. 305 Chang, R. 325 Crowder, R. G. 14–15, 210
Brown, R. T. 161, 254–255, Channon, S. 194 Cuddy, L. J. 125, 246
285 Charness, N. 321, 340 Cunitz, A. R. 177, 178
Bruce, V. 152 Chase, W. G. 321, 326 Cutler, S. J. 341
Brucker, B. S. 70 Chavaja, P. 353 Cutmore, T. R. H. 55
Bruckmeir, G. 128 Cheyne, J. A. 123, 123, 197 Cytowic, R. E. 343
Bruer, J. T. 20 Chi, M. T. H. 336
Bryant, K. J. 324 Chiba, A. A. 176, 176 Dadds, M. R. 55
Bryant, R. A. 45 Chorover, S. L. 274 Dalezman 206
Buchanan, M. 208 Choy, Y. 45 Damasio, Antonio R. 138, 320
Buchel, C. 251 Christal, R. E. 154 Damasio, H. 320
Budge, M. 347 Christianson, S.-A. 254, 256 Daneman, M. 228
Bullemer, P. 318, 318 Chua, Amy 353–354 Dark, V. J. 212, 213, 228, 260
Burger, J. M. 331 Chung, P. H. 108 Daum, I. 69, 194
Burgess, Paul 223, 229 Clark, D. M. 282 Davey, G. C. L. 56, 66, 76
Burns, H. J. 289 Clark, M. C. 159, 167 Davis, B. 344
Burns, T. E. 256 Claxton, C. S. 348 Davis, H. P. 156, 157
Bushman, B. J. 124 Cleeremans, A. 220 Davis, M. 37, 138
Butters, N. 157, 294, 311, Clifford, B. R. 206, 207 Davis, T. B. 246
321 Coane, J. H. 160 Dawson, M. E. 69
Byrne, J. H. 204 Cohen, G. 151, 268, 300 Day, R. O. 223
Byrne, M. D. 108 Cohen, L. J. 131 De Houwer, J. 71
Byrne, R. 308 Cohen, N. J. 181, 245, 255 de Wijk, R. A. 213
Cohen, T. E. 34 Deese, J. 286, 297
Cabeza, R. 19, 19, 287 Colcombe, S. J. 9 DeJager, C. 347
Caharack, G. 261 Cole, M. 353 Delgado, M. R. 138
 Author Index 411

Delis, D. C. 156 Enright, M. 86 Gais, S. 251

DeLoache, J. S. 76 Erber, J. T. 341 Gale, A. 27
Dempster, F. N. 245, 247 Erdelyi, M. H. 278, 279 Galluccio, L. 148
deQuervain, D. J. 284 Ericsson, K. Anders 287, 321, Garcia, J. 13, 65
Desmond, J. E. 234 322, 326 Gartman, L. 246
Detterman, D. K. 243, 244, Erikson, G. C. 14, 252 Gatchel, R. J. 31
332 Estes, W. K. 120, 121, 127, Gathercole, S. E. 217
Diamond, A. 220 150, 289, 290 Geer, J. H. 135
Dickinson, A. 8, 87 Evans, G. W. 326 Geller, J. 260
Diekelmann, S. 251, 252 Evans, H. 124 Gerler, D. 286
Dienes, Z. 320 Eysenck, Hans J. 55, 57 Gershman, S. J. 77
Dodson, J. D. 249 Eysenck, M. C. 236, 238 Gershoff, E. T. 125
Dollard, J. 92, 132 Eysenck, M. W. 236, 238 Gesell, A. 7
Domjan, M. 17, 41 Ghatala, E. S. 262, 337, 337
Donchin, E. 243 Fabiani, M. 243 Gibson, Eleanor J. 38, 38, 39,
Dorner, W. W. 73 Fagan, J. W. 86 40, 43, 47
Doty, K. 326 Faloon, S. 321 Gibson, J. J. 38, 38, 39
Dozier, C. L. 127 Farah, M. J. 252 Giles, D. K. 87
Drews, F. 223 Faulkner, D. 151 Gilligan, S. G. 282
Duckworth, A. L. 351 Fehr, S. 336 Gillihan, S. J. 45
Dulany, D. E. 98 Feingold, A. 345 Gilliland, K. 14, 252
Duncan, C. P. 119 Feld, G. B. 252 Gillin, J. C. 281
Dunlosky, J. 20 Fellows, J. 326 Girala, N. 197
Dunning, D. 261 Fendrich, D. W. 311 Gladwell, Malcolm 320
Dunsmuir, W. T. M. 223 Field, A. P. 71 Glanzer, M. 177, 178
Dweck, C. S. 55, 351 Finkel, D. 332 Glass, D. C. 135
Dwyer, D. M. 40 Fischer, J. 166 Glisky, E. L. 323
Dyer, H. W. 261 FitzGerald, P. 197 Glover, J. A. 234, 246
Fitzgerald, H. E. 333 Gluck, M. A. 254
Eacott, M. J. 339 Fivush, R. 250, 250, 338, 339 Gobet, F. 224, 241, 322
Eals, M. 347 Flavell, J. H. 335, 336, 337 Godden, D. R. 281, 295
Ebbinghaus, Herman 145, Foa, E. B. 45, 141 Goettl, B. P. 312
146, 147, 147, 170, 172, Foer, Joshua 169–170 Goff, L. M. 288
240, 267, 268, 296 Fogarty, S. J. 282 Gold, P. E. 95, 307
Ecker, U. 291 Folkard, S. 253, 253 Goldman, M. S. 294
Edwards, J. A. 27 Folkman, S. 131 Goldstone, R. L. 39
Eelen, P. 76 Ford, J. K. 5 Golkar, A. 77
Eggemeier, F. T. 206 Forrest, T. J. 287 Goodwin, K. A. 287
Ehrlinger, J. 261 Forzano, L. B. 350 Gordon, A. M. 310
Eich, E. 132, 281, 282, 283, Fosse, M. 251 Gorman, N. 252
294 Franks, J. J. 240 Gormezano, I. 53
Eikelboom, R. 72 Freedman, D. G. 124, 196 Goyeche, J. R. M. 123, 123
Eimas, P. D. 213 Frensch, P. A. 320 Graf, P. 164, 187, 248
Einstein, G. O. 168 Freud, Sigmund 132, 198 Grams, A. E. 341
Eisenberger, R. 128 Friedman, N. P. 219 Grayson, J. B. 141
Ellard, Colin 324 Friendly, M. 240, 241 Gredebäck, G. 76
Elliot, C. 236 Frings, M. 30, 30 Green, C. 132
Elliott, J. M. 336 Fritz, C. O. 152 Greene, D. 99
Ellis. N. C. 208, 268 Fujiwara, E. 197, 325 Greene, R. L. 157
Ellison, G. D. 57 Fyer, A. J. 45 Greene, S. L. 270
Emmons, W. H. 251 Greene, W. 34
Engen, T. 213 Gabrieli, J. D. E. 234, 310 Greenough, W. T. 273
Engle, R. W. 69, 216, 219, 336 Gage, F. H. 7 Greenspoon, J. 98
412 Author Index

Grigorenko, E. L. 348 Hertzog, C. 341 Inoue, S. 215

Grillon, C. 38 Heth, C. D. 309 Ishikawa, T. 325
Grimm-Thomas, K. 41 Heuer, F. 254 Israel, L. 286
Grogan-Kaylor, A. 125 Hiby, E. F. 126 Ition, A. 333
Gross, J. J. 351 Hicks-Pass, S. 125 Ivy, R. B. 69
Groves, P. M. 37 Higgs, E. 325
Growdon, J. H. 310 Hilgard, E. R. 52, 149 Jacklet, J. W. 34
Gulya, M. 148 Hill, W. F. 41 Jackson, J. 295
Gunder, L. 41 Hillis, A. E. 272 Jacoby, L. L. 161, 246, 291
Gunter, B. 206, 207 Hills, T. 331 Jaeggi, S. M. 224
Gunther, L. M. 55 Hinson, R. E. 55 James, William 107, 151, 203,
Gustavson, C. R. 13, 54 Hintzman, D. L. 176, 245, 278 204, 229, 285
Guthrie, J. P. 262 Hirnstein, M. 346 Jamieson, K. H. 291
Hiroto, D. S. 34, 136 Janet, Pierre 196
Haaland, K. Y. 310 Hirst, E. S. 255, 261 Janis, I. L. 195, 196
Haber, R. N. 239 Hirtle, S. C. 241 Jenkins, H. M. 67
Hall, L. K. 279 Hitch, G. 204, 215 Jenkins, J. J. 159, 247, 248
Hall, G. 73 Hobson, J. A. 251 Johansson, M. 76
Hall, R. V. 87, 279 Hoehl, S. 76 Johnson, J. C. 130
Hamann, S. B. 181, 182 Hoffman, H. 254 Johnson, K. 261
Hambrick, D. Z. 322 Hogervorst, E. 347 Johnson, M. K. 287, 288
Hamdin, M. 128 Holdsworth, M. 346 Johnson, N. F. 246
Hammerl, M. 70 Holinger, D. P. 284 Jones, C. R. 291
Hammond, D. 44 Hollis, K. L. 50, 54 Jones, D. M. 210
Hampe, E. 76 Holloway, F. A. 283 Jones, M. G. 307
Hampson, E. 346 Honey, R. C. 40 Jones, W. T. 73
Hanley, G. L. 326 Honzik, C. H. 8, 21, 96, 97, Josselyn, S. A. 19
Hanley-Dunn, P. 176 105, 112 Jou, R. L. 138
Hardwick, C. 334 Hopkins, R. O. 176, 176 Justice, L. V. 338
Harkness, A. R. 258 Horton, D. L. 236
Harlow, H. F. 92, 93 Hostetler, A. J. 199 Kail, R. 336
Harrington, D. L. 310 Hothersall, D. 44 Kalamarides, P. 293
Harris, B. 75 Houston, J. P. 15 Kalat, J. W. 66
Harris, J. E. 163, 170 Hovland, C. I. 311, 313 Kalish, H. I. 102, 103, 301
Hart, J. T. 285 Howe, M. L. 338 Kamin, Leon J. 62, 129
Hartley, J. T. 294 Hudson, J. A. 293 Kaminski, J. 166
Hasher, L. 9, 221, 257, 293 Hull, Clark L. 10, 10, 87, 92, Kandel, Eric R. 33, 34, 35, 37,
Hastie, R. 243 105, 113, 301, 327 47, 273
Hatano, G. 241 Hulme, C. 207, 225 Kang, S. H. K. 243
Hausmann, M. 346 Hume, David 73 Kao, S. F. 73
Hawkins, R. D. 34 Humphreys, M. S. 252 Kaplan, E. 156
Hayne, H. 335, 339 Hunt, R. R. 236, 243 Kapur, N. 343
Healy, A. F. 155, 227, 293, 311, Hunter, Walter 214 Kardes, F. R. 68
316 Hurley, D. 224 Karibian, D. 287
Hebb, D. O. 188, 273 Husain, M. 252 Karlin, M. B. 238
Heerdt, W. A. 128 Huston, J. P. 95 Karni, A. 251
Heindel, W. C. 311, 321 Huxley, Aldous 49, 74 Karpicke, J. D. 11, 12, 155, 242
Hellmer, K. 76 Hyde, T. S. 247, 248 Kasri, F. 222
Helmsley, D. R. 282 Kassam, K. S. 249
Hennelly, R. A. 208 Ilieva, I. 252 Kausler, D. H. 212
Herlitz, A. 345 Imai, O. 325 Kawecki, T. J. 331
Herrmann, D. J. 169, 346 Ince, L. P. 70 Kazdin, Alan E. 92, 127
Hertwig, R. 331 Inhoff, A. W. 310 Kearins, J. M. 353
 Author Index 413

Keenan, J. 209 Kunst-Wilson, W. R. 43 Locke, John 3, 21

Keeney, T. J. 336 Kurbat, M. A. 149 Lockhart, D. 187, 235
Kelley, C. M. 161, 291 Kvavilashvili, L. 132, 277 Lockhead, G. R. 153, 153
Kemp, S. 288 Ky, K. N. 20 Loftus, E. F. 254, 255, 256,
Kemperman, G. 7 Kyllonen, P. C. 154 266, 271, 272, 288, 289
Kenny, L. M. 132 Loftus, G. R. 254, 256, 266
Kensinger, E. A. 256 Lackey, S. 168 Logan, F. A. 54
Kesner, R. P. 176, 176 Laing, D. G. 41 Logue, A. W. 350
Kessen, W. 28 Lally, P. 107 Longman, D. J. A. 311, 312,
Ketcham, K. 255 Landauer, T. K. 2, 95 313
Kientzle, M. J. 313 Larkina, M. 339 Lopatka, C. 58
Kiewra, K. A. 261 Lashley, Karl 14, 18, 150 Lord, A. B. 351–352
Kim, J. 68 Lasko, N. B. 56 Lott, D. F. 106
Kimble, G. A. 7 Latimer, P. R. 141 Lucas, D. 86
Kimura, D. 346 Laughlin, J. E. 216 Lukowiak, K. 34
Kindt, M. 296 Lave, C. 353 Luria, A. R. 342, 356
Kirby, F. D. 91 Lawton, C. A. 345 Lynch, S. 211
Klatzky, R. L. 15 Lazarus, R. S. 131 Lyubomirsky, S. 222
Klein, D. C. 305 Lechner, H. A. 204
Kleinbard, J. 278, 279 LeDoux, J. E. 137, 138, 139, MacInnes, J.W. 305
Klima, E. S. 210, 214 274, 276 Mackay, D.G. 244
Klingsporn, M. J. 310 Lee, C. L. 150 Macken, W. J. 210
Knowlton, B. 184 Lee, S. Y. 208 Mackenzie, S. 207
Koelling, R. A. 65 Leichtman, M. D. 354 Mackintosh, N.J. 66
Kolb, B. 306 Lemonick, M. D. 13 Macklin, M. L. 56
Kolb, D. A. 348, 348, 356 Lennartson, E. R. 287 Madigan, S. 151
Kolss, M. 331 Lenneberg, E. H. 7 Magnussen, S. 15
Kopelman, M. D. 192, 221 Leon, M. 333 Maguire, E. A. 306, 307, 343
Koral, D. L. 307 Leonard, C. 337 Maia, T. V. 220, 320
Koriat, A. 259 Leonesio, R. J. 259 Maier, S. F. 133, 134, 135
Kornell, N. 118 LePort, A. K. R. 344 Malmi, R. A. 164
Korol, D. L. 95 Lepper, M. R. 99 Mandler, G. 159, 164, 277
Kozlowski, L. T. 324 Lesgold, A. M. 159, 228 Mangiulli, I. 197
Kraiger, K. 5 Lessac, M. S. 122 Manning, C. A. 307
Kramer, J. H. 156 Lettvin, J. Y. 189 Mantyla, T. 280
Krampe, R. T. 322 Levey, A. B. 70 Marchon, I. 326
Krank, M. D. 55 Levin, I. P. 262 Marder, E. 311
Kratzig, G. 349 Levine, M. 326 Markar, H. R. 282
Krause, J. A. 222 Levitt, M. 250, 250 Markova, I. S. 197
Kreutzer, M. A. 337 Levy, B. J. 132 Marler, P. 8
Kristensen, M. P. 138 Lewandowsky, S. 291 Marlin, D. W. 41
Kritchevsky, M. 197 Lewin, Kurt 227 Marquis, D. G. 52
Kroese, B. S. 25, 27 Lewis, D. J. 119, 225 Marshall, K. 336
Kroll, J. F. 155 Lewis, M. 117, 118 Marslen-Wilson, W. D. 269
Kross, E. 132 Lieberman, D. A. 90 Martin, E. 246
Krug, D. 246 Lieberwitz, K. J. 212 Martin, I. 70
Kruger, J. 261 Lin, Y.-G. 284 Martin, R. C. 218
Kuhara-Kojima, K. 241 Linton, M. 180, 258, 293 Maslow, Abraham 40
Kuhn, H. G. 7 Lipp, O. V. 36, 66 Massa, L. J. 347
Kuiack, M. 25, 27 Lipsitz, J. D. 45 Masserman, Jules 124, 125
Kuisma, J. E. 213 Lobb, H. 334 Matchett, G. 56
Kulhavy, R. W. 239, 261 LoBue, V. 76 Matsuzawa, T. 215
Kulick, J. 254, 255 Locke, J. L. 336 Maturana, H. R. 189
414 Author Index

Matute, H. 55 Miller, M. E. 294 Nicholson, D. E. 315

Mayer, R. E. 347 Miller, Neal E. 16, 92, 124, Nickerson, R. S. 39
Mayhew, D. 209 131, 132, 140 Nilsson, L. 248, 256, 345
Maylor, E. A. 295 Miller, P. L. 196 Nisbett, R. E. 99
McCarthy, R. A. 273 Miller, R. R. 55, 77 Nissen, M. J. 318, 318
McClelland, J. L. 320 Mills, C. G. 236 Nitsch, K. E. 240
McClosky, M. 255 Milner, Brenda 193, 194 Noble, E. P. 294
McConnell, A. R. 222 Milner, Peter 93, 95, 178, 306 Noice, H. 326
McCulloch, A. R. 189 Milton, F. 196 Noice, T. 326
McCully, J. 55 Mineka, S. 75, 141 Norman, G. R. 108, 204, 209,
McDaniel, M. A. 16, 168, 243, Mischel, Walter 350, 356 247
350 Mishkin, M. 19, 226 Nyberg, L. 19, 19
McDermott, K. B. 242, 243, Miyake, A. 219
286, 297, 333 Miyatsu, T. 16 Ober, B. A. 156
McDonald, R. J. 305 Moèand, A. 346 Oerter, R. 321
McGaugh, J. L. 254, 273, 284, Moffitt, T. E. 351 O’Hara, R. 151
343, 344 Mondadori, C. 95 Ohman, A. 75
McGeoch, J. A. 266 Monk, T. H. 253, 253 Okabe, A. 325
McGue, M. 332 Monteiro, K. P. 282 Olufs, M. 118
McGuire, W. J. 152 Montello, D. R. 326 Olds, James 93
McIntosh, J. L. 90, 176 Moore, T. E. 67, 98, 303, 309 Olson, G. M. 35
McIsaac, H. K. 132 Morgulis, S. 51 Olsson, A. 139
McKeachie, W. J. 284 Morris, P. E. 152, 163 Olton, D. S. 302, 302, 303,
McKeithan, K. B. 241 Morris, R. G. M. 305 303, 309
McLachlan, D. 318, 319 Morrison, C. M. 338 Orr, S. P. 55, 56, 57
McLenon, J. 45 Moscarello, J. 138, 139 Osborne, J. G. 110, 111
McMurtry, C. M. 45 Moscovith, M. 318, 319 Osgood, C. E. 50, 53
McNamara, D. S. 227 Mowrer, Hobart 129, 140 Otaka, S. R. 295
McNaughton, B. L. 251 Mudd, S. A. 238 Otgaar, H. 185, 186
McNeil, D. 161, 285 Mueller, K. D. 167 Overman, W. H. 309
McWhinney, B. 209 Muller, G. E. 204
Mehta, M. A. 252 Mundy, M. E. 40 Paivio, A. 240
Mehus, C. J. 125 Murphy, D. L. 282, 294 Paller, K. A. 252
Meinz, E. J. 322 Murphy, D. R. 199 Palombo, D. J. 345
Meissner, C. A. 287 Murrell, P. H. 348 Pandeirada, J. N. S. 248
Melby-Lervåg, M. 225 Musen, G. 184 Papka, M. 69, 70, 334
Melcher, J. M. 239 Park, D. C. 170, 290
Melton, A. W. 206 Nachtigall, P. E. 214 Parker, E. S. 294, 343
Mendoza, R. 333 Nadar, K. 276 Parker, J. F. 250, 250
Menzel, E. W. 159, 308 Nairne, J. S. 248 Parkes, K. R. 197
Merikle, P. M. 228 Naveh-Benjamin, M. 208, 284 Partick, M. E. 125
Mery, F. 331 Neisser, Ulric 14–15, 255, 261, Pashler, H. 246, 350
Merzenich, M. M. 104 292, 339, 343 Pate, B. J. 309
Messo, J. 256 Nelson, C. A. 229 Pavlov, Ivan P. 49, 55, 270
Metzger, L. I. 56 Nelson, D. L. 236, 237, 239 Payne, J. D. 127, 256
Metzler, J. 346 Nelson, K. 338, 339 Peake, P. 350
Mickel, S. F. 310 Nelson, T. O. 162, 248, 249, Pearce, J. M. 18, 304
Miller, R. R. 275 259, 286 Pearlstone, Z. 158, 161, 280
Miller, D. G. 289 Nemiah, J. C. 196 Pendlebury, W. W. 333
Miller, George A. 207, 209, Neumann, D. L. 66 Penfield, Wilder 18, 266–267
229 Neuringer, A. 118 Penner, R. H. 214
Miller, H. 282 Newport, E. L. 166 Pereira, J. K. 284
Miller, L. C. 76 Nguyen, K. 16 Perfetti, C. A. 228
 Author Index 415

Perkins, C. C. 68 Reason, James T. 107, 174 Rubin, D. C. 150, 240, 241,

Perl, T. 56 Reaves, C. C. 261 255, 288
Perlmutter, M. 340 Reber, A. S. 300, 317, 317, Rueter, H. H. 241
Peters, F. 167 318 Rundus, D. 158, 158, 211
Peterson, Lloyd R. 205 Reber, R. 43 Runger, D. 320
Peterson, Margaret 205 Redd, W. H. 55 Russell, W. A. 159
Peterson, S. 239 Reder, L. 239 Ryan, E. B. 260, 324
Petri, H. L. 19 Reed, G. 86, 87, 285 Ryan, J. D. 245
Petrill, S. A. 333 Reeves, R. V. 125 Ryan, R. S. 239
Petros, T. V. 252, 253 Regian, W. 312 Rydell, R. J. 222
Peuster, A. 309 Reid, A. K. 166
Pezdek, K. 258, 291 Reisberg, D. 254 Sachs, J. S. 209
Phelps, E. A. 138, 139 Reiser, B. J. 293 Sacks, Oliver 194, 195
Piaget, Jean 255, 288 Reitman, J. S. 241 Safer, M. A. 279
Pichert, J. W. 294 Rescorla, Robert A. 50, 57, 63, Saffran, J. R. 166
Pickrell, J. E. 288 64, 68, 106 Sala, G. 224
Pilley, J. W. 166 Restle, F. 302 Salas, E. 5
Pilllsbury, W. B. 205 Revelle, W. 252 Salmon, K. 284, 311, 321
Pinker, S. 4 Revusky, S. 55 Salthouse, T. A. 221, 340
Pitman, R. K. 56 Reyna, V. F. 287 Sampaio, C. 289
Pitts, W. H. 189 Riccio, D. D. 12 Samuelson, R. J. 302, 303
Pliner, O. 41 Richardson, R. 283 Sauda, E. 325
Plomin, R. 332 Rips, L. J. 149 Saufley, W. H. 295
Plous, S. 16 Ritchie, B. F. 301 Schab, F. R. 213
Pohl, R. W. 55 Rob, I. 223 Schacter, D. L. 164, 179, 181,
Pont, J. 331 Roberts, B. W. 303, 351 182, 182, 184, 187, 196,
Potts, H. W. 107 Robinson, E. S. 148, 149 256, 286, 287, 320, 323
Powers, R. B. 110, 111 Robinson, K. J. 286 Schafe, G. E. 138, 274, 276
Pradere, D. 287 Roby, T. B. 93 Scheier, M. F. 137
Premack, David 94, 95 Rodriguez, M. L. 350 Schell, A. M. 69
Pressley, M. 154, 262, 337, Roediger, H. L. III 11, 12, 155, Schiller, D. 274
337 169, 169, 182, 183, 183, Schmidt, R. A. 12, 247, 314,
Price, J. 344 240, 242, 243, 278, 279, 315
Price, J. L. 345 286, 288, 297, 333 Schneider, V. I. 155
Price, M. 284 Rogers, L. 45, 304 Schooler, J. W. 239
Purdy, J. E. 17 Rogoff, B. 353 Schreiber, T. A. 236, 237, 239
Rohrer, D. 246, 350 Schrepferman, L. 124
Quillian, M. R. 270, 271, 296 Roitblat, H. L. 214 Schugens, M. M. 69
Rolls, B. J. 42 Schwartz, B. L. 285
Rabbitt, P. M. A. 295 Rooney, N. J. 126 Schwartz, B. N. 129
Rachman, S. 58 Roozendaal, B. 284 Schwartz, N. 43
Rahhal, T. A. 9 Rosenbaum, R. S. 193 Schwartz, N. H. 239
Rajaram, S. 182 Rosenn, M. 131 Schwarz, N. 290, 291
Rakison, D. H. 76 Ross, M. 283, 354 Scoville, W. B. 178, 193
Ramaswami, M. 35 Roth, S. 131 Scrivener, E. 279
Rampon, C. 7, 333 Rothbaum, B. O. 45 Seamon, J. G. 286
Rankin, C. H. 29 Rothberg, S. T. 341 Sears, S. F. 138, 139
Rasch, B. 251 Rotter, J. B. 136 Sehulster, J. R. 148
Rasmussen, T. 18 Rovee, D. T. 85 Seifert, C. 291
Rauscher, F. H. 20 Rovee-Collier, Carolyn K. 85, Seligman, M. E. P. 66, 75, 130,
Raye, C. L. 288 86, 148 133, 134, 135, 136
Raymond, C. K. 88 Royer, J. M. 262 Semb, G. B. 268
Rayner, Rosalie 74, 75, 79 Rozin, P. 66 Shallice, T. 178, 211, 226, 229
416 Author Index

Shand, M. A. 210 Solomon, P. R. 333 Tesch-Romer, C. 322

Shanks, D. R. 87, 165 Solomon, R. L. 122, 133, 134 Teuber, H. L. 269, 306
Shapiro, D. C. 314, 315 Spear, N. E. 12 Thapar, A. 333
Sharp, D. 353 Spelke, E. 261 Thomas, E. 57
Sharps, M. J. 345 Spence, K. 88 Thomas, G. V. 90
Shaughnessy, J. J. 245 Spencer, W. A. 29, 33, 46 Thompson, D. M. 280, 281
Shaw, G. L. 20 Spivak, M. 100 Thompson, H. 7
Sheen, M. 288 Sporer, S. L. 238 Thompson, L. A. 332, 333
Sheffield, F. D. 93 Springer, A. D. 275 Thompson, M. A. 254
Shepard, R. N. 346 Squire, L. R. 157, 164, 181, Thompson, Richard F. 29, 33,
Sherwin, B. B. 347 182, 184, 196, 197, 204, 37, 46, 69
Shettleworth, S. 100 227, 267, 269, 305, Thompson, S. R. 248
Shevell, S. K. 149 320, 321 Thomson, N. 208
Shields, F. 91 St. Peter, C. 124 Thorndike, Edward Lee 82, 83,
Shiffrin, R. M. 157, 177, 204, Staddon, J. 121 105, 111, 113
205, 210, 230 Standing, L. 239 Thorndyke, P. W. 326
Shimamura, A. P. 184, 321 Stanford, L. 305 Thorpe, L. A. 279
Shimp, T. 69 Stanhope, N. 268 Tierney, J. 350
Shoda, Y. 350 Stark, C. E. L. 182 Timberlake, W. 95
Sholl, M. J. 308 Stasz, C. 326 Tinklepaugh, O. L. 214
Shore, D. I. 305 Steele, K. M. 20 Tirre, W. C. 154
Siddle, D. A. T. 25, 27, 27, 32, Steger, J. A. 240 Todd, R. D. 333
33, 36, 66 Steketee, G. 141 Tolman, Edward C. 8, 21, 84,
Siegel, S. 41, 55, 57, 67, 71, 72 Sternberg, R. J. 348 96, 97, 98, 105, 106, 112,
Silver, L. 261 Steubs, J. 346 301, 327
Silverman, I. 345, 347 Stevenson, H. W. 89, 208, 308 Tonegawa, S. 19
Silverthorne, C. P. 70 Stewart, G. 72 Tough, P. 351
Simcock, G. 335, 339 Stickgold, B. 251 Touron, D. R. 341
Simon, D. A. 241, 252, 315, Stigler, J. W. 208 Tranel, D. 320
334 Still, M. L. 260 Tremblay, S. 210
Simons, D. J. 15, 224, 251 Stillman, R. C. 281 Tryon, R. C. 330, 331
Singer, J. E. 135 Stolz, S. B. 106 Tuber, David S. 44, 101
Singh, D. 105 Strayer, D. 223 Tuholski, S. W. 216
Siple, P. 214 Stuart, E. W. 69 Tully, Tim 330
Skinner, B. F. 83, 84, 90, 92, Sullivan, M. W. 86, 117, 118 Tulving, E. 19, 19, 158, 159,
99, 100, 106, 111, 112, Sullivan, R. M. 333 161, 175, 178, 179, 182,
120, 121, 127, 198, 270 Summers, J. J. 245 184, 187, 196, 225, 236,
Skinner, N. F. 262 Sutherland, A. 101 280, 281, 320
Skurnik, I. 290 Swanson, H. L. 229 Turner, L. H. 122
Slamecka, N. J. 248 Swinnen, S. P. 314, 315 Turner, R. M. 141
Slater, L. 44, 196, 269 Sylvester, A. 205
Small, Willard S. 300 Umanath, S. 160
Smeets, T. 95, 185, 186, 283, Taborsky-Barbar, S. 333 Underwood, B. J. 146, 164,
284 Talarico, J. M. 255 225, 245
Smilek, D. 197 Talwar, S. K. 102 Urcelay, G. P. 77
Smith, A. D. 170, 347 Tam, L. 108
Smith, M. A. 155 Tang, Y.-P. 330 Valentine, E. R. 343
Smith, M. E. 252 Tangney, J. P. 351 van Jaarsveld, C. H. 107
Smith, S. M. 281, 283 Taylor, S. E. 137 Vance, A. 44
Snowman, J. 261 Teasdale, J. D. 136, 282 Vargas, P. 71
Snyder, J. 124 Tenenbaum, G. 148 Vaughn, W. 270
Soeter, M. 296 Terry, W. S. 14, 61, 93, Vela, E. 283
Sokolov, Y. N. 35, 37 151, 174 Verfaellie, M. 287
 Author Index 417

Verplanken, B. 108 Weiskrantz, L. 182, 186 Wolf, M. M. 87

Vervliet, B. 296 Weiss, J. M. 135 Wolf, O. T. 185, 186
Villeneuve, S. 167 Weldon, M. S. 183 Woloshyn, V. E. 161, 262, 291
Voith, V. L. 44 Welton, A. L. 345 Wolpe, J. 80
Volk, H. E. 333 Werker, J. F. 39 Wood, W. 262, 343
von Restorff, Helena 243 Wessel, G. 325 Woodruff-Pak, D. S. 70, 108,
Vul, E. 246 West, R. L. 341 334, 334
Westbrook, J. I. 223 Woods, A. 223
Waddill, P. J. 243 Wetherington, C. L. 26 Woollett, K. 307
Wagner, A. D. 234 Wheeler, M. A. 77, 278 Wright, A. A. 212, 214, 215
Wagner, Allan R. 35, 36, 37, Whishaw, I. O. 306 Wulf, G. 315
54, 55, 57, 58, 61, 63, 64, White, N. H. 95 Wynn, K. 28, 220, 221
68, 125 White, N. M. 305, 306
Wagner, D. 352–353 White, S. H. 354 Yadrick, R. M. 312
Wagner, I. C. 323 Whitlow, J. W. 35, 36, 245 Yarnell, P. R. 211
Walcott, C. 304 Wible, C. G. 255 Yates, F. A. 168
Walk, R. D. 39 Wickelgren, I. 204 Yeager, D. S. 351
Walker, M. P. 314 Wickelgren, W. A. 225 Yeo, R. A. 310
Walker, S. 25 Wickens, D. D. 206, 207 Yerkes, R. M. 51, 249
Walters, R. H. 123, 123 Wiggs, C. L. 42, 42 Yoon, C. 290
Wang, Q. 349 Wilding, J. M. 343 Young, M. E. 73
Wang, P. L. 196 Wilensky, A. E. 138 Young, A. 152
Wang, Q. 354 Wiley, J. G. 212 Young, D. E. 314
Wang, R. F. 289 Wilk, A. 148
Wardle, J. 107 Wilkins, S. M. 307 Zacks, R. T. 221
Waring, J. D. 256 Williams, L. M. 291–292 Zajonc, R. B. 40, 43
Warrington, E. K. 178, 182, Williamson, V. J. 278, 282 Zangwill, O. L. 289, 290
186, 211, 226, 273 Willoughby, T. 262 Zehm, K. 222
Waser, P. G. 95 Wilson, B. 251, 294 Zhao, Z. 234
Wasserman, E. A. 73 Winkielmanm, P. 43 Ziemkiewicz, C. 325
Watkins, M. J. 207, 225 Winocur, G. 318, 319 Zigler, E. F. 89
Watson, J. M. 223 Winograd, E. 238 Zimet, S. 128
Watson, John B. 74, 75, 79, Winzenz, D. 159 Zimmerman, J. 245
300 Witt, M. G. 108 Zinbarg, R. 75
Waugh, N. C. 204, 209 Wittlinger, R. P. 7, 268, 269 Zoladek, L. 303
Weingartner, H. 281, 282, 294 Wittman, W. T. 293 Zola-Morgan, S. 227
Weinstock, S. 118, 119 Wixted, J. T. 235, 246 Zorn, M. 326
Weiskopf, S. 28 Wolf, A. S. 294 Zubek, J. P. 331, 332
SUBJECT INDEX

Note: Page numbers in italics indicate figures; numbers in bold refer to tables.

abnormal memory 191–197 aggressive behavior 6, 6, 54, amnesic syndrome 193, 195,
absentmindedness 163, 197, 101, 124, 140 201, 226
198, 202, 204, 260, aging: and conditioning amygdala 69, 137–140, 139,
307, 341 333–334; and memory 139, 143–144, 254, 274,
abstract learning 356 capacity 340; and 276
abstract words 234 metamemory 340–341; animal studies 16–17, 22;
abuse 291–292, 298 and self-efficacy 341 aversive consequences
academic learning 251, alcohol: and amnesia 192, 115, 125; avoidance
261–262; and encoding 195, 201–202; effect on learning 129, 130; cats in
265 memory 294–295; as a puzzle box 82; classical
accessible memories 161– reinforcer 94; and state conditioning 49, 51, 52,
162, 172 dependency 282, 295 53–55; conditioning with
acoustic encoding 208–209, Alzheimer’s disease 167, 179, drug USs 71; CS-US
230 187–188, 192, 221–222, relevance 65; delay of
acoustic startle response 31 314, 320, 321, 328 punishment 122; discrimi-
acquisition 55, 78 amnesia 15, 186, 187–188, native stimulus control
acquisition training 85–86 191–192, 200–201; 102; electroconvulsive
acronym mnemonics 167 anterograde 192–193, shock therapy (ECT) 274;
acting 326–327 194, 196, 201, 323; fast mapping 166;
actions, memory for 212 childhood 338–339; fear-relevant CSs 75;
active experiencing 327 classification of 192–193, genetics of learning ability
active experimentation 348, 193; cryptomnesia 330–331; habituation 27,
348 198–199; functional 33, 34, 35, 100; instru-
adaptive control of thought (psychogenic) 192, mental learning 85;
(ACT) theory 323, 328 196–197, 202; and learned helplessness 135;
adaptive responses 26, 61 hippocampal lesions long-distance navigation
Adderall 252 305–306; infantile 304; memory storage
advertising: and classical 338–339; organic 192; 273; nonreward learning
conditioning 68; sublimi- retrograde 192–193, 195, 121; organization 159; pet
nal 98 201, 202; treatment for containment systems
age differences: conditioning 274, 275, 276; types of 140, 144; positive
333–335; memory 193–195; see also reinforcement 87, 88, 89,
development 335–342 forgetting 96; reconsolidation
aggression, shock-elicited 124, amnesiacs 323; and false blocking 276; response
140, 144 recall 287 learning 100; response-
 Subject Index 419

reinforcer learning 284, 297; treatment of avoidance coping 143

105–106; room-specific 44–46; and working avoidance learning 116,
tolerance 72; selective memory 222, 231 128–132, 134, 142–143,
breeding 330; short-term Aplysia Californica 33, 34, 35, 144; approach-avoidance
memory (STM) 214–215, 37, 47, 273 conflict 130; cognitive
226; singing of sparrows applied research 13–14, 22; theory of 129–130;
8; Skinner box 83–84, see also research conditioning theory of
83; spatial learning approach-avoidance conflict 129; functional approach
300–303, 305; 130–131 130; summary of 130
­stimulus-response theory approach-coping style 131 avoidant coping style 131
105–106; working arousal 235, 249–253, 264; awareness: and learning 98,
memory 220; see also and circadian rhythms 163, 165, 300; and
maze learning 253; encoding and remembering 179, 181,
animal training 100, 101 249–253; recall and stress 184; role of in condition-
animals (non-human) used in 250; and retrieval 297; ing 69–70, 79
studies: cats 53, 54, 82, and stimulant drugs 252;
125, 130; chimpanzees targeted memory backward association 171
159, 214–215, 308–309, reactivation (TMR) backward conditioning 61
327; coyotes 54; dogs 44, 251–252; Yerkes-Dodson behavior(s): adaptive 82, 111;
51, 53, 57, 58, 71, 100, law 249–250, 249; see addictive 110; aggressive
101, 122, 124, 126, 140, also emotional arousal 6, 6, 18, 54, 101, 124,
144, 166, 304; dolphins articulation hypothesis 208 140; automatic 107;
214; exotic animals 101; artificial grammar 317, 317, avoidance 129–130;
fruit flies 330–331; 328 compulsive 140–141,
hamsters 100, 113; assertiveness training 136 144; food-related 94,
marine snails 33, 34, 35, associations 280 100; frustration 117;
47, 273; mice 19, 27, 54, associative theory 165 habitual 82, 105,
129, 130, 273, 305, attention control 219 107–108, 310; instrumen-
330–331, 333; monkeys attention-deficit hyperactivity tal 82; learned 104, 134;
75, 93, 214, 215, 226; disorder (ADHD) 252 and learning 5–9, 21–22;
one-celled organisms 35; attention-deficit hypothesis 245 reinforcers as 94–95;
pigeons 67, 83, 102, 103, autobiographical memory 17, spatial 305–306; volun-
106, 214, 215, 270, 304; 196; in children 338; tary 50
pigs 100; rabbits 53, highly superior 343–345; behavior modification 98–99,
333–334; rats 18, 19, 27, Western vs. Eastern 109–110, 114, 117, 142;
37, 39, 41, 53, 54, 55, 57, cultures 354 number of swear words
65, 83, 87, 88, 89, 93, 96, available memories 161, 172 used 111; self- 110, 111; of
97, 98, 101, 103, 105, aversion therapy 55 taking extra food 109
105–106, 112, 120–121, aversive events: controllability Behavior of Organisms, The
121, 122, 124, 127, 129, of 133, 135, 143; and (Skinner) 100
130, 133, 138, 251, 274, learned helplessness behavioral approach to
276, 284, 300, 301–303, 133–135; predictability of learning 17–18, 23
305, 307, 330–331, 332, 133, 135, 143 behavioral medicine 16
355; shelter animals 101; aversive learning 137–141, behavioral repertoire 21
sparrows 8; wolves 304 143–144 behavioral research 16
anisomycin 274, 276 aversive stimulus 84, 116, 120, behavioral silence 8
anomalous forgetting phenom- 124–125, 129–131, 133, behaviorism 17–18
enon 198–199 142 behavior-punishment contin-
anterograde amnesia 192–193, avoidance: cognitive theory of gency 123
194, 196, 201, 323 142–143; and repression biases: learned associative 75;
anxiety: and conditioning 132 predisposing 291
55–57; and the startle avoidance conditioning biofeedback 96, 112
effect 37–38; test 8, 260, 138–140 biological preparedness 4
420 Subject Index

biological substrates 273 caffeine: effect on learning and reinforcement 92; see
biology: evolutionary develop- memory 14; as stimulant also infant development
mental 35; see also 252, 264 chunking 209, 229
nature vs. nurture California Verbal Learning Test circadian rhythms 253, 262,
biorhythms 264 156 264
blocking effect 62–63 capacity: in children 336; classical conditioning 49–50;
brain: differences between limited 209, 19, 252, 321, and advertising 68;
individuals 321–322; 209–210, 216, 229, 230, applications of 74–77,
learning in 69; left vs. 249, 254; in older adults 79–80; basic phenomena
right 347; and maze-learn- 340; of working memory/ of 55–60, 78; and
ing 305–308; see also STM 221, 222, 228, 230, configural learning
brain injuries; brain 231, 336 189–191, 190; defined
physiology capture errors 108 50–52, 77; extensions of
brain bugs 277–278 causal attribution theory 136, 70–74, 79; methods of
brain injuries 18, 175–176, 176, 143 studying 52–55, 77–78;
179–180, 191–195, 231; causal learning 73, 165 other factors affecting
the case of H. M. 178, causality 73–74, 74 61–67, 78–79; and the
186, 187, 193–194; the central executive 216, 217, role of contiguity 60–61,
case of K. F. 178, 200, 220, 222, 231–232 78; stimulus-response
211, 226, 230; and central nervous system (CNS) theory 67, 67, 68, 79;
executive dysfunction 73, 188, 252 stimulus-stimulus theory
229; frontal-lobe impair- cerebellum 69, 79 67–68, 67, 79; summary
ment 294; and maze chaining 113, 169; narrative 77–80; what is learned in
learning 305–307; and 172, 173 67–70, 79; see also
memory 226, 226 child prodigies 321 conditioning; Pavlovian
brain physiology: amygdala childhood abuse 291–292, 298 conditioning
69, 137–140, 139, childhood amnesia 338–339 clustering 159; see also
143–144, 254, 274, 276; children: amnesia in 338–339; organization
cerebellum 69, 79; and and approach-avoidance cognitive ability, and gender
classical conditioning conflict 130–131; capacity 345–347, 356
69–70; dopamine 94; and in 336; cultural differ- cognitive approach: defined
fear conditioning 139– ences in memory 18–19, 23; to conditioning
140; hippocampus 176, 352–353; encoding in 56; to habituation 35–36,
178, 191, 194–195, 201, 335; and false recall 287; 36; to learned helpless-
226, 251, 273, 305–306, fear of animals 75–76; ness 137
322, 327, 330, 339; hearing impaired 213– cognitive deficits 135, 137,
neural basis of reinforce- 214; and instrumental 143, 167, 320; see also
ment 93–94; parietal lobe conditioning (punishment) developmental disabili-
344; temporal lobe 123, 124, 125–126; and ties; learning disabilities
266–267; thalamus 195, instrumental learning cognitive effort 236
202; see also brain 85–86; learning to read cognitive elaboration 154, 171,
scanning; brain stimula- 153; memory and 187, 235, 249
tion; neurons unpleasantness 284; cognitive impairment see
brain scanning 18–19, 19, 252, memory development in cognitive deficits
321, 343 335–339, 355; metamem- cognitive maps 18, 305, 307,
brain stimulation 93–94, 95, ory in 336–338; overesti- 309; in animals 17, 84,
112, 266–267 mation of quiz scores 304; distortion in 308; vs.
brain training 224; and maze 337–338, 337; recall and route knowledge 300–
learning 305–307 stress 250, 250; self-con- 303, 301, 327
Brown-Peterson distractor trol in 350–351; and cognitive psychology 18,
task 205–207, 206, 209, spatial memory 148–149, 145–146, 192
216, 230 309; and speech therapy cognitive skills 316
building design 325–326 104; and the use of cognitive strategies 235, 260
 Subject Index 421

cognitive theory 47, 165; of causality 73–74; in simultaneous 60–61; skin

avoidance learning discrimination training 59; conductance response
129–130 and drug tolerance 71–73; (SCR) 53, 60, 77; stimu-
cognitive therapy 140 in evaluative conditioning lus-response 105; taste
cognitive training (CT) 224 70, 79; exteroceptive aversion 54; theory of
cold pressor stress 95, 185, 65–66, 78; and extinction phobias 74–77; trial and
284 57, 76–77, 78; in eyeblink error 52; Type S and Type
common knowledge 15–16 conditioning 53; generali- R 52; unaware 98; verbal
common sense 15–16 zation 58–59; interocep- 98; Watson-Mowrer
comparative cognition 215 tive 65; and magazine theory of 129, 140, 144;
comparative evolution 35 approach conditioning 54; see also classical
compensatory response neutral 124; and PTSD 56; conditioning; conditioned
71–72; see also condi- relevance of to the US response (CR); instrumen-
tioned response (CR) 65–66, 67, 70, 77–79; and tal conditioning
compound conditioned the Rescorla-Wagner conditioning theory 129
stimulus 61–62, 62, 64 model 63–64, 78; and the configural learning 189–191,
conceptual nervous system role of contiguity 60–61, 190
(CNS) 188 78; in second-order connectionism/connectionist
concrete learning 348, 356 conditioning 59–60; modeling 188–191
concrete words 234, 239 sequential arrangements conscious awareness 4,
concurrent reinforcement 122 with US 60; and skin 69–70, 79, 98, 107, 113,
conditioned inhibition 66, 79 conductance response 179, 316, 328
conditioned reinforcement 91 53, 56, 56; and spontane- consciousness 220
conditioned response (CR) ous recovery 57–58, 58; consolidation 273
51–52, 52, 57–58; and stimuli serving as 54–55, consolidation theory 273–275,
acquisition 55; and 78; in stimulus-reinforcer 274, 296
classical conditioning learning 106; and taste contextual learning 1, 55,
67–68, 67; as compensa- aversion learning 53–54; 281–282, 281, 283,
tory response 71; control test trials 52, 55; and the 295, 298
procedures 57; defined treatment of phobias contiguity 3, 271; role of in
51; and extinction 57–58; 74–76 conditioning 60–61,
eyeblink conditioning 53, conditioning: and aging 78, 86
54, 57, 69–70, 77, 333– 333–334; anxiety and contingency: analysis of 73,
335, 334; magazine 55–57; backward 61; 74, 83–84; CS-US 69, 79;
approach conditioning 52, basic phenomena of 78; in instrumental condition-
54, 77; and phobias 75; compound 78; control ing 116, 141; omission
salivary 49, 51–52, 52, 57, procedures 57; evaluative 116; response-contingent
58, 60, 77; skin conduct- 70–71, 79; evolutionary/ punishment 121–122;
ance response 27, 27, 53, adaptive approach to response-outcome 89,
56, 56, 57, 69–70, 77; see 68–69; excitatory 66; 98; response-to-reinforcer
also conditioning; taste extinction 57–58; eyeblink 86–87, 86, 102, 113, 117
aversion learning; 53, 54, 77; fear 77; control procedures: classical
evaluative conditioning forward 60–61; generality conditioning 57; instru-
conditioned stimulus (CS) of 334–335; in infants mental conditioning 86-87
51–52, 62, 165; and 333; inhibitory 66; control processes 211
acquisition 78; and the magazine approach 54; coping 131, 137, 143
blocking effect 62–63, 78; other factors affecting coronavirus pandemic 45
competing 61–62; 78–79; Pavlovian 52, 68, correlational approach
compound 61–62, 64; and 73, 84; problem solving 328–329
conditioned inhibition 66; 52; role of awareness CR see conditioned response
contextual stimuli 55, 70; in 69–70; and the role (CR)
control procedures 57; of contiguity 78; cryptomnesia 198–199
defined 51, 77; detecting ­second-order 59–60, 59; cued recall 161–162
422 Subject Index

cultural psychology 353–354; dissociation 175, 176, 200, “earworms” 277–278

and memory 353, 354, 226; double 175; experi- Ebbinghaus 3, 146–147, 150,
357; working memory mental 185–186; neu- 161, 170, 172, 233, 240,
and 220–221 ropsychological 272–273; 267, 268, 296
curve of forgetting 147, 147, neuropsychological 186, effect, law of 83
267, 268, 296 200–201 elaboration 20, 236, 261;
distinctiveness 278, 297 conclusions about 239
d-amphetamine (d-AMP) 252 distortion 198, 199, 257, 289, elaborative interrogation 262
Darwin, Charles 3, 21 290, 327; in cognitive elaborative processing 236,
decay 266, 274, 276, 279, 297 maps 308 238–239; limits of 239
declarative learning 184, 299, distractor tasks 79, 205, 219, elaborative rehearsal 187, 201,
314 226, 227; Brown-Peterson 235–236, 263
Deese-Roediger-McDermott 205–207, 206, 209, 216, electric shock 19, 37–38, 50,
procedure (DRM) 286, 230 55–56, 65, 68, 77, 78,
287, 297–298 distributed practice 245, 246, 120–121, 124, 140, 284,
defensive responses 26 312, 313 330
déjà vu 199 domain-specific knowledge electroconvulsive shock (ECS)
delay of reinforcement 87, 241 273
90–91, 314 dopamine 94 electroconvulsive shock
delayed knowledge of results double dissociation 175 therapy (ECT) 195–196,
(KOR) 315 Down syndrome (DS) 202, 274, 275
delayed matching-to-sample 334–335 elicited imitation 335, 339
procedure (DMTS) 214 dreams 254, 288, 297 emotional arousal 233, 235,
delayed responding 214 drive 87–88, 112 254, 256, 257, 264, 320,
dementia 167, 221–222; see drive reduction 92, 92–93 324; and eyewitness
also Alzheimer’s disease drugs: antidepressant 94; memory 256; and
depression 132, 135–136, 137, d-amphetamine (d-AMP) memory 254; and
143, 195, 197, 222, 282, 252; heroin 71, 72; retrieval 283–284, 297;
341, 355 morphine 67, 71–72; for see also arousal
depth of processing theory OCD 141; protein emotional blocking 283–284
187–188, 201 synthesis inhibitors 274; emotional deficits of helpless-
desensitization 55 psychoactive effects of ness 135, 143
developmental disabilities 16; research using 27, emotions: and arousal
334, 355; see also 223; side effects 215; and 256–257, 264; and
cognitive deficits; learning state dependency 282; encoding 254–257; and
disabilities stimulant 95, 235, 250, retrieval 283; and
differential reinforcement of 252, 264; tolerance of memory 254, 257,
other (DRO) 126–127 71–72, 72; as uncondi- 326–327; unpleasant 141;
direction of initial learning 155 tioned stimulus (US) see also anxiety; emo-
discrimination 39–40, 47; 71–73, 79 tional arousal
generalization and 102, dual coding 240, 263 empiricism 3, 12, 21
104; odor 330; stimulus dual n-back task 218–219, 218, encoding 184–185, 185, 186,
78, 152–153, 171; taste 224 188, 197, 200, 233, 326,
239 dual-memory theory 35–36 328; and academic
discrimination training 57, 59; dual-process theory of learning 261–262;
and generalization 102–103 habituation 37 acoustic 208–209, 230;
discriminative stimuli 84, 85, dual-store theory 177–178, age differences in 335;
95, 113; control of 102, 188, 227, 232 and alcohol 294; applica-
103, 104; labels for 104 dual-task performance 218, tions of 261–262, 265;
disfluency effect 259–260, 228, 230, 231, 236, 238, and arousal 249–253;
259 336 basic variables 234–241,
dishabituation 31–32, 33, 47 dysexecutive syndrome 229, 263; deficits 294;
dissociating stages 185–186 232 elaborative 237; and
 Subject Index 423

emotions 254–257; evolutionary theory 68–69 eyeblink conditioning 53, 54,

enhanced 243; incidental exceptional memory 342–345, 57, 69–70, 77, 333–335,
247–248; learner variables 356 334
247–261, 263–264; and excitatory conditioning 66 eyeblink startle response 30,
metamemory 258–261, executive functions (EFs) 30
264–265; of odors 213; 219–220, 227; dysfunc- eye-hand coordination 314
presentation variables tion of 229 eyewitness accounts 15,
241–247, 263; and expectancy 36 256–257, 264, 279
schemas 257–258, 264; experience(s): aversive 56–57, eyewitness memory 256, 264
separating from retrieval 76; concrete 348; direct
234; summary 262–265; 139–140; emotional 254; false fame effect 161
variability in 245–246 knowledge and 3, 4, 12, false memory 286–287,
encoding specificity 281, 297 39; learning and 6–8, 21, 297–298; generality of
encoding specificity principle 47, 95, 295; mental 267, 287
280 296; observational false retrieval 286–289,
encoding variability hypothesis 139–140; shared 254; 297–298
246 stimulus-reinforcer 106; fan effect 272
encoding-retrieval paradigm tip-of-the-tongue 161, fast mapping 166
280, 281; limitations of 172, 285, 297; traumatic fear: classical conditioning of
283 135, 192, 196, 255, 256, 137–138; learned/
engram 33–34, 273 292 conditioned 124, 139–140,
environment: effect on experiential context 283 142; and OCD 140–141;
learning 4, 273, 331, 355; experimental approach therapy for 298
and language develop- 328–329 fear potentiated startle 37–38,
ment 4, 39; spatial 300, experimental dissociations 47
301, 307, 308, 327, 328; 185–186 feedback 2, 40, 96, 299;
see also nature vs. experimental extinction 15 consistent 328; delayed
nurture experimental studies see 328; internal kinesthetic
environmental context 283 research 328; knowledge of results
epic memories 351–352 expertise 321–323, 328 (KR) 314–316, 328; partial
epilepsy 176, 178, 186, 193, explicit learning 164, 320, 323 328; vs. self-guidance 316
196, 266, 306 explicit memory 180, 181, feeling of knowing (FOK)
epinephrine 95, 254 184, 200, 323–324 285–286, 297
episodic buffer 217–218, 231 exposure therapy 44–45, 48, fight or flight response 25, 46
episodic memory 178–180, 295 first-letter mnemonics 167,
193, 196, 200, 339, 346, external locus of control 136, 170, 236
354, 355; in children 335; 143 first-order conditioning 59
retrieval from 278–279 extinction 57–58, 78, 80, 84, fixed-interval schedules 89, 90
epistemology 2–3, 21 116, 117–120; experi- fixed-ratio schedules 89, 90
evaluative conditioning 70–71, mental 15; gradual 77; flashbulb memories 254–256,
79 overview 120; partial 264, 278
everyday forgetting 197, 198, reinforcement extinction flexibility 219
202, 340 effect (PREE) 118–119, flooding 44
evolution 3–4, 17, 21, 100; 119, 127, 141; resistance food tastes: conditioning to 57,
comparative 35; and to 118–119; stimulus 75; and illness 65, 75;
conditioning 50, 68–69; 125, 144; as therapy memory of 239; and
and gender differences 76–77 perceptual fluency 43;
347; and habituation 35; extinction bursts 117, 118 and short-term memory
and learning ability 331, extinguish in multiple contexts 214; see also taste
355; and spontaneous 77 aversion learning; taste
memory 278; and extramaze cues 303, 303, 307 learning
working memory 220 eye fixations 27–28; in infants forgetting 174, 185; and aging
evolutionary preparedness 100 28–29, 29 156, 341; anomalous
424 Subject Index

198–199; as coping generation effect 248 helplessness training 133–136

mechanism 131–132; and genes 329, 330–333 hereditary component 332;
depth of processing 187, genetic differences: in ability see also genetic
201, 225, 227, 230, 231; and skills 321; in learning differences
everyday 197–199, 202, ability 330–333, 355 heroin 71, 72
340; interference theories glucose 95 hierarchical associations,
of 289; of names 151; of GPS (ground positioning theory of 150, 170
naturalistically acquired devices) 324–325 highlighting 261, 265
knowledge 196; rapid 187, gradual extinction 77 highly superior autobiographi-
201, 205, 206, 231, 268; growth mindset 351 cal memory (HSAM)
as retrieval failure 186, 343–345, 344
266, 276–277; retrograde habit slips 107–108, 114 hippocampus 176, 178, 191,
192, 274; from STM 177, Habit Strength 87, 89 194–195, 201, 226, 251,
191, 210, 212; temporary habits 107–108 273, 305–306, 322, 327,
1, 297, 307; of traumatic habituation 24, 25, 100; 330, 339
experiences 292; see also applications of 48; H. M. (amnesic subject) 178,
amnesia biological approach to 47; 186, 187, 193–194
forgetting curve 11, 11, 22, cognitive theories of homing mechanisms 304
147, 147, 170, 267, 35–36, 36, 37, 47; defined human memory see memory
296–297 26, 46; and dishabituation hunger drive 87, 92, 109
forward conditioning 60–61 31–32, 33, 47; dual-mem- Huntington’s disease 314,
free association 240, 277 ory theory 35–36; 321, 328
free recall 156–160, 164, 171, dual-process theory 37; hypermnesia 278–279, 279,
172, 353; age effects 157; effects of repeated 30; 297
and clustering 159; and evolutionary theories of hypnosis 15, 22, 267, 296
organization 158–159; and 35; explanations of 32–33,
rehearsal 157–158, 158, 34, 35–38, 47; and food illness: and memory 15, 355;
160; and serial-position tastes 41–42; frequency and optimism/pessimism
effects 156–157; and of repetition 29; of human 137; psychosomatic 135,
subjective organization eyeblink startle response 143; stress-related 143;
159; summary 158–159 30, 30; learning-based symptoms of 73; taste
frequency of repetition 29–32, 36–37; as measure of associated with 53–54,
241 infant perception and 60, 65, 66, 75
frustration hypothesis 119–120 cognition 28–29; and imagery 234, 239–240
fugue reaction 202, 306 memory 35; methods of imagery mnemonics 168–170
functional amnesia 192, studying 26–28; neurosci- imagination inflation 288–289,
196–197, 202 ence theory of 33, 37; 298
functional approach 17, 23, nonlearning explanations implicit learning 163–164,
130, 143 32–33; novelty recogni- 316–321, 339; applica-
functional magnetic resonance tion task 27; and the tions of 323–324;
imagery (fMRI) 18 orienting response (OR) dissociating categories of
46; parametric features of 320–321; grammar
gender differences: in 29–32, 46; and sensitiza- learning 317–318, 317,
aggressive behavior 6; tion 37–38; short-term vs. 328; inverted-letter text
and cognitive ability long-term 31, 46; skin 318–319, 319; serial-reac-
345–347, 356; and conductance response tion-time procedure 317,
memory 346–347; in (SCR) 27, 27; spontaneous 317; tasks 317–320; see
response to caffeine 252; recovery 29–30, 31; also implicit memory
in spatial memory 345, stimulus specificity of 32; implicit memory 172, 200; see
346; in verbal memory to tone 27 also implicit learning
345–346 hearing impairment 213–214 implicit memory tasks 180,
generalization 58–59, 78; and helplessness see learned 181
discrimination 102–103 helplessness implicit priming 184
 Subject Index 425

implicit testing 163, 164, 184, drive reduction 92–93; interference hypothesis 149
186 elements of 84; and interleaving 20, 155
implosion therapy 44 extinction 117–120; four intermixing 155
impulsiveness 91 basic contingencies 117; internal cues 280, 284
incentive motivation 93, 112 and incentive motivation internal locus of control 136,
incentives 235, 248–249, 93; and learned helpless- 143
263–264 ness 133–136, 143; neural internal representations 18
incidental learning 187, 235, basis of reinforcement intrinsic motivation 99, 113
247–248, 248, 263 93–94; nonreward 98, intuition 21, 266, 297, 320
incidental memory 248 112, 114, 115–144; and Inuit populations 325
individual differences 329– OCD 144; and persis- inverted-letter text 318–319,
330; age differences in tence 127–128, 142; and 319
learning and memory punishment 120–127, 142; involuntary autobiographical
333–342, 355; excep- reinforcement variables memories 277–278
tional memory 342–345, 87, 88, 89–92; reinforcers involuntary semantic memo-
344, 356; gender and as behaviors 94–95; ries 277–278
cognitive abilities 345– reinforcers as information Iowa Gambling Task (IGT) 320,
346, 356; genetics of 95–96; reinforcers as 328
learning ability 330–333, strengtheners 95; isolation effects 235, 243–244,
355; learning styles theories of reinforcement 244, 263
347–351, 356; social and 92; use of rewards in
cultural differences 86–87, 88, 89–91, 96, 97, K. F. (amnesic subject) 178,
351–355, 356–357; 98, 101, 112; what 200, 211, 226, 230
summary 355–357; see associations are required? knowing: feeling of 285–286,
also age differences; 104; see also instrumen- 297; vs. remembering
gender differences; tal learning 160–161
genetic differences instrumental learning 81–82; knowledge 4; in children 336;
infant development 7–8; and and animal training 101; declarative 299;
conditioning 333; eye awareness in 98, 113; domain-specific 241;
fixations 28–29, 29; contingent relationship existing/prior 262, 326;
habituation as measure of between action and explicit 320; factual 316,
perception and cognition outcome 86; criticisms of 323; procedural 299–300,
28–29; and infant the use of reinforcement 328; remembering
memory 148; instrumen- 98–99; definition and 268–269; route 300–303,
tal learning 85–86, 85; history 82–83; and 301, 327; spatial 324–325
language learning discriminative stimulus knowledge compilation 323
165–166; reaction to 102, 103, 104; methods of knowledge of results (KR)
perceived threats 76; see study 85–86; necessity of 314–316, 328;
also children reinforcement 96, 97, 98; delayed 315
infantile amnesia 338–339 and positive reinforce- Korsakoff’s syndrome 194,
information processing 18, ment 86–87; and self-­ 195, 201–202, 294
234, 295 control 91; see also Kruger-Dunning effect 261
inhibition 219 instrumental conditioning;
inhibitory conditioning 66 reinforcement; reinforc- landmarks 307, 327
instrumental conditioning 84, ers; response-learning language discrimination 39;
99, 111; applications of theory Japanese ideograms 42
107–111, 114, 144; and instrumental response 84, language learning 17, 153–155,
aversive learning 137– 104, 105 156, 165; and autobio-
141; and avoidance intellectual deficits see graphical memory 338;
learning 128–132, cognitive deficits, concrete vs. abstract
142–143; and causal developmental disabilities words 234; and false
attribution theory intentional learning 247–248, recall 287; grammar
136–137; defined 116; and 263 learning 317–318, 317,
426 Subject Index

328; using mnemonic affecting 261; functional learning ability: biological

imagery 262; neuropsy- approach 23; genetics of costs of 331; genetics of
chological dissociations learning ability 330–331, 330–331
272–273; and short-term 355; goal driven 84; learning curve 9, 10–11, 11,
memory 228; statistical incidental 187, 235, 327
165–166 247–248, 248, 248, 263; learning disabilities 207, 225;
latent learning 8, 22, 93, 96, instrumental-response see also cognitive
97, 98, 105, 113 112; intentional 247–248, deficits; developmental
law of effect 83, 111 263; language 153–155, disabilities
learned associative bias 75 156, 165–166; latent 8, learning styles 347–351;
learned helplessness 133–135, 22, 93, 96, 97, 98, 105, Kolb’s two-dimensional
133, 134; and causal 113; and maturation 7–8, organization 348, 348,
attribution theory 136; 22; measurement of 5; 356; self-categorization of
and depression 135–136; memory and 22; and 349–350; and self-control
and physical health 137; motivation 87; motor-skills 350–351, 356;
summary 143 309–316, 327–328; name Sternberg’s model
learned safety hypothesis 41 learning 151–152; neural 348–349; visualizer-verbal-
learner variables: arousal network models of 64; izer dimension 347
249–253; emotions and neuroscience approach levels of processing 187, 188,
encoding 254–257; 23; operant 83–84, 111, 235, 236, 247, 263
incentives 248–249; 114; paired-associate limited capacity, short-term
incidental vs. intentional 151–155, 164, 171, 172; memory 177, 200,
learning 247–248; vs. performance 8–9, 22, 209–210, 216, 229, 230
metamemory 258–261; 61, 68, 87, 112; perma- limited duration, short-term
schemas 257–258 nence of 7; place 304, memory 200, 210, 224,
learning: abstract 356; 327; practice-independent 230
academic 251, 261–262, 313–314; prepared 66, 75; Little Albert case study 74–75,
265; age differences in procedural 181, 226, 300, 129
333–342, 355; associa- 320, 339; propositional Locke, John 3, 21
tion 164; aversive 165; response 99–101, locus of control, internal vs.
137–141, 143–144; 113, 171, 301, 302, 327; external 136, 143
avoidance 116, 128–132, response-reinforcer 105, long-term memory (LTM) 35,
134, 142–143, 144; 113; resulting from 177, 184, 200, 226; in
behavioral approach experience 6; rote 146; animals 270; and arousal
17–18, 23; breadth of 5; sleep 250–251; social 264; consolidation of
causal 73, 165; circadian 139–140; state-dependent 275; encoding and
rhythms and 253; and 281, 281, 294, 297; retrieval from 217;
classical conditioning statistical 164–165, 172, episodic and semantic
67–70, 79; cognitive 339; stimulus 104; 178–180, 193, 200;
approach 23; concrete stimulus-reinforcer 106; fearful 276, 298; forma-
348, 356; configural stimulus-response tion of 274; for naturally
189–191, 190; context-­ 105–106; study of 22–23; learned material 268–
dependent 1; context-­ subliminal 98; trial and 270; organization of 184,
specific 281, 295, 298; error 82–83; unaware 296; permanent 204,
contextual 281–282; 319–320; vicarious 79; 209, 210, 296; and the
contingency 73–74; visual 347; see also role of rehearsal 235; vs.
declarative 314; deficits implicit learning; instru- short-term memory 18,
167; defined 4–8, 9, mental learning; maze 47, 157, 158, 177–180,
21–22; direction of initial learning; perceptual 204; transfer from
155; effects of additional learning; serial learning; short-term memory
study or practice test spatial learning; taste 210–211, 230; unstable
11–12, 12; explicit 164, aversion learning; verbal 273; working memory
320, 323; factors learning connecting to 216
 Subject Index 427

long-term personal memory meaningfulness 21, 153, 170, 323–324; eyewitness

148 171, 233, 234, 236, 239, 256, 264; for faces 238;
long-term potentiation 296 240–241, 262, 263 factors affecting 341,
long-term skill retention 148 memorists 342–345; see also 342, 355; false 286–287,
long-term store (LTS) 205 mnemonists 297–298; flashbulb
memory/memories: abnormal 254–256, 264, 278;
magazine approach condition- 191–197; and absent- forgotten 266–267, 275;
ing 52, 54, 77 mindedness 163, 197, free recall 171, 172; and
Magic Shrinking Box 335, 339 198, 202, 204, 260, 307, habituation 35; for high
maintenance rehearsal 187, 341; of abuse 291–292; school classmates 268,
201, 235, 263 accessible 161–162, 172; 269; immediate 335–336;
mania 282 acting and 326–327; for implicit 172, 180, 181,
map learners 326 actions 212; age differ- 200; improvement of
marking stimulus 90–91 ences in 333–342, 349, 15–16, 328, 336; inciden-
massed practice 1–2, 155, 355; aging and 340–342, tal 248; infant 148;
245, 312–313, 316 342; alcohol and 294– involuntary 277–278;
massed repetitions 20, 245, 295; autobiographical 17, lapses in 132, 174; modal
246, 247, 263, 311 196, 338, 343–345, 354; model 177, 200, 204–205,
maturation 7–8, 22 available 161, 172; 205, 216, 227; mood-con-
maze learning: adaptation to college professors’ gruent 282; mood-de-
punishment in 122; memory for their pendent 283, 297; myths
behavioral vs. cognitive students 163; consoli- about 22; odor 213;
approach 19; and brain dated 275; copy theory partitioning 175–176,
physiology 18; of 292; cued recall 199–200; permanent
­caffeine-enhanced 14; 161–162; decay of 266, 266–267, 273, 296; for
cognitive maps vs. 274, 276, 279, 297; pictures 239–240;
routes 84, 300–301, 327; declarative 184; decline primary 203, 229;
and discriminative in capacity 340; deficits priming 181–184, 182,
stimulus control 102; 167, 179–180; defined 9, 183; prospective 284–
habitual behavior in 22; déjà vu 199; depth of 285, 297; for public
105–106, 107; inability to processing theory events 254–255, 269–
learn 178; and incentive 187–188, 201; disputed 270, 278; recent 227;
motivation 93, 105–106; 288; distinctive 297; recognition 160–164,
and instrumental learning distortion in 198, 199, 171, 176; reconsolidation
85; and latent learning 257, 289, 290, 308, 327; of 298; reconstructive
22, 96, 97, 98, 112; Maze dual-store theory 292–293; recovered
Bright vs. Maze Dull rats 177–178, 188, 227, 232; 291–292, 298; remem-
330–331, 332, 355; and effects of stressors 186; bering vs. knowing
memory impairment and electroconvulsive 160–161, 171; repressed
284; Morris water maze shock therapy (ECT) 291; retrieval 185, 186,
305, 331; and motor 195–196, 202, 274, 275; 188; secondary 203, 229;
skills learning 309–311, for emotional vs. semantic 178–180, 193,
327–328; and perceptual non-emotional items 256; 194, 200, 264, 270–273,
learning 39; place vs. and emotions 257; 271, 277; social and
response studies encoding 184–185, 185, cultural differences
301–302, 302; practice 186, 188, 197, 200; epic 351–354, 356–357;
311–314; radial maze 351–352; episodic source 287–288, 297–
302–303, 303, 306, 309; 178–180, 193, 196, 200, 298; stages of 200–201;
sleep-enhanced 251; and 278–279, 335, 339, 346, storage 185, 186; stored
spatial learning 175, 327; 354, 355; everyday 294; strategies for
use of rats 7, 8, 300, 309, 14–15, 247; exceptional searching 293–294;
330–331; virtual reality 342–345, 356; explicit temporary 273; three
251 180, 181, 184, 200, approaches to 175; of
428 Subject Index

trauma 132, 255, mood-congruent memory 282 stages of memory

291–292, 298; for TV mood-dependent recall 281, 200–201; of two-memory
shows 269; two-memory 282–283 systems 226
systems 226–227; verbal Moonwalking with Einstein neuropsychology 187–188,
343, 345–346; visual 212, (Foer) 170 296, 298; cognitive
230, 353; word 176, 240; morphine 67, 71–72 deficits 191
see also long-term Morris water maze 305, 331 neuroscience 215; application
memory (LTM); mnemon- motivation: deficits in 135; to instruction 20; and
ics; recall; short-term incentive 93, 112; intrinsic aversive learning 137–
memory (STM); spatial 99, 113; and learning 5, 7, 140, 143–144; compara-
memory; working 8, 9, 21, 247; and tive 35; research in 27, 63,
memory memory 340, 341; see 79, 234
On Memory (Ebbinghaus) 146 also drive neuroscience approach 23;
memory blackouts (MBOs) motivational deficits 135, 143 defined 18–19, 33, 34, 35
295 motor-skills learning 309–316, nonsense syllables 146
memory consolidation 112, 316, 327–328; knowledge note-taking 261, 265
273–276 of results 314–316; novelty recognition task 27
Memory Olympics 169–170 practice 311–314, 328 nurture see nature vs. nurture
memory self-efficacy 260–261 Mozart effect 20, 23
memory span 207–208, 230, multitasking 222–223, 231 object locations 347, 353
340; in children 336 music: background 61, 283; obsessive-compulsive disorder
memory tests 344 earworms 277–278; (OCD) 140, 144
memory-attribute model 225 Mozart effect 20, 23; odor memory 146, 283;
mental rotation test 345–346, shock-avoidance experi- encoding 213; short-term
346 ment 131; talent in memory for 213
mere exposure effect 40, 41, 321–322, 323; as US 55 omission 116
42, 43, 48 operant conditioning chamber
metacognition 315–316 name learning 151–152 83–84, 83
metamemory 258–261, narrative story method operant learning 83–84, 111,
264–265, 297, 336–338; 167–168 114
and aging 340–341; and naturalistically learned material operant response 84
partial retrieval 285–286 196, 267–268 optimism 137
method of loci 168, 169, nature vs. nurture 2–3, 4, OR see orienting response
172–173, 343 330–333 (OR)
methylphenidate 252 n-back task 218–219, 218, organic amnesia 192
mild cognitive impairment 224 organization: in animal
(MCI) 167 near transfer 224 memory 159; categorical
mirror-tracing 310 needle phobia 45–46, 48 171; of knowledge 270,
misidentification disorders 199 negative emotional effects 296; of long-term
misinformation effect 289 128; see also anxiety memory 184; in free
missing stimulus effect 36, 47 negative reinforcement 116, recall 158–159, 296, 308,
mnemonics 167–170, 172– 124 352, 353; of semantic
173, 336; acronym 167; negative self-talk 222, 231 memory 272; in spatial
imagery 168–170, 262; neophobia 41 memory 308–309;
narrative story method neural network models 270 subjective 171
167–168; serial recall of neural networks 188–191, 201 orienting 27–28
20-word lists 169; verbal neurons 252; dopamine 94; orienting response (OR)
167–168 and spinal cord reflexes 25–26, 43, 46
mnemonists 342–345 37; synapses between On the Origin of Species
modal model 177, 200, 26, 33, 35, 50, 188, 270, (Darwin) 3
204–205, 205, 216, 227 273 overdose reactions 72
modeling causality learning neuropsychological dissocia- overshadowing 62, 239
73–74, 74 tions 186, 272–273; and overtraining 77
 Subject Index 429

paired-associate learning persistence of responding 127, prediction error 63

151–155, 164, 171, 172; 127–128, 142; generalized Premack principle 94–95, 112
analysis of 152–154; 128, 128 prepared learning 66, 75
response learning 153; personality 41, 76, 116, 136, preparedness hypothesis
stimulus discrimination 243, 249, 252, 329–330 75–76
152; stimulus-response pessimism 137 primacy 158, 171, 177, 200,
associating 153–154 pet containment systems 140, 214
pairing: compound CSs 61; 144 primacy effect 156, 157, 158,
CS-US 51, 52, 53, 57, phobias 74–77; conditioning 204, 291, 353
60–61, 69, 77, 78, 334; theory of 74–77, 76; primary memory 203, 229
stimulus-response 151; defined 74; desensitiza- primary reinforcers 91–92, 112
tone-shock 19, 61, 274, tion treatment 80; priming 181–184, 182, 183,
276 exposure therapy for 277, 320; implicit 184
panic disorder 38 44–46, 48, 295; extinction principle of reinforcement 83
parametric features 29–32, 46 as therapy 76–77; and prior exposure 39, 134
Parkinson’s disease 314 memory reconsolidation proactive interference 149,
partial punishment 128 298; needle 45–46, 48; 206, 207
partial reinforcement 89, 112, spontaneous recovery 58 problem-solving 228–229, 231
128, 314 phobic-type stimuli 45, 58, 75, procedural knowledge
partial reinforcement extinc- 76, 79, 80 299–300, 328
tion effect (PREE) phonological loop 216–217, 230 procedural learning 181, 226,
118–119, 127, 141 physical health 137 300, 320, 339
partial retrieval 297 picture memory 239–240 propositional learning 165, 172
Pavlovian conditioning 52, 68, place vs. response studies propranolol 296
73, 84; see also classical 301–305, 327 prospective memory 284–285,
conditioning placebo effect 2, 224 297
peg word system 168–169, positive reinforcement 84, protein inhibitors 274, 275,
169, 240 86–87, 111–112, 116–117, 276
perceptual fluency hypothesis 122, 124, 141 psychogenic amnesia 192,
43 positron emission tomography 196–197, 202; see also
perceptual identification 180, (PET) 18–19, 18 functional amnesia
181 postevent information psychology: behavioral 99;
perceptual learning 38–40, 289–290, 290, 298 biological 33; cognitive
47–48; factors affecting post-traumatic stress disorder 18, 145–146, 192;
39–40, 47–48; scribbles (PTSD) 38, 45, 48, 55–56, comparative 329;
used to assess 38 76, 132, 254; conditioning correlational approach
Perceptual Learning (Gibson) 38 in 56 328–329; cultural 353–
performance: academic 124, potentiated startle 37, 43, 47 354; experimental 14;
137, 222, 252; and aging power curve 11, 311 experimental approach
340–342; cognitive 223; power law 327 328–329; humanistic 41;
defined 8; dual-task 228, practice: distributed 245, 246, personnel 5; see also
230, 231, 236, 336, 238; 312, 313; and expertise neuropsychology
feedback about 40, 315, 322; massed 1–2, 155, psychopathology 135
328; inhibition of 313; vs. 245, 312–313, 316; and psychosomatic illness 135, 143
learning 8–9, 22, 61, 68, motor-skills learning psychotherapy 140
87, 112; long-term skill 311–314; schedules of public events 193–194, 196,
retention 316; motivation 311–312, 312, 313; and 269–270, 296
for 113; overestimation of talent 322; see also punishment 116, 120–127, 142;
337–338, 337; psychologi- rehearsal alternatives to physical
cal 135 practice testing 11–12, 12, 20, punishment 126–127; and
permanent-memory hypothe- 22 concurrent reinforcement
sis 266–267, 296 practice-independent learning 121; delay of 121;
perseveration 229 313–314 effectiveness of 121–124;
430 Subject Index

incompatible responses maintenance 187, 201, release from proactive

121; individual differences 235, 263; in short-term interference 206
124; intensity of 121, 125; memory 191; see also remembering 160; state-de-
vs. nonreward 121, practice pendent 202
125–126; paradoxical reinforcement: amount of 87; remembering vs. knowing
rewarding effects of concurrent 122; condi- 160–161, 171
124–125; partial 128; tioned 91; continuous 89; remote associations 150
question of use 126; criticisms of the use of renewal 58, 77, 80
response-contingent 121; 98–99; defined 96; delay repetition effects 241
schedule of 121, 122; side of 87, 90–91, 314; delayed repressed memories 291
effects of 124–125; 128; differential 126–127; repression 132
spanking 125–126; verbal and drive 87–88; food as Rescorla-Wagner model
rationale for 123, 123 100; and misbehavior 126; 63–64, 78
punishment-aggression necessity of 96, 97, 98, research: using animals 16–17,
hypothesis 124 112–113; negative 116, 22; applied 13–14, 22;
pursuit rotor 310–312, 310, 124; neural basis of basic 13, 22; behavioral
312, 320, 327 93–94, 112; in operant 16; brain 20; cross-cultural
learning 111; partial 89, 352–353, 357; real-world
radial maze 302–303, 303, 112, 128; partial (extinc- 15, 44; twin studies 7,
306, 309 tion effect, PREE) 288, 332–333, 335; see
radical behaviorism 17–18 118–119, 119, 127, 141; also animal studies; maze
random control procedures 57 primary/secondary 92–93, learning
reaction time 318, 318 112; principle of 83, 85; resistance to extinction
reading comprehension 228, schedule of 89–90, 90; 118–119; see also PREE
231 secondary 91, 92, 93; response cost (RC) 126–127
real-world research 15, 44 social 91–92, 112, 122; response learning 99–100, 153,
recall 162–163; mood-depend- theories of 92, 112–113; 171; limitations of 100
ent 282–283; postevent variables affecting response override 132
information on 289–291; acquisition 87, 88, 89–96, response prevention 141,
spontaneous 277–278; 92; see also positive 144
and stress 250, 284; see reinforcement; reinforcers response substitution 110
also encoding; free recall; reinforcement schedule: response-contingent punish-
memory/memories; serial fixed-interval 89, 90; ment 121
recall fixed-ratio 89, 90; response-learning theory 67,
recall test 162 intermittent 90; varia- 99–101, 113, 301, 302,
recency 156, 158, 171, 177, ble-interval 89, 90; 304, 327; summary 104
200, 214 variable-ratio 89, 90 response-outcome contin-
recency effect 156, 204 reinforcer priming 93 gency 98
recent memory 227 reinforcers 104; as behaviors response-reinforcer associa-
recognition 160–164, 171, 176; 94–95, 112; defined 112; tion 105, 106, 113, 117
remember vs. know devaluation of 106; as response-reinforcer learning
160–161 information 95–96, 112; 105, 113
reconsolidation hypothesis positive 92; primary/ restaurant schema 257
275–276, 295–296, 298 secondary 91–92, 112; in retention 145, 146, 147, 147,
reconstructive memory response-reinforcer 160, 187, 209; long-term
292–293, 298 learning 105; as stimuli 148, 155, 187, 211, 270,
recovered memory 291–292, 112; as strengtheners 95, 314, 316, 328; short-term
298 112; see also reinforce- 205, 210, 211, 219, 221,
reflection 348 ment; response-reinforcer 227, 230, 235; verbal
reflex responses 26, 100 learning overshadowing 239
rehearsal 160, 171, 211, 230; reinstatement 31, 58, 77, 80, retention intervals 9, 12, 176,
elaborative 187, 201, 282 206, 221, 226, 246, 270
235–236, 263; and free relearning 162–163; see also retention tests 147, 162, 171,
recall 157–158; savings in relearning 212, 268, 274
 Subject Index 431

retrieval 185, 186, 188, scribbles 38–39, 38, 40, 47 processes 211; correct
276–277, 297; and secondary memory 203, 229 recall in three span tasks
emotional arousal 297; secondary reinforcement 91, 213; defined 227; and
false 297–298; partial 92, 93 executive functions 227;
285–286, 297; practice second-order conditioning failures of 197; forgetting
278–279; vs. reconstruc- 59–60, 59 from 210, 225–226, 226,
tion 298; reconstructive selective breeding 330 230; in the hearing
process of 292; strategies self-behavior modification 110, impaired 213–214; history
for 293–294 111 of 204–205, 229; hypothe-
retrieval cues: associations self-control (S-C) 91, 350–351, sizing 225–227; and
280; changing 293–294; 356 language learning 228;
context-specific 281; self-distancing 132 limited capacity 209–210,
contextual learning self-efficacy 260–261, 262, 230; limited duration 210,
281–282; cues 279–283; 265; and aging 341 230; vs. long-term
encoding specificity 280, self-guidance hypothesis 315, memory 18, 47, 157, 158,
281; limitations of 316 177–180, 204; memory
encoding-retrieval self-referent effect 248 span 207–208, 230; for
paradigm effects 283; semantic memory 178–180, odors 213, 230; and
mood-dependent 281, 193, 194, 200, 264, problem solving 228–229;
282–283; state-depend- 270–273, 271; involuntary and the role of rehearsal
ent 281, 282; verbal 339 277 235; sensitivity to
retrieval failure 1, 266, semantic network theories disruption 210, 212; spatial
276–277, 286–287 270, 271 212; summary 229–232;
retrieval practice effect 242 sense of direction 324–325 tasks 230; transfer to
retrieval tasks 18–19 sensitization 37–38, 43, 47 long-term memory
retroactive interference sequential hypothesis 120 210–211, 296; verbal 216;
149–150 serial learning 146–151, 164, visual 212, 230; see also
retrograde amnesia 192–193, 170, 172, 173; and working memory
195, 201, 202 proactive/retroactive short-term store (STS) 205
reward training 84; see also interference 149–150; and sign language 210, 213–214
positive reinforcement remote associations 150; simultaneous conditioning
rotary pursuit 310–312, 310, and serial position 60–61
312, 320, 327 148–150; serial-position skilled memory theory 326
Rousseau, Jean-Jacques 3 curve 148–150, 149; skills: acquisition of 323;
route knowledge 300–303, summary 151 long-term retention of
301, 327 serial-position curve 177, 178 316; memory 326–327
rumination 56, 222 serial-position effect 156, 171 skin conductance response
serial-reaction-time procedure (SCR) 27, 27, 53, 56, 56,
salivary conditioning 49, 51, 317, 317 57, 69–70, 77
52, 52, 57, 58, 60, 77 shallow processing 187, 201, Skinner box 54, 83–84, 83;
savants 321 235–236, 238, 240, 261, see also operant condi-
savings in relearning 146, 147, 263, 265 tioning chamber
150, 161, 162, 170, 172 shaping 99–100, 100 sleep learning 250–251
schedule of punishment 121, short-term memory (STM) 35, slot machines 90
122 177, 191, 197, 200, social facilitation 77
schedule of reinforcement 203–204, 273; acoustic social learning 139–140
89–90, 90 encoding 208–209; for social reinforcement 91–92,
schemas 257–258, 262, 264, actions 212; in animals 112, 122
292–293; restaurant 214–215; applications of source amnesia 199
schema 257; spatial 327; 227–229; Brown-Peterson source memory 287–288,
in spatial memory distractor task 205–207, 297–298
307–308 206, 209, 216, 230; spaced repetition 20, 23, 263,
SCR see skin conductance characteristics of (verbal) 268, 311; and habituation
response (SCR) 208–211, 230; control 31, 47
432 Subject Index

spaced versus mass practice state-dependent learning 281, stimulus extinction 125, 144
effect 1–2, 155 281, 282, 294, 297 stimulus learning 104
spacing effects 245–247, 263; state-dependent remembering stimulus satiation 109, 114
attention-deficit 245; 202 stimulus selection 61
conclusions about statistical language learning stimulus-reinforcer learning
spacing 246–247; 165–166 106
encoding variability statistical learning 164–165, stimulus-response associating
245–246; optimal spacing 172, 339 153–154
interval 246 stereotype threat 9, 222 stimulus-response (S-R)
spacing extinction trials 77 stereotypes 192, 196, 300, theory 67, 67, 68, 79,
spanking 125–126 321, 354, 355; of age 174, 105–106
spatial information 204, 217, 340, 341, 342; gender stimulus-stimulus (S-S) theory
289, 299, 327 346, 356; negative 222 67–68, 67, 79
spatial learning 85, 300–309; stimulant drugs 95, 235, 250, storage 185, 186, 267–270;
long-distance navigation 252, 264 dual- 177, 200; nature of
by animals 304; maze stimulus/stimuli 24–25; 270–276, 296; in schemas
learning and the brain aversive 84, 116, 120, 257; short-term 187, 228;
305–307; rats, mazes, 124–125, 129–131, 133, temporary 215–216
and psychology 300; 142; conditioned 51, 52, strengtheners 95
routes vs. cognitive 54; contextual 55, 283; stress 101, 131, 137, 143, 185,
maps 300–303, 301, contrasting 39–40; 196–198, 200, 250,
305; schemas in spatial defined 25–26, 46; 283–284; emotional 233,
memory 307–309; discriminative 84, 85, 95, 250, 254
spatial memory in 102, 103, 104, 113; effects Stroop task 219
children 309; summary of exposure 48; studying, recommended
327; see also spatial environmental 114; methods 20–21, 155-156,
memory frequency of repetition 29; 261-262
spatial memory 148–149, 306, internal 282; marking subject factors 235
343, 353; gender differ- 90–91; missing stimulus subjective organization 159;
ences 345; and improved effect 36; neutral 112; and see also organization
building design 325–326; the orienting response subliminal perception 2, 98
mental rotation test (OR) 25–26, 43; successive approximations 99,
345–346, 346; organiza- perceptual learning of 113
tion in 308–309; schemas 38–39; phobic 45, 58, 75, suffix effect 210
in 307–308; and sense of 76, 79, 80; preference for summarizing 261, 265
direction 324–325; familiar 40–43; as supertaskers 223
short-term 212; see also reinforcers 112; repetition surprise 194, 256
spatial learning of 48; response to 34; survival processing 248
species-specific defense secondary reinforcers 91; swearing reduction 110, 111
responses (SSDRs) 130 selection of 61; synapses 26, 33, 35, 50, 188,
specificity: context 298; sensitization to 37–38; 270, 273
encoding 280, 297; spacing of 31, 36, 112; synesthesia 343, 344, 356
stimulus 32 specificity of habituation
spinal cord reflexes 37 32; Tone-Light (T-L) 36; talent 299, 321–322, 328
spontaneous recall 277–278 transfer from easy to targeted memory reactivation
spontaneous recovery 29–30, difficult 40; unconditioned (TMR) 251–252
31, 43, 57–58, 58, 76–77, 51, 55; as unconditioned taste aversion learning 13, 41,
78, 117–118 stimulus 55; see also 53–54, 55, 60, 61, 66, 66,
spreading activation 271–272 conditioned stimulus (CS); 69, 75, 76–78, 106; in fruit
S-R theory see stimulus-re- unconditioned stimulus (US) flies 331; stimulus
sponse (S-R) theory stimulus control 102, 103, 104, selectivity in 65–66, 65
S-S theory see stimulus-stim- 113 taste learning 41–42
ulus (S-S) theory stimulus discrimination taste neophobia 41–42
startle reaction 26, 36, 37, 47 152–153, 171 technology, behavioral 98–99
 Subject Index 433

television shows 196, 229, and phobias 76; relevance visuospatial sketchpad 216,
267, 269 of CS to 65, 78; and the 217, 230
temperature 71, 72, 283 role of contiguity 78; vocabulary learning 230; see
test anxiety 8, 260, 284, 297 sequential arrangements also language learning
testing effect 235, 241–243, with CS 60; in stimulus-re- von Restorff effect 243, 244,
263, 278, 297 inforcer learning 106; in 263; see also isolation
testosterone 347 taste aversion learning 53; effect
thalamus 195, 202 test trials 55, 66
thirst drive 87, 92 uncontrollability 135 warning signals (WS) 129,
time-outs 126 underlining 261 130, 131, 133, 138, 139,
time-of-day dependency 283 updating 219 141, 142, 144
tip-of-the-tongue (TOT) UR see unconditioned Watson-Mowrer theory 129,
experience 161, 172, 285, response (UR) 140, 144
297 US see unconditioned way finding 324
tolerance 71–72 stimulus (US) weapons focus 256, 264
trauma: amnesia following Wechsler Intelligence scales
192, 194–197; depression vanishing-cue method 207
caused by 135; memory 323–324 word memory 176, 240
of 132, 255, 291–292, variable-interval schedule 89, word-length effect 208
298; theory of phobias 76, 90, 100 working memory 204; and
79; unremembered 74; variable-ratio schedule 89, 90 aging 221, 340; and
von Restorff effect 256; verbal information 204, 239 anxiety 222; and the
see also brain injuries verbal learning 145–146, central executive 216, 217,
trial-and-error learning 82–83, 170–171, 175, 347; 231; cognitive training
111 applications of 167–170; (CT) 224; and conscious-
twin studies 7, 288, 332-333, the Ebbinghaus legacy ness 220; and culture
355 146–147; free recall 220–221; defined 227;
two-process learning theory of 156–160, 171; and and dementia 221–222;
avoidance 129, 130 mnemonics 167–170, and the episodic buffer
172–173; paired-associate 216, 217–218, 231; and
unaware learning 319–320; learning 151–155, 171; executive functions
see also awarewness in and recognition memory 219–220; impairment of
learning; implicit learning 160–164, 171–173; 222; individual differences
unconditioned response (UR) relationships among the in 221–224, 231; meas-
51, 57–58; and classical verbal-learning tasks 164; urement of 218–219;
conditioning 67, 67; serial learning 148–151, model of 215–216, 216;
defined 51; and drug 170; statistical learning and multitasking 222–223,
tolerance 70; and evalua- 164–166; summary 170; 231; and the phonological
tive conditioning 70; in see also verbal memory loop 216–217, 230; and
eyeblink conditioning 53; verbal memory 343, 345–346; problem solving 231; and
and magazine approach see also verbal learning reading comprehension
conditioning 54; and the verbal overshadowing 239 228, 231; susceptibility to
Rescorla-Wagner model verbal rationales 123 distraction 221; theoreti-
63; salivation 51–52; in verbal rehearsal 227, 230, 326 cal overview 215–216;
taste aversion learning 53 verbal-learning tasks 164 training 224–225; and the
unconditioned stimulus (US) video games 313 visuospatial sketchpad
51, 55, 165, 189–191; virtual reality exposure therapy 216, 217, 230; see also
aversive 75, 77, 124; and 45 short-term memory (STM)
classical conditioning visual encoding 239 worry 222; see also anxiety
67–68, 77; defined 51; visual learners 347
devaluation of 68; and visual memory 353 Yerkes-Dodson law 249–250,
extinction 76; pairing with visual short-term memory 249, 264
conditioned stimulus (CS) 212, 230 “you-are-here” (YAH) maps
57, 58, 60–61, 68, 70, 79; visual-spatial memory 217 326