A SOCIOCOGNITIVE MODEL

OF ATTITUDE STRUCTURE
AND FUNCTION

Anthony R. Pratkanis Anthony G. Greenwald

BOARD OF PSYCHOLOGY DEPARTMENT OF PSYCHOLOGY
UNIVERSITY OF CALIFORNIA UNIVERSITY OF WASHINGTON
SANTA CRUZ, CALIFORNIA 95064 SEATTLE, WASHINGTON 9819.5

I. Introduction

In 1935, Gordon Allport declared that attitude is “social psychology’s most

indispensable concept. That observation remains true today, as evidenced by
”

common practices within the discipline of social psychology. Attitudes receive

prominent treatment by social psychology textbook authors. They are often used
to explain such diverse phenomena as prejudice, interpersonal attraction, consumer
behavior, organizational behavior, human sexuality, mass communications, social
influence, and the self-concept. As McGuire (1969) points out, attitudes are
often a high-volume research area for psychology, accounting for large amounts
of psychological research at given times. And as Katz and Stotland (1959) noted,
the attitude concept finds employment in all the major theoretical schools of
social psychology including field, behavioral, cognitive, mathematical, and ex-
pectancy-value approaches.

A. BUT IS THE ATTITUDE CONCEPT

REALLY NEEDED?

The history of attitude research presents a paradox (see Katz & Stotland, 1959).
Despite the perceived importance and the popularity of the attitude construct, it
has been severely criticized, with some calling for it to be abandoned. Disillu-
sionment with the attitude concept stems from three basic concerns.
I . Attitudes appear to be poor predictors of behavior toward their objects.
In his definition of attitudes, Allport (1935) explicitly specified that attitudes
ADVANCES IN EXPERIMENTAL 24.5 Copyright 0 1989 by Academic I’rcar. Inc
SOCIAL PSYCHOLOGY. VOL. 22 All rights of reproduction in any torm reserved.
246 ANTHONY R . PRATKANIS AND ANTHONY G. GREENWALD

direct behavior. Not surprisingly, it is generally assumed that attitudes should

predict behavior. Weakness of empirical support for this behavior-prediction as-
sumption has been the basis for several prominent critiques, including those of
LaPiere (1934), Festinger (1964), Wicker (1969), and Abelson (1972).
2. Attitudes appear not to be a selective force in information processing. In
their 1948 textbook, Krech and Crutchfield advanced the proposition that attitudes
guide perceptual and cognitive processes. The two most often stated principles
regarding attitude-guided information processing are that persons selectively ( 1)
seek information that agrees with their attitudes while avoiding disagreeing in-
formation (e.g., Festinger, 1957) and (2) remember attitude-agreeable (as opposed
to disagreeable) information (e.g., Levine & Murphy, 1943). However, the em-
pirical evidence for these two principles has been consistently weak (see Freedman
& Sears, 1965; Frey, 1986; Pratkanis, 1984; Waly & Cook, 1966).
3. Attitudes may be epiphenomenal. Few would disagree with the observation
that attitudes are pervasive, as evidenced by ( 1 ) the ease with which people
report evaluative reactions to a wide range of objects, (2) the difficulty of iden-
tifying categories of objects within which evaluative judgments are not made,
and (3) the prominence of the evaluative dimension in judgments of meaning
(Osgood, Suci, & Tannenbaum, 1957). This pervasiveness of attitudes, however,
does not guarantee that attitudes are important in explaining social behavior. As
Bem ( 1967) suggested, attitudes might be epiphenomenal, cognitive illusions
that are constructed after the fact of behavior.

B. WHY THE PARADOX OF

ATTITUDE IMPORTANCE?

It is our position that social psychology’s fondness for the attitude construct
and the tenuous empirical status of that construct are related. The popularity of
attitudes has promoted an indiscriminate use of the term. As Greenwald (1989)
argues, almost any conceivable entity, from sensory qualities (such as odors and
tastes) through concrete objects (such as persons and things) and abstract thoughts
(such as social roles and political ideologies) to behavior (intentions to act), and
even attitudes (e.g. attitudes towards prejudice), can serve as an attitude objec;.
Furthermore, in survey contexts, researchers have used the term attitude even
more broadly to refer to almost any social or motivational disposition, including
personality constructs (such as self-efficacy, time orientation, and authoritari-
anism) and motives (such as need for power, affiliation, and achievement; see
Robinson & Shaver’s, 1973, collection of attitude scales).
As Blumer (1955; see also Bain, 1928; Symonds, 1927) argued, the attitude
concept fails to meet the criteria of a scientific concept. Its popularity has produced
a watered-down concept incapable of effective prediction. The root of this problem
began with early efforts at defining “attitude,” and it culminates today in the
 SOCIOCOCNITIVE MODEL OF ATTITUDES 247

two most popular ways of conceptualizing attitude structure and function-the

tripartite model and the principle of cognitive consistency.

I . Early Hisrory of the Definition of Attitude

The 1930s saw two prominent proposals for the definition of attitude: (1) the
oft-cited definition by Allport (1935) guided conceptual development of the at-
titude construct; (2) the other, by Thurstone (1931) guided the development of
attitude measures (cf., Edwards, 1957; see Ostrom, 1989, for a review of this
history).
For the 1935 Handbook of Social Psychology, Allport undertook to survey
the (then relatively brief) history of attitude research. He found, much as have
more recent reviewers (e.g., Fleming, 1967), that the early usage of the term
attitude was extensive and diverse, extending to such conceptions as adaptedness
(by evolutionary biologists such as Spencer), muscular preparedness (by Lange
and Baldwin), cognitive set (by the Wurzburg school), dynamic motivation (by
Freudians), and mental processes that determine response (by sociologists such
as Thomas and Znaniecki). Allport offered an integrative definition of attitude
as “a mental and neural state of readiness, organized through experience, exerting
a directive or dynamic influence upon the individual’s response to all objects
and situations with which it is related” (p. 810). Allport’s characterization of
attitude was more than a definition; it was an implicit theory of attitude structure
and function; it assumed that attitudes guide behavior by orienting and energizing
the organism to act.
In contrast, Thurstone (193 I ) defined attitude as “the affect for or against a
psychological object” (p. 261). This definition was minimally theoretical, and
left the conceptual formulation of attitude (i.e., attitude structure and function)
as a task for subsequent investigation. For similar reasons, we start from a position
that is much like Thurstone’s, defining attitude as a person’s evaluation of un
object of thought.

2 . The Tripartite Model of Attitude Structure

For the most part, attitude researchers have been reluctant to suggest that one
model of attitude is superior to others. Accordingly, many researchers have sup-
ported definitions that (1) permit a broad array of research operations for attitude
measurement and (2) put no apparent boundaries on the sort of entity that can
be regarded as an attitude (e.g., Allport, 1935; DeFleur & Westie, 1963; Green-
wald, 1968b). The model of attitude structure that has been most often advanced
by social psychologists (perhaps due to its breadth) is in terms of a tripartite
entity; in it, an attitude is conceived as having three components-affective,
cognitive, and conative (or behavioral). As Rosenberg and Hovland (1960) de-
scribe, “We here indicate that attitudes are predispositions to respond to some
248 ANTHONY R . PRATKANIS AND ANTHONY G. GREENWALD

class of stimuli with certain classes of responses and designate the three major
types of responses as cognitive, affective, and behavioral” (p. 3).
Although such a broad model of attitude may be of value in the early stages
of a concept’s development (see Smith, Bruner, & White, 1956, p. 341, the
tripartite now deters research progress for at least two reasons. First, operations
and measures have not been linked with theoretical constructs, resulting in little
consensus concerning the meaning of the terms of the tripartite model. For ex-
ample, affect has been variously described as an evaluation, subjective feelings,
and physiological correlates. The conative component has been variously assessed
using observed action, verbal self-report of past action, and self-report of inten-
tions. Any given research investigating the conative component can purportedly
be investigating the attitude-behavior relationship, the relationship of the conative
to other attitude components, or the relationship between behavior and the conative
component of attitude.
A second concern with the tripartite model is that its breadth severely limits
the model’s predictive and explanatory power. The triparite model is not strong
enough to specify the conditions under which expected attitude effects are most
likely to occur. As a consequence, most of the research on the tripartite model
has addressed the question of validation (cf. Breckler, 1984; Kothandapani, 1971;
Ostrom, 1969) as opposed to investigating the implications of the model for
predicting attitude effects (see the line of research beginning with Norman, 1975,
for an exception). (For further critical discussion of the tripartite model, see
Ajzen, 1989; Breckler & Wiggins, 1989; Cacioppo, Petty, & Geen, 1989; and
Zanna & Rempel, 1988).

3 . Consistency Theories of Attitude Functions

During the era of consistency theories in the 1950s and 1960s (culminating
in Abelson et a l . , 1968), social psychologists believed that an attitude should
induce a correspondence of response. Positive attitudes result in positive feelings,
thoughts, and behaviors toward an object, whereas negative attitudes engender
the opposite, negative response. Of course, the expectation that all attitude effects
are consistency effects may cause one to overlook other (possibly important)
attitudinal effects and processes.
Consistency theories also posit that individuals are motivated to maintain bal-
ance and consistency among cognitive elements and interpersonal relationships.
This “drive for consistency” postulate finds a parallel in Allport’s definition of
attitude-when an attitude is invoked, an individual is driven to act consistently
with the implications of the attitude. As research on consistency processes pro-
gressed, the original “motive for cognitive consistency” was gradually replaced
by one implicating the role of self-esteem maintainence and self-justification as
a guide for cognitive processes (see Aronson, 1969, 1984; Greenwald & Ronis,
1978). However, the attitude construct has yet to be updated to reflect these
developments.
 SOCIOCOGNITlVE MODEL OF ATTITUDES 249

C. OVERVIEW OF THE SOCIOCOGNITIVE MODEL

OF ATTITUDE

This chapter seeks to resolve the paradox of attitude importance (1) by reviewing
research to determine whether or when the pessimism concerning the predictive
power of attitudes is warranted and (2) by using the results of this review to
develop and support a sociocognitive model of attitudes. This model proposes
( I ) that attitudes have a cognitive representation (as opposed to a tripartite struc-
ture) and (2) that attitudes serve to relate a person to the social world (i.e., serve
a social function [Cooley, 1912; Katz, 1960; Smith, Bruner, & White, 1956;
Tetlock, 1988) rather than a cognitive consistency function).

I . The Cognitive Representution .fAttitude Structure

According to the sociocognitive model of attitude, an attitude is represented
in memory by (1) an object label and rules for applying that label, (2) an evaluative
summary of that object, and (3) a knowledge structure supporting that evaluation.
This representation operates in accordance with principles of social cognition
(cf. Wyer & Srull, 1984). The sociocognitive model of attitude is similar in
form to those proposed to account for findings obtained in the impression for-
mation literature (see in particular, Fiske & Pavelchak, 1986, and also Anderson
& Hubert, 1963; Carlston, 1980; Dreben, Fiske, & Hastie, 1979).

2 . The Functions of an Attitude

The sociocognitive model offers a two-part answer (see Herek, 1986 for a
similar proposal) to the functionalist’s (Smith, Bruner, & White, 1956) question,
“Of what value is an attitude to a person?” First, an attitude is used to make
sense of the world and to help the organism operate on its environment. In terms
of structure, the evaluative summary serves a heuristic function (a simple strategy
for appraising an object) and the knowledge structure supporting an evaluation
serves a sctiematic function (organizes and guides memory for events and complex
action toward an object). Second, an attitude is held in service of perhaps the
most important attitude object-the self. An attitude is used to define and maintain
self-worth. We attach different labels to this self-related function of attitude,
depending on the audience (public, private, or collective) that is observing the
attitude and its expression.

3 . Chupter Overview
Sections 11, 111, and IV describe the cognitive components of the sociocognitive
model and discuss its role in the prediction of attitude effects. Section V describes
various research findings indicating that attitudes serve self-related functions. In
anticipation of our conclusion, we find that the cause for concern about the attitude
250 ANTHONY R. PRATKANIS AND ANTHONY G . GREENWALD

concept (described previously) is not warranted. Attitudes are successful predictors

of a wide range of cognitive processes; they influence processing of episodic
information in a predictable fashion; and they are related to important social
behaviors under specifiable conditions.

11. Identification of the Attitude Object:

Engaging Attitude Functions

Attitude researchers and theorists implicitly assume that an attitude object is

represented in semantic memory. Earlier theorists (see Cantril, 1932; Newcomb,
Turner, & Converse, 1965; Scott, 1969; Scott, Osgood, & Peterson, 1979) at-
tempted to describe the properties of an attitude object by using such terms as
abstractness and inclusiveness. More recently, Greenwald ( 1 989) has used his
levels-of-representation framework (see Greenwald, 1988) to describe the rep-
resentation of attitude objects at one of four cognitive levels: (1) physical objects,
(2) categories, (3) propositions, and (4)schemata. Few research endeavors, how-
ever, have investigated these distinctions.
According to the sociocognitive model, an attitude object is represented in
memory by a label (e.g., “militant feminist,” “Ariel the dog,” “the pesticide
controversy”) and a set of rules and operations for applying that label (such as
categorization processes described by Smith & Medin, 1981). Object identification
is the first step to accessing the evaluative and knowledge components of an
attitude and thereby engaging the heuristic and schematic attitude functions.

A. THE DIFFICULTY OF (CORRECTLY)

IDENTIFYING ATTITUDE OBJECTS

An implicit assumption of much previous work on attitudes is that persons

consistently and consensually apply the same rules and operations for labeling
an attitude object. On the contrary, the labeling of an attitude object can be
highly variable across persons, places, and times. For any given situation, ( I )
objects may be difficult to identify in compact verbal labels, (2) multiple attitude
objects can be potentially salient, and (3) different object labels can be applied
simultaneously by the researcher and by the research subject. When these problems
occur in a research situation, the attitude measured by the researcher cannot be
expected to be more than weakly predictive of social behavior (see also, Kelman,
1974; Salancik, 1982).
For example, consider the classic attitude-behavior study published in 1934
by LaPiere (see also Dillehay’s 1973 critique). In this study, a young Chinese
couple travelled with LaPiere, seeking accommodation at many hotels and res-
taurants. LaPiere sought to relate the hotel and restaurant proprietors’ attitudes
 SOCIOCOGNITIVE MODEL OF ATTITUDES 25 I

toward Chinese (assessed with a mailed questionnaire) to their behavior of pro-

viding accommodations or service to the Chinese couple. LaPiere found little
relationship between his measure of attitude and the proprietors’ behavior toward
the Chinese couple.
LaPiere assumed that the salient attitude object was “members of the Chinese
race.” However, the couple (who were described as personable and charming)
could have been identified as customers, as middle-class people, as friendly ex-
amples of the Chinese race, as a young married couple, etc. LaPiere further
assumed that the proprietor’s behavior was dependent primarily on attitudes toward
Chinese. The proprietor’s behavior may have been intluenced as much (or more)
by attitudes toward other objects, such as the woman’s perfume, income generated
from the business, the reputation of the restaurant, or other patrons. It is also
possible that proprietors did not possess a representation of an object corresponding
to the label, “Chinese people.”
LaPiere’s research is not an isolated example of the observation that object
labels are labile. Object labels can be changed and modified as a function of
disconfirming evidence (Weber & Crocker, 1983), dissonance reduction (Abelson,
1959), decision structuring (Farquhar & Pratkanis, 1987; Plott & Levine, 1978;
Schattschneider, 1960). and social influence (Billig, 1987; Ries & Trout, 1981).
Different object labels can be applied to the same stimulus as a result of simple
priming manipulations (Berkowitz & Rogers, 1986; Higgins & King, 198 I;
Iyengar & Kinder, 1987).
Although variability in object identification hinders behavior prediction, it may
have functional value for the person. Greenwald (1980), in his description of
the rotulituriun ego, notes that rewriting history (of which modifying an object
label is one example) can serve to maintain an organization of knowledge by
eliminating the need for drastic restructuring in the face of disconfirming infor-
mation. The maintenance of an organization of knowledge is useful for preserving
access to stored information and for achieving a sense of coherence of perception.
The ease with which object labels are changed may also serve to protect and
maintain positive self-regard. Objects can be redefined for the self‘s benefit. For
example, an attack on a cherished opinion can be dodged by changing the topic
(e.g., “It’s not that I hate all Lithuanians, just the pushy ones.”) and a desired
reference group can be embraced by adopting their conceptual system.

B. OBJECT IDENTIFICATION: IMPROVING THE

PREDICTIVE UTILITY OF ATTITUDES

Much recent research on attitudes has been an attempt to specify the conditions
under which attitudes predict behavior. The success of these programs has been
achieved, in part, by proposing techniques for reducing the multiple and variable
meanings that can be associated with an attitude object in a given setting. Four
general classes of techniques for increasing the likelihood of engaging one
252 ANTHONY R. PRATKANIS AND ANTHONY G . GREENWALD

particular object label (over other potentially competing labels) can be identified,
all of which improve predictive power.

I . Increasing the Level of Object Specificity

Early attitude measurement theorists such as Thurstone and Likert stressed the
importance of precisely delimiting the attitude object for successful scale con-
struction (see Mueller, 1986). To assist in implementing this admonition, Payne
(195 1) proposed a number of suggestions (such as avoiding double-barrelled
questions and eliminating items with hidden meaning) for increasing the clarity
of an attitude object for survey respondents. Similarly, Ajzen and Fishbein (1977)
report evidence that attitude-behavior correlations are stronger when there is
correspondence in the level of specificity between attitude and behavior in terms
of the action to be performed, the target at which the action is directed, the
context in which it is to be performed, and the time at which it is to be performed.

2 . Increasing the Salience of an Attitude Object

Numerous research programs have developed manipulations for increasing the
salience and relevance of a given attitude object (in contrast to the salience of
competing objects). For example, Borgida and Campbell (1982) found higher
attitude-behavior consistency for a global attitude object (the environment) when
it was explicitly linked to a behavior (increasing the amount of campus parking).
Snyder (1982) reviews evidence showing that there is a stronger attitude-behavior
relationship ( 1 ) with low self-monitors than with high self-monitors who may
be influenced by competing concerns such as impression management, (2) when
the situation demands that the subject think about the attitude object, thus in-
creasing the salience of a given object, and (3) when the representation of the
attitude object is explicitly made relevant for the behavior. Granberg (1985) has
demonstrated the role of object salience for predicting social judgment. In his
study, subjects’ identification of Ted Kennedy’s position on abortion varied as
a function of their own attitude toward abortion and toward Kennedy, plus their
perception of the Democratic position on abortion, and their perception of Ken-
nedy as a liberal and as a member of the Catholic Church. Granberg found that
he could manipulate the strength of any of these factors by increasing their salience
prior to the judgment task.

3 . Specifiing the Nature of the Object Representation

Two studies by Lord, Lepper, and Mackie (1984) demonstrate the value of
specifying the object representation for increasing attitude-behavior correlations.
In their two studies, subjects first provided their attitudes toward specific Uni-
versity social clubs and toward homosexuals. In a later session, subjects were
given an opportunity to interact with prototypical (i.e., the best examplar of the
 SOCIOCOGNITIVE MODEL OF ATTITUDES 253

category) and nonprototypical members of these two social groups. The results
showed that attitudes were more predictive of behavior toward prototypical
members.

4 . Controlling for the Effects of Competing Objects

Davidson and Morrison ( 1983) have developed a within-subjects procedure
for attitude assessment, which takes into consideration some of the more salient
attitude objects that may be in competition with the object of investigative interest.
In a study demonstrating the technique, married couples completed an attitude
survey concerned with four contraceptive methods. One year later, subjects re-
ported which birth control methods they had used during the year. Davidson and
Morrison found that attitudes were moderately related to behavior (explaining
55.5% of the variance) when analyzed in the traditional manner of relating at-
titudes to behavior across subjects. However, considerably more variance in be-
havior (77.5%) was explained when the analysis compared a single person’s
attitude toward all objects and used the most positive attitude as a predictor of
behavior.

C. ATTITUDES AS PREDICTORS

The preceding four sets of research findings have identified a condition that
improves the predictive power of an attitude. An attitude is a better predictor
when an object label is held consistently across time and situations and when
that label is shared by both the researcher and the subject. Consistent object
identification allows consistent access to the evaluative and knowledge com-
ponents of an attitude.
Nevertheless, the difficulties in producing strong attitude-behavior correlations,
and the necessity for clearly specifying an attitude object, may prompt some to
believe that attitudes are only weakly connected to behavior and other dependent
variables. To the contrary, however, the need for well-controlled research settings
to demonstrate strong attitude-behavior relations may mean only that the influence
of attitudes on behavior is so pervasive that it is difficult to isolate the effects
of a single attitude. To paraphrase Shakespeare, perhaps attitude should be con-
sidered a construct that predicted “not wisely, but too well.”

111. Attitude as Heuristic

Historically, as Mueller ( 1986) notes, there has been substantial agreement

that evaluation (i.e., affect for or against) is the most essential component of
254 ANTHONY R . PRATKANIS AND ANTHONY G. GREENWALD

the attitude concept, distinguishing it from other mental constructs (see McGuire,
1989 as an exception). In the sociocognitive model of attitude, the evaluative
component is represented in memory by a simple evaluative summary of an
object. This summary, typically expressed by words such as “like/dislike,”
“agree/disagree,” and “good/bad,” plays an important role in the appraisal of
social objects.

A. ATTITUDE HEURISTIC DEFINED

Attitude theorists have long noted that attitudes serve a cognitive or knowledge
function. For example, Lippmann (1922) viewed public opinion as an economical
simplifier of a complex world. Smith, Bruner, and White (1957) posited that
opinions serve an object appraisal function: “An attitude provides a ready aid
in ’sizing up’ objects and events in the environment” (p. 41). Katz (1960) has
suggested that attitudes satisfy a knowledge function, providing adequate structure
to the social world. McGuire (1969) noted that attitudes serve as “a simplified
and practical manual of appropriate behavior toward specific objects” (p. 158).
Katona ( 1975) has specified a principle of affect generalization-good things
produce good consequences and bad things produce bad consequences. Pettigrew
( 1979) concluded that past research on prejudice indicates that people often com-
mit an ultimate attribution error-that is, the attribution of good and bad qualities
to liked and disliked groups of people (respectively).
Pratkanis (1989) used the term attitude heuristic to refer to the use of an attitude
in knowing. A heuristic is a simple, but often only approximate, strategy for
solving a problem. Some examples include, “If the experimental results are null,
then the design is inadequate.” and “If the person is from the South, he or she
must be intellectually backward.” Heuristics differ from detailed sets of pro-
cedures or algorithms, which invoke complex sets of rules for problem-solving
(see Sherman & Corty , 1984; Tversky & Kahneman, 1974).
An attitudinal heuristic uses the stored evaluation of an object as a cue to
solving the problem of how to act in relation to the object. Attitudes are used
to assign objects to a favorable class (for which strategies such as favoring,
approaching, praising, cherishing, and protecting are appropriate) or to an un-
favorable class (for which strategies such as disfavoring, avoiding, blaming,
neglecting, and harming are used). An example is: “I dislike Reagan, therefore
the federal deficit is the result of his ‘charge card’ economic policies.” Such
attitude heuristics operate according to the principles of balance theory (Heider,
1958; Zajonc & Burnstein, 1965), although it is not necessary to assume a mo-
tivation to maintain consistency (see Wyer, 1974).’
‘Wyer (1974) notes that balance principles describe social inference processes when they summarize
a subject’s prior experience with the type of objects about which inferences are made. Thus, an
attitude heuristic should be viewed as a summary of past experience as opposed to a motivation to
maintain balance among cognitive elements (as traditionally assumed by balance theory).
 SOCIOCOGNITIVE MODEL OF ATTITUDES 255

B. CONCEPTUAL MEMORY PROCESSES

According to Pratkanis (1989), conceptual processes are often affected by the

use of an attitude as a heuristic. Conceptual or semantic memory is knowledge
of the world, consisting of facts, ideas, and concepts. According to Tulving
(19831,

[Semantic memory] is a mental thesaurus, organized knowledge a person possesses about

words and other verbal symbols, their meaning and referents, about relationships among
them, and about rules, formulas, and algorithms for the manipulation of symbols, concepts,
and relations. (p. 21)

The contents of conceptual memory are relatively permanent, are accessed au-
tomatically, and are context free (i.e., not linked to a temporal event). Conceptual
memory tasks include the processes of comprehension, categorization, inference,
judgment, and reasoning. It can be contrasted with episodic memory which is
“a system that receives and stores information about temporally dated episodes
or events, and temporal-spatial relations among them.” (Tulving, 1983, p. 21).
Episodic memory is involved when an individual attempts to recall yesterday’s
happenings, a list of experimental nonsense syllables, or a persuasive message
presented in a laboratory setting.

C. ELEVEN USES OF ATTITUDE AS A HEURISTIC

IN CONCEPTUAL PROCESSING

Pratkanis (1 989) classified 1 1 reliable effects of attitudes on conceptual pro-

cesses. This list is reproduced here.

I . Interpretation and Explanation

Attitudes are used to interpret and explain social events. For example, Smith
(1947) found that pro- and anti-Soviet individuals rated the credibility of Soviet
news items in accord with their attitudes. Cooper and Jahoda (1947) and Kendall
and Wolf ( 1 949) demonstrated that prejudiced individuals misunderstood cartoons
presenting a bigoted person in an unfavorable manner. Hastorf and Cantril (1954)
found that interpretations of a Princeton-Dartmouth football game varied as a
function of support for the two opponents. Manis (1961) showed that attitudes
affected the interpretation of a message such that subjects mistakenly attributed
an own-attitude-consistent position to an admired person on the issue of college
fraternities. Regan, Straus, and Fazio (1974) found that subjects made internal
attributions when an actor behaves consistently with attitudinal expectations (i.e.,
liked actors perform positive behaviors and disliked actors perform negative be-
haviors), but the subjects made external attributions when an actor behaved in-
consistently with attitude-based expectations.
256 ANTHONY R . PRATKANIS AND ANTHONY G . GREENWALD

2 . Halo Effects: Expectations and Inferences

A favorable or unfavorable person impression biases expectations and inferences
about that person. For example, much research shows that people who are eval-
uated as attractive are expected to be “better” than those not so evaluated (Hat-
field & Sprecher, 1986) and that liked individuals are expected to possess pos-
itive traits, whereas disliked individuals are assumed to possess negative ones
(see Lott & Lott, 1972; Lott, Lott, Reed, & Crow 1970). Similarly, in sur-
veys of economic expectations, Katona (1975) finds that good events such as
the end of World War I1 and the end of the Cuban missile crisis are typically
associated with optimistic expectations concerning the economy, whereas
negative events such as the U-2 incident and the Berlin crisis induce economic
pessimism.

3 . Syllogistic Reasoning
Attitudes toward the conclusion of a syllogism can influence the ability to
determine whether the syllogism is logically valid (cf. Evans, Barston. & Pollard,
1983; Feather, 1964; Gordon, 1953; Janis & Frick, 1943; Lefford, 1946; Morgan,
1945; Morgan & Morton, 1943, 1944). For example, Thistlewaite (1950) asked
respondents to state whether syllogisms such as the following were valid:

G i v m : If production is important, then peaceful industrial relations are desirable. If pro-

duction is important, then it is a mistake to have Negroes for foreman and leaders over
Whites.
Th~r&w: If peaceful industrial relations are desirable, then it is a mistake to have Negroes
for foreman and leaders over Whites.

For this syllogism, prejudiced individuals (who agree with the conclusion) are
more likely to indicate (incorrectly) that the logic is valid compared to less prej-
udiced individuals.

4 . Responses to Persuasive Communications

Greenwald ( 1968a) has found that individuals with an unfavorable attitude
toward the topic of persuasive communication are more likely to counterargue
a message, whereas those with a favorable attitude are more likely to provide
consonant cognitive responses. This pattern of results has been frequently obtained
in studies of persuasion (see Petty, Ostrom, & Brock, 1981). Similarly, Waly
and Cook (1965) found that an argument was considered more effective and
plausible by those who agreed (as opposed to disagreed) with the position ex-
pressed in the argument (see Lord, Lepper, & Ross, 1979, for a more recent
example).
 SOCIOCOGNITIVE MODEL OF ATTITUDES 257

5 . Interpersonal Attraction
Byrne (1971) repeatedly finds that individuals with attitudes similiar to one’s
own are viewed as attractive. In a typical experiment, subjects receive information
about others’ attitudes. Attitude similarity is manipulated by varying the proportion
of shared attitudes between the subject and a stimulus person. The more shared
attitudes, the more attractive the subject rates the stimulus person. (See Rosen-
baum, 1986, for an alternative repulsion interpretation of these findings).

6 . Judgment of Social Stimuli

One’s own attitude provides a reference point for the judgment of social stimuli.
For example, Vroom (1960) found that individuals with a positive attitude toward
an organization viewed that organization’s goals as similar to their own. Granberg
and Jenks (1977) found that survey respondents perceived the position of their
preferred candidate in the 1972 Presidential election (McGovern or Nixon) on
nine sociopolitical issues to be highly similar to their own position on these
issues. Edwards (1941) and Vallone, Ross, Lepper (1985) obtained results in-
dicating that attitudes influence the judgment of bias in the source of a message.
Sherif and Hovland ( 1961) present additional evidence that attitudes bias social
judgments.

7 . False Consensus of Opinion

Wallen ( 1943) found that subjects believe their attitude position to be more
popular than it actually is. In this first “false-consensus” study, students were
asked to give both their opinions on, and their estimate of, the percentage of
students who supported the Selective Service, the St. Lawrence Seaway Project,
and war with Germany. The results showed that the respondents estimated the
attitudes of others so that their own opinion coincided with that of the estimated
majority. The false-consensus effect has been replicated often (cf. Fields &
Schuman, 1976; Granberg, Jefferson, Brent, & King, 1981; Ross, Greene, &
House, 1977; Sherman, Chassin, Presson, & Agostinelli, 1984). Goethals (1986)
presents evidence indicating that the false-consensus effect, like some of the
other effects listed in this section, may also serve the self-functions of attitude
described in Section V.

8 . Fact Identification
The use of an attitude can lead to the selective reconstruction of past events.
For example, 2% years after the event, Eberhart and Bauer (1941) assessed
memory for a riot involving the Chicago police and a crowd of striking employees
258 ANTHONY R . PRATKANIS AND ANTHONY G. GREENWALD

of the Republic Steel Company. Eberhart and Bauer found that subjects with a
prolabor attitude were more likely to remember that the crowd was unarmed and
that the police brutally shot peaceful citizens, whereas antilabor subjects recalled
the opposite. Similarly, Pratkanis ( I988a) found that subjects misidentified facts
consistent with their attitudes. Given pairs of statements such as ( I ) Ronald Reagan
maintained an “A” average at Eureka college and (2) Ronald Reagan never
achieved above a “C” average at Eureka college (a fact), subjects were most
likely to identify as true those statements that agreed with their attitudes (see
also Smith, 1968).

9 . Estimates of Personal Behavior

Past personal behavior is often revised to be consistent with current attitudes.
For example, Bem and McConnell (1970) changed attitudes on the student control
of their university, using a counterattitudinal essay procedure. The results showed
that subjects erred by overestimating the extent to which their premanipulation
attitudes (as assessed earlier by Bern and McConnell) were similar to their current
attitudes. Although there is disagreement over the psychological mechanisms
involved, this finding has been often replicated (cf. Aderman & Brehm, 1976;
Goethals & Reckrnan, 1973; Ross & Shulman, 1973; and Shaffer, 1975a,b).
Markus (1985) has found a similar effect in a national survey of teenagers and
parents, conducted in 1965, with follow-up collections in 1973 and 1982. He
found that current attitudes on policy issues, such as legalization of marijuana,
women’s roles, and rights of the accused biased the recall of past attitudes on
these issues.
Ross, McFarland, and Fletcher (1981) provide another example of the use of
attitudes in the selective estimation of the frequency of past personal behavior.
In their studies, subjects received persuasive messages that either derogated or
promoted daily toothbrushing and frequent bathing. Those who heard the anti-
toothbrushing and antibathing messages estimated that they toothbrushed and
bathed less often than those who heard the promessages. Ross, McFarland, Con-
way, and Zanna (1983) have replicated this effect, but they find that it is less
likely to occur when the domain of recall is objective, as opposed to ambiguous
(i.e., recall of the frequency of exercise versus how vigorous exercise was per-
ceived to be).

IO. Informution Error Technique

The fact that attitudes produce errors in judgments led Hammond (1948) to
suggest an information error technique as an indirect measure of attitudes (see
Krernan, 1949; Kubany, 1953; Parrish, 1948; Weschler 1950a,b). In this tech-
nique, respondents are asked, under the guise of an information survey, to select
which of two statements is true. In reality, both responses are incorrect-for
 SOCIOCOGNITIVE MODEL OF ATTITUDES 259

example (taken from Weschler, 1950a): “During the strike wave of April 1948,
the percentage of estimated working time lost was (1) 1.1% or (2) 2.2%?” (The
correct answer is 1.6%). Respondents frequently chose the error most consistent
with their attitudes. Given that lost working time is viewed as negative, individuals
with a probusiness attitude indicate that the strike produced considerable down-
time, whereas the prolabor supporter attributes less lost time to the strike.

I I , Prediction of Future Events

Hadiey Cantril (1940) once observed, “What people want to happen, they
tend to think will happen” (p. 406). The Gallup (1972) poll has asked the fol-
lowing question of nationally representative samples before the Presidential elec-
tions of 1944, 1948, 1952, 1960, and 1968: “If you were to guess at this time,
who do you think will win the next presidency, a Republican or Democrat?”
The results, averaged across the years, indicate that 63.3% of the Democrats
believe that a Democrat will win the next election (versus 19.5% predicting a
Republican victory, with the rest undecided), whereas 57.0% of the Republicans
believe a Republican will win (versus 26.3% predicting a Democratic win, with
the rest undecided). This partisan attitude effect held for all elections except the
Truman-Dewey Presidential election of 1948 (see also Granberg & Brent, 1983).

D. IMPLICATIONS OF THE ATTITUDE HEURISTIC

FOR BEHAVIOR

Fazio (1986) has proposed a model of the attitude-behavior relationship that

implicates the heuristic function of attitude. According to Fazio, attitudes are
used to selectively perceive and interpret an object. This selective perception of
the object then comprises part of the individual’s definition of the situation (in
conjunction with norms and other perceptions). This definition of the situation
subsequently determines the direction and nature of behavior toward the attitude
object. Fazio’s (1986) model suggests three factors that influence the magnitude
of the attitude-behavior relationship: ( 1 ) probability of retrieving the attitude
object from memory (see previous section on engaging attitude functions), (2)
strength of the attitude-object-evaluation relationship, and (3) opportunity and
ability to execute the attitudinal strategy (i.e., act in accordance with one’s at-
titude-based perceptions).

I . Attitude Strength Increases Attitude-Behavior

Consistenq
Attitude researchers have suggested many terms for the strength of an attitude,
including certainty, confidence, conviction, crystallization, extremity, intensity,
260 ANTHONY R. PRATKANIS AND ANTHONY G . GREENWALD

magnitude, salience, and stability (see Abelson, 1988; Raden, 1985). There is
a growing literature indicating that the stronger an attitude (as measured in a
variety of ways), the more predictive it is of behavior (Fazio & Zanna, 1978a,b;
Petersen & Dutton, 1975; Sample & Warland, 1973) and of conceptual cognitive
processes (Fazio, 1989; Pratkanis, 1988). Fazio (1989) has conceptualized attitude
strength as a continuum ranging from nonattitudes to highly accessible attitudes.
A nonattitude (Converse, 1970) is an attitude (evaluation) that is not stored in
memory, but is created (computed) when a person is asked to evaluate an object.
Highly accessible attitudes are those for which the evaluation is activated au-
tomatically and quickly (as evidenced by short evaluative judgment latencies)
by an encounter with the attitude object. Fazio (1986; 1989) has identified sit-
uational manipulations that increase the accessibility of weakly held attitudes
and reports that highly accessible attitudes are more predictive of behavior.

2 , Situational and Personal Constraints Decrease

Attitude-Behavior Consistency
Situational influences may overpower even the most highly accessible attitude
(see Schuman & Johnson, 1976). For example, organizational researchers have
found that job satisfaction is only mildly related to job performance and moderately
related to absenteeism and turnover (cf. Iaffaldano & Muchinsky, 1985), although
it is reasonable to assume that job attitudes should be highly accessible. In ex-
plaining these low correlations, researchers have identified numerous situational
factors, such as family responsibilities, other employment opportunities, personal
health, technological constraints, the task structure, and so on that restrict an
employee’s opportunity for action consistent with an attitude (see Baron, 1986;
Cotton & Tuttle, 1986; Schminke, 1986).
Similarly, an individual may not possess (or believe he or she possesses) the
abilities and skills for acting consistently with an attitude (see Abelson’s I982
discussion of attitudes, scripts, and behavior). For example, complex knowledge
structures are needed in order to interact with some attitude objects (e.g., the
oenophile’s knowledge of Napa Valley and Santa Cruz Mountains microclimates
or the gambler’s understanding of casino games; see next section). In some cases,
knowledge needed to act on an attitude may not just be missing, but may actually
contradict the attitude. Devine (1988; see also Goldberg, Gottesdiener, &
Abramson, 1975 for a similar example) has identified the interesting case of the
white liberal for whom an attitude (e.g., support for racial equality) conflicts
with knowledge (i.e., derogatory racial stereotypes taught by society), resulting
in a complex relationship between attitudes, knowledge, and social perceptions.
Recently, Ajzen (1988; see also Rogers, 1983) has pointed out that it may not
be enough to know how to act toward an attitude object; one must also believe
that one can perform the behavior (i.e., self-efficacy), if given the opportunity
to act on an attitude.
 SOClOCOGNlTIVE MODEL OF ATTITUDES 26 1

IV. The Schematic Function of Attitudes

Attitude theorists have suggested a number of formats for describing knowledge

structures supporting an attitude. For example, expectancy-value theorists (Fish-
bein & Ajzen, 1975; Rosenberg, 1956) describe attitudinal knowledge in terms
of a multiattribute belief structure typically expressed by the linear-additive
equation A, = C:=, b,e,, where A, is the attitude toward some object, 0; b, is
the subjective probability that 0 is related to attribute i; e, is the evaluation of
attribute i; and n is the number of beliefs. Cognitive-oriented theorists (e.g.,
Katz & Stotland, 1959; Newcomb, Turner, & Converse, 1965; Scott, 1969; Scott,
Osgood, & Peterson, 1979) have prepared lists of cognitive dimensions and
properties (e.g., differentiation, integration, coherence, ambivalence) on which
attitudinal knowledge structures are assumed to differ. Although such specifi-
cations are a logical first step for distinguishing attitudes, there has been far too
little successful empirical and theoretical work to distinguish the meaning of one
property from another and to indicate which properties are useful for predicting
attitude effects. For example, in a review of studies investigating differentiated
and integrated attitudinal knowledge, Ajzen ( 1989) finds that including such
properties does little to improve behavior prediction.
According to the sociocognitive model of attitude, attitudinal knowledge can
be quite varied, consisting of any (or all) of the following information: arguments
for and against a given proposition, esoteric and technical knowledge about the
domain, subjective beliefs, information on how to behave toward the object,
goals and wishes about the object, the social meaning of adopting a certain attitude
position, and personal episodes. The specific contents and organization of this
knowledge is an important determinant of attitude and memory effects.

A. THE SCHEMATIC FUNCTION DEFINED

One early proposition for relating attitudinal knowledge to attitude effects,

suggested both by Watson and Hartmann (1939) and by Levine and Murphy
(1943), was that attitudes serve as a frame of knowledge for encoding and re-
constructing a persuasive message. In support of this proposition, early studies
obtained what can be termed an “attitude and selective learning effect”-superior
recall for information that agrees with one’s attitude. For example, Levine
and Murphy (1943) found that a procommunist message was learned better
by communists, whereas an anticommunist message was learned better by
capitalists (see Watson & Hartmann, 1939, for a similar finding using religious
attitudes).
Pratkanis ( 1989) referred to the use of an attitude to organize and guide memory
for events as the attitude’s schematic function. A schema is an organization of
262 ANTHONY R . PRATKANIS AND ANTHONY G. GREENWALD

a subset of knowledge relevant to a limited domain. It consists of both content

(an organization of information) and procedure (the use of this information in
knowing). In contrast to a heuristic, which is one simple rule, a schema is an
organization of a set of procedures and data within a domain.
A frequent finding in learning and memory research is that persons with well-
developed schemas demonstrate superior episodic memory for terms and infor-
mation related to the schematized domain. (As just one example, see the research
of Voss and his colleagues on expertise and memory [Chiesi, Spilich, & Voss,
1979; Spilich, Vesonder, Chiesi, & Voss, 1979; Voss, Vesonder, & Spilich,
19801). The schema is presumably used both during encoding (to aid in com-
prehension, interpretation, and elaboration of to-be-learned materials) and during
retrieval (to provide internal cues that serve as covert mnemonics for recall and
reconstruction of an event [Hastie, 1981; Taylor & Crocker, 19811).

B. ATTITUDES AND EPISODIC MEMORY

Over four dozen studies spread over nearly 50 years of research have yielded
little reliable support for the attitude and selective learning effect obtained by
Watson and Hartmann (1939) and by Levine and Murphy (1943). In contrast to
the effects of attitudes on conceptual processes, the influence of attitudes on
episodic memory (the recall of temporally located and dated materials) has been
inconsistent (see Pratkanis, 1984; Roberts, 1985 for reviews). Some studies have
replicated the Watson-and-Hartmann-Levine-and-Murphyfinding that agreeable
information is learned best (cf. Edwards, 1941; Jones & Kohler, 1958; Weldon
& Malpass, 1981); others have demonstrated no attitude and learning relationship
(cf. Brigham & Cook, 1969; Greenwald & Sakumara, 1967; Smith & Jamieson,
1972; Waly & Cook, 1966); still others have found better recall of attitudinally
extreme compared to moderate information (cf. Doob, 1953; Judd & Kulik, 1980;
Postman & Murphy, 1943); and at least one study has yielded a negative attitude
and learning correlation (Cacioppo & Petty, 1979). Similarly, Greenwald (1968a)
reported that persuasion researchers also typically fail to find consistent corre-
lations between message agreeableness and recall of message content. Although
the pattern of attitude and episodic learning results is complex, there is growing
evidence that the knowledge structures supporting an attitude plays an important
role in attitudinal learning (see Pratkanis, 1989).*

'Attitudes can also influence learning via mechanisms other than the knowledge structure supporting
an evaluation. For example, Pratkanis (1988) discusses how the use of an attitude as a heuristic can
influence memory when recall is primarily an inference or guess about the past (see as examples,
Edwards, 1941; Lydon, Zanna, & Ross. 1988; Read & Rosson, 1982). Attitudes can also motivate
the biased processing of a persuasive communication, resulting in a selective recall of the com-
munication (see Cacioppo & Petty, 1979).
 SOClOCOGNITIVE MODEL OF ATTITUDES 263

C. BIPOLAR ATTITUDE STRUCTURE

Around 2500 years ago, the Sophist Protagoras observed, “There are two
sides to every question.” Any opinion or argument can be opposed by a coun-
terargument (see Billig, 1987, for an exposition of the Protagoras maxim). Much
more recently, Judd and Kulik (1980; see also Doob, 1953; Postman & Murphy,
1943) found that people, despite their own personal beliefs, can learn and recall
information on both sides of an issue, thereby providing empirical evidence for
Protagoras’s bipolar model of attitude.
In Judd and Kulik’s (1980) study, college students saw 54 Thurstone-scaled
attitude statements on the topics of women’s rights, capital punishment, and
majority rule in South Africa. Subjects stated their agreement with each statement
and then rated each one for the degree it reflected an extremely pro or extremely
anti position. A day later, subjects attempted to recall the statements. Items rated
at the extremes of either the agree-disagree or the pro-anti continuum were ( 1 )
most rapidly rated and (2) most easily recalled. These results suggested that

Attitudes thus may act as bipolar schemas that contain representations or expectations of
very agreeable and disagreeable points of view. Information that closely matches these
expectations is more easily judged and recalled than is information that, although relevant.
does not match as well. (Judd & Kulik, 1980, p. 570)

In other words, a knowledge structure supporting an attitude (evaluation) contains

not only arguments, beliefs, and expectations supporting one’s own position,
but may also contain opposing information (and perhaps counterarguments refuting
this opposing material).3Persons with such a bipolar structure should demonstrate
superior learning and faster encoding for information that fits the knowledge
frame-that is, information that is consistent with either a pro or an anti position.
rather than just agreeable information.
Further support for bipolar attitude schemas comes from a variety of sources.
Hymes (1986) reasoned that subjects who are neutral on abortion should be
aschematic (i.e., lacking well-defined knowledge on this issue) whereas pro-
and anti-abortion advocates should possess a bipolar schema for the issue. Con-
sistent with this supposition, Hymes found that pro- and anti-abortion advocates
were better in categorizing information as either pro- or anti-abortion than were
neutral subjects.
Pratkanis (1984) also provided data in support of the bipolar hypothesis for
some attitude domains. In one study, college students were asked to list arguments
for and against 1 of 10 controversial social issues (e.g., nuclear power, gun

‘A bipolar knowledge structure should not be confused with ambivalence of belief. Amhivrr/enc.e
refers to the endorsement of conflicting (opposing) beliefs about an issue. whereas the term h i p d m
refers to knowledge of supporting and opposing arguments (regardless of personal endorsement).
264 ANTHONY R. PRATKANIS AND ANTHONY G . GREENWALD

control, abortion, draft registration, death penalty). Although subjects’ opinions

on these issues tended to be distributed across the attitude continuum, the results
revealed that 82.7% of the subjects could generate at least one argument on both
sides of an issue. (However, subjects typically listed more arguments agreeing
with their own position, consistent with the findings of Feather, 1969.)
In two additional experiments, Pratkanis (1984) obtained recall, recognition,
and reaction-time data consistent with an assumed bipolar schematic structure.
In those studies, subjects first stated their agreement with pro, neutral, and anti
statements on the issues of nuclear power, defense spending, and welfare. Results
for free recall and recognition of the statements revealed no attitude-selective
learning effect-the correlation between recall and statement agreement was - .04
in Experiment I and .02 in Experiment 2. However, subjects tended to recall
better, evaluate faster, and recognize more quickly those items rated at either
extreme of the agreement scale.
McGuire’s (1964) work on resistance to persuasion can be viewed as showing
that bipolar attitude structures are better “inoculated” against attitude change
than unipolar ones. McGuire found that attitudes toward cultural truisms (issues
such as toothbrushing and X-ray exams) could be easily changed, presumably
because subjects had little in the way of an organized structure containing either
attitude-supporting or (especially) attitude-opposing knowledge. By providing
exposure to opposing arguments, McGuire created bipolar knowledge structures.
These inoculated (bipolar) structures effectively resisted the effects of otherwise
highly persuasive attacking messages (see also Wood, 1982).

D. UNIPOLAR ATTITUDE STRUCTURE

Pratkanis (1984) reasoned that some attitudes may be supported by unipolar

knowledge structures. A positive unipolar structure consists exclusively of
knowledge favorable to (supporting) its ~ b j e c tFor
. ~ example, a sports fan typically
possesses an elaborate knowledge structure containing technical and esoteric in-
formation. With unipolar attitudes, persons with positive attitudes have much
more knowledge than those with less favorable (or neutral) attitudes. Conse-
quently, with unipolar attitudes, an attitude and selective learning effect should
be expected. A person with a positive attitude possesses knowledge structures
that are useful for encoding and recalling domain-related information, whereas
those persons with a neutral attitude lack such structures and should process
domain-related information poorly.

‘Unipolar structures may also he associated with negative attitudes (although such structures have
not yet been the focus of research). In such cases, knowledge associated with the attitude is unfavorable
toward and derogatory of the object. Negative unipolar structures may arise under conditions of
strong societal censorship (such as attitudes toward birth control during the Middle Ages: see Ranum
& Ranum, 1972).
 SOCIOCOGNITIVE MODEL OF ATTITUDES 265

Three criteria distinguish unipolar (versus bipolar) attitude structures: ( 1 ) lack

of controversy (versus controversy); (2) population distribution of attitude ranging
from neutral to just one extreme (versus extreme-to-extreme distribution); and
(3) linear relationship between attitude and amount of attitude-relevant knowledge
(versus a tendency to find curvilinear relationships).
Three studies have compared the roles of unipolar versus bipolar knowledge
structures in the learning of attitude-related materials. Pratkanis ( 1984) asked
subjects to complete a relationship judgment task similar to the one used in
self-reference research (see Rogers, Kuiper, & Kirker, 1977). In this task,
subjects answered questions with the stem, “Is this word or phrase related to
?” Subjects answered such questions for 72 words, 12 of which
were related to each of six topics (welfare, defense spending, nuclear power,
sports, music, and religion). For example, words related to defense spending
included pentagon, freeze, and domino; words related to sports included
screen, touchdown, and forward. After a filler task, subjects attempted free
recall of the words and completed a survey to assess their attitudes and knowl-
edge concerning the six domains.
The results revealed that the topics of sports and music were unipolar in nature.
Few subjects endorsed negative statements concerning these activities, and the
correlation between attitude and knowledge measures was linear and positive.
In contrast, the topics of welfare, nuclear power, and defense spending showed
the bipolar pattern. Attitudes were distributed across the full range of the scale
and subjects with extreme attitudes professed to know more about the issue.
Knowledge structures supporting religion were complex, not clearly fitting either
the unipolar or bipolar pattern.’ The recall results showed that subjects with more
knowledge of bipolar and unipolar topics demonstrated superior free recall of
topic-related terms (an “attitude-reference’’ effect). This effect took the form
of an attitude-selective learning effect for unipolar topics-that is, a positive
attitude-recall correlation. However, for bipolar topics, subjects tended to recall
better information at both extremes of the attitude scale, thus yielding no linear
attitude-learning relationship.
Pratkanis ( 1987) replicated these unipolar-bipolar results using persuasive
communications, rather than word-and-phrase stimuli. In this study, subjects
completed an attitude-knowledge survey and then attempted to learn and recall
messages that were anti-nuclear-power, pro-defense-spending, promusic, and
prosports. Each message consisted of a series of arguments in support of the
message conclusion (e.g. “nuclear power plants have a high risk of accidental

’Religion may be a superordinate category spanning multiple categories corresponding to denom-

inations and sects (e.g., Catholic, Mennonite, Zen Buddhist). Unipolar attitudes may exist at the
denominational level. Similarly. sports and music can be viewed as superordinate categories. Unlike
religion, however, attitudes toward various types of sports and music are typically positively correlated
( e . g . , the sports fan enjoys baseball, football, and basketball. whereas the religious person rarely
identifies with more than one denomination).
266 ANTHONY R . PRATKANIS AND ANTHONY G. GREENWALD

meltdown;” “the simple Bach chorale moves the soul”). For sports and music,
few subjects endorsed the negative end of the attitude continuum and there were
positive correlations between attitudes and knowledge. For the issues of defense
spending and nuclear power, subjects were distributed along the attitude contin-
uum, and subjects with attitudes at the extremes of the continuum professed the
most knowledge about the domain. For the two unipolar messages (sports and
music), there was an attitude-selective learning effect-subjects who held positive
attitudes toward sports and music showed superior learning of the message. For
the bipolar messages (defense spending and nuclear power), subjects with extreme
attitudes on the topic demonstrated a tendency to recall the information best.
Pratkanis, Syak, and Gamble (1987) investigated the relationship between at-
titude toward social drinking and recall of persuasive communications concerning
drinking. The domain of social drinking is interesting because, in regards to
social policy, it is bipolar (i.e., most people can provide arguments for and
against drinking). However, knowledge concerning drinking activities is dis-
tributed in a unipolar pattern. Persons who have prodrinking attitudes are more
likely to know technical details such as how to fix exotic drinks, the number of
gallons in a keg of beer, and the names of various liquors. Pratkanis, Syak, and
Gamble asked subjects to learn three communications: one favoring social drink-
ing, one opposing social drinking, and one of a technical nature on how to make
an exotic drink called a Pousse-cafe. The results showed that attitudes were not
related to the learning of the pro- or anti-social-drinking message. However,
subjects with prodrinking attitudes demonstrated superior learning of the technical
Pousse-cafe message. Subjects with prodrinking attitudes also possessed greater
technical knowledge concerning drinking (as assessed by an objective test), and
this greater knowledge moderated the attitude-learning results.
Previous research has identified other unipolar attitude objects. For example,
Greenwald and Pratkanis ( 1984) reviewed evidence that a subject’s judgments
about self tend to be overly positive and are made rapidly when they set the
subject apart from others in a favorable direction and that self-relevant information
(which is usually positive) is recalled efficiently. Gustafson (1957) found that
members of various ethnic groups were better at learning the accomplishments
of in-group members in American history. Liben and Signorella (1980) and others
found that children with traditional gender-role attitudes (compared to those who
have not as yet developed these attitudes) demonstrate superior memory for ma-
terials that protray traditional sex roles. Becker and Byrne (1985) reviewed re-
search on sexual attitudes, and they presented a pattern of memory results that
are similar to those obtained with attitudes toward social drinking-erotophiles
obtained higher test scores in a college sexuality course, retained more information
from a birth-control lecture, and were better at recalling erotica than erotophobes.
Saegert and Young (1983) find that health food enthusiasts score well on general
nutrition tests.
 SOCIOCOGNITIVE MODEL OF AITITUDES 267

E. ADDITIONAL USES OF ATTITUDINAL

KNOWLEDGE

The foregoing research indicates that attitudes can be used to guide and organize
memory for events. However, attitude knowledge structures may serve other
purposes. As discussed in Section 1II.D. attitudinal knowledge can guide usage
of (or complex behavior toward) an object. Such knowledge structures should
contain technical and action-oriented information, possibly organized in a script-
format (Abelson, 1976, 1982). As illustrated by McGuire’s (1964) inoculation
research, attitudinal knowledge, especially that of a bipolar nature, can be of
use in arguing the merits of a proposal. Attitudinal knowledge can also by used
to describe the utility of an object, resulting in a structure listing the benefits
and drawbacks of an object similar to that proposed by expectancy-value theorists
(e.g., Fishbein & Ajzen, 1975). The varied usage of attitudinal knowledge sug-
gests that the type of knowledge structure supporting an attitude may be not just
a function of the topic domain (as demonstrated by Pratkanis), but also a function
of personality variables (see, for example, Olson & Zanna’s, 1979 research on
repressor’s and selective exposure), an understanding of intent to persuade (see
Adler et al., 1980). and the roles and tasks a person must perform in a given
situation (see, for example, Tetlock’s 1988 research on accountability and cog-
nitive structure, Sidanius’s, 1988 discussion of attitude function and political
sophistication, and Bosso’s, 1987, account of the life cycle of the pesticide issue).

V. The Self-Functions of Attitudes

The functional theories of Smith, Bruner, and White (1956) and of Katz (1960)
explicitly linked the possession of an attitude to the definition and appraisal of
the self-concept. Smith et al. (1956) suggested that attitudes serve to defend the
self through two functions: externalization (holding an attitude in order to protect
the self) and social adjustment (holding an attitude in order to maintain rela-
tionships with others). Katz suggested the self-related functions of ego-drf;nsive
(similar to Smith et d . ’ s externalization) and value-expressive (holding an attitude
as an act of self-expression).
The functional approach to attitudes has not yet generated an impressive volume
of research. After a brief flurry of research, mostly addressing the ego-defensive
functions (e.g.. Katz, McClintock, & Sarnoff, 1957; Stotland, Katz, & Pdtchen,
1959), the functional approach has been relegated to a position as a passing
remark in the attitude chapters of many social psychological texts. One reason
for this empirical neglect is that attitude functions were never explicitly linked
with empirical operations (Shavitt, 1989, and Snyder & DeBono, 1989), perhaps
268 ANTHONY R . PRATKANIS AND ANTHONY G . GREENWALD

as a general reflection of the weakness of the self metaphors of the 1950s and
1960s at linking conceptual definitions to research procedures (Prdtkanis &
Greenwald, 1985).
The sociocognitive perspective agrees with early functional models on the
importance of the self in determining attitude structure and functioning. In this
section, we use the Greenwald and Pratkanis (1984) view of the self to specify
the self-functions of an attitude. Unlike some critics of functional theories, we
believe that considerable evidence has accumulated in support of the proposition
that attitudes are held in service to the self, although much of this evidence is
linked to other theoretical perspectives and is typically not recognized as functional
research.

A. THE THREE FACETS OF THE SELF

AND THEIR ATTITUDE FUNCTIONS

The self has come to be viewed as a major factor in motivation as persons

seek to establish and maintain positive self-identities. In a recent analysis,
Greenwald and Breckler (1985; see also Breckler & Greenwald, 1986; Greenwald,
1982; Greenwald & Pratkanis, 1984) identified three classes of strategies for
establishing and maintaining self-esteem, termed the ego tasks, of the public,
private, and collective facets of the self. The public self‘s strategy is to maintain
and establish self-worth by earning favorable evaluations from others (a public
audience); the private self achieves self-worth by meeting or exceeding internalized
evaluative standards (the approval of a private audience); and the collective self
establishes self-worth by seeking to attain the goals of reference groups (a col-
lective audience). Greenwald (1989) has suggested that a wide variety of attitudes
readily participate in these strategies for establishing and maintaining self-regard.
When the public facet of the self is emphasized, the person should display
attitudes that are agreeable to others. Such attitudes can be used to earn the
approval of others and, via this public-self strategy, self-regard. The public self‘s
strategy corresponds to Smith et al.’s social adjustment function (“[Olne will
more readily and forthrightly express acceptable attitudes while inhibiting or
modulating the expression of less approved ones” [pp. 4 1-42]).
The private facet of the self earns self-regard by meeting or exceeding inter-
nalized criteria of success. One strategy for achieving this goal is to maintain
attitudes that are consistent with other attitudes and with one’s self-concept. Katz’s
( 1960) value-expressive function (‘‘the individual derives satisfactions from ex-
pressing attitudes appropriate to his personal values” [p. 1701) can thus be in-
terpreted as a manifestation of the private facet of the self.
The collective facet of the self establishes self-worth by helping to achieve
the goals of important reference groups (family, church, profession, work group,
 SOCIOCOGNITIVE MODEL OF ATTITUDES 269

etc.) An obvious strategy toward that end is to value objects that are identified
with those groups. Attitudes that are shaped by this strategy may be said to serve
a group solidarity or sociul identification function. This last function is not one
that appears in either the Smith et ul. or the Katz original list. Smith (1980) later
extended the meaning of social adjustment to include mediation of self-other
relationships and social identification. However, the present list of self-functions
was anticipated by Kelman’s ( 1961) three influence processes of compliance
(public), internalization (private), and identification (collective).

B. EVIDENCE FOR THE SELF-RELATED

FUNCTIONS OF ATTITUDES

Much evidence exists in social psychology to demonstrate that attitudes are

used to define, appraise, and maintain the self-concept. We review eight domains
of empirical inquiry. (See also Herek, 1987; Shavitt, 1989; Snyder & DeBono,
1989, for recent discussions on operationalizing attitude functions).

I . Self-Definition and Verification

Research from a number of theoretical perspectives indicates that attitudes are
used to define one’s self. Cialdini et al. (1976) find that people tend to BIRG
(bask in reflected glory) by expressing positive attitudes toward winning football
teams (e.g., donning team insignia). Conversely, Tesser and Campbell (1983)
report that subjects express negative attitudes toward persons whose performance
is threatening to their own self-esteem. Swann’s self-verification ( 1983) and
Wicklund and Gollwitzer’s ( 1982) symbolic self-completion theories both suggest
that attitudes toward physical objects and people can be used to maintain one’s
self-concept. C. Snyder and Fromkin (1980) review studies indicating that at-
titudes can be adopted as a strategy for establishing uniqueness (see also Abelson
& Prentice, 1989). Goethals (1986) finds that strategies for maintaining a positive
self-image can result in both a false-consensus and a false-uniqueness effect.

2 . The Self-Justification of Dissonant Acts

In a reformulation of dissonance theory, Aronson ( 1969, 1984; see also Green-
wald & Ronis, 1978) stated that dissonance is most powerful as a motivator of
human behavior when the self-concept is threatened. In such cases, attitudes can
be adopted and expressed as one possible means of reconciling undesirable acts
to a positive self-concept. Aronson (1984) interprets the literature on dissonance
as indicating that attitudes are adopted as a means of justifying the self‘s in-
volvement in ( 1 ) escalating commitment to a failing course of action (Brockner
270 ANTHONY R . PRATKANIS AND ANTHONY G. CREENWALD

& Rubin, 1985); (2) the performance of a self-discrepant act for insufficient
justification (Festinger & Carlsmith, 1959); (3) the overexpenditure of effort
(Aronson & Mills, 1959); and (4) cruelty (Lerner, 1980).

3 . Selective Exposure
As noted previously, early research, based on cognitive dissonance theory,
has revealed that the selective-exposure effect is not particularly easy to obtain.
Recently, Frey (1986) has reviewed evidence showing that selective exposure is
more likely to occur under conditions of (1) perceived free choice and (2) com-
mitment. For example, Sweeney and Gruber (1984) found that committed Nixon
supporters (who supposedly chose their attitudes freely) reportedly paid less at-
tention and knew less information about the Watergate proceedings. In the ter-
minology of a revised dissonance theory (Aronson, 1969; Greenwald & Ronis,
1978). selective exposure occurs when the possession of an attitude has impli-
cations for self-evaluation.

4 . Impression Management
According to Schlenker ( 1980), attitudes are used for impression management
purposes; people express attitudes in order to claim desirable self-images and to
reject undesirable ones. As evidence for this proposition, Schlenker points to
research indicating that subjects ( I ) are differentially persuaded by a commu-
nication as a function of the nature of an audience observing the attitude expression
(Braver, Linder, Corwin, & Cialdini, 1977); (2) express attitude similarity and
liking for others as an ingratiation tactic (Jones, 1964); and (3) modify their
opinions in anticipation of a persuasive communication, to gain strategic self-
presentational advantage (Cialdini & Petty, I98 I ) .

5 . Self Monitoring
Snyder and DeBono (1989) use the Self-Monitoring scale (Snyder, 1974; Sny-
der, 1987) to operationalize attitude functions. They reasoned that for high self-
monitors (who strive to be the person called for by the situation), attitudes will
serve primarily a social adjustment function, whereas for low self-monitors (who
attempt to use an internal standard as a guide for behavior), attitudes will serve
a value-expressive purpose. In a program of research, Snyder and DeBono (1989)
found that high self-monitors are most persuaded by appeals targeted to the social
adjustment function (e.g., image advertising and messages describing widespread
social agreement for a position) and that low self-monitors are most persuaded
by appeals targeted to the value-expressive function (e.g., reason-why advertising
and messages relating an issue to underlying values). In a similar vein, Scheier
and Carver ( 1983) review evidence showing that when public self-consciousness
is high, people modify the expression of their attitudes to fit the situation, whereas
 SOCIOCOGNITIVE MODEL OF AITITUDES 27 1

when private self-consciousness is high, people act in accordance with their at-
titudes.

6 . Symbolic Attitudes
Recent research in the area of AIDS and racial integration have found that
attitudes are often held for symbolic reasons (i.e., related to values that are im-
portant for self-definition), in contrast to instrumental concerns (i.e., direct and
immediate consequences of interacting with an object). Pryor, Reeder, Vinacco,
and Russo (1987) found that attitudes toward a heterosexual AIDS victim were
a function of both the instrumental value of interacting with the victim (as mea-
sured by procedures developed by Fishbein & Ajzen, 1975) and symbolic reasons
(e.g., the belief in AIDS as a homosexual affliction and one’s own level of
homophobia). Similarly, Kinder and Sears (1981) found that whites’ support for
a black mayoral candidate was largely determined by symbolic factors (e.g.,
traditional white Protestantism morality), as opposed to direct racial threats (e.g.,
loss of job or property-value). Although racial attitudes may serve symbolic
(value-expressive)purposes, they can also result, as Pettigrew (1958, 1959) found,
from conformity to group norms, thus serving social adjustment and identification
functions. (For a further discussion of symbolic attitude research, see Bobo,
1983; Herek, 1986; Sears & Kinder, 1985; Sniderman & Tetlock, 1986).

7. Social Status, Social Role, and Attitudes

Attitudes often serve to maintain the social status and role occupancy of the
self. For example, Lieberman (1956) found that workers who became company
foremen developed promanagement attitudes (i.e., positive attitudes toward the
company as a place to work, toward top management, and toward the company’s
incentive system), whereas workers who became union stewards developed pro-
union attitudes (i.e., positive attitudes toward the labor union, toward top union
officers, and toward seniority, as opposed to ability, as a criterion for advance-
ment). When the foremen and stewards returned to their original rank-and-file
worker roles, their attitudes tended to revert back to their previous positions.
Similarly, Frederickson (1969) found that the role occupancy of local government
officials (in particular, legislators, elected executives, appointed executives, labor
representatives, and arbitrators) was an important factor in determining attitudes
toward various labor relations issues. Herman and Hulin (1972) found that various
job-related attitudes were a function of status and departmental membership within
an organization.

8 . The Maintenance of In-Groups and Out-Groups

Attitudes can serve as a means of maintaining boundaries between groups and
of defining one’s self within the collective (see Tajfel, 1981, and Turner, 1987
272 ANTHONY R . PRATKANIS AND ANTHONY G . GREENWALD

for a theory of self-categorization into a social group). Abelson and Prentice

(1989) refer to this as the “badge value of attitudes”-the adoption of highly
visible attitudes, important for maintaining social identification with the group.
Social groups often expect members to adopt certain attitudes (e.g., acceptance
of distinctive religious garb, ethnic food preferences, shared beliefs about out-
groups) as a means of defining the group’s distinctiveness and for identifying
the individual with the group. For example, Hurtado and Gurin (1987) found
that support for bilingualism (i.e., using Spanish and English as means of com-
munication) among individuals of Mexican descent living in the United States
served to promote ethnic identity. Approval of bilingualism was more strongly
associated with a Chicano (i.e., mestizo-mestiza, la raza) as opposed to an
“upwardly mobile” identity.

C . THE IMPORTANCE OF THE SELF FOR

UNDERSTANDING ATTITUDES

The self plays an important role in establishing the contents and organization
of attitude structure and in specifying the conditions under which certain attitude
effects will occur.

I . The Self and Attitude Structure

The pursuit of positive self-regard can affect the label, evaluation, and knowl-
edge associated with an attitude object. As noted in Section II,A, objects can
be labelled and identified to the self‘s advantage. Functional theories of attitude
change (see Insko, 1967; Kiesler, Collins, & Miller, 1969, for reviews) have
suggested, and much of the foregoing research has demonstrated, that object
evaluations are created and maintained in service to the self. Attitudinal knowledge
structures may develop as a result of the persistence of an ego task. For example,
social-adjustive attitudes should contain information about how other people value
the object; value-expressive attitudes should be associated with knowledge of
the relationship between an attitude and other attitudes and the self; and social-
identification attitudes should be associated with knowledge about how reference-
group members value the attitude object.

2 . The Self and Attitude Prediction

The research on self-functions has identified another condition that improves
the predictive power of an attitude. People are more likely to act in accord with
their attitudes when those attitudes are important (either situationally or chron-
ically) for their self-concepts. Furthermore, people are likely to act on these
attitudes, not solely in an evaluatively consistent manner, as suggested by cog-
 SOCIOCOGNITIVE MODEL OF ATTITUDES 273

nitive consistency theories, but in a fashion that increases self-worth (which in

some cases may give the appearance of inconsistency).

VI. Concluding Remarks

A. WHAT IS AN ATTITUDE?

According to a sociocognitive model, an attitude is represented in memory

by ( I ) an object label and procedures for applying that label, (2) a summary
evaluation of that object, and (3) a knowledge structure supporting that evaluation.
This cognitive representation relates a person to the social world by serving to
make sense of the environment (heuristic and schematic functions) and to es-
tablish, maintain, and enhance positive self-regard (social adjustment, value-
expressive, and social identification functions).

B. HOW WARRANTED IS THE PESSIMISM

CONCERNING ATTITUDES'?

Contrary to popular belief in social psychology, attitudes are nor epiphenomena1

but are predictive of large classes of social behavior. Attitudes are found to
predict: ( 1 ) conceptual cognitive processes such as inference, judgment, and rea-
soning in a wide variety of circumstances; (2) behavior when the evaluation of
an object is readily used to interpret the situation and the person has the ability
(knowledge) and opportunity to act in accord with these perceptions; ( 3 ) memory
for a persuasive communication via the knowledge structure supporting an eval-
uation; and (4) selective exposure to information when the self is implicated in
the further processing of that information. In general, attitudes are better predictors
when there is consistent and consensual identification of the attitude object.

C. WHY THEN ALL THE CONCERN

ABOUT ATTITUDES'?

Approximately 20 years ago, there was a flourishing of books and articles on

attitudes (Abelson et ul., 1968; Fishbein, 1967; Greenwald, Brock, & Ostrom.
1968; McGuire, 1969, to list a few). At this time, there was broad acceptance
of a definition of attitude that was stated in terms of the venerable tripartite of
affect, behavior, and cognition. Attitude theory was strongly dominated by cog-
nitive-consistency principles that were associated with the concepts of balance,
congruity, and dissonance.
274 ANTHONY R . PRATKANIS AND ANTHONY G . GREENWALD

As we have seen in this article, attitudes are complexly related to behavior,

to selective exposure for information, and to memory for a persuasive com-
munication. The major theories of attitudes, popular 20 years ago, are unable
adequately to describe and summarize the conditions under which attitudes pro-
duce such effects. It is a mistake to assume that attitudes are poor predictors of
social processes merely because our theories of attitude have not succeeded in
describing those relationships.

D. WHITHER THE COURSE OF

ATTITUDE RESEARCH?

Greenwald, Pratkanis, Leippe, and Baumgardner ( 1986) view scientific prog-

ress as both the search for qualifying conditions on existing effects and the de-
velopment of increasingly powerful theories to describe those conditions. In this
sense, the course of attitude research can certainly be viewed as scientific progress.
In 1935, Allport proclaimed the importance of the attitude concept for social
psychology. After approximately three decades of attitude research, multiple
commentaries were published announcing that attitudes were apparently not as
predictive of social behavior as originally assumed and that it was time to consider
abandoning the attitude concept. Research in the 1970s responded to these cri-
tiques by addressing the question, “Under what condition does an attitude predict
a given effect?” (see Zanna & Fazio, 1982). This later research has achieved
rather promising results, as previously unruly attitude effects are increasingly
brought under experimental control.
The sociocognitive model of attitude represents our attempt to resolve the
paradox of attitude importance and to integrate research findings by specifying
the conditions under which significant attitude effects occur. As a summary of
attitude effects, it is hoped that the sociocognitive model of attitude will generate
further formulations about the relationship of attitudes to social behavior and
will serve to maintain the conception that attitudes function powerfully to guide
interaction in a complex social world.

VII. Summary

The history of attitude research presents a paradox. As early as 1935, attitude

was proclaimed as social psychology’s most indispensable construct. That faith
in the attitude construct remains strong today. At the same time, the predictive
utility of attitudes has been widely questioned, as researchers have had difficulty
demonstrating strong positive relationships of attitudes ( I ) to behavior, (2) to
memory for persuasive messages, and (3) to selection among items of controversial
information.
 SOCIOCOGNITIVE MODEL OF ATTITUDES 275

The present thesis is that pessimism concerning the predictive utility of the
attitude construct is unwarranted. We review evidence from diverse areas of
social psychology to demonstrate that attitudes frequently serve three functions:
heuristic (attitudes provide a simple strategy for appraising an object), schematic
(attitudes organize and guide complex behavior towards an object and memory
for events), and self-related (attitudes are used to define and maintain self-worth).
These functions are used to construct a sociocognitive model of attitude (as a
replacement of previous approaches). According to this model, an attitude is
represented in memory by ( I ) an object label and procedures for applying that
label (which needs to be engaged for attitude effects to occur), (2) an evaluation
associated with that structure (which guides the heuristic function), and (3) a
knowledge structure supporting that evaluation (which serves the schematic
function). As a cognitive representation, attitudes are used to relate an individual
to the social world.

Acknowledgments

This article is an elaboration of two previous chapters appearing as Greenwald ( 1989) and Pratkania
(1989). The authors thank Cheryl Boglarsky, David Douglass. Rebecca Slaton, M. Brewster Smith,
and Marlene E. Turner for comments on an earlier draft.

References

Abelson, R. P. (1959). Modes of resolution of belief dilemmas. Journul of Conflict Resolurion. 3,

343-352.
Abelson, R. P. (1972). Are attitudes necessary'? In B. T. King & E. McGinnies (Eds.), Attitudes,
conflict. and social change (pp. 19-32). New York: Academic Press.
Abelson, R. P. (1976). Script processing in attitude formation and decision-making. In I . S. Carroll
& J . W. Payne (Eds.), Cognition and social behavior (pp. 33-45). Hillsdale, NJ: Erlbaum.
Abelson, R. P. (1982). Three modes of attitude-behavior consistency. In M. P. Zanna, E. T. Higgins,
& C. P. Herman (Eds.), Consisfency in sociul behavior (pp. 131-146). Hillsdale, NJ: Erlbaum.
Abelson, R. P. (1988). Conviction. American Psychologist. 43, 267-275.
Abelson, R. P., Aronson, E., McGuire, W. J.. Newcomb, T. M., Rosenberg, M. J., & Tannenbaum,
P. H. (Eds.) (1968). Theories ofcognitive corisistenc:vc A sourcebook. Chicago: Rand-McNally.
Abelson, R. P., & Prentice, D. A. (1989). Beliefs as possessions: A functional perspective. In A.
R. Pratkanis, S. J . Breckler. & A. G. Greenwald (Eds.). Attitudes structure andfitncrion (pp.
361-381). Hillsdale, NJ: Erlbaum.
Aderman. D., & Brehm, S. S . (1976). On the recall of initial attitudes following counterattitudinal
advocacy: An experimental reexamination. Personolity and Social Psychology Bulletin. 2 , 59-
62.
Adler, R. P., Friedlander. B. Z., Lesser, G. S., McRingoff. L.. Robertson, T. S., Rossiter. J . R . ,
& Ward. S . (1980). Reseurrh on the efects of television advertising on children. Lexington,
MA: Lexington Books.
Ajzen, 1. (1989). Attitude structure and behavior. In A. R . Pratkanis, S. J. Breckler, & A. G.
Greenwald (Eds.), Attitudes sfructure undJunction (pp. 241-274). Hillsdale, NJ: Erlbaum.
276 ANTHONY R. PRATKANIS AND ANTHONY G. GREENWALD

Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of
empirical research. f‘svcho/ogicu/ Review. 84, 888-9 18.
Allport, G. W. (1935). Attitudes. In C. Murchison (Ed.), The hundbook c~fsocicrlpsychology (pp.
798-844). Worchester. MA: Clark University Press.
Anderson, N . H., & Hubert. S . (1963). Effects of concomitant verbal recall on order effects in
personality impression formation. Journal of Verlxrl Leuming and Verbal Behavior. 2, 379-
391.
Aronson, E. (1969). The theory of cognitive dissonance: A current perspective. In L. Berkowitz
(Ed.), Advances in e.xperimm/nl . s o c k ~ pl ~ . ~ c / i ~ l o(Vol.
g V 4, pp. 1-34). New York: Academic
Press.
Aronson. E. (1984). The socicil uniniul. New York: W. H. Freeman and Co.
Aronson, E.. & Mills, I . (1959). The effect of severity of initiation on liking for a group. Journcrl
of Abtiormirl and Socicrl Psvchologv. 59, 177-18 I .
Bain, R. (1928). An attitude on attitude research. Antericm Journul of Sociology. 33, 940-957.
Baron, R. A. (1986). Behuvior in orgunizufions. Boston: MA: Allyn & Bacon.
Becker. M. A , , & Bryne, D. ( 1985). Self-regulated exposure to erotica, recall errors, and subjective
reactions as a function of erotophobia and Type A coronary-prone behavior. Journul of’Persoritility
nnd Soc~ialPsychvlugv. 48, 76Ck.767.
Bem, D. J. (1967). Self-perception: An alternative interpretation of cognitive dissonance phenomena.
Psychological Rrview, 74, 183-200.
Bem, D. J., & McConnell, H. K . (1970). Testing the self-perception of dissonance phenomena: On
the salience of premanipulation attitudes. Journal of Personality und Socinl Psychology. 14,
23-3 I .
Berkowitz, L.. & Rogers. K. H. (1986). A priming effect analysis of media influences. In J. Bryant
& D. Zillnian (Eds), Perspectives on muss medicr (pp. 57-81). Hillsdale, NJ: Erlbdum.
Billig, M. ( 1987). Arguing mid thinking: A rhetorical approach to social p s . ~ ~ h ~ l Cambridge:
~gV.
Cambridge University Press.
Blunier. H. (1955). Attitudes and the social act. Sociril Problems, 3, 59-65.
Bobo, L. ( 1983). Whites’ opposition to busing: Symbolic racism or realistic group conflict’?Joirrnul
of Personuli/y and Social P.svchologv, 45, I 196- I2 10.
Borgida, E. & Campbell, B. (1982). Belief relevance and attitude-behavior consistency: The mod-
erating role of personal experience. Journal of Persoiia/ify and Sociid PS!.ChOkJKv, 42, 239-
247.
Bosso. C. J. ( 1987). Pesticides und poliricst The rifr cycle ($ti public issue. Pittsburgh. PA: University
of Pittsburgh Press.
Braver, S. L., Linder, D. E., Corwin, T. T., & Cialdini, R. B. (1977). Some conditions that affect
admissions of attitude change. Journal qfE.xperinten/al Socinl P S ~ C . / I O / O13, 565-576.
~.V .
Breckler. S . J. (1984). Empirical validation of affect, behavior, and cognition as distinct components
of attitude. Journal of Personality and Social Psvchology. 47. I 19 1-1 205.
Breckler, S. I., & Greenwald, A. G. (1986). Motivational facets of the self. In R. M. Sorrentino
& E. T. Higgins (Eds.). Handlmok of’ motivation mnd cognition (pp. 145-164). New York:
Guilford Press.
Breckler, S. J . . & Wiggins. E. C. (1989). On defining attitude and attitude theory: Once more with
feeling. In A. R. Pratkanis. S. J. Breckler, & A. G. Greenwald (Eds.), Attitude structure and
function (pp. 407-427). Hillsdale. NJ: Erlbaum.
Brigham, I . C., & Cook, S. W. (1969). The influence of attitude on the recall of controversial
materials: A failure to confirm. Journal o j Experirnrntal Social Psychology. 5 , 24-243.
Brockner, J., & Rubin, J. 2. ( 1985). Entrapnients in escalu/itig conJicts: A social p.sycholugicu/
analysis. New York: Springer-Verlag.
Byrne, D. (1971). The attruction parudigm. New York: Academic Press.
Cacioppo. J. T., & Petty. R. E. (1979). Effects of message repetition and position on cognitive
response, recall, and persuasion. Journul of Personnliry und Soc i d Psychology, 37, 97- 109.
 SOCIOCOGNITIVE MODEL OF ATTITUDES 277

Cacioppo, J . T., Petty. R. E.. & Geen. T. R. (1989). Attitude structure and function: From the
tripartite to the homeostasis model of attitudes. In A. R. Pratkanis, S. J. Breckler. & A. G.
Greenwald (Eds.). Attitude sfructure and function (pp. 275-309). Hillsdale. NJ: Erlbauni.
Cantril, H. (1932). General and specific attitudes. Psychologicul Monographs, 42, ( 5 , Whole No.
192).
Cantril. H. (1940). America faces the war: A study in public opinion. Public Opinion Quarterly.
4, 387407.
Carlston, D. E. (1980). Events, inferences, and impression formation. In R . Hastie, T. M. Ostrom,
E. B. Ebbesen, R . S. Wyer. D. L. Hamilton. & D. E. Carlston (Eds.), Person memory: The
cognitive hasis of sociul perception (pp. 89-1 19). Hillsdale, NJ: Erlbaum.
Chiesi. H. L.. Spilich. G. J . , & Voss. J . F. (1979). Acquisition of domain-related information in
relation to high and low domain knowledge. Journcil of Verhul Lecrrning and Verbal Behnvior.
18, 257-273.
Cialdini. R. B.. Borden. R . J.. Thorne. A,. Walker. M. R.. Freeman, S., & Sloan, L. R . (1976).
Basking in reflected glory: Three (football) field studies. Journril of Personulih atid Sociul
Psycliolop. 34, 366-375.
Cialdini. R . 8.. & Petty, R . E. (1981). Anticipatory opinion effects. In R. E. Petty. T. M. Ostrom,
& T. C. Brock (Eds.), Cognitive respouses in per.sutision (pp. 217-235). Hillsdale. NJ: Erlbaum.
Converse, P. E. (1970). Attitudes and non-attitudes: Continuation of a dialogue. In E. R . Tuite
(Ed.), The yuatititative uncilysis cf social prohlerns (pp. 168-189). Reading. MA: Addison-
Wesley.
Cooky. C. H. (1912). Valuation as a social process. Ps.vclzoIogicu1Bulletin, 9, 441450.
Cooper, E., & Jahoda, M. (1947). The evasion of propaganda: How prejudiced people respond to
anti-prejudice propaganda. Joumcrl qf Sociul Psv:vc~holog.v.23, 15-25,
Cotton. J. L.. & Tuttle. J . M. (1986). Employee turnover: A meta-analysis and review with implications
for research. Academy of Manugement Review. 11, 55-70.
Davidson. A. R., & Morrison. D. M. (1983). Predicting contraceptive behavior from attitudes: A
comparison of within- versus across-subjects procedures. Jourtial of Personalify mid Social
P.yvcholog.v, 45, 997- 1009.
DeFleur. M. L., & Westie, F. R. (1963). Attitude as a scientific concept. Social Forces, 42, 17-
31.
Devine, P. G. (1989). Automatic and controlled processes in prejudice: The role of stereotypes and
personal beliefs. In A. R. Pratkanis, S. 1. Breckler, & A. G. Greenwald (Eds.). Attitude structure
and,function (pp. 181-212). Hillsdale, NJ: Erlbaum.
Dillehay. R. C. ( 1973). On the irrelevance of the classical negative evidence concerning the effect
of attitudes on behavior. Americtm Psvchologist. 28, 887-891.
Doob, L. W. (1953). Effects of initial serial position and attitude upon recall under conditions of
low motivation. Journal of Abnormal mid Sociul Psycholugv. 48, 199-205.
Dreben, E. K . , Fiske, S. T.. & Hastie, R . (1979). The independence of evaluative and item infor-
mation: Impression and recall order effects in behavior-based impression fomiation. Journul of
Pcvx)nality und Soc~ialPs~chologv,31, 1758- 1768.
Eberhart. J . C.. & Bauer. R. A. (1941). An analysis of the influences on recall of a controversial
event: The Chicago Tribune and the Republic Steel strike. Jountul of’Sociul Psvchology. S.P.S.S.I.
Bitlletin. 14, 21 1-228.
Edwards. A. L. (1941). Political frames of reference as a factor influencing recognition. Journal of
Abnormal cind Soikrl Psychology, 36, 34-50.
Edwards, A. L. ( 1957). Techniques of attitude scale consrruction. New York: Appleton-Century-
Croft.
Evans. J . St. B. T.. Barston, J. L.. & Pollard, P. (1983). On the conflict between logic and belief
in syllogistic reasoning. Metnory and Cognilion. 11, 295-306.
Farquhar. P. H., & Pratkanis, A. R . (1987). Decision structuring with uriuvailuble options. Un-
published manuscript. Carnegie-Mellon University. Pittsburgh, PA
27 8 ANTHONY R. PRATKANIS AND ANTHONY G. GREENWALD

Fazio, R. H. (1986). How do attitudes guide behavior? In R. M. Sorrentino & E. T. Higgins (Eds.).
Handbook of motivation and cognition (pp. 204-242). New York: Guilford Press.
Fazio, R. H.(1989). On the power and functionality of attitudes: The role of attitude accessibility.
In A. R. Pratkanis, S. J . Breckler, & A. G . Greenwald (Eds.), Attitude structure and function
(pp. 153-179). Hillsdale, NJ: Erlbaum.
Fazio, R. H., & Zanna, M. P. (1978a). On the predictive validity of attitudes: The roles of direct
experience and confidence. Journal of Personality, 46, 228-243.
Fazio, R. H., & Zanna, M. P. (1978b). Attitudinal qualities relating to the strength of the attitude-
behavior relationship. Journal of Experimental Social Psychology, 14, 398A08.
Feather, N. T. (1964). Acceptance and rejection of arguments in relation to attitude strength, critical
ability, and intolerance of inconsistency. Journal of Abnormal and Social Psychology, 69, 127-
136.
Feather, N. T. (1969). Attitude and selective recall. Journal of Personality and Social Psychology.
12, 31&319.
Festinger, L. (1957). Theory of cognitive dissonance. Stanford, CA: Stanford University Press.
Festinger, L. (1964). Behavioral support for opinion change. Public Opinion Quarterly, 28, 404-
417.
Festinger, L., & Carlsmith, J . M. (1959). Cognitive consequences of forced compliance. Journal
of Abnormal and Social Psychology, 58, 203-210.
Fields, J . M., & Schuman, H. (1976). Public beliefs about the beliefs of the public. Public Opinion
Quarterly. 40,4 2 7 4 8 .
Fishbein, M. (Ed.) (1967). Readings in attitude theory and measurement. New York: Wiley.
Fishbein, M., & Ajzen, 1. (1975). Belief, attitude. intention and behavior: An introduction to theory
and research. Reading, MA: Addison-Wesley.
Fiske, S. T., & Pavelchak, M. A. (1986). Category-based versus piecemeal-based affective responses:
Developments in schema-triggered affect. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook
of motivation and cognition. New York: Guilford.
Fleming, D. (1967). Attitude: The history of a concept. Perspectives in American history, 1, 287-
365.
Frederickson, H.G. (1969). Role occupancy and attitudes toward labor relations in government.
Administrative Science Quarterly, 14, 595-606.
Freedman, J . L., & Sears, D. 0. (1965). Selective exposure. In L. Berkowitz (Ed.), Advances in
experimental social psychology (Vol. I , pp 57-97). New York: Academic Press.
Frey, D. (1986). Recent research on selective exposure to information. In L. Berkowitz (Ed.), Advances
in experimental social psychology (Vol. 19, pp. 41-80). New York: Academic Press.
Gallup, G. H. (1972). The Gallup Poll: Public opinion 1935-1971. New York: Random House.
Goethals, G. R. (1986). Fabricating and ignoring social reality: Self-serving estimates of consensus.
In J . M. Olson, C. P. Herman, & M. P. Zanna (Eds.), Relative deprivation and social comparison
(pp. 135-157). Hillsdale. NJ: Erlbaum.
Goethals, G. R., & Reckman, R. F. (1973). The perception of consistency in attitudes. Journal of
Experimental Social Psychology, 9, 491-501.
Goldberg, P. A., Gottesdiener, M., & Abramson, P. R . (1975). Another put-down of women?
Perceived attractiveness as a function of support for the feminist movement. Journal of Personality
and Social Psychology, 32, 113-1 15.
Gordon, R. L. (1953). The effect of attitude toward Russia on logical reasoning. Journal of Social
Psychology, 37, 103- I 1 I .
Granberg, D. (1985). An anomaly in political perception. Public Opinion Quarterly, 49,
504-516.
Granberg, D., & Brent, E. (1983). When prophecy bends: The preference-expectation link in U.S.
Presidential elections, 1952-1980. Journal of Personalify and Social Psychology, 45, 477-
491.
 SOCIOCOGNITIVE MODEL OF ATTITUDES 279

Granberg, D., Jefferson, N. L. Brent, E. E., & King, M. (1981). Membership group, reference
group, and the attribution of attitudes to groups. Journal of Personality and Social Psychology,
40,833-842.
Granberg, D., & Jenks, R. (1977). Assimilation and contrast effects in the 1972 election. Human
Relations. 30,623-640.
Greenwald, A. G. (1968a). Cognitive learning, cognitive response to persuasion, and attitude change.
In A. G. Greenwald, T. C. Brock, & T. M. Ostrom (Eds.), Psychological foundations of
attitudes (pp. 147-170). New York: Academic Press.
Greenwald. A. G. (l968b). On defining attitude and attitude theory. In A. G. Greenwald, T. C.
Brock, & T. M. Ostrom (Eds.), Psychological foundations of attitudes (pp. 361-388). New
York: Academic Press.
Greenwald, A. G . (1980). The totalitarian ego. American Psychologist. 35, 603-618.
Greenwald, A. G. (1982). Ego task analysis: An integration of research on ego-involvement and
self-awareness. In A. Hastorf & A. M. lsen (Eds.), Cognitive socialpsychology (pp. 109-147).
New York: Elsevier North-Holland.
Greenwald, A. G. ( 1988). Levels of representation. Unpublished manuscript. University of Wash-
ington, Seattle, WA.
Greenwald, A. G. (1989). Why attitudes are important: Defining attitude and attitude theory 20
years later. In A. R. Pratkanis, S. J. Breckler, & A. G. Greenwald (Eds.), Attitude structure
and function (pp. 429-440). Hillsdale, NJ: Erlbaum.
Greenwald, A. G., & Breckler, S. J. (1985). To whom is the self presented? In B. R. Schlenker
(Ed.), The selfand social life (pp. 126145). New York: McGraw-Hill.
Greenwald, A. G., Brock, T. C . , & Ostrom, T. M. (Eds.). (1968). Psychological foundations of
artitudes. New York: Academic Press.
Greenwald, A. G., & Pratkanis, A. R. (1984). The self. In R. S. Wyer & T. K. S ~ l (Eds.). l The
handbook of social cognition (pp. 129-178). Hillsdale, NJ: Erlbaum.
Greenwald, A. G., Pratkanis, A. R . , Leippe, M. R., & Baumgardner, M. H. (1986). Under what
conditions does theory obstruct research progress? Psychological Review, 93, 2 1 6 2 2 9 .
Greenwald, A. G., & Ronis, D. L. (1978). Twenty years of cognitive dissonance: Case study of
the evolution of a theory. Psychological Review, 85, 53-57.
Greenwald, A. G., & Sakumara, J. S. (1967). Attitude and selective learning: Where are the phe-
nomena of yesteryear? Journal of Personality and Social Psychology, 7 , 387-397.
Gustafson, L. (1957). Relationship between ethnic group membership and the retention of selected
facts pertaining to American history and culture. Journal of Educational Sociology, 31, 49-
56.
Hammond. K. B. (1948). Measuring attitudes by error-choice: An indirect method. Journal of Ab-
normal and Social Psychology, 43, 3 8 4 8 .
Hastie, R. (1981). Schematic principles in human memory. In E. T. Higgins, C. P. Herman, & M.
P. Zanna (Eds.), Social cognition (pp. 39-88). Hillsdale, NJ: Erlbaum.
Hastorf, A. H., & Cantril, H. (1954). They saw a game. Journal of Abnormal and Social Psychology.
49, 129-134.
Hatfield, E., & Sprecher, S. (1986). Mirror, mirror: The importance of looks in everyday life.
Albany, NY: State University of New York Press.
Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley.
Herek, G. M. (1986). The instrumentality of ideologies: Toward a neofunctional theory of attitudes.
Journal of Social Issues, 42, 99-1 14.
Herek, G. M. (1987). Can functions be measured? A new perspective on the functional approach
to attitudes. Social Psychology Quarterly. 50, 285-303.
Herman, 1. B., & H u h , C. L. (1972). Studying organizational attitudes from individual and or-
ganizational frames of reference. Organizational Behavior and Human PerJormance. 8, 84-
108.
280 ANTHONY R. PRATKANIS AND ANTHONY G. GREENWALD

Higgins, E. T., & King; d. (1981). Accessibility of social constructs: Information-processing con-
sequences of individual and contextual variability. In N. Cantor & J. F. Kihlstrom (Eds.),
Personality. cognition. and social interaction (pp. 69-121). Hillsdale, NJ: Erlbaum.
Hurtado, A., & Gurin, P. (1987). Ethnic identity and bilingualism attitudes. Hispanic Journul of
Behavioral Sciences, 9, 1-18,
Hymes, R. W. (1986). Political attitudes as social categories: A new look at selective memory.
Journal of Personality and Social Psychology, 51, 233-241.
laffaldano, M. T., & Muchinsky, P. M. (1985). Job satisfaction and job performance: A meta-
analysis. Psychological Bulletin, 97, 251-273.
Insko, C. A. (1967). Theories of attitude change. New York: Appleton-Century-Crofts.
Iyengar, S . , & Kinder, D. R. (1987). News that matters. Chicago: University of Chicago Press.
Janis, 1. L., & Frick, F. (1943). The relationship between attitudes toward conclusions and
errors in judging the logical validity of syllogisms. Journal of Experimental Psychology. 33,
73-77.
Jones, E. E. (1964). Ingratiation. New York: Appleton-Century-Crofts.
Jones, E. E., & Kohler, R. (1956). The effects of plausibility on the learning of controversial state-
ments. Journal of Abnormal and Social Psychology, 53, 27-33.
Judd. C. M., & Kulik, J. A. (1980). Schematic effects of social attitudes on information processing
and recall. Journul of Personality and Social Psvchology, 38, 569-578.
Katona, G . (1975). Psychological economics. New York: Elsevier.
Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion Quurterly. 24,
163-204.
Katz, D., McClintock, C., & Sarnoff, I. (1957). The measurement of ego defense as related to
attitude change. Journal of Personality. 25, 465-474.
Katz. D., & Stotland. E. (1959). A preliminary statement to a theory of attitude structure and change.
In S. Kock (Ed.). Psychology: A study ? f a science (Vol. 3, pp. 423475). New York: McCraw-
Hill.
Kelman, H. C. (1961). Processes of opinion change. Public Opinion Quarterly, 25, 57-78.
Kelman, H.C. (1974). Attitudes are alive and well and gainfully employed in the sphere of action.
Americun Psychologist, 29, 310-324.
Kendall, P. L., & Wolf, F. M. (1949). The analysis of deviant cases in communications research.
In P. F. Lazarsfeld & F. N. Stanton (Eds.), Communicutions reseurch 1 9 4 8 4 9 (pp. 152-179).
New York: Harper & Brothers.
Kiesler, C. A,, Collins, B. E., & Miller, N. (1969). Attitude chunge: A critical ana/ysis of theoretical
approaches. New York: Wiley.
Kinder, D. R., & Sears, D. 0. (1981). Prejudice and politics: Symbolic racism versus racial threats
to the good life. Journal of Personality and Social Psychology. 40, 414431.
Kothandapani, V. (1971). Validation of feeling, belief, and intention to act as three components of
attitude and their contribution to prediction of contraceptive behavior. Journal qf Personcrlity
and Social Psychology. 19, 321-333.
Krech, D., & Crutchfield, R. S. (1948). Theories and problems of socialpsychology. New York:
McGraw-Hill.
Kreman. E. 0. (1949). An artempt to ameliorate hostility towards the Negro through role playing.
Unpublished master’s thesis, The Ohio State University, Columbus, OH.
Kubany, A. J . (1953). A validation study of the error-choice technique using attitudes on national
health insurance. Educational and Psychological Measurement, 13, 157- 163.
LaPiere, R. T. (1934). Attitudes versus action. Social Forces. 13, 230-237.
Lerner, M. J. (1980). The belief in a just world. New York: Plenum.
Lefford, A. (1946). The influence of emotional subject matter on logical reasoning. Journul of
General Psychology, 34, 121- I5 I .
Levine, J. M., & Murphy, G . (1943). The learning and retention of controversial statements. Journul
of Abnormal and Social Psychology, 38, 507-5 17.
 SOCIOCOGNITIVE MODEL OF ATTITUDES 28 I

Liben, L. S., & Signorella. M. L. (1980). Gender-related schemata and constructive memory in
children. Child Derteloptnetir, 51, 11-18.
Liebernian. S. (1956). The effects of changes in roles on the attitudes of role occupants. Human
Relufiotis. 9, 385-402.
Lippmann. W. (1922). Public Opinion. New York: Harcourt, Brace & Co.
Lord, C. G.. Lepper, M. R.. & Mackie, D. (1984). Attitude prototypes as determinants of attitude-
behavior consistency. Journal of Personulil?, mid Sociul Psychology, 46, 1254- 1266.
Lord. C. G . . Lepper. M. R . , & Ross. L. (1979). Biased assimilation and attitude polarization: The
effects of prior theories on subsequently considered evidence. Journul ufPersonaliry und So(iu1
Psvchology. 37, 2098-2 109.
Lott. A . 1.. & Lott. B. E. (1972). The power of liking: Consequences of interpersonal attitudes
derived from a liberalized view of secondary reinforcement. In L. Berkowitz (Ed.). Aduunces
in e.rperimentu/ social p.sychology (Vol. 6, pp. 109-148). New York: Academic Press.
Lott, A. I . . Lott, B. E., Reed. T.. & Crow. T. (1970). Personality-trait descriptions of differentially
liked persons. Journul of Personality and Sociul Psychology. 16, 284-290.
Lydon. J . . Zanna, M. P.. & Ross, M. (1988). Bolstering attitudes by autobiographical recall: Attitude
persistence and selective memory. Personalil?, and Social Psychology Bulletin. 14, 78-86.
Manis, M. (1961).The interpretation of opinion statements as a function of message ambiguity and
recipient attitudes. Journal of Abnormal und Social Psychology. 63, 76-8 I ,
Markus. G . B. ( 1985). Sfabilify und change in politic-ul attitudes: Observed, recalled. arid e.rplained.
Unpublished manuscript. University of Michigan.
McGuire. W. J. (1964). Inducing resistance to persuasion. In L. Berkowitz (Ed.). Aduunces in
experirnentul social psychology (Vol. I , pp. 191-229). New York: Academic Press.
McGuire, W . J . (1969). The nature of attitudes and attitude change. In G. Lindzey & E. Aronson
(Eds.), The handbook oj’sociul psychology (Vol. 3 , pp. 136314). Reading, MA: Addison-
Wesley.
McGuire, W . J. (1989). The structure of individual attitudes and of attitude systems. In A. R.
Pratkanis, S. J. Breckler. & A. G. Greenwald (Eds.). Artirude structure andfunction (pp. 37-
69). Hillsdale, NJ: Erlbauni.
Morgan. J. J. B. (1945). Attitudes of students toward the Japanese. Journal of Social Psychology,
21, 219-221.
Morgan, J . J . B., & Morton. J. T. (1943). Distorted reasoning as an index of public opinion. School
titid Society, 51, 333-335.
Morgan, J. J . B., & Morton. J. T. (1944). The distortion of syllogistic reasoning produced by
personal convictions. Journul of Sociul Psychology, 20, 39-59.
Mueller. D. J. ( 1986). Measuring sociul attifudes: A hundbook for researchers and pructitioners.
New York: Teachers College Press.
Newcomb, T. M., Turner, R. H.. & Converse, P. E. (1965). Sociul psychology: A study ofhurnati
interuction. London: Routledge & Kegan.
Norman, R . ( 1975). Affective-cognitive consistency, attitudes. conformity. and behavior. Journal
of Persotiulitv utid Sociul Psychology. 32, 83-9 I .
Olson, J. M.. & Zanna. M. P. (1979). A new look at selective exposure. Journal ofE.rl,c,ritnc,tifcil
SOCiUl PSVCh(J/ogy. 15, 1 - 1.5.
Osgood. C. E.. Suci. G. J.. & Tannenbaum, P. H. (1957). The meusurettient ofmeaning. Urbana,
IL: University of Illinois Press.
Ostrom, T. M. ( 1969). The relationship between the affective, behavioral. and cognitive components
of attitude. Journul of Experirnental Social Psychology, 5, 12-30,
Ostroni. T. M. (1989). Interdependence of attitude theory and measurement. In A. R. Pratkanis, S.
J . Breckler. & A . G . Greenwald (Eds.). Attitude structure andfuncfioti (pp. 11-36). Hillsdale.
NJ: Erlbaum.
Parrish, J. A. ( 1948). The direct and indirect assessnietit of uttitudes us influenced by proptrguiidized
radio trutiscripfs. Unpublished master’s thesis. The Ohio State University, Columbus, OH.
282 ANTHONY R. PRATKANIS AND ANTHONY G. GREENWALD

Payne, S. L. (1951). The art of asking questions. Princeton, NJ: Princeton University Press.
Petersen, K . K., & Dutton, J. E. (1975). Centrality, extremity, intensity: Neglected variables in
research on attitude-behavior consistency. Social Forces, 54, 393414.
Pettigrew, T. F. (1958). Personality and sociocultural factors and intergroup attitudes: A cross-
national comparison. Journal of Conflict Resolution. 2, 2 9 4 2 .
Pettigrew, T. F. (1959). Regional differences in anti-Negro prejudice. Journal of Abnormal and
Social Psychology, 59, 28-36.
Pettigrew, T. F. (1979). The ultimate attribution error: Extending Allport’s cognitive analysis of
prejudice. Personality and Social Psychology Bulletin, 5, 461-476.
Petty, R. E., Ostrom, T. M., & Brock, T. C. (1981). Cognitive responses in persuasion. Hillsdale,
NJ: Erlbaum.
Plott, C. R., & Levine, M. E. (1978). A model of agenda influence on committee decisions. American
Economic Review, 68, 146-160.
Postman, L., & Murphy, G. (1943). The factor of attitude in associative memory. Journal of Ex-
perimental Psychology. 33, 228-238.
Pratkanis, A. R. (1984). Attitudes and memory: The heuristic and schematic functions of attitudes.
Unpublished dissertation, Ohio State University.
Pratkanis, A. R. (1987). [Unipolar and bipolar attitudes in learning persuasive communications].
Unpublished raw data. Camegie-Mellon University, Pittsburgh, PA.
Pratkanis, A. R. (1988). The attitude heuristic and selective fact identification. British Journal of
Social Psychology 27, 257-263.
Pratkanis, A. R. (1989). The cognitive representation of attitudes. In A. R. Pratkanis, S. J. Breck-
ler, & A. G. Greenwald (Eds.), Attitude structure and function (pp. 71-98). Hillsdale, NJ:
Erl baum .
Pratkanis, A. R., & Greenwald, A. G. (1985). What is the self! Journal for the Theory of Social
Behavior, 15, 31 1-329.
Pratkanis, A. R., Syak, P., & Gamble, E. (1987). The role of technical knowledge in attitudinal
learning. Paper presented at the annual meeting in May of the Midwestern Psychological As-
sociation, Chicago, IL.
Pryor, J. B., Reeder, G. D., Vinacco, R., & Russo, T. K. (1987). The instrumental and symbolic
functions of attitudes towards AIDS victims. Unpublished manuscript, Illinois State University.
Raden, D. (1985). Strength-related attitude dimensions. Social Psychology Quarterly, 48, 3 12-330.
Ranum, O., & Ranum, P. (Eds.). 1972. Popular artitudes toward birth control in pre-industrial
France and England. New York: Harper and Row.
Read, S. J., & Rosson, M. B. (1982). Rewriting history: The biasing effects of attitudes on memory.
Social Cognition, 1, 385-397.
Regan, D. T.. Straus, E., & Fazio, R. (1974). Liking and the attribution process. Journal of Ex-
perimental Social Psychology. 10, 385-397.
Ries, A., & Trout, J. (1981). Positioning: The battle for your mind. New York: Warner.
Roberts, J. V. (1985). The attitude-memory relationship after 40 years: A meta-analysis of the
literature. Basic and Applied Social Psychology, 6, 221-241.
Robinson, J. R., & Shaver, P. R. (1973). Measures of social psychological attitudes. Ann Arbor,
MI: Institute for Social Research, University of Michigan.
Rogers, R. W. (1983). Cognitive and physiological processes in fear appeals and attitude change:
A revised theory of protection motivation. In J . T. Cacioppo & R. E. Petty (Eds.), Social
psychophysiology: A sourcebook (pp. 153-176). New York: Guilford Press.
Rogers, T. B., Kuiper, N. A,, & Kirker, W. S. (1977). Self-reference and the encoding of personal
information. Journal of Personality and Social Psychology, 35, 677-688.
Rosenbaum, M. E.( 1986). The repulsion hypothesis: On the nondevelopment of relationships. Journal
of Personality and Social Psychology. 51, 1156-1 166.
Rosenberg, M. J. (1956). Cognitive structure and attitudinal affect. Journal of Abnormal and Social
Ps~vchology.53, 367-372.
 SOCIOCOGNITIVE MODEL OF ATTITUDES 283

Rosenberg, M. J., & Hovland, C. 1. (1960). Cognitive, affective, and behavioral components of
attitudes. In C. 1. Hovland & M. J . Rosenberg (Eds.), Afritude organization und change (pp.
1-14). New Haven, CT: Yale University Press.
Ross, L., Greene, D., & House, P. (1977). The “false-consensus effect”: An egocentric bias in
social perception and attribution process. Journal of Experimenral Social Psychology. 13, 279-
301.
Ross, M., McFarland, C., Conway, M., & Zanna, M. P. (1983). Reciprocal relation between attitudes
and behavior recall: Committing people to newly formed attitudes. Journal of Personality and
Social Psychology, 45, 257-267.
Ross, M.. McFarland, C., & Fletcher, G . J. 0. (1981). The effect of attitude on the recall of personal
history. Journal of Personality and Social Psychology. 40, 627-634.
Ross, M., & Shulman, R. F. (1973). Increasing the salience of initial attitudes: Dissonance versus
self-perception theory. Journal of Personality and Social Psychology, 28, 138-144.
Saegert. J., & Young, E. A. (1983). Nutrition knowledge and health food consumption. Nutrition
and Behavior, 1, 103-1 13.
Salancik, G . R. (1982). Attitude-behavior consistencies as social logics. In M. P. Zanna, E. T.
Higgins, & C. P. Herman (Eds.), Consistency in social behavior (pp. 51-73). Hillsdale, NJ:
Erl baum.
Sample. J . , & Warland, R. (1973). Attitude and prediction of behavior. Social Forces, 51, 292-
304.
Schattschneider. E. E. (1960). The semi-sovereign people. New York: Holt, Rinehart & Winston.
Scheier. M. F., & Carver, C. S. (1983). Two sides of the self One for you and one for me. In J.
Suls & A . G . Greenwald (Eds), Psychologicalper.spectives on the self(Vol. 2 , pp. 123-157).
Hillsdale, NJ: Erlbaum.
Schlenker, B. R. ( 1980). lmpression managemenr. Monterey, CA: BrooksKole.
Schminke, M. (1986). Power and opportunity in the turnover decision. Unpublished doctoral dis-
sertation. Carnegie-Mellon University, Pittsburgh, PA.
Schuman. H.. & Johnson, M. P. (1976). Attitudes and behavior. In A. Inkeles, J. Coleman, & N.
Smelser (Eds.), Annuul review of sociology (Vol. 2, pp. 161-207). Palo Alto, CA: Annual
Reviews.
Scott, W. A. (1969). Attitude measurement. In G. Lindzey & E. Aronson (Eds.), The handbook of
sociul psychology (Vol. 2, pp 204-273). Reading, MA: Addison-Wesley.
Scott. W. A.. Osgood, D. W., & Peterson, C. (1979). Cognitive structure: Theory andmeasurement
of individuul di3erence.s. Washington, DC: V. H. Winston & Sons.
Sears, D. O., & Kinder, D. R. (1985). Whites’ opposition to busing: On conceptualizing and op-
erationalizing group conflict. Journal qf Personality and Social Psychology. 48, 1141-1 147.
Shaffer, D. R. (1975a). Some effects of consonant and dissonant attitudinal advocacy on initial
attitude saliency and attitude change. Journal of Personality and Social Psychology, 32, 160-
168.
Shaffer, D. R. (1975b). Another look at the phenomenological equivalence on pre- and post-ma-
nipulation attitudes in the forced compliance experiment. Personality and Social Psychology
Bulletin, 1, 497-500.
Shavitt, S. (1989). Operationalizing functional theories. In A . R. Pratkanis, S. J . Breckler, & A.
G. Greenwald (Eds.), Atritude strucrure and function (pp. 31 1-337). Hillsdale, NJ: Erlbaum.
Sherif, M., & Hovland, C. I. (1961). Socialjudgment. New Haven, CN: Yale University Press.
Sherman, S. J., Chassin, L., Presson, C. C., & Agostinelli, G. (1984). The role of evaluation and
similarity principles in the false consensus effect. Journal of Personality and Social Psychology,
47, 1244-1262.
Sherman, S. J . , & Corty, E. (1984). Cognitive heuristics. In R. S. Wyer & T. K. Srull (Eds.). The
handbook of social cognifion (Vol. I . pp. 189-286). Hillsdale, NJ: Erlbaum.
Sidanius, J . ( 1988). Political sophisticationand political deviance: A structural equation examination
of context theory. Journal of Personality and Social Psvchology. 55, 37-51,
284 ANTHONY R. PRATKANIS AND ANTHONY G. GREENWALD

Smith, D. D. (1968). Cognitive consistency and the perception of others' opinions. Public Opinion
Quarterly, 32, 1-15.
Smith, E. E., & Medin, D. L. (1981). Categories and concepts. Cambridge, MA: Harvard University
Press.
Smith, G. H. (1947). Beliefs in statement labelled fact and rumor. Journal ofAbnorma1 and Social
Psychology, 42, 8 6 9 0 .
Smith, M. B. (1980). Attitudes, values, and selfhood. In M. M. Page (Ed.), Nebraska symposium
on motivation 1979 (pp. 305-350). Lincoln, NB: University of Nebraska Press.
Smith, M. B., Bruner, J. S . . &White, R. W. (1956). Opinionsandpersonality. New York: Wiley.
Smith, S. S., & Jamieson. B. D. (1972). Effects of attitude and ego-involvement on the learning
and retention of controversial material. Journal of Personality and Social Psychology. 22, 303-
310.
Sniderman. P. M., & Tetlock, P. E. (1986). Symbolic racism: Problems of motive attribution in
political analysis. Journal of Social Issues, 42, 129-150.
Snyder, C. R., & Fromkin, H. L. (1980). Uniqueness: The human pursuit of diflerence. New York:
Plenum.
Snyder, M. (1974). The self-monitoring of expressive behavior. Journal of Personalify and Social
Psychology. 30, 526-537.
Snyder, M. (1982). When believing means doing: Creating links between attitudes and behavior.
In M. P. Zanna, E. T. Higgins, & C. P. Herman (Eds.). Consistency in socia[ behavior (pp.
105-130). Hillsdale, NJ: Erlbaum.
Snyder, M. (1987). Public appearanceslfrivate realities: The psychology of self-monitoring. New
York: W. H. Freeman and Co.
Snyder, M., & DeBono, K. G . (1989). Understanding the functions of attitudes: Lessons from
personality and social behavior. In A. R. Pratkanis, S . J. Breckler, & A. G. Greenwald (Eds.),
Attitude structure and function (pp. 339-359). Hillsdale, NJ: Erlbaum.
Spilich, G. J., Vesonder, G. T., Chiesi, H. L., & Voss, J . F. (1979). Text processing of domain-
related information for individuals with high and low domain knowledge. Journal of Verbal
Learning and Verbal Behavior, 18, 275-290.
Stotland, E., Katz, D., & Patchen, M. (1959). The reduction of prejudice through the arousal of
insight. Journal of Personality. 27, 507-53 1.
Swann, W. B. (1983). Self-verification: Bringing social reality into harmony with the self. In J.
Suls & A. G. Greenwald (Eds), Psychological perspectives on the self (Vol. 2, pp. 33-66).
Hillsdale, NJ: Erlbaum.
Sweeney, P. D., & Gruber, K. L. (1984). Selective exposure: Voter information preferences and
the Watergate affair. Journal of Personality and Social Psychology, 46, 1208-1221,
Symonds. P. M. (1927). What is an attitude? Psychological Bulletin, 24, 200-201.
Tajfel, H. (1981). Human groups and social categories. Cambridge: Cambridge University Press.
Taylor, S. E., & Crocker, J. (1981). Schematic bases of social information processing. In E. T.
Higgins, C. P. Herman, & M. P. Zanna (Eds), Social cognition (pp. 89-134). Hillsdale, NJ:
Erlbaum.
Tesser, A.. & Campbell, J. (1983). Self-definition and self-evaluation maintenance. In J. Suls &
A. G. Greenwald (Eds), Psychological perspectives on the self(Vo1. 2 , pp. 1-31). Hillsdale,
NJ: Erlbaum.
Tetlock, P. E. (1989). Structure and function in political belief systems. In A. R. Pratkanis, S. J.
Breckler, & A. G. Greenwald (Eds.), Artitude structure andfunction (pp. 129-151). Hillsdale,
NJ: Erlbaum.
Thistlethwaite. D. (1950). Attitude and structure as factors in the distortion of reasoning. Journal
of Abnormal and Social Psychology, 45, 442-458.
Thurstone, L. L. (1931). The measurement of social attitudes. Journal of Abnormal and Social
Ps~chology,26, 249-269.
 SOCIOCOGNlTIVE MODEL OF ATTITUDES 285

Tulving, E. (1983). Elements of episodic memory. New York: Oxford Press.

Turner. J. C. ( 1987). Rediscovering the sociul group: A self--ruregorizciriontheory. Oxford: Basil
Blackwell Ltd.
Tversky, A,. & Kahneman. D. (1974). Judgments under uncertainty: Heuristics and biases. Science,
185, 1124-1131.
Vallone, R. P.. Ross, L.. & Lepper, M. R. (1985). The hostile media phenomenon: Biased perception
and perceptions of media bias in coverage of the Beirut massacre. Journal of Personali/y and
Social Psychology. 49, 577-585.
Voss. J. F.. Vesonder, G . T., & Spilich, G . J. (980). Text generation and recall by high-knowledge
and low-knowledge individuals. Journul of Verbul Learning and Verbal Behavior. 19, 65 1-
661.
Vroom, V. H. (1960). The effects of attitudes on perception of organizational goals. Human Relutions.
13, 229-240.
Wallen. R. (1943). Individuals' estimates of group opinion. Journal of Social psycho log^. 17, 269-
274.
Waly. P., & Cook, S. W. (1965). Effects of attitude on judgment of plausibility. Journal ofPersonality
and Social Psychology, 2, 745-749.
Waly, P., & Cook, S. W. (1966). Attitude as a determinant of learning and memory: A failure to
confirm. Journal of Personality und Social Psychology, 4, 280-288.
Watson, W. S.. & Hartmann. G. W. (1939). The rigidity of a basic attitudinal frame. Journal qf
Abnormal and Social Psychology, 34, 3 1 4 3 3 5 .
Weber, R.. & Crocker. J. (1983). Cognitive processes in the revision of stereotypic beliefs. Journal
of Personality and Social Psychology, 45, 96 1-971.
Weldon, D. E., & Malpass, R. S. (1981). Effects of attitudinal, cognitive and situational variables
on recall of biased communications. Journal of Personality and Social Psychology. 40, 39-
52.
Weschler, I. R. (1950a). An investigation of attitudes toward labor and management by means of
the error-choice method: 1. Journal of Social Psychology, 32, 51-62.
Weschler, I . R. (1950b). A follow-up on the measurement of attitudes toward labor and management
by means of the error-choice method: 11. Journal of Social Psychology, 32, 63-69.
Wicker, A. W. (1969). Attitudes versus actions: The relationship of verbal and overt behavioral
responses to attitude objects. Journal of Social Issues. 25, 41-78.
Wicklund, R. A . , & Gollwitzer, P. M. (1982). Symbolic self-complerion. Hillsdale, NJ: Erlbaum.
Wood, W. (1982). Retrieval of attitude-relevant information from memory: Effects on susceptibility
to persuasion and on intrinsic motivation. Journal of Personality and Social Psychology. 42,
798-8 i a.
Wyer, R. S. (1974). Cognitive organizarion and change: An information processing approach. Hills-
dale, NJ: Erlbaum.
, K. (1984). Handbook of social cognirion. Hillsdale, NJ: Erlbaum.
Wyer. R. S., & S ~ l l T.
Zajonc, R. B., & Bumstein, E. (1965). The learning of balanced and unbalanced social structures.
Journul of Personality. 33, 153- 163.
Zanna, M. P., & Fazio, R. H. (1982). The attitude-behavior relation: Moving toward a third generation
of research. In M. P. Zanna, E. T. Higgins, & C. P. Herman (Eds.), Consisrency in social
behavior (pp. 283-301). Hillsdale, NJ: Erlbaum.
Zanna, M. P., & Rempel, J. K. (1988). Attitudes: A new look at an old concept. In D. Bar-tal &
A. W. Kruglanski (Eds.), The socialpsycho1og.v ofknowledge (pp. 315-334). New York:Cam-
bridge University Press.