COGNITIVE CONSISTENCY THEORIES

Cognitive consistency theories have their origins in the principles of Gestalt psychology, which suggests that people seek to perceive the environment in ways that are simple and coherent (Köhler 1929). Cognitive consistency theories have their beginnings in a number of seemingly unrelated research areas (Eagly and Chaiken 1993). Early consistency theorists drew upon theories of conflict (Lewin 1935; Miller 1944), memory (Miller 1956), and the intolerance for ambiguity by those with an authoritarian personality (Adorno, Frenkel-Brunswick, Levinson, and Stanford 1950). According to Newcomb (1968a), social scientists should not have been surprised at the rise of cognitive consistency theories. He points to a truism that in any field of scientific inquiry, there is an inevitable movement from description of the elements of the field, to understanding the relationships between them. At the heart of cognitive consistency theories is the assumption that people are motivated to seek coherent attitudes, thoughts, beliefs, values, behaviors, and feelings. If these are inconsistent, they will produce a "tension state" in the individual, and motivate the individual to reduce this tension. Individuals reduce this tension, according to consistency theories, by making their relevant cognitions consistent.

Cognitive consistency theories gained tremendous popularity in the social sciences in the 1950s, and generated hundreds of studies. Toward the end of the 1960s, however, research interest waned. In 1968, Abelson and collegues published a massive handbook, entitled Theories of Cognitive Consistency: A Sourcebook. The book was a thorough chronicle of cognitive consistency theories, and it addressed these theories from virtually any angle the reader could imagine. Ironically, the scholarly detail in which the editors and authors carefully described their research seemed to have been the death knell of cognitive consistency theories. Virtually no research on cognitive consistency theories took place during the 1970s.

With the end of the 1960s, theories of behavior that centered around motivational and affective forces (which certainly described cognitive consistency theories) were "out of vogue" with researchers. Many point to the simultaneous rise of social cognition approaches in general, and attribution theory in particular as helping to divert interest from cognitive consistency theories. Research on cognitive consistency theories was supplanted by more complex (some believed) "cold cognition" approaches that strictly dealt with how cognitive processes work together, not accounting for "hot" forces such as feelings and motivation.

Theories of cognitive consistency theory did not die, they just went away for a while. Abelson (1983) noted the reemergence of cognitive consistency theories in the early 1980s with the observation that authors were beginning to write about social cognition theories in light of the renewed interest in the nature of affect (e.g., Fiske 1982; Hamilton 1981). Toward the end of the 1980s, researchers began to take a closer look at the influence of affect on cognitive processes (e.g., Forgas 1990; Isen 1987; Schwarz 1990). This change was precipitated by the development of several theoretical perspectives concerning the nature and structure of emotion (e.g., Frijda 1988; Ortony, Clore, and Collins 1988). In an influential article, Zanna and Rempel (1988) argued that attitudes toward different attitude objects may be more or less determined by affective, rather than cognitive, sources. Consistency theories were not only back, they were thriving (Harary 1983).

This chapter will discuss five major theories of cognitive consistency that have had the most impact on the behavioral sciences. They are (in no particular order) balance theory (Heider 1946, 1958), strain toward symmetry (Newcomb 1953, 1968b), congruency theory (Osgood and Tannenbaum 1955), affective-cognitive consistency model (Rosenberg 1956), and finally, what Eagly and Chaiken (1993) refer to as the "jewel in the consistency family crown," (p. 456) cognitive dissonance theory (Festinger 1957).

BALANCE THEORY

The earliest consistency theory is Heider's balance theory (1946, 1958). This approach is concerned with an individual's perceptions of the relationships between himself (p) and (typically) two other elements in a triadic structure. In Heider's formulation, the other elements are often another person (o) and another object (e.g., an issue, object, a value). The attitudes in the structure are designated as either positive or negative. The goal of assessing the structure of a triad is to ascertain whether the relationships (attitudes) between the actors and the other elements are balanced, or consistent. According to Heider (1958), a balanced triad occurs when all the relationships are positive, or two are negative and one is positive (i.e., two people have a negative attitude toward an issue, but they like each other), and the elements in the triad fit together with no stress. Imbalance occurs when these outcomes are not achieved (i.e., all three relationships are negative, or you have a negative attitude toward an issue that your friend favors). Heider assumed that people prefer balanced states to imbalanced ones, because imbalance results in tension and feelings of unpleasantness. Balance, according to Heider, is rewarding.

Interestingly, imbalanced states can also be rewarding and exciting. Heider said that sometimes balance can be "boring" and that "The tension produced by unbalanced situations often has a pleasing effect on our thinking and aesthetic feelings" (1958, p. 180). In other words, imbalance stimulates us to think further, to solve the problem, to imagine, and to understand the mystery of the imbalance. According to balance theory, there are three ways to restore balance to an imbalanced triad: (1) one may change one's attitude toward either the object or the other person, in order to restore balance; (2) one might distort reality to perceive that the relationships are balanced (e.g., your friend doesn't really favor something you dislike, she really dislikes it); and (3) one might cognitively differentiate the relationship one has with a friend, so that the friend's opposing attitude toward something one favors is separated from one's positive attitude toward the friend as a person (e.g., you might compartmentalize a friend's opposite political views apart from your attitude toward her, in order to maintain your friendship and maintain balance, in most other areas where she is concerned) (Eagly and Chaiken 1993).

A limitation of Heider's balance theory is that it did not account for the strength of attitudes between persons and objects in the triad. It merely categorized the relationships as either positive or negative, and it therefore assumed that tension that is produced by imbalance was objectively of the same strength and effect on the individuals in the triad. Because some attitudes are held with more conviction and are more meaningful and important to us, it stands to reason that triads that involve imbalance with such strongly-held attitudes ought to evoke more tension (Eagly and Chaiken 1993). Another shortcoming of the theory is that it only deals with relationships between three entities. To address this latter concern, Cartwright and Harary (1956) published a paper that nicely generalized Heider's theory to account for structures of any size.

STRAIN TOWARD SYMMETRY MODEL

Newcomb (1953, 1968b) suggested that there are three, rather than two types of balance relationships in a triad. First, a structure that does not motivate modification (or acceptance) is termed a "nonbalanced" structure. These situations are characterized by indifference. Here, disagreement with another individual about an issue or object does not arouse tension if that other individual is devalued (or otherwise not important). However, when the other person is valued (i.e., a friend or significant other), then agreement with him or her about an object results in a "positively balanced" structure, while disagreement results in a "positively imbalanced" structure. The term "positive" in the latter two types of structure denotes the valence of the relationship between p and o, who is a valued other. The important focus in Newcomb's approach is the relationship of o to p, and p's view of o as a valued person, and "suitability as a source of information, or support, or of influence concerning the object" (Newcomb 1968b, p. 50). Newcomb's experiments supported his idea that the tension that is aroused when p and o have strong attitudes in the structure is much greater than when their attitudes are held with little conviction. He also found that positively balanced situations are the most preferred structures, followed by nonbalanced structures, with the positively imbalanced situations being the least preferred.

CONGRUENCY THEORY

A particular advantage of Osgood and Tannenbaum's (1955; Tannenbaum 1968) congruency theory is its precision in assessing: 1) the strength of the relationships between p and o, 2) the strength of the motivation to change an incongruent triad, and 3) the degree of attitude change that is necessary to balance a triad. Another advantage of this theory is that, like Newcomb's approach, it takes into account the strength of the attitudes of p and o in evaluating the degree of incongruity in the structure. Osgood and Tannenbaum discuss the Heider triad in terms of p, another individual, termed the source (s) and s's attitude (termed an "assertion") toward another object or concept (x). According to the theory, attitudes can be quantified along a seven-unit evaluative scale, from extremely negative (−3) to neutral (0) to extremely positive (+3).

When p's attitude toward s and x are positive, and s's assertion is equally strong and of the same valence, there is a "congruous" structure to the triad. There is no motivation to change one's attitude toward the object or toward the source. When p's attitude toward s is positive, and p has an equally positive attitude toward x that s later negatively evaluates, an incongruous structure is established. In this situation, p is motivated to change his or her attitude toward s, or x, or both, in the direction of congruity. Consider the following example. If p's attitude toward s is a +2, and p's attitude toward x is a −2, the structure would be congruent if s's assessment is a −2. If, however, the assessment is a +2, the structure is imbalanced. In this case, p's attitude toward either x or s needs to change four units to make the triad congruent. Of course, if the relationships are weaker, the degree of attitude change to make the triad congruent is that much less (by the exact amount denoted in the quantitative calculation of all the relations of p, s, and x). Osgood and Tannenbaum also argued that strongly held attitudes would be less likely to be modified in incongruent triads. This was supported in subsequent research (Tannenbaum 1968).

AFFECTIVE-COGNITIVE CONSISTENCY MODEL

This approach suggests that people seek consistency in order to satisfy a general motivation toward simplicity in cognition, and/or to adhere to norms, traditions, customs, or values that reinforce consistency in one's cognitions and behavior (Rosenberg 1956, 1968). Another interesting twist on the consistency approach is that in the affective-cognitive consistency model, Rosenberg (1956, 1968) proposed that people are more motivated to maintain cognitive consistency so that other people perceive that they are consistent. In other words, while the individual may occasionally feel some tension as a result of inconsistency, other people find the inconsistency more aversive, because it represents a conflict for those around the individual. Specifically, if o has a positive attitude toward p, but p dislikes x, which o likes, o is caught between being friendly with, and avoiding, p. In this model, o feels tension at this conflict, and must reduce the tension by changing attitudes toward p (e.g., increasing attraction toward p, which would thereby outweigh any conflict with p's negative attitude toward x) or toward x (e.g., o devalues x, so that p's dislike of x does not result in o feeling conflicted).

Rosenberg's model also considers the relationship between the individual, his or her values, and an attitude object. For example, consider that p also has various other important values, denoted as y1, y2, y3, etc. Rosenberg suggests that the p-x-y triad is just as important in understanding cognitive consistency as the traditional p-o-x triad. In the affective-cognitive consistency approach, we must consider p's attitudes toward each of his or her values, how p feels about x, and p's perception of the relationship between x and each of the values. When all or most of the p-x-y triads are consistent, the individual has achieved cognitive consistency. When most or all of the p-x-y triads are inconsistent, the individual experiences cognitive inconsistency.

The reason Rosenberg's approach is called the affective-cognitive consistency model is that it proposes that inconsistency results when one's feelings are inconsistent with one's beliefs. That is, when the way we think and feel about an object or person are at odds, we will modify one or both to make the attitude consistent. Thus, this model attempts to address consistency within one's own attitudes toward other people and objects, but also consistency in how one's value system relates to other people and objects. The model is also unique in suggesting that other people experience more tension as a result of one's own inconsistency. Generally speaking, the model has been supported by experiments (Rosenberg 1964), and is considered a very useful addition to the family of cognitive consistency theories.

COGNITIVE DISSONANCE THEORY

Of all the cognitive consistency theories, none has had more influence on researchers and subsequent theories than cognitive dissonance theory (Festinger 1957). A conservative estimate suggests that at least 1,000 articles have been published in which researchers present data bearing upon the theory and their own revisions of the theory (Cooper and Fazio 1984). Many agree with Jones's (1976) assessment that cognitive dissonance theory is "the most important development in social psychology to date" (p. x). Along the way, the theory has been hailed for its elegant simplicity, and its powerful range of utility (Collins 1992). It has also been criticized for its lack of specificity (Lord 1992; Schlenker 1992).

In formal terms, Festinger's theory states that two elements (behaviors or thoughts, or both) "…are in a dissonant relation if, considering these two alone, the obverse of one element would follow from the other" (1957, p. 13). Dissonance, then, refers to a negative arousal brought about by one's inconsistent thoughts or actions, or both. Essentially, this translates into the following assumptions. If one has opposing thoughts or behaviors, or both, this brings about an aversive state of tension, akin to a drive state like hunger or thirst. This tension motivates the individual to seek relief by eliminating the tension. The tension can be dissipated by changing: 1) either a thought or attitude to make it consonant with the opposing thought or behavior, or 2) one's behavior, to make it consonant with the opposing behavior or thought. Because it is often much easier to change one's thoughts rather than one's behaviors, these are typically the elements that get modified by the person in dissonance reduction.

As an example, Festinger (1957) talked about the dissonance experienced by most smokers at some point in their lives. Smokers engage in behavior (smoking) that is harmful to their health. This is at odds with our desire to avoid harming ourselves. This arouses tension in the individual. The smoker could reduce it by changing his or her behavior (quit smoking) or changing the way he or she thinks about the smoking behavior. As mentioned above, changing behavior is often more difficult than changing cognitions, and, as most smokers will affirm, quitting smoking is certainly no exception to this axiom. In this instance, Festinger suggests, smokers eliminate their dissonance by changing their thoughts about smoking. They may: 1) disbelieve the validity of the health consequences of smoking, or distort the information about smoking by thinking that smoking is only harmful if you smoke so many packs a day, or if you inhale cigar smoke, etc., or more fatalistically, 2) convince themselves that "we all die of something, and I might as well die doing something I enjoy." All of these changes in thoughts eliminate the dissonance for the smoker.

It should be noted that Festinger was not talking about logical inconsistencies. There are certainly conditions under which people think and do logically inconsistent things, yet feel no dissonance, or they feel dissonance, yet are not in a situation where a logical inconsistency is present. Festinger recognized what has become a truism in psychology, that a person's reaction to a stimulus is not a function of the objective properties of the stimulus itself, but rather the individual's construal, or perception of, that stimulus. This explains why the presence or absence of logical inconsistencies may or may not be accompanied by dissonance in an individual. The most important and reliable way to predict a person's behavior in a dissonance situation is to understand how he or she construes the potential dissonance arousing thoughts or behaviors, or both.

History. Cognitive dissonance theory came onto the scene in the 1950s when reinforcement theories of behavior were very dominant in virtually all areas of inquiry in psychology. According to reinforcement principles, behavior that is followed by a reward is more likely to be repeated. Behavior that is followed by a strong reward should be more likely to be learned and repeated than behavior followed by a weak (or no) reward. Reinforcement theory was such a simple yet very powerful principle that it seemed to explain virtually all behavior in any context. For that reason, it was extremely popular among behavioral scientists. An experiment by Festinger and Carlsmith (1959) showed that reinforcement theory was not the all-purpose theory it appeared to be. In their experiment, Festinger and Carlsmith (1959) had participants do boring tasks (i.e., turning pegs one-quarter turn on a cribbage board) for an hour. Participants randomly assigned to a control group were then given a short questionnaire in which they were asked to rate how much they enjoyed the task. In other conditions, the experimenter then told participants that his research assistant had not yet arrived for a different version of the experiment, and he asked the participant if he would do the research assistant's job of telling the next participant (in the hall, who was in reality a confederate) that he enjoyed the experiment tasks. This was, of course, a lie, because the tasks were boring. These participants were assigned to one of two conditions. Some were given $1 to tell the lie, and others were given $20 to tell the lie. After participants had told the lie and were leaving, the experimenter ran up to the participant, explaining that he forgot to have the participant complete the ratings of the attitudes toward the experiment tasks.

The experiment pitted reinforcement theory against the predictions made from dissonance theory. Reinforcement theory suggests that the participants who were given $20 should find the tasks more rewarding (pleasant), and should have a more positive attitude toward the tasks (and the experiment) than those only given a weak (or no) reward. Dissonance theory suggests that those in the control condition would feel no dissonance because they did a boring task, and would rate the tasks as such on the questionnaire. However, counter to intuition (and reinforcement theory) those in the $20 condition should feel little (or no) dissonance, because although they did boring tasks, and disliked the tasks, saying that the tasks were fun is not an inconsistent behavior if one has adequate justification ($20) for doing so. They could attribute their lying to the incentive, and they would not feel hypocritical. The $1 participants experienced significant dissonance because they did boring tasks, but yet they said they thought the tasks were fun. As Festinger and Carlsmith predicted, the $1 was an insufficient justification for the lie, so the dissonance remained unless the participants changed their attitudes toward the task, and convinced themselves (as shown in their ratings of the tasks) that maybe the tasks were not boring, and in fact, they rather enjoyed them! The results were precisely as predicted, and this paved the way for a flurry of research that tested the exciting, often dramatic, and counterintuitive predictions that arose from cognitive dissonance theory.

Alternate Versions of Dissonance Theory. Very soon after the publication of Festinger's theory, research revealed that the theory might need to be revised somewhat, to account for more of the data that were being published, which didn't quite fit with the theory. In one notable revision, one of Festinger's protegés, Aronson (1969) posited that the theory would be strengthened if it stated that dissonance would be most clearly aroused when the self-concept of the person is engaged. In other words, dissonance is stronger and more clearly evoked when the way we think about ourselves is at odds with our cognitions or behavior. This modification was supported by much subsequent research (Aronson 1980). Less an alternate version and more of a theoretical competitor, Bem's (1967) self-perception theory was the first major theory that offered a plausible account of the dissonance data, and pointed to different causal mechanisms. Unlike cognitive dissonance theory, Bem's approach did not invoke reference to hypothetical motivational processes, but rather tried to account for the person's behavior in terms of the stimuli present in the individual's environment and his or her related behavior. Bem's theory proposed that attitudinal change in dissonance experiments happens not due to an aversive tension (or other motivation), but due to a person's perceptions of his or her own behavior. Specifically, Bem said that people infer their attitudes from their actions, in much the same way that observers of our behavior infer the nature of our attitudes from our behavior. Attitude change occurs when their most recent behavior is different from their previous attitudes.

This behavioral approach to dissonance phenomena recasts the Festinger and Carlsmith experiment in a very different light. In a replication of the Festinger and Carlsmith study, Bem asked participants to listen to a tape describing a person named Bob, who did some boring motor tasks. Control condition participants then were asked to assess Bob's attitude toward the tasks. Other subjects then learned that Bob was given $1 or $20 to say to the next participant that the motor tasks were fun. Participants then listened to a recording of Bob enthusiastically telling a subsequent woman participant how enjoyable the motor tasks were. Participants were then asked to evaluate Bob's attitude toward the motor tasks. Those who were told that Bob was given $20 to tell the lie inferred that the only reason he told the lie was because he was paid a lot of money. They assumed that he didn't really have a positive attitude toward the motor tasks. Those who were told that Bob received $1 didn't think Bob had a good reason for lying, so his behavior (lying) told participants that Bob must really feel positively about the motor tasks. Control condition participants inferred that Bob negatively evaluated the motor tasks. As can be seen, these results are virtually identical to those obtained in the Festinger and Carlsmith experiment. Thus, according to self-perception, the Festinger and Carlsmith participants inferred their attitudes toward the boring tasks based on their recent behavior.

Subsequent research on self-perception theory was aimed at testing the self-perception theory contention that no arousal exists as a result of the dissonance situation. Zanna and Cooper (1976) found that arousal did indeed accompany counterattitudinal advocacy, so it was apparent that self-perception did not apply to all dissonance situations. Fazio, Zanna, and Cooper (1977) suggested that dissonance accounted for attitude change when behavior is truly counterattitudinal, but that self-perception can account for situations where behavior is only mildly counterattitudinal. For most researchers, this seems to have settled the debate about the situations to which each theory may be applied (Abelson 1983).

A final major revision was proposed by Cooper and Fazio (1984). They suggested that dissonance does not result from mere cognitive inconsistency, but only is evoked when the person feels personally responsible for causing an aversive event. The theory suggests, then, that aversive consequences are necessary for dissonance to occur. In subsequent experiments, however, Aronson and his colleagues (Aronson, Fried, and Stone 1991) induced participants to make an educational video advocating the use of condoms for safe sex, to be shown in high schools. Then, the participant was subsequently reminded of situations in which he or she had not used condoms in the past. According to Cooper and Fazio, there should be no dissonance because the participant had not produced an aversive event, but rather, a positive one (advocating safe sex in an educational video). According to Aronson, however, the participant should feel dissonance because he or she was saying one thing and doing another (acting like a hypocrite). Aronson predicted that if dissonance was aroused in those hypocrites, they should be more likely to (if given the opportunity) take more free pamphlets and condoms at the end of the study, as a way of regaining consistency with their advocated position (i.e. they were making an effort to change their behavior to be consistent with their advocated message). Results showed that this is precisely what occurred. Subsequent research has demonstrated strong support for Aronson and colleagues' (1991) contention that the production of aversive consequences is not necessary to create dissonance (Harmon-Jones, Brehm, Greenberg, Simon, and Nelson 1996).

CURRENT STATUS OF COGNITIVE CONSISTENCY THEORIES

In the 1990s, cognitive consistency theories experienced a rebirth, primarily through renewed interest in cognitive dissonance theory (Aronson 1992). With the renewed interest in motivation, and the interaction of cognition and affect, researchers are once again taking up the questions (and there are many) left unanswered by earlier cognitive dissonance researchers. For example, while research has shown that dissonance evokes (as Festinger theorized) psychological discomfort (in comparison to physiological arousal), researchers know little about what occurs between the onset of the psychological discomfort and the start of the discomfort reduction process (Elliot and Devine 1994). Some have called for an effort to more fully explicate the conditions under which dissonance occurs, and when it does not occur (Aronson 1992), and researchers are beginning to do just that (Shultz and Lepper 1996). Cognitive consistency theories have been a cornerstone of psychology for over four decades, and while they receded into the background in the 1970s, they are experiencing a strong resurgence of empirical and theoretical interest.

references

Abelson, R. P. 1983 "Whatever Became of Consistency Theory?" Personality and Social Psychology Bulletin 9:37–54.

——, E. Aronson, W. J. McGuire, T. M. Newcomb, M. J. Rosenberg, and P. H. Tannenbaum (eds.) 1968 Cognitive Consistency Theories: A Sourcebook. Chicago, Ill.: Rand McNally.

Adorno, T. W., E. Frenkel-Brunswick, D. J. Levinson, and R. N. Sanford 1950 The Authoritarian Personality. New York: Harper and Row.

Aronson, E. 1969 "The Theory of Cognitive Dissonance: A Current Perspective." In L.Berkowitz, ed., Advances in Experimental Social Psychology, Vol. 4. New York: Academic Press.

——1980 "Persuasion via Self-Justification: Large Commitments for Small Rewards." In L. Festinger, ed., Retrospections on Social Psychology. New York: Oxford University Press.

——1992 "The Return of the Repressed: Dissonance Theory Makes a Comeback."Psychological Inquiry 3 (4):303–311.

——, C. Fried, and J. Stone 1991 "AIDS Prevention and Dissonance: A New Twist on an Old Theory." American Journal of Public Health 81:1636–1638.

Bem, D. J. 1967 "Self-perception: An Alternative Interpretation of Cognitive Dissonance Phenomena." Psy-chological Review 74 (3):183–200.

Cartwright, D., and F. Harary 1956 "Structural Balance: A Generalization of Heider's Theory." Psychological Review 63:277–293.

Collins, B. 1992 "Texts and Subtexts." Psychological Inquiry 3 (4):315–320.

Cooper, J., and R. H. Fazio 1984 "A New Look at Dissonance Theory." In L. Berkowitz, ed., Advances in Experimental Social Psychology, Vol. 17. New York: Academic Press.

Eagly, A. H., and S. Chaiken 1993 The Psychology of Attitudes. New York: Harcourt, Brace, Jovanovich.

Elliot, A. J., and P. G. Devine 1994 "On the Motivational Nature of Cognitive Dissonance: Dissonance as Psychological Discomfort." Journal of Personality and Social Psychology 67(3):382–394.

Fazio, R. H., M. P. Zanna, and J. Cooper 1977 "Dissonance and Self-Perception: An Integrative View of Each Theory's Proper Domain of Application." Journal of Experimental Social Psychology 13:464–479.

Festinger, L. 1957 A Theory of Cognitive Dissonance. Palo Alto, Calif.: Stanford University Press.

——, and M. Carlsmith 1959 "Cognitive Consequences of Forced Compliance." Journal of Abnormal and Social Psychology 58:203–210.

Fiske, S. T. 1982 "Schema-Triggered Affect: Applications to Social Perception." In M. S. Clark and S. T. Fiske, eds., Affect and Cognition. Hillsdale, N.J.: Erlbaum.

Forgas, J. P. 1990 "Affect and Social Judgments: An Introductory Review." In J. P. Forgas, ed., Emotion and Social Judgments. Oxford: Pergamon Press.

Frijda, N. H. 1988 "The Laws of Emotion." American Psychologist 43:349–358.

Hamilton, D. L. 1981 "Illusory Correlation as a Basis for Stereotyping." In D. L. Hamilton, ed., Cognitive Processes in Stereotyping and Intergroup Behavior. Hillsdale, N.J.: Erlbaum.

Harary, F. 1983 "Consistency Theory is Alive and Well." Personality and Social Psychology Bulletin 9:60–64.

Harman-Jones, E., J. Brehm, J. Greenberg, L. Simon, and D. E. Nelson 1996 "Evidence that the Production of Dissonance is Not Necessary to Create Cognitive Dissonance." Journal of Personality and Social Psychology 70:5–16.

Heider, F. 1946 "Attitudes and Cognitive Organizations." Journal of Psychology 21:107–112.

——1958 The Psychology of Interpersonal Relations. Hillsdale, N.J.: Erlbaum.

Isen, A. M. 1987 "Positive Affect, Cognitive Processes, and Social Behavior." In L. Berkowitz, ed., Advances in Experimental Social Psychology, Vol. 20. New York: Academic Press.

Köhler, W. 1929 Gestalt Psychology. New York: Liveright. Lewin, K. 1935 A Dynamic Theory of Personality. New York: McGraw-Hill.

Lord, C. G. 1992 "Was Cognitive Dissonance Theory a Mistake?" Psychological Inquiry 3 (4):339–342.

Miller, G. A. 1956 "The Magical Number Seven Plus or Minus Two: Some Limits on Our Capacity for Processing Information." Psychological Review 63:81–97.

Miller, N. E. 1944 "Experimental Studies of Conflict." In J. Hunt, ed., Personality and the Behavioral Disorders, Vol. 1. New York: Ronald Press.

Newcomb, T. M. 1953 "An Approach to the Study of Communicative Acts." Psychological Review 60:393–404.

——1968a "Introduction." In R. P. Abelson, E. Aronson, W. J. McGuire, T. M. Newcomb, M. J. Rosenberg, and P. H. Tannenbaum, eds., Cognitive Consistency Theories: A Sourcebook. Chicago, Ill.: Rand McNally.

——1968b "Interpersonal Balance." In R. P. Abelson, E. Aronson, W. J.McGuire, T. M. Newcomb, M. J. Rosenberg, and P. H. Tannenbaum, eds., Cognitive Consistency Theories: A Sourcebook. Chicago, Ill.: Rand McNally.

Ortony, A., G. L. Clore, and A. Collins 1988 The Cognitive Structure of Emotions. New York: Cambridge University Press.

Osgood, C. E., and P. H. Tannenbaum 1955 "The Principle of Congruity in the Prediction of Attitude Change." Psychological Review 62:42–55.

Rosenberg, M. J. 1956 "Cognitive Structure and Attitudinal Affect." Journal of Abnormal and Social Psychology 53:367–372.

——1964 "A Structural Theory of Attitude Dynamics." In E. E. Sampson, ed., Approaches, Contexts, and Problems of Social Psychology. Englewood Cliffs, N.J.: Prentice-Hall.

——1968 "Hedonism, Inauthenticity, and Other Goads Toward Expansion of a Consistency Theory." In R. P. Abelson, E. Aronson, W. J. McGuire, T. M. Newcomb, M. J. Rosenberg, and P. H. Tannenbaum, eds., Cognitive Consistency Theories: A Sourcebook. Chicago, Ill.: Rand McNally.

Schlenker, B. R. 1992 "Of Shape Shifters and Theories." Psychological Inquiry 3 (4):342–344.

Schultz, T. R., and M. R. Lepper 1996 "Cognitive Dissonance Reduction as Constraint Satisfaction."Psychological Review 103 (2):219–240.

Schwarz, N. 1990 "Feelings as Information: Informational and Motivational Functions of Affective States." In R. Sorrentino and E. T. Higgins, eds., Handbook of Motivation and Cognition, Vol. 2. New York: Guilford.

Tannenbaum, P. H. 1968 "The Congruity Principle: Retrospective Reflections and Recent Research." In R. P. Abelson, E. Aronson, W. J. McGuire, T. M. Newcomb, M. J. Rosenberg, and P. H. Tannenbaum, eds., Cognitive Consistency Theories: A Sourcebook.Chicago, Ill.: Rand McNally.

Zanna, M. P., and J. Cooper 1976 "Dissonance and the Attribution Process." In J. H. Harvey, W. J. Ickes, and R. F. Kidd, eds., New Directions in Attribution Research. Hillsdale, N.J.: Erlbaum.

Zanna, M.P., and J. K. Rempel, 1988 "Attitudes: A New Look at an Old Concept." In D. Bar-Tal and A. Kruglanski, eds., The Social Psychology of Knowledge. Cambridge, Eng.: Cambridge University Press.

Todd D. Nelson