Cognitive Dissonance: Reexamining a Pivotal Theory in Psychology - Eddie Harmon-Jones 2019
Understanding the Motivation Underlying Dissonance Effects
Perspectives Employing the Original Version of the Theory
The Action-Based Model
Eddie Harmon-Jones and Cindy Harmon-Jones
Starting in the 1960s, researchers began to challenge the original theory of cognitive dissonance and proposed that cognitive discrepancy (as defined by the original version of the theory; see Chapter 1, this volume) was not the cause of the cognitive and behavioral changes that were observed in experiments testing dissonance theory. Several revisions to the original theory emerged (Aronson, 1969; Cooper & Fazio, 1984; Cooper & Worchel, 1970; Steele, 1988). For the most part, these revisions aimed to address one of the most fundamental and important questions for dissonance theory and research—it concerns the underlying motivational force driving dissonance effects. In the present chapter, we provide a brief overview of the original version of the theory of cognitive dissonance (for more complete descriptions, see Festinger, 1957; Chapter 1, this volume) and one of these revisions that attracted much attention, the aversive consequences model (Cooper & Fazio, 1984). We then present our action-based model of dissonance, which proposes why dissonance and dissonance reduction occur. We then review evidence obtained in a variety of experimental paradigms that support predictions derived from this model. We conclude by suggesting that this extension of the original theory assists in understanding the function of dissonance processes.
THE ORIGINAL VERSION
The original statement of cognitive dissonance theory (Festinger, 1957) proposed that discrepancy between cognitions creates a negative affective state that motivates individuals to attempt to reduce or eliminate the discrepancy between cognitions (see Chapter 1, this volume, for a more complete description of this process). Several paradigms have been used to test predictions derived from dissonance theory. In each of these paradigms, the availability of the cognitions that serve to make the entire set of relevant cognitions more or less discrepant is manipulated. The most commonly used of these is the induced-compliance paradigm, in which participants are induced to act contrary to an attitude, and if they are provided few consonant cognitions (few reasons or little justification) for doing so, they are hypothesized to experience dissonance and reduce it, usually by changing their attitude to be more consistent with their behavior. In one of the first induced-compliance experiments, Festinger and Carlsmith (1959) paid participants either $1 (low justification) or $20 (high justification) to tell a fellow participant (confederate) that dull and boring tasks were very interesting and to remain on call to do it again in the future. After participants told this to the confederate, they were asked how interesting and enjoyable the tasks were. As predicted, participants given little justification for performing the counterattitudinal behavior rated the tasks as more interesting than did participants given much justification. Festinger and Carlsmith posited that participants provided low justification (just enough justification to say the counterattitudinal statement) experienced dissonance and changed their attitudes because of the inconsistency between their original attitude (they believed that the task was boring) and their behavior (they had said that the task was interesting). Participants provided with high justification, on the other hand, experienced little dissonance, because receiving $20 to perform the behavior justified the behavior or was consonant with the behavior.
In later research (Brehm & Cohen, 1962), the degree of dissonance was manipulated by means of perceived choice rather than by the magnitude of an incentive to engage in the counterattitudinal behavior. Having low choice (i.e., being forced) to behave counterattitudinally can be considered a cognition consonant with the counterattitudinal behavior; in contrast, the high choice induction would lack this consonant cognition (or at least have less of it). Experiments found that participants who were given high choice, as opposed to low choice, to write counterattitudinal essays changed their attitudes to be more consistent with their behavior.
REVISIONS TO THE ORIGINAL THEORY
Within the decades after the publication of the original theory of dissonance (Festinger, 1957) and its early experiments (e.g., Festinger & Carlsmith, 1959), researchers offered alternative theoretical and experimental accounts. One revision that received the most empirical attention was based on the idea that individuals needed to feel personally responsible for producing aversive consequences in order to experience dissonance and dissonance-related attitude change. We briefly review this revision below, and then review evidence that has challenged this revision. Afterward, we briefly review two other revisions and evidence that challenged those revisions.
The Aversive-Consequences Revision
The aversive-consequences revision suggested that low-justification participants in the Festinger and Carlsmith (1959) experiment changed their attitudes not because of cognitive discrepancy, but because their actions brought about an aversive event (convincing another person to expect boring tasks to be interesting). In one of the first experiments testing this explanation, Cooper and Worchel (1970) replicated and extended the Festinger and Carlsmith study. Cooper and Worchel found that low-justification participants changed their attitudes to be consistent with their behavior when the confederate believed their statement, but not when the confederate did not believe their statement.
Using a slightly different procedure, other research has suggested that, when the counterattitudinal actions do not cause aversive consequences, attitude change does not occur (e.g., Collins & Hoyt, 1972; Goethals & Cooper, 1975; Hoyt, Henley, & Collins, 1972; Scher & Cooper, 1989). According to the aversive-consequences revision, a sufficient cognitive discrepancy is neither necessary nor sufficient to cause dissonance and discrepancy reduction. Instead, feeling personally responsible for the production of foreseeable aversive consequences is necessary and sufficient. Aversive consequences are events that one would not want to occur (Cooper & Fazio, 1984).
Alternative Explanations for the Evidence Produced by the Aversive-Consequences Revision
The aversive-consequences revision was supported by evidence obtained in the induced-compliance paradigm. More specifically, the support for the aversive-consequences revision comes from the absence of measurable attitude change in the conditions in which aversive consequences were not produced. There are several explanations for the absence of this attitude-change effect in the no-aversive-consequences conditions, and these alternative explanations must prevent us from concluding that cognitive discrepancy is not necessary or sufficient to create dissonance. First, this is a null effect. It is difficult to draw clear inferences from null effects, as they may be produced by a variety of factors (e.g., failure to manipulate the constructs of interest, insufficient power to detect the effects). Had these past theorists and researchers drawn the conclusion that feeling personally responsible for producing an aversive outcome intensifies dissonance, we would be in complete agreement, for it is likely that feeling personally responsible for such will intensify dissonance and dissonance-produced attitude change. Feeling personally responsible for an aversive consequence is an important discrepant cognition, and importance increases the magnitude of dissonance. However, these past theorists and researchers instead proposed that feelings of personal responsibility for aversive outcomes were necessary to produce dissonance effects.
Two sets of alternative explanations can be offered for the lack of attitude change in the no-aversive-consequences conditions. The first set of alternative explanations argues that the level of dissonance was not large enough to generate dissonance sufficient to produce attitude change and that the addition of the production of aversive consequences was necessary to produce dissonance sufficient to cause attitude change. Several of the past induced-compliance experiments that included a no-aversive-consequences and an aversive-consequences condition used attitudinal issues that were not extremely negative or positive, that is, control-condition participants reported moderately negative or positive attitudes (e.g., Calder, Ross, & Insko, 1973; Nel, Helmreich, & Aronson, 1969). Moreover, the lack of extremity might have reflected ambivalence or a mix of positive and negative attitudes toward the issues. Because the attitudes used in past experiments were not extremely positive or negative and might have been held with ambivalence, they were likely not to arouse much dissonance when behavior counter to them occurred. In essence, the magnitude of dissonance aroused may have been too small to generate attitude change.
In the past experiments, the researchers often encouraged participants to generate lengthy counterattitudinal statements. This may increase the likelihood of finding no attitude change in the no-aversive-consequences conditions. Research has shown that the length of the counterattitudinal statement relates inversely with the amount of attitude change that occurs (e.g., Rabbie, Brehm, & Cohen, 1959; see also Chapter 3, this volume), that is, longer essays are likely to produce less attitude change. This inverse relationship between essay length and attitude change may occur because participants may provide their own justifications and hence more cognitions consonant with the behavior in these lengthy essays. As the number of consonant cognitions increases, the magnitude of dissonance decreases.
In addition, because of the salience of the audiences in these experiments, the participants’ attention may have been focused more on the audience and whether they were convinced or could affect a disliked policy than on the nature of their own counterattitudinal actions or their own attitudes. As a result of this, the magnitude of dissonance may have been determined in large part by what the audience did or would do as a result of the counterattitudinal advocacy. Thus, the unconvinced audience, in contrast to the convinced audience, may have reduced the importance of the dissonant cognitions, to the point of making the counterattitudinal action seem trivial. If the perceived importance of dissonant cognitions is low, dissonance may not reach a magnitude that requires reduction.
Another possible explanation is that participants in these past experiments may have been provided too much justification (too many consonant cognitions) for producing the counterattitudinal statement, and the production of aversive consequences may have been necessary to elicit enough dissonance to produce measurable attitude change. This explanation seems very reasonable when one considers the high compliance rates observed in most if not all of this past research.1 Typically, 100% of the participants have complied with the experimenter’s request to write the counterattitudinal statement. As Festinger (1957) explained, for attitude change to result from dissonance, the person should be offered “just enough reward or punishment to elicit the overt compliance” (p. 95, italics in original). Thus, the past experiments on the necessity of aversive consequences may have had inducing forces (the friendliness of the experimenter, the benefits to science) that were so great that little or no dissonance was produced, and the addition of feeling personally responsible for producing aversive consequences may have been necessary to produce sufficient dissonance to cause measurable discrepancy reduction (e.g., attitude change).
Another set of alternative explanations for the lack of attitude change in the no-aversive-consequences conditions argues that dissonance may have been aroused in participants in the no-aversive-consequences conditions of the past experiments but was not detected. The sole method of detecting dissonance in the experiments testing the aversive-consequences model against the original version of the theory has been assessment of attitude change. Because no assessments of dissonance were obtained in experiments testing the aversive consequences model, it is impossible to know whether dissonance was aroused in the no-aversive-consequences conditions. The only conclusion that can safely be drawn is that measurable attitude change did not occur. On the other hand, attitude change may have occurred in the no-aversive-consequences conditions but may have been small, and it would not have been detected if one had only 10 to 12 persons per condition, as was done in much of the past research (e.g., Calder et al., 1973; Cooper & Worchel, 1970). Research has found that effects in personality and social psychological research are typically small to medium (Fraley & Marks, 2007; Richard, Bond, & Stokes-Zoota, 2003). According to power analysis using G*Power (Faul, Erdfelder, Buchner, & Lang, 2009), 139 participants per condition would be required to detect an effect of small-to-medium size (Cohen’s d = 0.3), suggesting that these studies were seriously underpowered. In addition, the dissonance may have been reduced in a route other than attitude change. Persons whose counterattitudinal actions had no undesired effects may have reduced dissonance by reducing the importance (Simon, Greenberg, & Brehm, 1995) or the perceived effectiveness (Scheier & Carver, 1980) of the counterattitudinal behavior.
It is unlikely that one of these possible alternative explanations accounts for all of the nonsignificant effects that have been found in the past no-aversive-consequences conditions. However, given the number of plausible alternative explanations for the null effects produced in the past experiments that had been used to support the aversive-consequences revision, we thought it was premature to abandon the original version of the theory.
Induced-Compliance Experimental Results Inconsistent With the Aversive-Consequences Revision
All of the research on the aversive-consequences revision has been conducted using the induced-compliance paradigm, which is the focal paradigm in which predictions derived from dissonance theory and its revisions have been tested. We have conducted several induced-compliance experiments to test the hypothesis that feeling personally responsible for producing aversive consequences is not necessary to produce dissonance and that cognitive discrepancy is sufficient to produce dissonance even in the induced-compliance paradigm. In conducting these experiments, we created a situation in which participants would write counterattitudinal statements but not produce aversive consequences. We designed the experiments so that conditions present in previous induced-compliance experiments that might have prevented attitude change from occurring were not present. We took special care to ensure that the inducing force was “just barely sufficient to induce the person” to behave counterattitudinally (Festinger & Carlsmith, 1959, p. 204), to reduce the number of consonant cognitions to a bare minimum, so that the dissonance aroused after the action would be at high levels. In addition, we had participants write short counterattitudinal statements about objects toward which they held attitudes that were highly salient, strongly negative (or positive), simple, and not ambivalent. Also, in some of the experiments, we assessed negative affect and arousal, to provide measures of dissonance arousal as well as dissonance reduction.
In each experiment, under the guise of an experiment on recall, participants were exposed to a stimulus, were given low or high choice to write a counterattitudinal statement about that stimulus, threw away the statement they wrote, and then completed questionnaires that assessed their attitudes toward the stimulus. We assured participants that their counterattitudinal statements and their responses to the questionnaires would be made in private and would be anonymous. We did so to create a situation in which the counterattitudinal behavior would not lead to aversive consequences, because, as Cooper and Fazio (1984) argued, “making a statement contrary to one’s attitude while in solitude does not have the potential for bringing about an aversive event” (p. 232). We predicted that participants provided high choice for engaging in the counterattitudinal behavior would change their attitudes to be more consistent with their behavior, whereas participants provided low choice would not.
Dissonance and an Unpleasant Beverage. In the first experiment testing this idea (E. Harmon-Jones, Brehm, Greenberg, Simon, & Nelson, 1996, Experiment 1), participants were told that the research concerned factors that influenced the recall of characteristics of products, and that this study would test how writing a sentence evaluating a product would affect recall of the characteristics of the product. Participants would drink one of a variety of beverages, write a sentence about it, and then recall characteristics of it. The experimenter explained that participants should not let him know what type of drink they received. He also explained that all of their responses would be anonymous and that he would not see their responses to questionnaires but that an assistant would enter them into a computer.
The experimenter gave the participant a lid-covered cup containing fruit-punch flavored Kool-Aid mixed either with the amount of sugar suggested (a pleasant-tasting drink), or with white vinegar and no sugar (an unpleasant-tasting drink). Because the experimenter was unaware of whether participants were given a pleasant- or an unpleasant-tasting drink, he was unaware of whether participants experienced dissonance.
After the participant drank some of the beverage, the experimenter returned to the participant’s cubicle and induced the choice manipulation. He told participants in the low-choice condition that they were randomly assigned to write a statement saying they liked the beverage. He told participants in the high-choice condition that they could write a statement saying they liked or disliked the beverage and that it was their choice. The experimenter explained that he needed some more persons to write that they liked the beverage, and he asked the participant if she or he would write that she or he liked the beverage. Once the experimenter gained compliance from the participant, he reminded her or him that it was her or his choice.
The experimenter then asked both low-choice and high-choice participants to write one sentence saying they liked the beverage. He also told participants that he did not “need the sheet of paper you will write your sentence on; we just need for you to go through the process of writing the sentence. So when you are done, just wad it up and throw it in the wastebasket.” He did this to ensure that the participants perceived that they had anonymity and that there would be no consequences to their behavior. The experimenter then left the participants alone to write the sentence.
After the participant discarded the sentence, the experimenter gave the participant an envelope and said that previous research had indicated that the characteristics a person recalls about a product may be affected by whether they liked the product and that to take this into account, he needed them to answer a questionnaire that assessed their thoughts about the drink. The questionnaire assessed how much the drink was liked. The experimenter left the participant alone to answer this questionnaire. After the participant finished with this questionnaire, the experimenter had the participant complete a questionnaire that assessed the effectiveness of the manipulation of choice. After assessing suspicion and debriefing, the experimenter collected the participants’ statements from the trash can, to assess whether participants complied with the request to write the counterattitudinal statement.
Approximately 15% of the participants did not write counterattitudinal statements. This effect suggests that we had designed a situation in which there was just enough but not too much external justification to write the counterattitudinal statement. Results indicated that participants in the unpleasant-tasting drink/high-choice condition reported more positive attitudes toward the drink than did participants in the unpleasant-tasting drink/low-choice condition. This effect was significant when both compliers and noncompliers were included in the analysis, as well as when only compliers were included. Thus, the results of the first experiment suggested that dissonance can be created in induced-compliance situations void of aversive consequences.
Dissonance, Boring Passages, and Electrodermal Activity. Subsequent experiments were designed to conceptually replicate the effects of the first experiment using a different manipulation of choice, a different attitudinal object, and measures of dissonance arousal and affect. Using similar procedures as used in the first experiment, we had participants read a boring passage and gave them low or high choice, by means of written instructions, to write a statement saying that the passage was interesting. Because choice was induced by means of written instructions, the experimenter was unaware of when dissonance was expected. Results from this experiment replicated those of the first experiment, with high-choice participants, as compared with low-choice participants, rating the boring passage as more interesting (E. Harmon-Jones et al., 1996, Experiment 2).
In another experiment (E. Harmon-Jones et al., 1996, Experiment 3), we measured nonspecific skin conductance responses (NS-SCRs) that occurred in the 3 minutes after the writing of the counterattitudinal statement but before the assessment of attitude. Previous research has indicated that increased NS-SCRs are associated with increased sympathetic nervous system activity, which is increased during emotional arousal. If our experimental procedure evoked dissonance, we would observe increased NS-SCRs. Results indicated that participants given high choice to write the counterattitudinal statement evidenced more NS-SCRs and reported that the passage was more interesting than did participants given low choice to write the statement.
Dissonance and a Hershey’s Kiss. The previous results demonstrate that the production of aversive consequences is not necessary to arouse dissonance and dissonance-related attitude change. However, they may be subject to an alternative explanation: Perhaps the manipulation of choice to write the statement influenced individuals’ reconstructive construal of the situation, so that high-choice participants felt as though they had high choice to engage with the unpleasant stimulus, whereas low-choice participants felt as though they had low choice to engage with the unpleasant stimulus. If this were so, then the aversive-consequences revision could explain these results as being due to feeling personally responsible for inflicting a negative event on oneself (choosing to drink the unpleasant Kool-Aid or read the boring passage). To eliminate this alternative explanation, another experiment was conducted (E. Harmon-Jones, 2000a, Experiment 1). In this experiment, instead of writing a counterattitudinal statement about a negative stimulus, participants wrote a counterattitudinal statement about a positive stimulus.
Under the same cover story as in the previous experiments, participants ate a pleasant-tasting Hershey’s Kiss and then were given high or low choice to write that they did not enjoy it. Consistent with predictions, high-choice participants reported that they disliked the Hershey’s Kiss more than did low-choice participants.
In addition, in this experiment, state self-reported affect was measured immediately after the writing of the counterattitudinal statement or after the attitude-change opportunity. Time between counterattitudinal action and attitude assessment was controlled by having participants complete an affect questionnaire or a filler questionnaire, comparable in length. From the state affect measure, four indexes of affect were derived. Discomfort was measured with the scale developed by Elliot and Devine (1994). State social self-esteem and appearance self-esteem were measured with the subscales of the State Self-Esteem Scale (Heatherton & Polivy, 1991). Positive affect was measured with the items happy, proud, and enthusiastic.
Results indicated that participants who were given high choice and who completed the affect questionnaire before the attitude measure reported significantly more discomfort than did participants in the other conditions. Positive affect, state social self-esteem, and state appearance self-esteem did not differ significantly among conditions. Another experiment reported in this article conceptually replicated these results (E. Harmon-Jones, 2000a, Experiment 2). These results suggest that the cognitive discrepancy evoked in this situation increased discomfort. The results also suggest that the cognitive discrepancy evoked in this situation was more likely to increase discomfort than to decrease state self-esteem or positive affect.2
Summary. The experiments presented thus far were all conducted using the induced-compliance paradigm. The results from these experiments support the original theory of dissonance and are inconsistent with the aversive-consequences revision. These experiments are important because they show that dissonance arousal, dissonance affect, and dissonance-produced attitude change can occur in situations in which a sufficient cognitive discrepancy is present but feeling personally responsible for the production of aversive consequences is not present. The present evidence convincingly demonstrates that dissonance effects can be generated by a cognitive discrepancy that does not produce aversive consequences. Indeed, these results suggest that the original version of the theory was abandoned prematurely (see Chapter 3, this volume).
Aversive Consequences and Attitude Change
Earlier we argued that producing aversive consequences might intensify dissonance. Why would this occur? The production of aversive consequences may intensify dissonance because aversive consequences are a cognition dissonant with one’s preexisting attitude. If the attitude were the generative cognition, the cognition about the counterattitudinal behavior and the cognition of producing aversive consequences would be dissonant cognitions. Thus, the magnitude of dissonance aroused would be greater in psychological situations where counterattitudinal behavior and aversive consequences were produced than in situations where only counterattitudinal behavior was produced, because there are more dissonant cognitions in the former than in the latter situation. However, if the counterattitudinal behavior were the generative cognition (Beauvois & Joule, 1996), then the cognition of producing aversive consequences would be a consonant cognition (it follows from the behavior), and thus it would decrease the magnitude of dissonance aroused. Beauvois and Joule (1996) reported results consistent with this latter interpretation, but the past research on the aversive-consequences model is consistent with the former. This inconsistency between these two sets of results can be resolved by positing that in the induced-compliance paradigm, both the attitude and the behavior can serve as generative cognitions, and there may be a potential dissonance associated with each. That is, there is a potential dissonance associated with the attitude and a potential dissonance associated with the behavior. In general, the greater dissonance would be the one reduced. The generative cognition associated with that greater dissonance would not be altered, whereas the generative cognition associated with the lesser dissonance would be altered. However, the reduction of dissonance depends also on the availability of discrepancy-reduction routes. Hence, when the potential dissonances are not very different in magnitude, the discrepancy may be reduced by means of the most available route, which has been attitude change in most previous dissonance experiments.
An additional possibility is that the production of aversive consequences may increase the dissonance, because it increases the commitment to the behavior, making the behavior more resistant to change and attitude change more likely to result. In addition, when behavior produces important consequences, it will be regarded as a more important cognition and thus has the potential to create more dissonance. Although these explanations are more elegant in their simplicity, they do not fit with the large body evidence presented by Beauvois and Joule (1996), whereas the explanation offered in the previous paragraph does.
The Belief-Disconfirmation Paradigm
Research on the aversive-consequences revision has focused exclusively on the induced-compliance paradigm. However, other paradigms have been used to test predictions derived from dissonance theory, and evidence obtained in these paradigms is difficult to explain with the aversive-consequences revision (see also Berkowitz & Devine, 1989).
One such paradigm is the belief-disconfirmation paradigm. This paradigm is based on Festinger, Riecken, and Schachter’s (1956) observations of belief intensification among members of a group whose belief that a flood would destroy the continent was disconfirmed. This evidence suggests that the cognitive discrepancy that occurs when an important and highly resistant to change belief is disconfirmed produces dissonance, leading to the use of dissonance-reducing strategies such as belief intensification. Results obtained in this paradigm are not subject to an aversive-consequences alternative explanation, because individuals involuntarily exposed to belief-discrepant information have not produced an aversive consequence and thus cannot feel responsible for having done so.
Past Evidence From the Belief-Disconfirmation Paradigm Inconsistent With the Aversion Consequences Revision
In an experiment by Brock and Balloun (1967), committed churchgoers were confronted with audiotaped information that did or did not support their religious values. When the information was inconsistent with their religious values, individuals were less likely to press a button to eliminate white noise from the communication and thus make it easier to comprehend. Other research has replicated these findings (e.g., Schwarz, Frey, & Kumpf, 1980), further suggesting that dissonance effects occur even when inconsistencies are produced by outside information, not from actions that produce aversive consequences.
In a quasi-experiment by Batson (1975), girls attending a church youth program were asked to declare publicly whether they believed in the divinity of Jesus. After completing a measure of Christian orthodoxy, the girls were then presented with belief-disconfirming information (i.e., information that indicated that Jesus was not the son of God). Orthodoxy was once again assessed. As expected, those who believed in the divinity of Jesus and accepted the truthfulness of the disconfirming information intensified their belief in Jesus’ divinity, whereas those who were not believers or who believed but did not accept the truthfulness of the disconfirming information did not.
It is difficult to explain the results obtained in the belief-disconfirmation paradigm as resulting from the motivation to avoid feeling personally responsible for producing aversive consequences. That is, the person exposed to belief-inconsistent information has not acted in a manner to produce aversive consequences and thus cannot feel responsible for having done so. In this paradigm, persons are exposed to information from an external source; they have not done anything for which to feel responsible. Note that Cooper and Fazio (1989), two of the main proponents of the aversive-consequences revision, stated that according to the aversive-consequences revision, evidence obtained in the belief-disconfirmation paradigm is not the result of dissonance. Cooper and Fazio stated that exposure to belief-discrepant information “will not necessarily create an unwanted consequence and will not necessarily arouse dissonance” (p. 525). In our view, dissonance does occur in response to belief-discrepant information, and the aversive-consequences revision has unfortunately excessively narrowed the range of application of dissonance theory. One way to demonstrate that evidence obtained in the belief-disconfirmation paradigm is the result of dissonance processes is to show that the negative affect that motivates the cognitive effects occurs as a result of belief disconfirmation and is reduced after reconciliation of the cognitive discrepancy.
Recent Evidence From the Belief-Disconfirmation Paradigm Inconsistent With the Aversion Consequences Revision
Two belief-disconfirmation experiments have done exactly that: they assessed whether dissonance-related emotive pressures drove the cognitive effects (Burris, Harmon-Jones, & Tarpley, 1997). The design was based on Allport’s (1950) idea that “the suffering of innocent persons is for most people the hardest of all facts to integrate into religious sentiment” (p. 81). In the experiments, Christian participants were exposed to a newspaper article that highlighted the discrepancy between belief in a loving, protecting, just, and omnipotent God and knowledge of the gratuitous suffering humans often experience. The newspaper article reported the drive-by shooting death of an infant boy in his grandmother’s arms, as she and the child’s father prayed for protection because a similar incident had occurred two nights earlier. The article concluded with a quote from the infant’s grandfather that expressed his continued faith in God. The cognitive discrepancy between the participants’ religious beliefs (God is a good God who protects the innocent and answers prayers) and this tragic outcome (the infant dies during a prayer for protection) was highlighted by having participants read, “Some people would think that the grandfather’s continued belief and trust in a good God is naive and misguided.”
Belief Disconfirmation and Transcendence Experiment. The first experiment tested the hypothesis, offered by Abelson (1959), that “the theosophical dilemma of God’s presumed permissiveness toward evil is sometimes resolved by appeal to transcendent concepts” (p. 346). If participants exposed to this belief-discrepant story were allowed to engage in transcendence (i.e., allowed to reconcile dissonant cognitions under a superordinate principle), they would experience less negative affect. Moreover, the more they engaged in transcendence, the less their negative affect would be.
Participants were randomly assigned to one of two conditions. In the transcendence-opportunity condition, they read the newspaper article and then completed a questionnaire that allowed them to engage in transcendence. These participants were given the explicit opportunity to reconcile the cognitive discrepancy after the dissonance had been aroused, when they were most in need of reducing the dissonance brought about by exposure to the belief-inconsistent information. They then completed a measure of self-reported state negative affect. In the no-transcendent-opportunity condition, participants completed the transcendence measure, read the tragic newspaper article, and then completed the measure of negative affect. These participants were thus not given an explicit opportunity to engage in transcendence after the dissonance had been aroused, when they were most in need of reducing the dissonance. The transcendence measure included questions such as, “How much does God intervene in persons’ lives?” and “How often do things happen to persons because of God’s greater purpose?” The measure of negative affect included items to measure discomfort (uncomfortable, uneasy, bothered; Elliot & Devine, 1994) and agitation (angry, frustrated, distressed, and threatened).
Based on dissonance theory, the endorsement of higher levels of transcendence subsequent to reading the newspaper article should relate to lower levels of dissonance-related affect. In contrast, higher levels of transcendence before the reading should not relate to lower levels of dissonance-related affect. Consistent with predictions, results indicated that higher endorsement of transcendence predicted decreased agitation in the transcendence-opportunity condition, whereas it did not in the no-opportunity condition. A similar pattern emerged for discomfort. Moreover, individuals in the transcendence-opportunity condition engaged in more transcendence than did individuals in the no-transcendence-opportunity condition, who completed the transcendence measure prior to reading the belief disconfirming story.
As expected, more extreme endorsement of transcendent beliefs after exposure to a belief-discrepant article was associated with reduced dissonance-related affect. In contrast, belief transcendence before exposure to the belief-discrepant article did not relate to reduced dissonance-related affect. This evidence strongly suggests that exposure to belief-discrepant information arouses dissonance that motivates persons to engage in discrepancy reduction, which then reduces the dissonance. Because persons in the belief-disconfirmation paradigm do not produce aversive consequences for which to feel responsible, this evidence suggests that the production of aversive consequences is not necessary to create dissonance.
Belief Disconfirmation and Belief Affirmation Experiment. In the previous experiment, time and transcendence opportunity were confounded, making it difficult to infer what caused the observed effects. That is, those who completed the transcendence scale after reading the article and before responding to the affect measures may have been distracted by this intervening task compared with those who completed the transcendence scale before the article. This explanation does not seem plausible, given that there was no main effect of transcendence-opportunity condition on dissonance-related affect but only an interaction with level of transcendence endorsement. However, to eliminate this explanation, a second study was conducted to conceptually replicate the first. In addition, instead of assessing transcendence in response to belief disconfirmation, belief affirmation was manipulated, much like Batson (1975) and Festinger et al. (1956) did. This was done to test the hypothesis that religious individuals would “rigidly maintain or even intensify” (Batson, 1975, p. 178) their beliefs when faced with disconfirming evidence. Religiously interested participants completed religious belief measures either after (belief-affirmation condition) or before (no-affirmation condition) reading the belief-discrepant article or completed comparable-length nonreligious belief measures after the article (distraction condition). All then completed self-report affect measures. Results revealed that, as predicted, agitation was lower in the religious-affirmation condition than in either the no-affirmation or the distraction condition; agitation levels in the latter two conditions did not differ.
Past results from the belief-disconfirmation paradigm cannot be interpreted in terms of the aversive-consequences revision. This past research, however, suffers from an important limitation because no measures of dissonance-related affect were obtained either during the experience of dissonance or after discrepancy reduction, rendering it difficult to know whether the effects generated in this belief-disconfirmation paradigm were caused by the mechanisms proposed by dissonance theory. However, the research by Burris et al. (1997) demonstrates that the effects produced in the belief-disconfirmation paradigm are due to dissonance processes.
Other Experimental Results Inconsistent With the Aversive-Consequences Revision
Other experiments provide evidence of dissonance in situations void of the production of aversive consequences. For instance, Aronson and Carlsmith (1962) found that individuals with experimentally-created expectancies for failure reacted with dissonance to behaving successfully. In the experiment, after individuals had repeatedly failed at a task, they were given feedback indicating that they were succeeding at the task. Then the individuals were given an opportunity to change their responses. Results indicated that these individuals changed their responses from being correct to being incorrect. Because failure is regarded as negative and success as positive, the behavior of these individuals would be difficult to interpret in the aversive-consequences formulation. Other research has demonstrated that dissonance can occur even when participants engage in proattitudinal behavior that has positive consequences (see Chapter 6, this volume). Finally, additional experimental evidence indicates that dissonance can occur in the absence of the production of aversive consequences (e.g., Beauvois & Joule, 1996; see also Chapters 3 and 6, this volume).
Summary of Evidence Concerning Aversive-Consequences Revision
Results from experiments using several methodologies suggest that a cognitive discrepancy in the absence of feeling personally responsible for the production of aversive consequences can cause increased dissonance-related negative affect and discrepancy reduction. The results of these experiments suggest that a cognitive discrepancy is enough to generate dissonance and discrepancy reduction. Feeling personally responsible for producing aversive consequences is not necessary to generate dissonance and discrepancy reduction, but it may enhance the magnitude of dissonance effects because producing aversive consequences is an important cognition that is dissonant with one’s preexisting attitude or because the aversive consequences increase the commitment to behavior.
A cognitive discrepancy, however, will produce an aversive state, in keeping with Festinger’s speculations and the results presented in this chapter. This aversive state is not equivalent to the aversive consequences that dissonance theory revisionists discussed. Two other prominent revisions of dissonance theory posited that dissonance responses were responses to self-threats (see Chapters 7 and 8, this volume). Other research and theoretical arguments have suggested that these revisions do not fully explain research generated by dissonance theory, and therefore, the original theory, with its focus on cognitive discrepancy, is still viable (E. Harmon-Jones, 2000b, 2000c, 2001, 2002; E. Harmon-Jones, Amodio, & Harmon-Jones, 2009; see also Chapters 3 and 5, this volume). However, the original theory never clearly specified why cognitive discrepancy creates a negative affective state and motivates discrepancy reduction. Below, we outline a conceptual model and some research that addresses these issues.
THE MOTIVATION UNDERLYING DISSONANCE EFFECTS—AN ACTION-BASED MODEL
The reviewed research cogently demonstrates that the motivation to avoid the production of aversive consequences is not the motivation underlying dissonance effects. The studies reviewed above affirmed Festinger’s (1957) original version of the theory as being a more inclusive and elegant explanation for the evidence. However, the original version left several questions unanswered.
According to the original version of the theory, a sufficient cognitive discrepancy is the source of the motivation underlying dissonance and its effects. But why would a cognitive discrepancy evoke such a motivation? What function does the capacity to experience negative affect in response to a sufficient cognitive discrepancy and then be motivated to reduce it have for the organism? Is this set of psychological mechanisms adaptively beneficial? That is, is the dissonance mechanism functional or beneficial for the organism?
The action-based model was designed to address these questions. It extends dissonance theory by providing an explanation of why cognitive inconsistency arouses negative affect and how and why this negative affect motivates the cognitive and behavior adjustments. The model begins with the assumption that cognitions (broadly defined) can serve as action tendencies. The idea that cognition for action is seen not only in the 1890 writings of William James (1890/1950), but also in ecological approaches to perception (Gibson, 1979; McArthur & Baron, 1983) and in the study of attitudes (Bain, 1868; Spencer, 1865). In this sense, the cognitions that are of primary concern for this theoretical model are those that provide useful information, and usefulness of information is defined by its relevance to actions and goals. When information inconsistent with cognitions that guide action is encountered, negative emotion (dissonance) is aroused because the dissonant information has the potential to interfere with effective and unconflicted action. For the present model, effective behavior can occur in the absence of consciousness; in other words, effective behavior can be produced automatically. Thus, the present model does not propose that cognitive consistency is necessary for effective behavior. It only proposes that cognitive inconsistency interferes with effective behavior.
Thus, cognitive discrepancy may create negative affect because discrepancy among cognitions undermines the requirement for effective and unconflicted action (Beckmann & Irle, 1985; Jones & Gerard, 1967). Research on the theory of dissonance has identified commitment as an important, if not necessary, condition for the arousal of dissonance (Beauvois & Joule, 1996; Brehm & Cohen, 1962; Festinger, 1964). For most dissonance theorists, the notion of commitment implies that the person has engaged in a behavior for which he or she feels responsibility and that he or she has a definite understanding of the consequences of the behavior. However, persons can regard cognitions that may not involve an immediate behavioral commitment as true or certain (Mills, 1968) and would experience dissonance if information were presented that was inconsistent with these cognitions. A good example of this type of cognition is a person’s belief in the law of gravity. Information that violates the law of gravity would probably arouse dissonance in most persons because it guides behavior in a general sense, regardless of whether the individual has made a specific, recent commitment to this knowledge. Therefore, we propose that a commitment occurs when a person regards a behavior, belief, attitude, or value as a meaningful truth. Defining commitment in this way allows for viewing commitment as a continuous variable. When commitment is defined as overt behavior, as with previous dissonance theorists, commitment is reduced to a categorical variable, which may not be an accurate representation of psychological states involving more and less commitment. The degree of psychological commitment to the cognition guides information processing, which serves the ultimate function of producing and guiding behavior.
If dissonant information is encountered, negative emotion may result and cause the person to engage in cognitive work to support the commitment. However, if dissonant information continues to mount, the negative emotion that results may motivate the person to disengage from the commitment and accept the dissonant information. Whether the person’s cognitive work is aimed toward supporting the commitment or discontinuing the commitment is determined by the resistance to change of each cognition. If the commitment is more resistant to change than the dissonant information, then cognitive work would be aimed at supporting the commitment. If, however, the dissonant information is more resistant to change than the commitment, then the cognitive work would be aimed at discontinuing the commitment. Resistance to change of cognitions is determined by the responsiveness of the cognitions to reality (e.g., the grass is green), the extent to which the cognitions are in relations of consonance to other cognitions, the difficulty of changing the cognition, and so on. From the present view, resistance to change is ultimately determined by the degree to which individuals believe the information assists them in controlling and predicting outcomes and thus behaving effectively. When knowledge about the environment, about oneself, or about one’s actions, beliefs, or attitudes is in a dissonant relation, the sense of being able to control and predict outcomes may be threatened, and ultimately, the need to act effectively would be undermined.
From the current perspective, the proximal motivation to reduce cognitive discrepancy stems from the need to reduce negative emotion, whereas the distal motivation to reduce discrepancy stems from the requirement for effective action. When the maintenance of true and certain knowledge, and thus the potential for effective action, is threatened by information that is sufficiently discrepant from the psychological commitment, negative emotion results, which prompts attempts at the restoration of cognitions supportive of the commitment (i.e., discrepancy reduction). Thus, negative emotion works much like pain in that it provides the information and motivation that prompts the person to engage in cognitive action aimed at resolving the discrepancy. These speculations about the adaptive function of dissonance processes suggest interesting avenues of research.
Many of the situations that evoke dissonance can be conceptualized as situations that involve a commitment to a specific action (e.g., after a difficult decision or chosen counterattitudinal behavior). The commitment to a specific action engages psychological processes that essentially prime one to act and thus motivate the organism to convert their commitment or intention into an effective action. This motivational state should be approach oriented and is revealed in changes in attitude that are consistent with and ultimately support the commitment.
Most dissonance research puts individuals in situations that involve a behavioral commitment. The action-based model proposes that making a commitment to a behavior primes the individual to translate that intention into action. Thus, a commitment should motivate individuals to engage in cognitive work to bring their cognitions into alignment, to engage in dissonance reduction. This should be an approach-motivated process that functions to facilitate effective action because it relieves the competing action tendencies of conflicting cognitions.
Imagine a person who makes the difficult decision of temporarily leaving the workforce to obtain an advanced degree. The potential benefits of the degree, such as greater potential for earnings and advancement, are consonant cognitions. The drawbacks, such as temporarily reduced income and the effort that will need to be expended, are dissonant cognitions. According to the action-based model, the more the individual can increase consonant cognitions and/or their importance, while decreasing dissonant cognitions and/or their importance, the greater the likelihood of success in seeking the degree. Conversely, if the individual continued to regret the temporary loss of income and increased workload, the action tendencies of these cognitions may interfere with taking the steps necessary to complete the degree. By increasing consonant cognitions and decreasing dissonant cognitions, the individual not only reduces the negative emotion of dissonance but also is better able to carry out a difficult decision.
Testing the Action-Based Model
This action-based model of dissonance has been tested in a variety of ways. We review these tests here.
Action Orientation and Spreading of Alternatives
The above example illustrates that, following a behavioral commitment, persons should be in a state of “getting things done” (Kuhl, 1984). This approach-motivated state is consistent with what other theorists have called an action-oriented or implemental mindset (Gollwitzer & Bayer, 1999). This state enhances goal accomplishment and the implementation of decisions (Gollwitzer & Sheeran, 2006).
The above reasoning suggests that an action-oriented state should make cognitive discrepancy reduction more efficient. To test this prediction, participants made a difficult decision (between performing two experiments they had rated as similar in attractiveness). They were also induced to experience an action-oriented state, by thinking of an important life goal and the steps they would need to take to accomplish it, compared with a positive, non-action-oriented state or neutral state. In support of the action-based model, participants in the action-oriented condition engaged in more spreading of alternatives than those in the other conditions (E. Harmon-Jones & Harmon-Jones, 2002).
Embodied Manipulation of Approach Motivation and Discrepancy Reduction
The action-based model proposes that dissonance reduction is an approach-motivated process. From this, it follows that reducing approach motivation should also inhibit discrepancy reduction. Approach motivation is an embodied process and may be influenced by changes in body posture. For example, the supine position (lying on one’s back) leads to less approach motivation toward appetitive stimuli such as photographs of desirable individuals or delicious desserts (E. Harmon-Jones, Gable, & Price, 2011; Price, Dieckman, & Harmon-Jones, 2012). Thus, it would be expected that the supine position should reduce discrepancy reduction.
This hypothesis was supported by studies using both effort justification and the difficult decision paradigms. In one experiment, participants exerted unpleasant effort in order to obtain a goal. Those who did this while in a supine position rated the goal less positively compared to those who sat upright. That is, they engaged in less effort justification. In another experiment, participants made a difficult decision, and those who assumed a supine position engaged in less spreading of alternatives than those who sat upright (E. Harmon-Jones, Price, & Harmon-Jones, 2015). These studies suggest that dissonance reduction is an approach-motivated process.
Evidence That Dissonance Reduction Reduces Behavioral Conflict
Evidence that dissonance reduction assists in behavior comes from a study showing that meat-eaters are motivated to deny the mental capabilities of animals that are raised as food. Participants were assigned to taste either fruit or meat, and presented with a sample of that food. They were then instructed to write about the process of bringing meat from the farm to market, to include the slaughter of the animal. They rated the mental capacity of the animal they had written about. Participants in the meat-eating condition, who expected to eat that animal, rated the animal as having less mental capabilities. Additionally, greater denial of mind to the animal was related to less negative affect, suggesting that individuals who successfully reduced dissonance felt better about their intended behavior of eating meat (Bastian, Loughnan, Haslam, & Radke, 2012). In line with the action-based model, these results suggest that dissonance reduction lessens resistance toward behaviors for which individuals feel ambivalent.
Trait Approach Motivation and Discrepancy Reduction
The idea that dissonance reduction is approach motivated has also been supported at the trait level. In one study, trait approach motivation was positively correlated with spreading of alternatives in a difficult decision paradigm (C. Harmon-Jones, Schmeichel, Inzlicht, & Harmon-Jones, 2011). In a second study, individual differences in trait approach motivation were related to attitude change in an induced compliance paradigm in the high choice condition (where dissonance is expected to be high) but not in the low choice condition (where dissonance is expected to be low; C. Harmon-Jones et al., 2011). These results add support to the prediction that dissonance reduction is approach-motivated.
Neural Activity Involved in Dissonance Processes
The action-based model together with advances in cognitive and affective neuroscience suggested some patterns of neural activity that should be involved with various dissonance processes. Below, we review research that has revealed patterns of neural activity involved in dissonance processes.
Dissonance Arousal, Conflict Monitoring, and the Anterior Cingulate Cortex. The action-based model proposes that the reason for dissonance processes is that inconsistent cognitions lead to conflicted behaviors. Thus, it would be expected that discrepant cognitions would activate the brain regions that are activated by behavioral conflicts. The anterior cingulate cortex (ACC) is a brain region that is activated by cognitive tasks that involve behavioral conflicts, such as the incongruent Stroop task. The ACC may function to detect behavioral conflicts, to inhibit undesired behavior and allow other brain regions to substitute an appropriate response (Botvinick, Braver, Barch, Carter, & Cohen, 2001). Similar to dissonance, these simple cognitive conflicts also produce negative affect (Hajcak & Foti, 2008).
Activation of the ACC has been found for conflicts related to important values, as well as for low-level cognitive tasks. For example, when participants low in racial prejudice become aware that they have inadvertently behaved in a racist manner, they show ACC activation (Amodio et al., 2004). This activation is even greater for individuals who are strongly motivated to avoid being prejudiced (Amodio, Devine, & Harmon-Jones, 2008). When individuals behave contrary to their values, they experience dissonance, which is associated with activation in the ACC, a region that reflects conflicts.
More recently, ACC activation has also been found with paradigms typically used in dissonance research. One study used a within-subjects design difficult decision paradigm and measured functional magnetic resonance imaging (fMRI). Participants rated 160 foods on their desirability and then made choices between foods that were rated similarly (difficult decisions) and dissimilarly (easy decisions). They made these choices either before or after rating the foods for a second time. As the original theory of dissonance would predict, participants spread alternatives more when the decision was difficult and when they made their choices before rerating the foods. In addition, as the action-based model would predict, ACC activity was greater during the difficult decisions (Izuma et al., 2010; see also Chapter 11, this volume). Other studies have shown that individuals evidence increased ACC activity in induced compliance experiments (van Veen, Krug, Schooler, & Carter, 2009). Consistent with the action-based model, these results suggest both that dissonance is associated with cognitive conflict and that dissonance reduction affects motivation.
Dissonance Reduction and the Prefrontal Cortex. According to the action-based model, dissonance reduction should involve approach motivation so as to facilitate behaviorally following through with decisions and commitments. According to models of cognitive control, the prefrontal cortex functions to promote intended responses over unintended responses (Kerns et al., 2004). This suggests that whereas the unpleasant arousal due to a discrepancy should evoke ACC activation, dissonance reduction should involve the prefrontal cortex.
More specifically, as the left prefrontal cortex is activated during approach motivation (E. Harmon-Jones & Gable, 2018), the action-based model predicts that dissonance reduction should be associated with greater left frontal cortical activity. The results of three EEG experiments support this prediction. Individuals in the high choice conditions of induced compliance paradigms showed greater left frontal activation, as well as attitude change, compared to those in the low choice conditions (E. Harmon-Jones, Gerdjikov, & Harmon-Jones, 2008; E. Harmon-Jones, Harmon-Jones, Serra, & Gable, 2011).
Subsequently, experimental manipulation of left frontal cortical activity has been shown to influence the degree of dissonance reduction. Neurofeedback was used to increase or decrease left frontal cortical activation in one experiment (E. Harmon-Jones, Harmon-Jones, Fearn, Sigelman, & Johnson, 2008), and transcranial direct current stimulation was used in another (Mengarelli, Spoglianti, Avenanti, & di Pellegrino, 2015). In both cases, greater induced left frontal cortical activity led to more dissonance reduction. Similarly, an fMRI study showed that activity in the left lateral prefrontal cortex after a difficult decision predicted greater spreading of alternatives (Qin et al., 2011). Furthermore, experimentally inducing an action-oriented state after a difficult decision increased both spreading of alternatives and left prefrontal cortical activity (E. Harmon-Jones, Harmon-Jones, et al., 2008).
The ventral striatum is another brain region associated with approach motivation, so the action-based model predicts that activity in this region should be associated with dissonance reduction. In support of this, fMRI studies have found that activity in the ventral striatum during difficult decisions predicts spreading of alternatives (Jarcho, Berkman, & Lieberman, 2011; Kitayama, Chua, Tompson, & Han, 2013). Conversely, stimuli that were rejected during in a difficult decision produce less activity in the ventral striatum, suggesting the dissonance reduction assists in motivationally disengaging from unchosen alternatives (Izuma et al., 2010).
Dissonance Reduction in Nonhuman Animals
Evidence that nonhuman animals engage in dissonance reduction is also consistent with the action-based model of dissonance. The dissonance revisions such as the aversive consequences model (Cooper & Fazio, 1984), as well as self-consistency (Aronson, 1969) and self-affirmation (Steele, 1988), propose that dissonance reduction is motivated by complex, high-level concepts about the self. It follows that organisms that lack a well-developed self, such as nonhuman animals and young children, would not experience dissonance affect or engage in dissonance reduction. In contrast, the action-based model proposes that dissonance affect and discrepancy reduction are basic, approach-motivated responses to inconsistency that need not be conscious nor related to the self-concept. Thus, the action-based model would presume that dissonance processes may occur in nonhuman animals.
Indeed, recent research has found evidence consistent with dissonance in pigeons, capuchin monkeys, and human preschoolers. Pigeons have been shown to prefer signals of reward that they have had to work harder to obtain (Zentall & Singer, 2007). Although Zentall (2016) interpreted this as a contrast effect, it is equally consistent with effort justification (C. Harmon-Jones, Haslam, & Bastian, 2017; E. Harmon-Jones, 2017). Similarly, both capuchin monkeys and preschool children have demonstrated spreading of alternatives (Egan, Santos, & Bloom, 2007). This effect occurs even in a blind-choice paradigm, when the subjects choose the rewards without knowing what they are until after making the choice (Egan, Bloom, & Santos, 2010). Consistent with the action-based model, these results suggest that dissonance does not rely on a well-developed self, but is a more basic motivational process.
Although the action-based model is consistent with the original theory of cognitive dissonance, it goes beyond in providing a functional explanation for why these effects occur. The model proposes that, in most situations, dissonance reduction promotes effective action. The discomfort that organisms experience when their cognitions are in opposition motivates changes that make action tendencies less conflicted.
Although most of the research on dissonance has employed situations in which participants are required to engage in behavior, dissonance may also result from conflicts between both high-level values and low-level perceptions (Proulx, Inzlicht, & Harmon-Jones, 2012). The action-based model presumes that these are important cognitions with strong implications for action, even when the individual may not be acting on them in the moment. For example, religious beliefs serve as an overarching guide to behavior in many situations, even if one is not sitting in church at that moment. Similarly, a perceptual illusion that causes one to misperceive the depth of a body of water could cause injury if one were to dive into it. According to the action-based model, important cognitions are those with strong action implications, whether or not they are being acted upon at a particular time. This perspective on dissonance has restored the breadth of the original theory, as well as integrated the theory with other research on motivation and emotion, cognitive conflict and control, and affective and cognitive neuroscience.
Abelson, R. P. (1959). Modes of resolution of belief dilemmas. The Journal of Conflict Resolution, 3, 343—352. http://dx.doi.org/10.1177/002200275900300403
Allport, G. W. (1950). The individual and his religion. New York, NY: Macmillan.
Amodio, D. M., Devine, P. G., & Harmon-Jones, E. (2008). Individual differences in the regulation of intergroup bias: The role of conflict monitoring and neural signals for control. Journal of Personality and Social Psychology, 94, 60—74. http://dx.doi.org/10.1037/0022-35188.8.131.52
Amodio, D. M., Harmon-Jones, E., Devine, P. G., Curtin, J. J., Hartley, S. L., & Covert, A. E. (2004). Neural signals for the detection of unintentional race bias. Psychological Science, 15, 88—93. http://dx.doi.org/10.1111/j.0963-7214.2004.01502003.x
Aronson, E. (1969). The theory of cognitive dissonance: A current perspective. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 4, pp. 1—34). New York, NY: Academic Press.
Aronson, E., & Carlsmith, J. M. (1962). Performance expectancy as a determinant of actual performance. Journal of Abnormal and Social Psychology, 65, 178—182. http://dx.doi.org/10.1037/h0042291
Bain, A. (1868). Mental science: A compendium of psychology, and the history of philosophy. New York, NY: Appleton-Century-Crofts.
Bastian, B., Loughnan, S., Haslam, N., & Radke, H. R. (2012). Don’t mind meat? The denial of mind to animals used for human consumption. Personality and Social Psychology Bulletin, 38, 247—256. http://dx.doi.org/10.1177/0146167211424291
Batson, C. D. (1975). Rational processing or rationalization?: The effect of disconfirming information on a stated religious belief. Journal of Personality and Social Psychology, 32, 176—184. http://dx.doi.org/10.1037/h0076771
Beauvois, J.-L., & Joule, R.-V. (1996). A radical dissonance theory. London, England: Taylor & Francis.
Beckmann, J., & Irle, M. (1985). Dissonance and action control. In J. Kuhl & J. Beckmann (Eds.), Action control: From cognition to behavior (pp. 129—150). Berlin, Germany: Springer-Verlag. http://dx.doi.org/10.1007/978-3-642-69746-3_7
Berkowitz, L., & Devine, P. G. (1989). Research traditions, analysis, and synthesis in social psychological theories: The case of dissonance theory. Personality and Social Psychology Bulletin, 15, 493—507. http://dx.doi.org/10.1177/0146167289154002
Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108, 624—652. http://dx.doi.org/10.1037/0033-295X.108.3.624
Brehm, J. W., & Cohen, A. R. (1962). Explorations in cognitive dissonance. New York, NY: Wiley. http://dx.doi.org/10.1037/11622-000
Brock, T. C., & Balloun, J. L. (1967). Behavioral receptivity to dissonant information. Journal of Personality and Social Psychology, 6, 413—428. http://dx.doi.org/10.1037/h0021225
Burris, C. T., Harmon-Jones, E., & Tarpley, W. R. (1997). “By faith alone”: Religious agitation and cognitive dissonance. Basic and Applied Social Psychology, 19, 17—31.
Calder, B. J., Ross, M., & Insko, C. A. (1973). Attitude change and attitude attribution: Effects of incentive, choice, and consequences. Journal of Personality and Social Psychology, 25, 84—99. http://dx.doi.org/10.1037/h0034264
Collins, B. E., & Hoyt, M. F. (1972). Personal responsibility-for-consequences: An integration and extension of the “forced compliance” literature. Journal of Experimental Social Psychology, 8, 558—593. http://dx.doi.org/10.1016/0022-1031(72)90080-7
Cooper, J., & Fazio, R. H. (1984). A new look at dissonance theory. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 17, pp. 229—264). Orlando, FL: Academic Press.
Cooper, J., & Fazio, R. H. (1989). Research traditions, analysis, and synthesis: Building a faulty case around misinterpreted theory. Personality and Social Psychology Bulletin, 15, 519—529. http://dx.doi.org/10.1177/0146167289154005
Cooper, J., & Worchel, S. (1970). Role of undesired consequences in arousing cognitive dissonance. Journal of Personality and Social Psychology, 16, 199—206. http://dx.doi.org/10.1037/h0029830
Cooper, J., Zanna, M. P., & Goethals, G. R. (1974). Mistreatment of an esteemed other as a consequence of affecting dissonance reduction. Journal of Experimental Social Psychology, 10, 224—233.
Egan, L. C., Bloom, P., & Santos, L. R. (2010). Choice-induced preferences in the absence of choice: Evidence from a blind two choice paradigm with young children and capuchin monkeys. Journal of Experimental Social Psychology, 46, 204—207. http://dx.doi.org/10.1016/j.jesp.2009.08.014
Egan, L. C., Santos, L. R., & Bloom, P. (2007). The origins of cognitive dissonance: Evidence from children and monkeys. Psychological Science, 18, 978—983. http://dx.doi.org/10.1111/j.1467-9280.2007.02012.x
Elliot, A. J., & Devine, P. G. (1994). On the motivational nature of cognitive dissonance: Dissonance as psychological discomfort. Journal of Personality and Social Psychology, 67, 382—394. http://dx.doi.org/10.1037/0022-35184.108.40.2062
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41, 1149—1160. http://dx.doi.org/10.3758/BRM.41.4.1149
Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson.
Festinger, L. (1964). Conflict, decision, and dissonance. Palo Alto, CA: Stanford University Press.
Festinger, L., & Carlsmith, J. M. (1959). Cognitive consequences of forced compliance. Journal of Abnormal and Social Psychology, 58, 203—210. http://dx.doi.org/10.1037/h0041593
Festinger, L., Riecken, H. W., & Schachter, S. (1956). When prophecy fails. Minneapolis: University of Minnesota Press. http://dx.doi.org/10.1037/10030-000
Fraley, R. C., & Marks, M. J. (2007). The null hypothesis significance testing debate and its implications for personality research. In R. W. Robins, R. C. Fraley, & R. F. Krueger (Eds.), Handbook of research methods in personality psychology (pp. 149—169). New York, NY: Guilford Press.
Gibson, J. J. (1979). The ecological approach to visual perception. Boston, MA: Houghton Mifflin.
Goethals, G. R., & Cooper, J. (1972). Role of intention and postbehavioral consequence in the arousal of cognitive dissonance. Journal of Personality and Social Psychology, 23, 293—301. http://dx.doi.org/10.1037/h0033123
Goethals, G. R., & Cooper, J. (1975). When dissonance is reduced: The timing of self-justificatory attitude change. Journal of Personality and Social Psychology, 32, 361—367. http://dx.doi.org/10.1037/0022-35220.127.116.111
Gollwitzer, P. M., & Bayer, U. (1999). Deliberative versus implemental mindsets in the control of action. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 403—422). New York, NY: Guilford Press.
Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 38, pp. 69—119). San Diego, CA: Elsevier Academic Press. http://dx.doi.org/10.1016/S0065-2601(06)38002-1
Hajcak, G., & Foti, D. (2008). Errors are aversive: Defensive motivation and the error-related negativity. Psychological Science, 19, 103—108. http://dx.doi.org/10.1111/j.1467-9280.2008.02053.x
Harmon-Jones, C., Haslam, N., & Bastian, B. (2017). Dissonance reduction in nonhuman animals: Implications for cognitive dissonance theory. Animal Sentience: An Interdisciplinary Journal on Animal Feeling, 1(12), 4.
Harmon-Jones, C., Schmeichel, B. J., Inzlicht, M., & Harmon-Jones, E. (2011). Trait approach motivation relates to dissonance reduction. Social Psychological & Personality Science, 2, 21—28. http://dx.doi.org/10.1177/1948550610379425
Harmon-Jones, E. (2000a). Cognitive dissonance and experienced negative affect: Evidence that dissonance increases experienced negative affect even in the absence of aversive consequences. Personality and Social Psychology Bulletin, 26, 1490—1501. http://dx.doi.org/10.1177/01461672002612004
Harmon-Jones, E. (2000b). A cognitive dissonance theory perspective on the role of emotion in the maintenance and change of beliefs and attitudes. In N. H. Frijda, A. R. S. Manstead, & S. Bem (Eds.), Emotions and beliefs (pp. 185—211). Cambridge, England: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511659904.008
Harmon-Jones, E. (2000c). An update on cognitive dissonance theory, with a focus on the self. In A. Tesser, R. Felson, & J. Suls (Eds.), Psychological perspectives on self and identity (pp. 119—144). Washington, DC: American Psychological Association. http://dx.doi.org/10.1037/10357-005
Harmon-Jones, E. (2001). The role of affect in cognitive dissonance processes. In J. Forgas (Ed.), Handbook of affect and social cognition (pp. 237—255). Mahwah, NJ: Lawrence Erlbaum.
Harmon-Jones, E. (2002). A cognitive dissonance theory perspective on persuasion. In J. P. Dillard & M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. 99—116). Mahwah, NJ: Lawrence Erlbaum.
Harmon-Jones, E. (2017). Clarifying concepts in cognitive dissonance theory: Commentary on Zentall on cognitive dissonance. Animal Sentience, 12(5).
Harmon-Jones, E., Amodio, D. M., & Harmon-Jones, C. (2009). Action-based model of dissonance: A review, integration, and expansion of conceptions of cognitive conflict. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (Vol. 41, pp. 119—166). San Diego, CA: Academic Press.
Harmon-Jones, E., Brehm, J. W., Greenberg, J., Simon, L., & Nelson, D. E. (1996). Evidence that the production of aversive consequences is not necessary to create cognitive dissonance. Journal of Personality and Social Psychology, 70, 5—16. http://dx.doi.org/10.1037/0022-3518.104.22.168
Harmon-Jones, E., & Gable, P. A. (2018). On the role of asymmetric frontal cortical activity in approach and withdrawal motivation: An updated review of the evidence. Psychophysiology, 55, e12879. http://dx.doi.org/10.1111/psyp.12879
Harmon-Jones, E., Gable, P. A., & Price, T. F. (2011). Leaning embodies desire: Evidence that leaning forward increases relative left frontal cortical activation to appetitive stimuli. Biological Psychology, 87, 311—313. http://dx.doi.org/10.1016/j.biopsycho.2011.03.009
Harmon-Jones, E., Gerdjikov, T., & Harmon-Jones, C. (2008). The effect of induced compliance on relative left frontal cortical activity: A test of the action-based model of dissonance. European Journal of Social Psychology, 38, 35—45. http://dx.doi.org/10.1002/ejsp.399
Harmon-Jones, E., & Harmon-Jones, C. (2002). Testing the action-based model of cognitive dissonance: The effect of action orientation on postdecisional attitudes. Personality and Social Psychology Bulletin, 28, 711—723. http://dx.doi.org/10.1177/0146167202289001
Harmon-Jones, E., Harmon-Jones, C., Fearn, M., Sigelman, J. D., & Johnson, P. (2008). Left frontal cortical activation and spreading of alternatives: Tests of the action-based model of dissonance. Journal of Personality and Social Psychology, 94, 1—15. http://dx.doi.org/10.1037/0022-3522.214.171.124
Harmon-Jones, E., Harmon-Jones, C., Serra, R., & Gable, P. A. (2011). The effect of commitment on relative left frontal cortical activity: Tests of the action-based model of dissonance. Personality and Social Psychology Bulletin, 37, 395—408. http://dx.doi.org/10.1177/0146167210397059
Harmon-Jones, E., Price, T. F., & Harmon-Jones, C. (2015). Supine body posture decreases rationalizations: Testing the action-based model of dissonance. Journal of Experimental Social Psychology, 56, 228—234. http://dx.doi.org/10.1016/j.jesp.2014.10.007
Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60, 895—910. http://dx.doi.org/10.1037/0022-35126.96.36.1995
Hoyt, M. F., Henley, M. D., & Collins, B. E. (1972). Studies in forced compliance: Confluence of choice and consequence on attitude change. Journal of Personality and Social Psychology, 23, 205—210. http://dx.doi.org/10.1037/h0033034
Izuma, K., Matsumoto, M., Murayama, K., Samejima, K., Sadato, N., & Matsumoto, K. (2010). Neural correlates of cognitive dissonance and choice-induced preference change. Proceedings of the National Academy of Sciences of the United States of America, 107, 22014—22019. http://dx.doi.org/10.1073/pnas.1011879108
James, W. (1950). The principles of psychology. New York, NY: Dover. (Original work published 1890)
Jarcho, J. M., Berkman, E. T., & Lieberman, M. D. (2011). The neural basis of rationalization: Cognitive dissonance reduction during decision-making. Social Cognitive and Affective Neuroscience, 6, 460—467. http://dx.doi.org/10.1093/scan/nsq054
Johnson, R. W., Kelly, R. J., & LeBlanc, B. A. (1995). Motivational basis of dissonance: Aversive consequences or inconsistency. Personality and Social Psychology Bulletin, 21, 850—855. http://dx.doi.org/10.1177/0146167295218008
Jones, E. E., & Gerard, H. B. (1967). Foundations of social psychology. New York, NY: Wiley.
Kerns, J. G., Cohen, J. D., MacDonald, A. W., III, Cho, R. Y., Stenger, V. A., & Carter, C. S. (2004). Anterior cingulate conflict monitoring and adjustments in control. Science, 303, 1023—1026. http://dx.doi.org/10.1126/science.1089910
Kitayama, S., Chua, H. F., Tompson, S., & Han, S. (2013). Neural mechanisms of dissonance: An fMRI investigation of choice justification. NeuroImage, 69, 206—212. http://dx.doi.org/10.1016/j.neuroimage.2012.11.034
Kuhl, J. (1984). Volitional aspects of achievement motivation and learned helplessness: Toward a comprehensive theory of action-control. In B. A. Maher (Ed.), Progress in experimental personality research (Vol. 13, pp. 99—171). New York, NY: Academic Press.
McArthur, L. Z., & Baron, R. M. (1983). Toward an ecological theory of social perception. Psychological Review, 90, 215—238. http://dx.doi.org/10.1037/0033-295X.90.3.215
Mengarelli, F., Spoglianti, S., Avenanti, A., & di Pellegrino, G. (2015). Cathodal tDCS over the left prefrontal cortex diminishes choice-induced preference change. Cerebral Cortex, 25, 1219—1227. http://dx.doi.org/10.1093/cercor/bht314
Mills, J. (1968). Interest in supporting and discrepant information. In R. P. Abelson, E. Aronson, W. J. McGuire, T. M. Newcomb, M. J. Rosenberg, & P. H. Tannenbaum (Eds.), Theories of cognitive consistency: A source book (pp. 771—776). Skokie, IL: Rand McNally.
Nel, E., Helmreich, R., & Aronson, E. (1969). Opinion change in the advocate as a function of the persuasibility of his audience: A clarification of the meaning of dissonance. Journal of Personality and Social Psychology, 12, 117—124. http://dx.doi.org/10.1037/h0027566
Price, T. F., Dieckman, L. W., & Harmon-Jones, E. (2012). Embodying approach motivation: Body posture influences startle eyeblink and event-related potential responses to appetitive stimuli. Biological Psychology, 90, 211—217. http://dx.doi.org/10.1016/j.biopsycho.2012.04.001
Proulx, T., Inzlicht, M., & Harmon-Jones, E. (2012). Understanding all inconsistency compensation as a palliative response to violated expectations. Trends in Cognitive Sciences, 16, 285—291. http://dx.doi.org/10.1016/j.tics.2012.04.002
Qin, J., Kimel, S., Kitayama, S., Wang, X., Yang, X., & Han, S. (2011). How choice modifies preference: Neural correlates of choice justification. NeuroImage, 55, 240—246. http://dx.doi.org/10.1016/j.neuroimage.2010.11.076
Rabbie, J. M., Brehm, J. W., & Cohen, A. R. (1959). Verbalization and reactions to cognitive dissonance. Journal of Personality, 27, 407—417. http://dx.doi.org/10.1111/j.1467-6494.1959.tb02363.x
Richard, F. D., Bond, C. F., Jr., & Stokes-Zoota, J. J. (2003). One hundred years of social psychology quantitatively described. Review of General Psychology, 7, 331—363. http://dx.doi.org/10.1037/1089-26188.8.131.521
Scheier, M. F., & Carver, C. S. (1980). Private and public self-attention, resistance to change, and dissonance reduction. Journal of Personality and Social Psychology, 39, 390—405. http://dx.doi.org/10.1037/0022-35184.108.40.2060
Scher, S. J., & Cooper, J. (1989). Motivational basis of dissonance: The singular role of behavioral consequences. Journal of Personality and Social Psychology, 56, 899—906. http://dx.doi.org/10.1037/0022-35220.127.116.119
Schwarz, N., Frey, D., & Kumpf, M. (1980). Interactive effects of writing and reading a persuasive essay on attitude change and selective exposure. Journal of Experimental Social Psychology, 16, 1—17. http://dx.doi.org/10.1016/0022-1031(80)90032-3
Simon, L., Greenberg, J., & Brehm, J. W. (1995). Trivialization: The forgotten mode of dissonance reduction. Journal of Personality and Social Psychology, 68, 247—260. http://dx.doi.org/10.1037/0022-3518.104.22.168
Spencer, H. (1865). First principles. New York, NY: Appleton.
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 261—302). San Diego, CA: Academic Press. http://dx.doi.org/10.1016/S0065-2601(08)60229-4
van Veen, V., Krug, M. K., Schooler, J. W., & Carter, C. S. (2009). Neural activity predicts attitude change in cognitive dissonance. Nature Neuroscience, 12, 1469—1474. http://dx.doi.org/10.1038/nn.2413
Zentall, T. R. (2016). Cognitive dissonance or contrast? Animal Sentience, 12(1).
Zentall, T. R., & Singer, R. A. (2007). Within-trial contrast: Pigeons prefer conditioned reinforcers that follow a relatively more rather than a less aversive event. Journal of the Experimental Analysis of Behavior, 88, 131—149. http://dx.doi.org/10.1901/jeab.2007.27-06
1An examination of the past experiments that manipulated whether participants produced an aversive consequence revealed extremely high compliance rates in the high-choice and low-justification conditions. These experiments are listed with the number of noncompliers indicated in parentheses after the date of publication: Nel et al., 1969 (1); Cooper and Worchel, 1970 (1); Collins and Hoyt, 1972 (0); Goethals and Cooper, 1972 (2 experiments; 0); Hoyt et al., 1972 (1); Calder et al., 1973 (1); Cooper, Zanna, and Goethals, 1974 (0); Goethals and Cooper, 1975 (1); Scher and Cooper, 1989 (1); Johnson, Kelly, and LeBlanc, 1995 (0).
2In the three induced-compliance experiments by E. Harmon-Jones et al. (1996) that are described in this chapter, compliance was approximately 85%. However, in the reported experiment by E. Harmon-Jones (2000a), in which participants ate a piece of chocolate and then wrote that they did not enjoy it, only 1 participant did not comply. In the experiments by E. Harmon-Jones et al. (1996), persons were asked to write that they believed a boring passage they had just read was interesting or that an unpleasant-tasting beverage was pleasant tasting. The attitudes toward these simple stimuli were quite negative, as evidenced by low-choice-condition and control-condition participants. In contrast, in the experiment by E. Harmon-Jones (2000a), persons’ attitudes toward the chocolate were not extremely positive. Thus, although this latter experiment had compliance rates similar to the ones discussed above, it still produced dissonance. Thus, compliance rates are not an inviolable assessment of amount of justification within an experiment. Other factors, such as the size and importance of the discrepancy between attitude and behavior and number and importance of justifications for the behavior (promised rewards or punishments, which are probably largely social in nature), need to be taken into account. Attempts to measure size and importance of discrepancy and justifications would aid tremendously in specifying the magnitude of dissonance.