What Is Cognitive Dissonance?

Take our Do You Still Believe These Myths About The Mind?

Find a therapist near me

A recent New Yorker article raised criticisms of research derived from cognitive dissonance theory.

The article seemed to conclude that cognitive dissonance is not a thing.

This post addresses those criticisms by reviewing additional evidence ignored by the New Yorker.

Reporters may behave as the theory predicts when they selectively avoid information.

A recent New Yorker article, “Is Cognitive Dissonance Actually a Thing?”, strongly implies that the answer to that question is “No.” The article arrived at this answer by considering a 1956 participant-observer study1 that was based on highly subjective methods that are now rarely used in psychological science. In this old study, Festinger and other researchers infiltrated a doomsday cult. After the world did not end as the cult expected, Festinger proposed the theory of cognitive dissonance. However, at that point it was just a theory and needed to be empirically tested. Since then, thousands of experiments have tested the theory in a wide variety of ways.

The evidence base of cognitive dissonance theory

The New Yorker article omitted literally thousands of pieces of evidence. Many, many studies have found solid evidence consistent with cognitive dissonance theory in myriad different situations in which organisms make difficult decisions, engage in effort, are exposed to belief-inconsistent information, or act counter to their values.2

Clarifying predictions regarding belief intensification

In Festinger’s account of the doomsday cult, the members intensified their beliefs after the world failed to end. The New Yorker article countered that the cult eventually split up. However, nothing in the original theory predicts that belief intensification after belief disconfirmation always occurs or that it will last forever; our cognitive systems are quite responsive to reality. In 1956, Festinger and colleagues laid out five precise conditions under which belief intensification would occur after belief disconfirmation.

Controlled tests of belief disconfirmation

Other, more controlled studies have revealed support for the dissonance theory prediction of belief intensification following belief disconfirmation. In one study, Christians who believed the veracity of an article that indicated that Jesus was not the son of God later intensified their beliefs in Jesus’ divinity3. In addition, studies have revealed several other ways individuals respond when their important beliefs are disconfirmed4,5.

A multi-lab replication of one dissonance effect

The New Yorker article also leaned heavily on a 2024 multilab replication failure that included only one method of testing the theory6. This replication tested two predictions: 1) that participants assigned to write an essay that was counter to their attitudes would change their attitudes to agree with the essay, and 2) that participants who felt they had a choice over whether to write the essay would change their attitudes more. The results confirmed the first prediction but“failed” with the second (moderation by choice).

The 2024 replication failure regarding choice has been heavily criticized. This study manipulated perceived choice (low vs. high) over engaging in a counterattitudinal behavior (writing for a tuition increase at one’s university). As noted elsewhere7, the replication study had a moderately high level of perceived choice even in the low-choice condition, so it is no surprise that low-choice participants also changed their attitudes to be consistent with their behavior. Moreover, two separate groups of researchers re-analyzed the replication study’s data and found that ratings of perceived choice correlated with more attitude change, as would be predicted by dissonance theory. 8,9

Ways to not replicate an effect

The authors of the replication study seemed to dismiss these criticisms10. However, they had no words for what might be the biggest concern: Why try to replicate a study that worked only 1 out of 2 times in the original article from 1983? That is, the original article11 reported 2 studies and only 1 of the 2 showed the predicted attitude change effect. Apparently, some things in the original studies made the attitude change effect somewhat unreliable and difficult to replicate. As Bornstein12 noted long ago, if researchers enter into a replication attempt expecting not to find an effect, there are many subtle things they can do, “consciously or unconsciously," that may produce a null effect. Getting rid of messy methods helps.13

What Is Cognitive Dissonance?

Take our Do You Still Believe These Myths About The Mind?

Find a therapist near me

Dissonance in Chat GPT

Dissonance effects certainly exist. In fact, studies have revealed that even Chat GPT changes its attitudes to be consistent with its behavior; when GPT is given a choice over its behavior, it changes its attitudes even more.14 This may be the best proof that humans show this dissonance effect, because GPT got this behavioral tendency from humans.

When journalism reflects dissonance processes

The reporting behind the story of how dissonance may not be a thing itself actually provides support for predictions derived from the theory. Once I learned of the New Yorker article, I communicated with two colleagues who have much expertise with dissonance theory and research. Both had been contacted by the author of the article and both responded with several criticisms of the questions about the controversy surrounding Festinger and colleagues’ work as well as the induced compliance replication. However, neither of these experts’ concerns were included in the New Yorker article. Thus, consistent with dissonance theory predictions, the author selectively avoided information that was inconsistent with the goal of discrediting dissonance theory and its research.

Cindy Harmon-Jones co-authored this post. Thanks to her and several colleagues with whom I discussed these issues: Mahzarin Banaji, Bertram Gawronski, Steven Lehr, and Travis Proulx.

Love, S. Is Cognitive Dissonance Actually a Thing? The New Yorker (2025).

1) Festinger, L., Riecken, H. W. & Schachter, S. When Prophecy Fails. vii, 257 (University of Minnesota Press, Minneapolis, MN, US, 1956). doi:10.1037/10030-000.

2) Harmon-Jones, E. Cognitive Dissonance: Reexamining a Pivotal Theory in Psychology, 2nd Ed. (American Psychological Association, 2019).

3) Batson, C. D. Rational processing or rationalization? The effect of disconfirming information on a stated religious belief. Journal of Personality and Social Psychology32, 176–184 (1975).

4) Burris, C. T., Harmon-Jones, E. & Tarpley, W. R. ‘By Faith Alone’: Religious Agitation and Cognitive Dissonance. Basic and Applied Social Psychology 19, 17–31 (1997).

5) Harmon-Jones, E., Harmon-Jones, C. & Denson, T. F. A novel way of responding to dissonance evoked by belief disconfirmation: making the wrongdoing of an opponent salient. Social Influence 15, 34–45 (2020).

6) Vaidis, D. C. et al. A Multilab Replication of the Induced-Compliance Paradigm of Cognitive Dissonance. Advances in Methods and Practices in Psychological Science 7, 25152459231213375 (2024).

7) Harmon-Jones, E. & Harmon-Jones, C. Dissonance in the Induced-Compliance Paradigm: A Commentary on Vaidis et al. (2024). Advances in Methods and Practices in Psychological Science 7, 25152459241268308 (2024).

8) Lishner, D. A. But Did They Really Perceive No (Low) Choice? Comment on Vaidis et al. (2024). Advances in Methods and Practices in Psychological Science 7, 25152459241267915 (2024).

9) Pauer, S., Linne, R. & Erb, H.-P. From the Illusion of Choice to Actual Control: Reconsidering the Induced-Compliance Paradigm of Cognitive Dissonance. Advances in Methods and Practices in Psychological Science 7, 25152459241265002 (2024).

10) Sleegers, W. W. A. et al. When Replication Fails: What to Conclude and Not to Conclude? Advances in Methods and Practices in Psychological Science 7, 25152459241268197 (2024).

11) Croyle, R. T. & Cooper, J. Dissonance arousal: Physiological evidence. Journal of Personality and Social Psychology 45, 782–791 (1983).

12) Bornstein, R. F. Publication Politics, Experimenter Bias and the Replication Process in Social Science Research. Journal of Social Behavior and Personality 5, 71–81 (1990).

13) Harmon-Jones, E., Harmon-Jones, C., Amodio, D. M., Gable, P. A. & Schmeichel, B. J. Valid replications require valid methods: Recommendations for best methodological practices with lab experiments. Motivation Science 11, 235–245 (2025).

14) Lehr, S. A., Saichandran, K. S., Harmon-Jones, E., Vitali, N. & Banaji, M. R. Kernels of selfhood: GPT-4o shows humanlike patterns of cognitive dissonance moderated by free choice. Proceedings of the National Academy of Sciences 122, e2501823122 (2025).

There was a problem adding your email address. Please try again.

By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy

© Psychology Today