Status: Complete; potentially interested in promising follow-up work
Together with many collaborators, particularly dr. David Vaidis, I conducted a large-scaled multi-lab replication study of the induced compliance paradigm based on cognitive dissonance theory.
Cognitive dissonance refers to a state of aversive arousal that is experienced when people realize they possess mutually inconsistent cognitions. This state is the foundation of cognitive dissonance theory (CDT)—a theory developed by Leon Festinger in 1957. One of the most popular paradigms to test this theory is the induced compliance paradigm. In this paradigm, participants are induced to do something that is counterattitudinal for most people (e.g., students being asked to argue in favor of increasing their tuition). They are then asked for their attitude towards the counterattitudinal topic and, if cognitive dissonance theory is correct, it should result in the participants changing their attitude to be more in line with the position they argued for (e.g., raising the tuition). To make sure the change in attitude is driven by cognitive dissonance and not something else (e.g., self-persuasion), experiments usually manipulate something that should cause more or less dissonance. Frequently, this favor is choice freedom. Participants are given either more or less choice over choosing to perform the counterattitudinal behavior. The idea is that those with more choice should experience more dissonance, as those with less choice can blame their behavior to the fact they were not given any other choice but to perform the counterattitudinal behavior.
We tried to replicate this idea using a multi-lab approach. We, David and I, worked together with many other collaborators to run a typical induced compliance study. In total, we managed to convince 39 labs to do this. Together we recruited a total of 4,898 participants, making this the largest cognitive dissonance to date. Unfortunately, we did find that the counterattitudinal behavior of participants in the high-choice condition led to more attitude change compared to the participants in the low-choice condition. This is the crucial effect that cognitive dissonance theory predicts.
The theory of cognitive dissonance can explain many different phenomena that we should understand so that we may intervene and improve the lives of others. For example, cognitive dissonance theory has been used to explain religious beliefs, unhealthy behaviors, and people's attitude towards animals.
Before applying the theory, we should be relatively confident that the theory holds up. We need to have sufficient evidence to believe in the theory. Surprisingly, the social psychological evidence we have for the theory is quite weak. A lot of the research that caused cognitive dissonance to become textbook material stems from relatively old research, mostly conducted in the 50s, 60s, and 70s. While this is not necessarily a problem, it is a problem in the case of social psychology. The original studies were conducted with extremely low sample sizes and without pre-registration, or other tools that limit p-hacking. This means that past findings may be false positives.
The possibility that something as popular as cognitive dissonance theory might not be true is not as unlikely as you might think. There have now been several high-profile examples of social psychological findings that have failed to replicate.
Our results cast significant doubt, in my opinion, on the credibility of cognitive dissonance theory. This means more work should be conducted to test if and when cognitive dissonance processes occur.