Conspiracy believers tend to overrate their cognitive abilities and think most others agree with them

A series of eight studies has uncovered a consistent pattern among people who believe in conspiracy theories: they tend to be overconfident in their cognitive abilities and significantly overestimate how much others agree with them. The findings, published in Personality and Social Psychology Bulletin, suggest that conspiracy belief may be fueled less by active motivations and more by a mistaken sense of certainty.

This research was conducted in response to a growing need to understand why people come to believe in conspiracy theories, especially those that are widely rejected by experts and the broader public. Prior work has emphasized motivations like the desire to feel unique, or to explain social and political events in a way that aligns with personal identity. But the researchers behind this new work proposed a different explanation. They hypothesized that people who consistently overestimate their own intellectual performance might be more likely to hold onto fringe beliefs and to falsely assume that others share their views.

“One of the thing that seems to distinguish (at least some) conspiracy theorists is not just that their beliefs seem to be based on poor evidence, but also that they appear so confident in their beliefs,” explained study author Gordon Pennycook, an associate professor at Cornell University

“The question we ask here is whether this tendency to be overconfident is more general than that. It’s obvious that people are overconfident in their beliefs; the question we ask here is whether people who tend to be overconfident in general (across domains) are more likely to believe conspiracies.”

The researchers carried out eight separate studies with a combined sample of 4,181 participants from the United States. Across the studies, participants completed various cognitive tasks designed to measure things like numerical reasoning, risk understanding, and perceptual accuracy. Crucially, after each task, participants were asked to estimate how well they thought they had performed.

This approach allowed the researchers to compare people’s actual test scores to their self-assessments, creating a measure of overconfidence. A person who scored poorly but believed they did well would be considered overconfident. The researchers then compared these overconfidence scores to how strongly participants endorsed conspiracy theories, including well-known but widely debunked claims like the moon landing being faked or vaccines being part of a government control plot.

The results showed a strong and consistent relationship between overconfidence and belief in false conspiracy theories. People who thought they had done well on tasks, even when they had not, were more likely to endorse conspiracies. Importantly, this relationship held even after accounting for other known predictors of conspiracy belief, such as low analytical thinking ability, narcissistic personality traits, and a desire to feel unique.

To ensure this wasn’t just a reflection of performance on specific types of tasks, the researchers included a perception-based test in which participants were shown difficult-to-identify images and asked to guess what they saw. Since the task was extremely hard and essentially relied on guessing, actual performance was unrelated to skill. Yet, participants who thought they had done well on this task were also more likely to believe in conspiracy theories. This made the perception test a particularly useful way of measuring dispositional overconfidence—that is, a general tendency to overrate one’s own abilities.

Another striking finding emerged when participants were asked to estimate how many other people agreed with them about the conspiracies they endorsed. Although most conspiracy theories were believed by only a small minority of participants—just 12% on average—those who did believe in them thought they were in the majority 93% of the time. This “false consensus effect” was strongest among the most overconfident individuals. They not only believed incorrect things, but also wrongly assumed that most others shared their views.

This sense of being in the majority held even when participants were asked to estimate agreement among people who likely disagree with them, such as members of the opposing political party. While their expectations of agreement were somewhat lower in those cases, conspiracy believers still significantly overestimated how much others—regardless of group—agreed with their beliefs.

The studies also explored how this overconfidence and belief pattern compares to other potential drivers of conspiratorial thinking. Grandiose narcissism and the need for uniqueness were both related to conspiracy beliefs, but they were not as consistent or strong as overconfidence. When all variables were included in the same analysis, overconfidence—especially as measured by the perception task—remained a significant predictor, while the need for uniqueness did not.

In one of the later studies, the researchers tested the robustness of the false consensus effect by assigning participants to rate how much they thought their political allies and opponents agreed with them about various conspiracy theories. Even under these more targeted conditions, conspiracy believers still dramatically overestimated the level of agreement. This suggests that their sense of consensus is not just a rhetorical posture, but a genuine misperception about the social reality of their beliefs.

Across all eight studies, the researchers found that overconfidence consistently predicted belief in conspiracies that are not supported by evidence. Interestingly, this was not the case for conspiracy theories that have been verified by history, such as government surveillance of civil rights leaders or unethical medical experiments. Belief in those conspiracies was not linked to overconfidence, suggesting that the pattern is specific to conspiracies that are actually false or unverified.

“The tendency to be overconfident in general may increase the chances that someone falls down the rabbit hole (so to speak) and believe conspiracies,” Pennycook told PsyPost. “In fact, our results counteract a prevailing narrative about conspiracy theorists: that they know that they hold fringe beliefs and revel in that fact.”

“Even people who believed very fringe conspiracies, such as that scientists are conspiring to hide the truth about the Earth being flat, thought that their views were in the majority. Conspiracy believers – particularly overconfident ones – really seem to be miscalibrated in a major way. Not only are their beliefs on the fringe, but they are very much unaware of how far on the fringe they are.”

There are some limitations to the study. All participants were drawn from online platforms, such as MTurk and Prolific, which may not capture the most extreme or isolated conspiracy believers. People who are highly distrustful of institutions may not be inclined to participate in university-run research studies, meaning the results could underestimate the strength of the patterns among the most committed conspiracy theorists.

The study, “Overconfidently Conspiratorial: Conspiracy Believers are Dispositionally Overconfident and Massively Overestimate How Much Others Agree With Them,” was authored by Gordon Pennycook, Jabin Binnendyk, and David G. Rand.