A new batch of online experiments suggests that simply encountering a conspiracy theory can inflate a person’s perception of how many others share the belief, and that providing hard statistics about public scepticism can reduce belief in those theories. The research, published in the European Journal of Social Psychology, involved six studies with 2,906 participants from the United States and the United Kingdom and was led by Jan-Willem van Prooijen of Vrije Universiteit Amsterdam.
Descriptive norms—our sense of how common a belief is within a group—play a key role in how conspiracy theories spread. The researchers note that people often look to others for cues, a phenomenon known as social proof, to decide what to think or do when events feel uncertain.
Even though conspiracy theories often attract massive online attention, believers typically form a minority of the wider population. However, those who endorse such narratives frequently overestimate how widely they are shared. The study cites a 2021 British survey showing overestimation of peer belief and a 2023 Turkish study finding similar misperceptions among ordinary people and experts.
In prior work, correcting these misperceptions has been shown to alter thinking. A 2021 experiment with British parents found that revealing that most other parents rejected anti-vaccine conspiracy theories reduced personal endorsement. The new project builds on that foundation, testing whether merely encountering a conspiracy theory can spur inflated views of its popularity and whether presenting sceptical data can dampen belief across multiple theories.
Van Prooijen explained that extreme voices online can create the impression that conspiracies are more widely believed than they actually are, and he emphasised the aim of examining whether conspiracy beliefs depend on perceived norms and whether those insights could inform interventions.
The researchers conducted six online studies with 2,906 participants from the United States and the United Kingdom, all preregistered to publicly document their hypotheses and analysis plans before data collection.
In the first pair of studies, Study 1a with 300 American participants and Study 1b with 299 British participants explored how people rated ten popular conspiracy claims on a 0 to 100 plausibility scale and then estimated the average rating of others in the group. The actual group averages were relatively low—33.09 for the United States and 34.16 for the United Kingdom—while participants guessed higher averages of 39.85 and 42.30 respectively. Strong believers tended to think others believed the claims too, with correlations of 0.65 in the United States and 0.71 in the United Kingdom. Estimates by believers were less accurate than those by skeptics.
The team also examined whether strong believers saw themselves as distinctive. Comparing each person’s own rating with their estimate of the group rating showed that the belief gap did not appear larger for stronger believers than for skeptics.
These findings align with earlier PsyPost coverage noting that overconfidence in thinking ability correlates with stronger conspiracy beliefs and a greater overestimation of public agreement.
To probe causality, the researchers then randomised participants to read either a conspiratorial or a non-conspiratorial blog post in Studies 2a (America) and 2b (Britain). After reading and summarising the text, participants rated their own belief in four statements on a 0 to 100 scale and estimated how others would rate them.
In the American sample, those exposed to the conspiratorial post believed the claims more strongly—an average of 61.78 compared with 49.46 in the control group—and also overestimated others’ beliefs (62.64 vs 51.00). A similar pattern emerged in Britain, with personal belief at 59.78 after reading the conspiratorial post versus 41.99 in the control group, and estimates of others’ beliefs at 60.28 versus 46.41.
Mediation analyses indicated that the rise in personal belief tracked with the heightened sense that others believed the claim, though the reverse pattern—that stronger personal belief shapes guesses about others—also fit, so the data could not establish a clear direction of causality.
Van Prooijen summarised the implications: “One reason why people believe conspiracy theories is because they assume many others believe them as well. Conspiracy beliefs thus spread via social influence processes. Understanding these processes can help design effective interventions to reduce conspiracy beliefs.”
To test whether shifting perceived norms could directly alter beliefs, the researchers designed two norm-information interventions in Studies 3 and 4. In Study 3, 501 American participants were shown six popular conspiracy theories and told that either “especially many” people found the statements plausible or that “only a few” did. This vague framing did not yield a statistically significant difference in personal belief, with averages of 44.48 in the “many” condition and 41.65 in the “few” condition. The authors cautioned that terms like “many” and “few” can be interpreted differently and that the study did not measure whether the norm perception actually shifted.
Study 4, conducted with 800 American participants, added concrete data. Participants were shown eight conspiracy theories from the first study, with half seeing a chart displaying the actual distribution of ratings from Study 1a, including the median and a note that most participants did not find the statement plausible. The other half saw the conspiracy statements without data. The statistical pictures mattered: those who saw the charts estimated the public average belief at 20.95 out of 100, compared with 38.34 in the control group, while personal endorsement fell to 24.06 versus 33.70 in the control group.
“We have done two studies testing the intervention, but it only worked if we gave people factual statistical information of how few people believed a conspiracy theory,” van Prooijen said. “If we provided more abstract information, only stating that ‘few people’ believe the theory without providing concrete numbers, the intervention did not work. It remains speculative why that is the case, but it is nevertheless informative for future interventions based on this research.”
The researchers stress that the practical significance lies in the potential for interventions that reveal how widespread belief in a conspiracy theory actually is, noting these findings could reduce the likelihood that people will endorse the theory themselves. They also acknowledge limitations common to online experiments: participants may respond to expectations, and those in the studies were online participants rather than the broader population. Social media environments can amplify local norms, complicating real-world applicability.
Additional caveats include the possibility that such interventions might not work for conspiracies that are genuinely widespread, such as theories surrounding the Kennedy assassination, or in certain groups where conspiracy thinking is entrenched. The researchers call for further work, including real-world network contexts and analyses in more collectivist cultures, to determine long-term effectiveness and applicability.
The study is titled Everyone Knows It’s a Hoax: The Role of Perceived Descriptive Norms in Conspiracy Beliefs and is authored by Jan-Willem van Prooijen, Paul K. Bergmann, Vladimira Čavojová, and Jakub Šrol.
