What prompts people to change their minds about conspiracy theories?

© 2014 Mouldy Sponge. Licensed under CC-BY.

For example, in 1985, French intelligence operatives really did conspire to sink the Rainbow Warrior. The fact that the public are willing to consider the possibility of people in power conspiring to cause harm is a good thing. Nevertheless, members of the public do sometimes endorse conspiracy theories that are not warranted by evidence, or even strongly contradicted by evidence. For example, Marques et al. (2022) found that approximately 8% of a large sample of New Zealanders agreed with the claim that “vapor trails left by aircraft are deliberately sprayed in a clandestine program directed by government officials” (the chemtrails theory, which is not at all supported by evidence).

The fact that people often believe conspiracy theories that are unwarranted by evidence is an interesting puzzle, and has attracted a great deal of interest from psychological researchers over the last few years. Researchers have collected lots of information about how many people agree with conspiracy theories, which groups of people are more or less likely to believe conspiracy theories, and what factors are correlated with such beliefs. But what prompts a person to change their mind about a conspiracy theory – to reject a theory they previously embraced, or endorse a theory they previously rejected?

A group of researchers led by Associate Professor Matt Williams from Te Kunenga ki Pūrehuroa Massey University set out to answer this question. They conducted a longitudinal study with a sample of just under a thousand participants from Aotearoa New Zealand, Australia, and the United Kingdom. The participants were surveyed monthly for two years. This provided a unique capacity to examine how often – and why – people change their minds about conspiracy theories. It also allowed the team to test hypotheses about the causes of changes in belief with more confidence than is possible with one-off surveys (albeit longitudinal studies still aren’t as conclusive as true experiments).

One popular hypothesis is the idea that experiencing stress or anxiety can make people more likely to embrace conspiracy theories. The premise being, that feeling distressed by threatening world events provokes us to find explanations for those events, and conspiracy theories can fill that void. A good deal of research has shown that stress and anxiety are correlated with belief in conspiracy theories. But correlation is not causation. In a preregistered study using seven waves of data from the longitudinal study, Fox, Williams and Hill (2025) found almost no evidence that stress, anxiety or even depression lead to subsequent belief in conspiracy theories. This provided crucial reason for doubt about a key model seeking to explain conspiracy theories.

Another popular hypothesis is that developing a belief in one theory can make you more likely to believe others; that developing a belief in one conspiracy theory can lead you down a “rabbit hole”. This idea has been enormously influential in the academic literature, but its key causal claims haven’t previously been subjected to a credible test. Williams et al. (2025) were able to do so, using thirteen waves of data from the longitudinal study. This time, the team did find evidence in favour of the hypothesis. However, the causal effects were very small. It seems that if conspiracy theorists fall down a rabbit hole, it might often be one with a rather gradual slope – perhaps a bit more like a real rabbit warren than the rabbit hole in Alice’s Adventures in Wonderland!

Williams and his colleagues are continuing to conduct research on conspiracy theories, using more data from the longitudinal study as well as other sources. For example, they’ve recently been examining how the apparent prevalence of belief in conspiracy theories might be distorted by insincere survey responses (e.g. ‘trolling’). Despite the attention conspiracy theories have received both inside and outside the academic literature, there is still lots more to learn!