Reflection predicts and leads to decreased conspiracy belief

Abstract

Recent research indicates a generally negative relationship between reflection and conspiracy beliefs. However most of the existing research relies on correlational data on WEIRD (Western Educated Industrialized Rich Democratic) populations. The few existing experimental studies are limited by weak manipulation techniques that fail to reliably activate cognitive reflection. Hence questions remain regarding (1) the consistency of the negative relationship between conspiracy beliefs and cognitive reflection (2) the extent of cross-cultural variation and potential moderating factors and (3) the presence of a causal link between cognitive reflection and conspiracy beliefs. In two preregistered studies we investigated the association between cognitive reflection and conspiracy beliefs. First we studied the correlation between two variables across 48 cultures and investigated whether factors such as WEIRDness and narcissism (personal and collective) moderate this relationship. In the second study we tested the causal effect of reflection using a reliable and effective manipulation technique—debiasing training—on both generic and specific conspiracy beliefs. The first study confirmed the negative association between reflection and belief in conspiracy theories across cultures with the association being notably stronger in non-WEIRD societies. Both personal and collective narcissism played significant moderating roles. The second study demonstrated that debiasing training significantly decreases both generic and COVID-19 conspiracy beliefs in a non-WEIRD context with more pronounced effects for general conspiracy beliefs. Our research supports that reflection is a consistent cross-cultural predictor of conspiracy beliefs and that activating reflection can reduce such beliefs through rigorous experimental interventions. © 2025 Elsevier B.V. All rights reserved.