Proceedings of the Annual Meeting of the Cognitive Science Society

Repeated exposure to information increases its perceived truth, a phenomenon known as the illusory truth effect. Beyond influencing truth judgments, repetition also increases individuals' confidence judgments. Existing theoretical models generally assume that confidence is lowest when statements are maximally ambiguous and highest when their truth status is known. However, it is currently unclear how the relationship between perceived truth and confidence can be formally modeled and whether such models are consistent with empirical data. In this study, we developed a Bayesian model to characterize the link between confidence and truth judgments, with several model variants capturing alternative assumptions about this relationship. The model also accounts for the effect of repetition on both variables. Fitting these models to two datasets provided evidence for an asymmetric, V-shaped relationship between truth and confidence. Specifically, individuals showed higher confidence when judging highly true statements than when judging highly false ones, with confidence reaching a minimum for ambiguous statements.