Proceedings of the Annual Meeting of the Cognitive Science Society

Refutation texts are effective for revising beliefs in neutral, science misconceptions. We investigated if refutation could be similarly effective for revising identity-based misconceptions (i.e., learning styles). In two studies, participants took a pretest assessing misconception endorsement, read a refutation text, and completed a posttest. In Study 1, refutation reduced belief in the misconception, but did not impact identification with a learning style. However, participants who revised their misconception demonstrated better memory for the text. Study 2 added a feedback manipulation to improve memory. Surprisingly, feedback condition did not predict memory for the text or belief change. However, Study 2 replicated key findings from Study 1 about belief, identification, and memory for the text. Open-response data from both studies suggest personal experience plays a role in the persistence of the misconception while learning explicit evidence facilitates belief revision. We discuss these findings and their implications for belief change in identity-based misconceptions.