Ease of Retrieval Effects in Persuasion: A Self-Validation Analysis

Abstract
Three studies are reported examining a new explanation for ease of retrieval effects in persuasion. In each study, participants read a persuasive communication and were induced to generate either a low or high number of favorable or unfavorable thoughts in response. In conflict with the assumptions of most previous studies, the authors predicted and found that ease of retrieval effects occur primarily under high rather than low-elaboration conditions. Under high-elaboration conditions, people were more influenced by their thoughts when few rather than many were retrieved (ease of retrieval effect), and this was mediated by the confidence participants had in those thoughts. These findings are consistent with the self-validation hypothesis. Under low-elaboration conditions, participants based judgments more on the actual number of thoughts generated, reflecting a numerosity heuristic.