Abstract
Base-rate sensitivity is known to develop over time through direct experience in classification tasks without explicit presentation of frequency information. Sensitivity is commonly assumed to develop through implicit learning, and recent empirical work supports this claim (Bohil & Wismer, 2015). However, the relative contributions of implicit and explicit learning to base-rate sensitivity have yet to be addressed empirically. The goal of the present work was to dissociate the roles of the implicit and the explicit systems by using a factorial design that included both implicit and explicit learning disruptions. We tested the hypothesis that implicit learning underlies base-rate sensitivity from experience (and that explicit learning contributes little). Participants classified simple stimuli (bar graph heights) with a 3:1 base-rate ratio. Participants learned from either "observational" training known to disrupt implicit learning or from "response" training which supports implicit learning. Category label feedback was followed either immediately or after a 2.5 second delay by a working memory task designed to disrupt explicit reasoning about feedback. Decision criterion values were more conservative after observational training, suggesting that implicit learning underlies base-rate sensitivity. Disrupting explicit processing had no effect on base-rate learning as long as implicit learning was supported. A follow-up study that presented half of the participants with explicit base-rate frequency information (in place of the dual working memory task) again found more conservative decision criterion values when the implicit system was disrupted. Additionally, explicit base-rate information did little to promote a more liberal criterion. Taken together, these results suggest that base-rate sensitivity develops primarily through implicit learning, as predicted by the COVIS theory of categorization (Ashby et al., 1998).
Meeting abstract presented at VSS 2016