Abstract
The theory of category learning called COVIS (COmpetition between Verbal and Implicit Systems) posits that separate brain systems mediate perceptual category rule acquisition. Verbalizable rules are learned explicitly; non-verbalizable rules are learned implicitly. Although the tenets of COVIS have been supported by behavioral and neuroimaging data, few studies have assessed subjective awareness of the strategy learners use in explicit and implicit category rule learning conditions. We conducted a study to examine perceptual classification of visual stimuli (Gabor stimuli varying in spatial frequency and angle of orientation) while also assessing subjective experience of rule learning and use. Participants completed either rule-based (RB; explicit) or information-integration (II; implicit) category learning. After every 40-trial block, participants were asked whether they knew the category rule and to try to describe it in their own words. By the final block, individuals who indicated that they knew the rule over the final 3 blocks of training achieved high accuracy (84% in both conditions). Participants in the RB condition who reported not knowing the rule had accuracy slightly above chance (60%). Many participants in the II condition indicated they did not know the rule, yet nonetheless achieved high accuracy (68%). The results clearly indicate limited ability to articulate the information-integration rule, despite high classification accuracy, as predicted by COVIS for visual stimulus categories.
Meeting abstract presented at VSS 2018