Purchase this article with an account.
Audrey Hill, Andrew Wismer, Corey Bohil; A Subjective Measure of Explicit and Implicit Category Rule Learning . Journal of Vision 2016;16(12):401. doi: https://doi.org/10.1167/16.12.401.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The neuropsychological theory of categorization known as COVIS (COmpetition between Verbal and Implicit Systems), postulates that distinct brain systems compete during learning of category rules (Ashby, Alfonso-Reese, Turken, & Waldron, 1998). The explicit learning system, mediated by the prefrontal cortex, involves conscious hypothesis testing about easily-verbalizable rules, while the implicit learning system, mediated primarily by basal ganglia structures, relies on procedural learning of rules that are difficult to verbalize. Although an enormous amount of behavioral data supports COVIS, it is unclear what participants understand about the category rule when it is learned implicitly. The current study was designed to gain a deeper understanding of the nature of implicit category rule learning. Using simple two-dimensional perceptual stimuli – lines of varying length and orientation -- participants were trained on either explicitly learnable rule-based (RB) or procedurally learned information-integration (II) category structures. Using an adaptation of Dienes & Scott's (2005) measure of unconscious knowledge, participants made trial-by-trial assessments attributing each categorization response to guessing, intuition (a marker of implicit learning), or rule use. Categorization accuracy for learners (>60% accuracy by end of training) was high in both conditions (~80%). Participants attributed substantially more responses to "rule" or "intuition" use than to guessing. Participants in the RB condition overwhelmingly gave the "rule" response compared to the "intuition" response. Participants in the II condition provided the "intuition" attribution significantly more than in the RB condition. These results provide a new form of support for the predictions of COVIS. The measurement scale provides valuable insight into learners' subjective understanding of implicitly-learned classification rules.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only