Using COVIS to investigate interactions between memory systems

Poster Session B - Sunday, March 30, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

Mitchell Mcturk¹ (mmcturk@uwo.ca), Priya Kalra¹, J. Paul Minda¹; ¹Western University

Categorization is a fundamental component of higher cognition, facilitated by the brain's procedural or declarative learning systems. While these systems have been studied independently, the proposed project aims to investigate their interaction using the Competition between Verbal and Implicit Systems (COVIS) model (Ashby et al., 1998), The COVIS model posits that procedural learning, supported by the dorsal striatum, relies on implicit, trial-and-error processes. In contrast, declarative learning, supported by the hippocampus and prefrontal cortex, involves explicit, rule-based reasoning. These systems interact and compete for access to response production during category learning (Ashby & O’Brien, 2005). According to COVIS, the category decision of the system with the greatest value for confidence will be selected, regardless of whether the two systems agree or disagree. However, Kalra et al. (2024) demonstrated that reaction times were slower when the two systems disagreed and faster when they agreed, suggesting that both systems’ category decisions contribute to response production. In that study, confidence in the procedural system varied while confidence in the declarative system was fixed. In the proposed study, healthy adult participants will learn to categorize stimuli based on a declarative shape rule while simultaneously learning procedurally through colour cues, but we will systematically vary confidence in each system as well as agreement between the two systems. We hypothesize that reaction times will be slowest in disagreement trials with high confidence in both systems. These findings will expand our growing understanding of the interaction between procedural and declarative learning systems.

Topic Area: LONG-TERM MEMORY: Other