Dissociating explicit and procedural-learning based systems of perceptual category learning.

Area of Science:

• Cognitive Psychology
• Neuroscience
• Machine Learning

Background:

• A core debate in cognitive science concerns whether category learning relies on a single system or multiple specialized systems.
• Existing theories often propose distinct explicit and implicit systems, though the precise nature of the implicit system remains debated.

Purpose of the Study:

• To review and evaluate the Competition Between Verbal and Implicit Systems (COVIS) dual systems theory of category learning.
• To examine empirical evidence supporting key predictions derived from the COVIS framework.

Main Methods:

• Review of nine empirical studies designed to test specific, a priori predictions of the COVIS model.
• Analysis of findings related to the neural mediation and learning characteristics of explicit and implicit category learning systems.

Main Results:

• The explicit system, associated with rule-based learning and frontal brain regions, effectively learns verbalizable categories.
• The implicit system, linked to procedural learning and subcortical structures (caudate nucleus, dopamine reward), excels at information-integration categories.
• All six tested predictions derived from the COVIS theory were consistently supported by the reviewed data.

Conclusions:

• The COVIS dual systems theory provides a robust framework for understanding category learning, differentiating explicit and implicit processes.
• Neurobiological evidence supports the distinct neural substrates underlying rule-based versus information-integration category learning.
• The findings underscore the importance of considering multiple, interacting systems in human learning.