Dissociation of Category-Learning Mechanisms via Brain Potentials

Behavioral, neuropsychological, and neuroimaging evidence indicate categories can be learned either via an explicit rule-based mechanism dependent on medial temporal and prefrontal brain regions, or via an implicit information integration mechanism relying on the basal ganglia and occipital cortex. In this study, participants viewed Gabor patches that varied on two dimensions, and learned categories via feedback. Different stimulus distributions can encourage participants to favor explicit rule-based or implicit information integration mechanisms. We monitored brain activity with scalp encephalography while participants (1) passively observed Gabor patches, (2) categorized patches from one distribution, and, one week later, (3) categorized patches from another distribution. Categorization accuracy was matched across the two learning conditions, which nevertheless elicited several distinct event-related potentials. These results demonstrate the efficacy of real-time neural monitoring during category learning and provide additional evidence implicating different neurocognitive mechanisms in explicit rule-based versus implicit information integration category learning.