Schedule of Events | Search Abstracts | Invited Symposia | Symposia | Poster Sessions | Data Blitz

Categorical Judgments: Examining the Learned Structure Derived from Implicit Statistical Learning

Poster Session D - Monday, March 9, 2026, 8:00 – 10:00 am PDT, Fairview/Kitsilano Ballroom

Alexander N. Lawriw¹ (), Christopher R. Cox¹; ¹Louisiana State University

Implicitly tracking transitional probabilities in our environment, termed statistical learning, facilitates early language acquisition and efficient perceptual processing. Recent work suggests that transitional probabilities may also aid rapid category formation and thereby inform explicit inferences about shared, non-temporal features. After viewing a sequence of items constructed from a random walk along a graph with community structure, participants tend to group items from the same temporal community, particularly when oriented toward some novel, invisible feature that items might share. While this suggests that implicitly learned structure can shape explicit similarity judgements, such behavior could reflect either inferring categories or simply tracking transitional probabilities. As a stronger test of category inference, we had participants learn a different fact about one member from each community before viewing the sequence. Then, rather than grouping pairs of items, we directly asked if the fact applied to other items from the same or different communities. Early results (n = 60) suggested a general bias towards selecting the item from the same temporal community as the relevant trained item, consistent with categorical generalization. However, post-hoc power analyses revealed that the study was underpowered to detect potential effects in this modified task, particularly across individual conditions. To address this limitation, we are collecting data from a new, larger sample (n = 180) to better assess the reliability and magnitude of the observed effects.

Topic Area: LONG-TERM MEMORY: Semantic