Category learning induces transfer of perceptual learning by steering Feature-Based Attention

Poster Session F - Tuesday, April 16, 2024, 8:00 – 10:00 am EDT, Sheraton Hall ABC

Luke Rosedahl¹ (luke_rosedahl@brown.edu), Thomas Serre¹, Takeo Watanabe¹; ¹Brown University

Visual Perceptual Learning (VPL; often defined as long-term performance increase due to visual experience) is highly specific to trained features. Previous work found that category learning can cause VPL to transfer across features to stimuli from the same category as the trained stimulus (Category-learning Induced Transfer of VPL or CIT-VPL; Wang et al., 2018, Current Biology). However, the mechanism of transfer is unknown. Based on work showing that Feature-Based Attention (FBA) can increase within-category stimulus similarity (Brouwer and Heeger, 2013), here we postulate that CIT-VPL occurs through FBA. We test this hypothesis utilizing two category structures that differ in the optimal FBA allocation strategy: Rule-Based (RB) and Information-Integration (II). RB structures benefit from targeting FBA to specific feature values while II structures do not. We use these structures to test three predictions from the theory: 1) RB structures will cause greater VPL transfer than II structures, 2) changes in visual cortex GABA concentration will correlate with VPL transfer, and 3) changes in functional connectivity between visual cortex and the inferior frontal junction (a region involved in controlling FBA) will correlate with VPL transfer. Across two experiments, the results confirmed the predictions of the theory. We then implement a neural network model that learns to apply feature-specific feedback (gain) modulation during category learning. We demonstrate that feedback connections enable the network to show the same behavior as human participants. Overall, this work provides computational, neural, and behavioral evidence for feature-based attention being the mechanism for category-learning induced transfer of VPL.

Topic Area: PERCEPTION & ACTION: Vision