Multidimensional Feature Tuning in Category Selective Areas of Human Visual Cortex

Area of Science:

• Neuroscience
• Cognitive Science
• Visual Perception

Background:

• Human high-level visual cortex is described by categorical (category-selective areas) and dimensional (continuous feature maps) views.
• The relationship between these two organizational accounts remains unclear.

Purpose of the Study:

• To investigate if categorical and dimensional views represent complementary aspects of a single organizational principle in the human visual cortex.
• To identify underlying functional organization that reconciles these two prominent accounts.

Main Methods:

• Used functional magnetic resonance imaging (fMRI) to record brain activity in participants viewing natural images.
• Applied data-driven decomposition of fMRI responses within face-, body-, and scene-selective areas.
• Analyzed spatially overlapping activity patterns shared across individuals.

Main Results:

• Identified interpretable dimensions within category-selective areas that capture both within-category and between-category distinctions.
• These dimensions were present even in highly category-selective voxels.
• Dimensions formed distinct clusters within areas but extended as distributed maps across the visual cortex.

Conclusions:

• The findings reveal an underlying functional organization that integrates category-selective areas with continuous feature maps.
• This reconciles the categorical and dimensional accounts of high-level visual cortex organization.
• Explains how visual cortex balances specificity and flexibility for perception and cognition.