Differential patterns of 2D location versus depth decoding along the visual hierarchy.

Area of Science:

• Neuroscience
• Visual Perception
• Cognitive Science

Background:

• The human brain transforms 2D retinal images into a 3D perception of the environment.
• Previous research often studied 2D spatial representation and depth perception separately.
• The interaction between 2D and 3D spatial processing in the visual cortex is not well understood.

Purpose of the Study:

• To investigate the relationship and interaction between 2D location and position-in-depth information processing in the human visual cortex.
• To map how these spatial representations change across different visual areas.

Main Methods:

• Functional magnetic resonance imaging (fMRI) and multi-voxel pattern analysis (MVPA) were employed.
• Participants viewed stimuli at different 3D locations (defined by horizontal, vertical, and depth) while fixating.
• Red/green anaglyph glasses were used to present stereoscopic depth information.

Main Results:

• 2D location information (horizontal, vertical) was strongly decodable in early visual areas, decreasing in later areas.
• Position-in-depth information decoding increased from intermediate to higher visual and category regions.
• Depth decoding showed an inverse relationship with 2D location decoding patterns, becoming more tolerant to 2D location changes in higher areas.

Conclusions:

• Visual cortex exhibits a systematic transition in spatial information processing along its hierarchy.
• Early visual areas prioritize 2D location, while later areas increasingly integrate depth information.
• Spatial representations shift from 2D-dominant to a balanced 3D representation (integrating 2D and depth) in higher visual areas.