Texture discriminability in monkey inferotemporal cortex predicts human texture perception.

Area of Science:

• Neuroscience
• Visual Perception
• Computational Neuroscience

Background:

• Shape and texture are crucial visual object properties, yet texture perception is less understood.
• Classic human texture discrimination findings include element dependency, asymmetries, and reduced discriminability for random orientations.

Purpose of the Study:

• To characterize neuronal responses to discrete textures in monkey inferotemporal (IT) cortex.
• To determine if these neuronal responses can explain established findings in human texture perception.

Main Methods:

• Recorded neuronal activity from monkey IT cortex while presenting various textures.
• Measured human texture discrimination performance for the same set of textures.
• Correlated neuronal selectivity and firing rate variations with human perceptual data.

Main Results:

• Monkey IT neurons exhibited congruent texture selectivity across different array sizes.
• Easily discriminable human textures elicited distinct neuronal activity patterns in monkeys.
• Asymmetric human texture discrimination corresponded to asymmetric firing rate variations in IT neurons.
• Reduced discriminability of random-oriented textures by IT neurons predicted human performance.

Conclusions:

• Neuronal representations in monkey IT cortex align with human texture discrimination capabilities.
• Findings suggest that human texture perception relies on neural mechanisms similar to those in monkey IT cortex.
• The study provides a neural basis for understanding texture perception phenomena.