Task learning increases information redundancy of neural responses in macaque visual cortex

TL;DR

This study shows that learning new visual discrimination tasks in macaques increases neural response redundancy in V4, enhancing information capacity and supporting a Bayesian inference model of sensory processing.

Contribution

It provides empirical evidence that task learning increases neural redundancy, aligning with Bayesian inference rather than efficiency-based hypotheses.

Findings

Redundancy in neural responses increases with learning.

Increased redundancy enhances information carried by individual neurons.

Supports Bayesian inference as a model of sensory processing.

Abstract

How does the brain optimize sensory information for decision-making in new tasks? One hypothesis suggests learning reduces redundancy in neural representations to improve efficiency, while another, based on Bayesian inference, predicts learning increases redundancy by distributing information across neurons. We tested these hypotheses by tracking population responses in macaque cortical area V4 as monkeys learned visual discrimination tasks. We found strong support for the Bayesian predictions: task learning increased redundancy in neural responses over weeks of training and within single trials. This redundancy did not reduce information but instead increased the information carried by individual neurons. These insights suggest sensory processing in the brain reflects a generative rather than discriminative inference process.