Explaining V1 Properties with a Biologically Constrained Deep Learning Architecture

TL;DR

This paper enhances CNN models with neuroscience-inspired components to better explain primary visual cortex (V1) neural properties, achieving state-of-the-art alignment with biological neural activity and advancing NeuroAI.

Contribution

It systematically integrates biologically constrained architectural features into CNNs, significantly improving their ability to replicate V1 neural properties.

Findings

Incorporation of center-surround antagonism improves model-V1 alignment

Architectural enhancements yield state-of-the-art explanation of V1 tuning properties

Models with these components better predict neural responses in V1

Abstract

Convolutional neural networks (CNNs) have recently emerged as promising models of the ventral visual stream, despite their lack of biological specificity. While current state-of-the-art models of the primary visual cortex (V1) have surfaced from training with adversarial examples and extensively augmented data, these models are still unable to explain key neural properties observed in V1 that arise from biological circuitry. To address this gap, we systematically incorporated neuroscience-derived architectural components into CNNs to identify a set of mechanisms and architectures that comprehensively explain neural activity in V1. We show drastic improvements in model-V1 alignment driven by the integration of architectural components that simulate center-surround antagonism, local receptive fields, tuned normalization, and cortical magnification. Upon enhancing task-driven CNNs with a collection of these specialized components, we uncover models with latent representations that yield state-of-the-art explanation of V1 neural activity and tuning properties. Our results highlight an important advancement in the field of NeuroAI, as we systematically establish a set of architectural components that contribute to unprecedented explanation of V1. The neuroscience insights that could be gleaned from increasingly accurate in-silico models of the brain have the potential to greatly advance the fields of both neuroscience and artificial intelligence.