The Topology and Geometry of Neural Representations

TL;DR

This paper introduces topological representational similarity analysis (tRSA), a novel method that characterizes brain and neural network representations by their topology, offering robustness to noise and individual differences while distinguishing different functional regions.

Contribution

The paper presents tRSA, an extension of RSA that incorporates topological summaries to better characterize neural representations beyond geometry, enhancing robustness and specificity.

Findings

Topology-sensitive metrics are robust to noise and individual variability.

tRSA effectively distinguishes neural representations in both simulations and fMRI data.

The method improves model and brain region comparison sensitivity.

Abstract

A central question for neuroscience is how to characterize brain representations of perceptual and cognitive content. An ideal characterization should distinguish different functional regions with robustness to noise and idiosyncrasies of individual brains that do not correspond to computational differences. Previous studies have characterized brain representations by their representational geometry, which is defined by the representational dissimilarity matrix (RDM), a summary statistic that abstracts from the roles of individual neurons (or responses channels) and characterizes the discriminability of stimuli. Here we explore a further step of abstraction: from the geometry to the topology of brain representations. We propose topological representational similarity analysis (tRSA), an extension of representational similarity analysis (RSA) that uses a family of geo-topological summary statistics that generalizes the RDM to characterize the topology while de-emphasizing the geometry. We evaluate this new family of statistics in terms of the sensitivity and specificity for model selection using both simulations and fMRI data. In the simulations, the ground truth is a data-generating layer representation in a neural network model and the models are the same and other layers in different model instances (trained from different random seeds). In fMRI, the ground truth is a visual area and the models are the same and other areas measured in different subjects. Results show that topology-sensitive characterizations of population codes are robust to noise and interindividual variability and maintain excellent sensitivity to the unique representational signatures of different neural network layers and brain regions. These methods enable researchers to calibrate comparisons among representations in brains and models to be sensitive to the geometry, the topology, or a combination of both.