Quantifying dynamical high-order interdependencies from the O-information: an application to neural spiking dynamics

TL;DR

This paper introduces a dynamic, conditioned version of the O-information to quantify high-order interdependencies in neural systems, revealing complex synergistic and redundant interactions during neural activity.

Contribution

It develops a novel dynamical, conditioned O-information framework that separates synergistic and redundant effects in high-order neural interdependencies.

Findings

Identifies synergistic neuron groups not evident from individual analysis

Tracks changes in interdependence structure during neural task performance

Separates synergistic and redundant information contributions

Abstract

We address the problem of efficiently and informatively quantifying how multiplets of variables carry information about the future of the dynamical system they belong to. In particular we want to identify groups of variables carrying redundant or synergistic information, and track how the size and the composition of these multiplets changes as the collective behavior of the system evolves. In order to afford a parsimonious expansion of shared information, and at the same time control for lagged interactions and common effect, we develop a dynamical, conditioned version of the O-information, a framework recently proposed to quantify high-order interdependencies via multivariate extension of the mutual information. We thus obtain an expansion of the transfer entropy in which synergistic and redundant effects are separated. We apply this framework to a dataset of spiking neurons from a monkey performing a perceptual discrimination task. The method identifies synergistic multiplets that include neurons previously categorized as containing little relevant information individually.