The Structured `Low Temperature' Phase of the Retinal Population Code

TL;DR

This paper investigates the collective structure of neural population activity, revealing a robust phase transition linked to attractor-like states and clustering, which suggests a generic, structured collective neural state.

Contribution

It demonstrates that the observed phase transition in neural data models indicates a structured, collective state with stable correlations and attractor-like energy landscape features.

Findings

Phase transition is robust to stimulus and state changes.

Neural correlations are strong and do not require fine tuning.

Emergence of attractor-like structures correlates with phase transition.

Abstract

Recent advances in experimental techniques have allowed the simultaneous recording of populations of hundreds of neurons, allowing more comprehensive investigation into the nature of the collective structure of population neural activity. While recent studies have reported a phase transition in the parameter space of maximum entropy models describing the neural probability distributions, the interpretation of these findings has been debated. Here, we suggest that this phase transition may be evidence of a `structured', collective state in the neural population. We show that this phase transition is robust to changes in stimulus ensemble and adaptive state. We find that the pattern of pairwise correlations between neurons has a strength that is well within the strongly correlated regime and does not require fine tuning, suggesting that this state is generic for populations of 100+ neurons. We find a clear correspondence between the emergence of a phase transition, and the emergence of attractor-like structure in the inferred energy landscape. A collective state in the neural population, in which neural activity patterns naturally form clusters, provides a consistent interpretation for our results.