Hierarchical models for neural population dynamics in the presence of non-stationarity

TL;DR

This paper introduces a hierarchical statistical model for neural population activity that captures multiple sources of variability, including non-stationarities in firing rates and correlations, improving understanding of neural dynamics.

Contribution

The authors develop a hierarchical model with variational inference to capture non-stationarities in neural population dynamics and correlations, advancing analysis of neural variability.

Findings

Model recovers non-stationarities in firing rates

Model better fits neural data than stationary models

Captures correlation structure changes over time

Abstract

Neural population activity often exhibits rich variability and temporal structure. This variability is thought to arise from single-neuron stochasticity, neural dynamics on short time-scales, as well as from modulations of neural firing properties on long time-scales, often referred to as "non-stationarity". To better understand the nature of co-variability in neural circuits and their impact on cortical information processing, we need statistical models that are able to capture multiple sources of variability on different time-scales. Here, we introduce a hierarchical statistical model of neural population activity which models both neural population dynamics as well as inter-trial modulations in firing rates. In addition, we extend the model to allow us to capture non-stationarities in the population dynamics itself (i.e., correlations across neurons). We develop variational inference methods for learning model parameters, and demonstrate that the method can recover non-stationarities in both average firing rates and correlation structure. Applied to neural population recordings from anesthetized macaque primary visual cortex, our models provide a better account of the structure of neural firing than stationary dynamics models.