Unsupervised Bayesian Ising Approximation for revealing the neural dictionary in songbirds

TL;DR

This paper introduces an unsupervised Bayesian Ising approximation method to identify precisely timed neural spike patterns that encode specific behaviors in songbirds, advancing understanding of neural codes and sensorimotor functions.

Contribution

The authors develop a novel Bayesian Ising approximation approach capable of detecting complex spike patterns from small datasets, accounting for dependencies, and revealing neural code dictionaries.

Findings

Successfully identified spike codewords associated with motor behaviors

Detected differences in neural encoding during exploration versus typical behavior

Method applicable to various biological and nonbiological datasets

Abstract

The problem of deciphering how low-level patterns (action potentials in the brain, amino acids in a protein, etc.) drive high-level biological features (sensorimotor behavior, enzymatic function) represents the central challenge of quantitative biology. The lack of general methods for doing so from the size of datasets that can be collected experimentally severely limits our understanding of the biological world. For example, in neuroscience, some sensory and motor codes have been shown to consist of precisely timed multi-spike patterns. However, the combinatorial complexity of such pattern codes have precluded development of methods for their comprehensive analysis. Thus, just as it is hard to predict a protein's function based on its sequence, we still do not understand how to accurately predict an organism's behavior based on neural activity. Here we derive a method for solving this class of problems. We demonstrate its utility in an application to neural data, detecting precisely timed spike patterns that code for specific motor behaviors in a songbird vocal system. Our method detects such codewords with an arbitrary number of spikes, does so from small data sets, and accounts for dependencies in occurrences of codewords. Detecting such dictionaries of important spike patterns --- rather than merely identifying the timescale on which such patterns exist, as in some prior approaches --- opens the door for understanding fine motor control and the neural bases of sensorimotor learning in animals. For example, for the first time, we identify differences in encoding motor exploration versus typical behavior. Crucially, our method can be used not only for analysis of neural systems, but also for understanding the structure of correlations in other biological and nonbiological datasets.