Computational model of avian nervous system nuclei governing learned song

TL;DR

This paper presents a computational model of the avian song system that links neural activity in HVC and downstream nuclei to song production, reproducing observed neural and acoustic correlations and making testable predictions.

Contribution

It introduces a novel functional architecture model of the songbird's neural pathway that explains song timing and structure, aligning with experimental data.

Findings

Reproduces correlations between neural activity and song features

Predicts electrophysiological properties of song system neurons

Suggests specific connectivity patterns in the song motor pathway

Abstract

The means by which neuronal activity yields robust behavior is a ubiquitous question in neuroscience. In the songbird, the timing of a highly stereotyped song motif is attributed to the cortical nucleus HVC, and to feedback to HVC from downstream nuclei in the song motor pathway. Control of the acoustic structure appears to be shared by various structures, whose functional connectivity is largely unknown. Currently there exists no model for functional synaptic architecture that links HVC to song output in a manner consistent with experiments. Here we build on a previous model of HVC in which a distinct functional architecture may act as a pattern generator to drive downstream regions. Using a specific functional connectivity of the song motor pathway, we show how this HVC mechanism can generate simple representations of the driving forces for song. The model reproduces observed correlations between neuronal and respiratory activity and acoustic features of song. It makes testable predictions regarding the electrophysiology of distinct populations in the robust nucleus of the arcopallium (RA), the connectivity within HVC and RA and between them, and the activity patterns of vocal-respiratory neurons in the brainstem.