Sensorimotor Integration in Parallel Cortico-Basal Ganglia Loops During Vocal Learning

University Of Southern California, Los Angeles CA

Investigators

Linked publications & trials

Abstract

? DESCRIPTION (provided by applicant): Motor skill learning during development is a process of sensorimotor integration that involves iterative comparisons between a current behavior and a goal behavior. Although cortico-basal ganglia circuits are known to mediate motor skill learning, little is known about the precise mechanisms underlying comparisons of current and goal behavior during development. Vocal learning in songbirds provides an excellent model for investigating how such evaluations are carried out: similar to speech acquisition in humans, juvenile songbirds evaluate feedback of their own immature song against those of an adult tutor (goal behavior) and gradually refine their song to achieve an accurate imitation of the tutor song. In juvenile songbirds, neural control of the vocal learning process is vested in two parallel circuits that traverse basal ganglia and cortex: a motor circuit drives vocalizations, while an evaluative circuit is necessary for developing an accurate copy of the tutor song. Projection neurons in the motor circuit that drive vocal motor output respond to a variety of vocal stimuli and can thus serve as broadly tuned auditory filters that encode auditory feedback of any self-produced sounds. These neurons send a corollary projection into a cortical target region, AId, of the evaluative circuit, thereby conveying feedback of the juvenile's vocalizations or efference copy of vocal output to AId. AId also receives input from a population of neurons within the evaluative circuit that respond selectively to playback of the tutor song. AId thus has access to a neural representation of the goal behavior in addition to information about the juvenile's self-produced vocalizations (from the motor circuit) specifically during the vocal learning period, making it well-suited to evaluate self-produced vocal sounds against the tutor song template to direct refinement of the bird's own song. This proposal will test the hypothesis that AId is essential for comparing the juvenile bird's immature vocalizations against the goal behavior during sensorimotor learning. A crucial first step to understanding how information about current and goal behavior is compared is to assess how the information is represented at the site of convergence in AId. Aim 1 will therefore assess the neuronal response properties of AId in juvenile birds in two conditions: (1) extracellular recordings in AId during playback of various song stimuli to investigate the auditory response to vocal stimuli that are important for learning, and (2) extracellular recordings in AId in singing juvenile birds to test whether comparative signals in AId are activated by efference copy input. Aim 2 will further test this hypothesis by altering normal activity in AId of singing juvenile birds via syllable-targeted optogenetic stimulation. If AId is a necessary site of evaluation, such disruption during vocal learning should prevent accurate imitation of the tutor song. These experiments will inform our understanding of basic mechanisms of sensorimotor integration in cortico-basal ganglia circuits - a fundamental process that applies to a wide range of learning processes throughout development.