Delineation of auditory-motor population dynamics underlying sensorimotor integration in the birdsong system

PROJECT SUMMARY Speech is a closed-loop behavior which requires the brain to continuously perceive and produce acoustic signals in real time. Current neurobiological theories of speech posit that neural population activity across auditory and motor regions is dynamically coupled during speech production, but that speech perception relies on auditory processing alone. This sensorimotor integration hypothesis would allow the brain to exploit immediate auditory feedback to fine-tune the motor actions that elicit speech. Rigorous neurobiological tests of sensorimotor integration require (1) a model system that enables the control and measurement of sensorimotor behaviors, (2) the experimental expertise to conduct large-scale neural recordings simultaneously in sensory and motor regions, and (3) the computational abilities to develop population scale analyses that assess coordination in the distributed dynamics of individual neurons. This proposal presents a synergistic combination of experiments and analyses that meet these requirements: Simultaneous recordings and perturbations of both auditory and motor regions in European starlings during birdsong production and perception are combined with novel topological data analyses (TDA) to uncover the population mechanisms that instantiate sensorimotor integration. European starlings are an ideal organism for understanding neurobiological mechanisms that support sensorimotor integration; they produce and rely on complex vocal communication signals and have a long history of use in invasive electrophysiology studies. The overarching goal of this proposal is to investigate how distributed neuronal population activity integrates auditory and motor information during closed-loop behavior—specifically birdsong. The central hypothesis of this proposal is that auditory and motor population activity is uniquely coupled when birds sing, in contrast to when birds listen to song. This hypothesis will be tested through the following specific aims: In Aim 1, simultaneously recording auditory and motor regions while birds sing and listen to song will enable an understanding of how population activity is coordinated across regions. In Aim 2, recordings from auditory regions with concurrent optogenetic inhibition of motor regions while birds sing and listen to song will enable a delineation of causal interactions between regions. Novel TDA will be used to quantify the coordination of neural activity across regions and through time, enabling mechanistic insight into how population dynamics structure song behavior. Contrasting population activity across auditory and motor areas between singing and listening will allow for the identification of dynamics unique to sensorimotor integration. In the near-term, this proposal provides a mechanistic understanding of how neuronal populations coordinate to perform sensorimotor integration in the songbird system. In the long-term, this approach will enable future research into how brain network dynamics support closed-loop behaviors, such as speech. Ultimately, this proposal will enable the training and development of a unique and synergistic combination of skills that has the potential to provide novel insight into the neurobiological mechanisms of sensorimotor integration.