Seminar Series

Population activity in the cerebellar cortex that mediates behavior and learning

Court Hull

Associate Professor, Department of Neurobiology, Duke University

host: Greg Horwitz

Abstract: The cerebellum is a key brain region involved in associative learning, and in particular for generating predictive sensorimotor associations.  To mediate such learning, convergent input from two main pathways is thought to be required; the climbing fiber and granule cell pathways.  Using multiphoton imaging in awake behaving mice, we have investigated how each of these pathways encodes the sensory and motor information necessary for learning.  These studies have revealed surprising results that extend current views of cerebellar learning.  Specifically, we have found that cerebellar climbing fibers can exhibit reward-related responses that are consistent with many of the predictions of reinforcement learning, in contrast with the long-held view that the cerebellum operates exclusively according to supervised learning principles.  In addition, we have found that granule cells generate sparse population codes that rely on local synaptic inhibition to enable pattern separation and learned sensorimotor discriminations.  I will discuss the implications of these results in the context of cerebellar associative learning.

Awake perception is associated with dedicated neuronal assemblies in cerebral cortex

Alain Destexhe

Saclay Paris and European Institute for Theoretical Neuroscience

Host: Adrienne Fairhall

Spiking codes for skilled motor control

Sam Sober, PhD

Director of the Simons-Emory International Consortium on Motor Control

Emory University

Abstract: Neurons coordinate patterns of muscle activity to produce an astonishing variety of behaviors. However, the biological and computational bases of sensorimotor control remain mysterious, in part due to a lack of experimental hardware and computational frameworks for examining motor signals. To address these challenges, my group combines physiological, computational, and engineering approaches to understand motor control across species and behaviors. My talk will provide an overview of three projects. First, physiological and computational studies of vocal production in songbirds reveal that neurons in the motor system employ millisecond-resolution spike timing codes to regulate vocal behavior, highlighting the need to examine spiking codes across cortical, basal ganglia, and spinal motor circuits. Second, to examine spiking codes across motor circuits, we have developed electrode arrays for examining spiking patterns in muscle tissue during natural behaviors. These “Myomatrix” arrays reveal the spatiotemporal structure of muscle activity at motor-unit resolution across effectors (forelimb, trunk, orofacial, respiratory, and vocal muscles) and species – including humans – during unconstrained behavior. Third, in-progress studies examining muscle spike trains in mice (locomotion) and monkeys (reaching movements) reveal how complex patterns of motor unit activity shape skilled forelimb control. https://scholarblogs.emory.edu/soberlab/  

seminar:

Speaker

Institution

Abstract: website: