David DiGregorio: sparse encoding in granule cells via dynamic synapses

The cerebellar cortex is a prototypical brain circuit important for fine-tuning precise motor and cognitive behaviors on the subsecond time scale. Synaptic connections between neurons change their strength dynamically during brief bouts of activity, and we hypothesize that they could, therefore, act as a cellular substrate for encoding time within neural networks. I will summarize the theoretical underpinnings of how diverse forms of short-term synaptic plasticity can serve as a substrate for a biological clock, and I will present experimental evidence supporting the role of dynamic synapses in sculpting neural dynamics within the cerebellum. We found that prolonged mossy fiber activity is transformed into temporally sharpened, sparse granule cell population sequences that tile sensory events in time. Using high-speed multiphoton calcium and glutamate imaging together with synaptic recordings and modeling, we identified heterogeneous mossy fiber–granule cell synaptic strength and short-term plasticity as the mechanisms underlying region-specific temporal sparsification. These findings establish heterogeneous synaptic dynamics as a biological substrate for shaping population activity in time, setting the temporal precision of sensorimotor associations underlying adaptive behavior. David DiGregorio is Professor and Chair of the Department of Physiology and Biophysics at the University of Colorado Anschutz Medical Campus. He trained in biology at Stanford, earned his Ph.D. in neuroscience at UCLA, and then completed postdoctoral work at University College London before establishing an independent research program at the CNRS and Institut Pasteur in Paris, where he was Director of the Neuroscience Department. He has made foundational contributions to understanding how short-term synaptic plasticity and dendritic integration generate temporal basis functions that support precisely timed behavior. His laboratory has pioneered optical and biophysical strategies for monitoring voltage, glutamate, and calcium dynamics, including generative-model approaches for high-fidelity spike inference from calcium imaging data.