Department of Neurobiology, University of California at Los Angeles
The stochastic nature of single neurons
I will discuss an integrated approach to better understanding retinal circuitry through closely coordinated computational modeling and physiology. Numerical simulations of a circuit model of salamander retina predicted a hitherto unsuspected dynamic interaction in the inner retina resulting from time-lagged inhibition of a transiently laterally-expanding representation of an object. Motivated by this prediction, we developed a method for reconstructing the spatiotemporal patterns of neural activity present at any layer of the retina from single-unit recordings. This method can be extended to multiunit recording in order to take into account possible correlations between neurons.