Complex/Dynamical Systems Seminar - Kevin K Lin

About

Kevin Lin, Department of Mathematics, University of Arizona

A multiscale approach to modeling cortical circuits

This talk introduces a class of multiscale, coarse-grained (CG) models of neuronal networks aimed at balancing biological fidelity and computational efficiency. As is typical of CG models, rather than individual cells we focus on the interaction between groups of cells. Dynamics are alternately updated at intra- and inter-group scales until equilibrium is achieved on both scales. A novelty is how we exploit the underlying biology: taking advantage of the similarity in local anatomical structures across large regions of cortex, we model intra-group dynamics as a *single* dynamical system driven by "external" inputs. This lets us precompute and tabulate all potential local responses, thereby speeding up the updating procedure significantly compared to direct multiscale simulations. This methodology is illustrated using a model of the primate visual cortex.