Large-scale model of primary visual cortex provides single neuron resolution on a macro scale

Armen Kherlopian, Cornell University

Photo of Armen Kherlopian

A key difficulty in systems neuroscience is relating neural activity to functional connectivity in a cortical circuit. In the primary visual cortex, neuroimaging provides important population information that can be used to construct maps such as ones for orientation selectivity and spatial frequency preference, while electrophysiology provides vital high-temporal resolution information such as individual spike times and local field potentials. However, both neuroimaging and electrophysiology rely on knowledge from post mortem anatomical studies for connectivity information. In understanding the activity of networks of neurons, a computational approach has a unique role, namely, to link disparate functional elements and local populations characterized by electrophysiology, and bounding them by larger population properties gleaned from neuroimaging. In our current study we implement a large-scale network model to investigate the effects of neural connectivity in the context of non-traditional stimuli targeting features in natural scenes. With this approach we aim to discern the implications of neural connectivity in enabling computations for visual processing.

Abstract Author(s): Armen R Kherlopian<sup>1,3</sup> and Jonathan D Victor<sup>2,3</sup><