Structural Statistical Properties of the Connectivity Could Underlie the Difference in Activity Propagation Velocities in Visual and Olfactory Cortices

Abstract

We show experimentally that the properties of the propagation of activity in cortical slices depend critically on the cortical area explored. Thus, olfactory cortex slices present a much faster speed of propagation than neocortical slices. In order to explore the possibility that this reveals different statistical properties of the underlying synaptic connectivity, we study the small-world properties of a computational network model of slow oscillatory activity that we have previously shown to replicate closely the activity in the slice. We show that for the Gaussian probability connectivity used, progressive reduction of the Gaussian spread makes the network transition from a random, to a small-world and to an ordered network. We then relate the small-world parameters of the connectivity to the velocity of activity propagation in the model. We conclude that the locality parameter C, and not the mean path length L, determines primarily the velocity of propagation.