Information propagation in recurrent neuronal populations with mixed excitatory–inhibitory synaptic connections

Abstract

Information propagation in the cortical neural network has attracted a large amount of attention. In this paper, a feedforward network is constructed with each subnetwork being a recurrent neuronal population, consisting of neurons coupled by excitatory and inhibitory synapses, in which neurons have both types of outgoing synapses. Namely, neurons in the constructed feedforward neuronal networks have mixed excitatory–inhibitory synapses. Here, propagation of population firing rate and pulse packets are investigated. For the propagation of population firing rate, it is found that proper feedforward connection probability and strength could promote the propagation of population firing rate. Properly larger feedforward synaptic time constant could considerably enhance the fidelity of the propagating population firing rate. For the propagation of pulse packet, we find suitably adjusting the relative strength, recurrent probability, and feedforward synaptic connection can promote propagation of pulse packet. Notably, different from each other, larger feedforward synaptic time constant can considerably promote the propagation of population firing rate while hindering the propagation of pulse packets.