A Neural Network with a Background Level of Excitation in the Cat Hippocampus

Abstract

A model based on a limited number of properties of pyramidal cells and interneurons in the CA3 region of the hippocampus has been developed. Elements in this model were selected to correspond to conventional inhibitory and excitatory postsynaptic potentials, neuron thresholds, and conduction delays. Additionally, a signal over pathways making synaptic connections with pyramidal cells was included to provide an ongoing level of excitatory activity. The calculated output of the model was then compared with PST histograms recorded from hippocampal pyramidal cells. The model appeared to account for multiple peaks previously noted in PST histograms recorded over a period of several hundred milliseconds following dorsal fornix stimulation. To test the model, the response of the network was calculated for paired shock stimulation of the fornix and compared with experimental data from the cat. The model was written for CSMP (an IBM program designed to facilitate digital stimulation of continuous processes) and the relation of the model to a state-space representation briefly outlined.