A neuronal model with inhibitory and excitatory synapses based on a Poisson process

Kerstin Lenk (1), Olaf H.-U. Schroeder (2), Barbara Priwitzer (1)

(1) Department of Information Technology/Electronics/Mechanical Engineering, Lausitz University of Applied Sciences, Senftenberg, Brandenburg, Germany

(2) NeuroProof GmbH, Rostock, Mecklenburg-Vorpommern, Germany

Neuronal models exhibit a simplified image of the reality in different detailedness. Our aim was to design a model which demonstrates inhibitory and excitatory effects with different dynamics which are observed in neuronal networks cultivated on microelectrode array (MEA) neurochips. Our spiking neuron model is a cellular automaton whose cells are neurons with two possible states: ON or OFF. In order to model spontaneous activity without external input which is observed in MEA experiments, the probability for a single neuron whether a spike occurs or not was calculated using a Poisson process. The neurons are connected by either inhibitory or excitatory synapses with varying strength. In order to examine the responses of the neurons to excitatory and inhibitory inputs, we ran a network with 10 neurons over 10 seconds with different structures. In the generated spike train, we detected bursts as known from experiments with MEA neurochips. Our assumption is that without inhibitory synapses characteristic burst patterns do not occur. For comparison with network activity in MEA experiments we calculated several features for the model spike train which describes the network activity.