The stochastic properties of the basic neuron populations as information processing system

摘要

This paper deals with the stochastic properties of the simplest neuron population consists of two neurons. Two modes of neural coupling are discussed. One is the forward inhibition mode FI and the other is the backward inhibition mode BI. In the forward inhibition mode, the two neurons are assumed to be independent of each other. In the backward inhibition mode, the two neurons interact, but the inputs to the neurons are assumed to be independent of each other. In the analysis, we first obtain the probability density function p.d.f. of the interspike intervals of the output impulse trains in FI and BI. Then we define the mean rate transfer function from the mean rate of these p.d.f.'s. Finally, by comparing our analytical results with the physiological experimental data, it is clear that the difference in the stochastic properties can be accounted for by the difference in the coupling mode (i.e. FI or BI).