Modeling of Spontaneous Synchronized Periodic Activity Observed in Networks

Recently Segev et al. [1,2] have done long-term measurements of spontaneous activity of cortical cell neural networks placed on multi-electrode arrays. Their observations differ from predictions of current neural network models in many features. The aim of this paper is to show that the same EI network model introduced in a previous paper [3] by one of us Z.Li and J. Hertz, to model driven activity and spike-timing-dependent-plasticity in cortical areas, is able to reproduce the experimental results of spontaneous activity of [1,2] (and the observed Power Spectrum Density (PSD) features), when we consider the model in isolation with intrinsic noise terms. Using preliminary analytical results as a guide line, we perform numerical simulations of the stochastic equations for the instantaneous firing rates. In one regime of parameters the network shows spontaneous synchronous periodic activity, and the PSD shows two peaks at the first and second harmonics, and a broad band at low frequency (indicating positive long range time correlations), in agreement with experiments. The two high peak in the PSD fades away when we increase the level of noise