Epilepsy as a self-organization process: a computational model

Chaos in the human brain and artificial neural networks is explained with a view to an understanding of human brain functions. Different chaotic solutions are known in neural network modelling, but the comparison of their quantitative characteristics with the human or animal EEGs can not be performed. In this work a model of the start and spread of epilepsy, based on neural nets, is presented. It is shown that the epilepsy-like phenomena can occur in the neural networks with increasing neuronal excitability. The dynamics of the quantitative EEG characteristics (correlation dimension, amplitude, largest Lyapunov exponent) is similar to one at the onset of epilepsy