Control and `anticontrolʹ of chaos in an analog neural network with time delay

The results of study of an analog neural network model with time delay which produces chaos similar to the human and animal EEGs are presented. It is found that the time delay plays an important role in the control of chaos and production of different chaotic and non-chaotic outputs. An increase in neural network asymmetry (increase of the neuron excitability) leads to the phenomenon similar to the epilepsy. Two cases of the chaos control by the action of the external low-dimensional chaotic force are considered: (1) on the neural network with developed non-chaotic activity (anticontrol); (2) on the neural network with developed low-dimensional chaos. In the first case, we obtained suppression of pathological non-chaotic activity at the optimal amplitude of external low-dimensional chaotic force. In both cases, the chaotic neural network demonstrates the phenomenon similar to a stochastic resonance for internal spontaneous oscillations under the action of external low-dimensional chaotic force instead of random noise, as studied previously.