Controlled transition from chaos to periodic oscillations in a neural network model

Neurobiological studies have indicated that rapid transitions between chaotic and relatively more ordered states may be the key towards understanding how the brain performs cognitive tasks. This immediately suggests that methods of controlling chaos put forward by Ott et al. and Hunt may be used to study similar phenomena in neural network models. In the work described in this paper, an oscillatory neural network that can exhibit chaos under certain conditions was used for controlling purpose. On imposition of control, transition of the network behavior from chaos to periodicity was observed. This has implications for both the explanation of observed neurobiological phenomena (e.g., during epileptic seizures) as well as a more dynamic interpretation of associative recall performed by neural network models.