Synchronized oscillation and dynamical clustering in chaotic PCNN

Abstract

Chaotic synchronization of pulse-coupled neural network (PCNN) is studied from the viewpoint of flexible information coding. Using extended Eckhorn's PCNN model, we numerically analyze a generation of chaotic synchronized clusters in a network with locally-excitatory-globally-inhibitory connection. When multiple clusters are generated in the chaotic network, there is almost no cross-correlation between spike sequences of neurons in different clusters.