A Neural Network System To Discover A Relevant Figure In Indefinite Information

A new type of oscillating neural network is presented. This system has an autonomic ability to separate a figure from its background as a coherent pattern of oscillation self-organized in the network. The network system is composed of two types of networks, a sensory and semantic subsystems, and has a flexible ability to discover a relevant figure in indefinite inputs from environments by the creative dynamics of the semantic subsystem.