Oscillatory model of novelty detection

Abstract

A model of novelty detection is developed which is based on an oscillatory mechanism of memory formation and information processing. The frequency encoding of the input information and adaptation of natural frequencies of network oscillators to the frequency of the input signal are used as the mechanism of information storage. The resonance amplification of network activity is used as a recognition principle for familiar stimuli. Application of the model to novelty detection in the hippocampus is discussed.