Coherent oscillatory responses as a feature linking mechanism in an artificial neural network model of visual perception

Recent discoveries in neurophysiology suggest that information about sensory stimuli can be encoded in the brain as patterns of oscillation. The discovery of such behaviour in the visual cortex is crucial in as much as it has generated a new approach to the explanation of how the processing and co-ordination of visual sensory information can be modelled in artificial systems. This has led to the development of a new type of artificial neural network which attempts to capture this form of dynamical behaviour, in particular stimulus evoked coherent oscillations and their role as a feature linking mechanism. The paper describes such a neural network model and demonstrates its behaviour though (sic) analysis and simulation experiments.