SENSORY INFORMATION PROCESSING BY NOISY BISTABLE NEURONS

Abstract

Many neurons in the central nervous system and at the sensory periphery receive periodic input. The neural activity of interest here is one which may be called "statistical phase-locking", where the neuron fires preferentially near a given phase of the stimulus, but skips a random number of stimulus cycles between successive firings. It is shown that intervals between successive firings are distributed according to the same density as that of residence times in bistable noisy systems, provided these times are computed from suitably chosen interwell transitions. This brief review further makes this connection plausible by relating properties of neurons to those of bistable systems. Further it looks at the question of what role noise and possibly stochastic resonance, which can arise in this context, may play in information processing at the sensory neuron level.