Noise-enhanced information transmission in rat SA1 cutaneous mechanoreceptors via aperiodic stochastic resonance

Abstract

1. Aperiodic stochastic resonance (ASR) is a phenomenon wherein the response of a nonlinear system to a weak aperiodic input signal is optimized by the presence of a particular, nonzero level of noise. Our objective was to demonstrate ASR experimentally in mammalian cutaneous mechanoreceptors. 2. Experiments were performed on rat slowly adapting type 1 (SA1) afferents. Each neuron was subjected to a perithreshold aperiodic stimulus plus noise. The variance of the noise was varied between trials. The coherence between the aperiodic input stimulus and the response of each SA1 afferent was computed. 3. Of the 12 neurons tested, 11 showed clear ASR behavior: as input noise variance was increased, the stimulus-response coherence rapidly increased to a peak and then slowly decreased. These findings were in contrast with those for the average firing rate, which increased monotonically as a function of input noise variance. 4. This work shows that noise can serve to enhance the response of a sensory neuron to a perithreshold aperiodic input signal. These results suggest a possible functional role for input noise in sensory systems. These findings also indicate that it may be possible to introduce noise artificially into sensory neurons to improve their abilities to detect arbitrary weak signals.