Information processing in noisy burster models

摘要

The detection of external stimuli by peripheral sensory neurons and the downstream processing by cortical neurons quite often involves the operational mode of bursting activity, i.e., periods of rapid firing alternating with intervals of quiescence. In some systems, e.g. the electrosense of the paddlefish (Polyodon spathula), a transition from tonic firing to the bursting regime can be induced by external noise. Under the hypothesis that information about the external stimulus is encoded in the statistics of the interspike intervals (ISIs) we quantify the information content of these noisy bursters and the sensitivity to weak stimuli. Our analysis is based on analytic methods and numerical simulations of effective burster models.