Randomness of spontaneous activity and information transfer in neurons

Abstract

The analysis of information coding in neurons requires methods that measure different properties of neuronal signals. In this paper we review the recently proposed measure of randomness and compare it to the coefficient of variation, which is the frequently employed measure of variability of spiking neuronal activity. We focus on the problem of the spontaneous activity of neurons, and we hypothesize that under defined conditions, spontaneous activity is more random than evoked activity. This hypothesis is supported by contrasting variability and randomness obtained from experimental recordings of olfactory receptor neurons in rats.