What is a role of spiking variability in coding rate information?
CBN (Computational Biology and Neurocomputing) seminars
Monday 29 October 2012
to 12:00 at
Shinsuke Koyama (Department of Statistical Modeling, The Institute of Statistical Mathematics, Tokyo, Japan)
In many cortical areas neural spike trains do not follow a Poisson process. In this study, we investigate a possible benefit of non-Poisson spiking for information transmission by studying the minimal rate fluctuation that can be detected by a Bayesian estimator. For this purpose, we introduce a relative entropy for quantifying the amount of information on rate fluctuations. We show that the relative entropy entirely determines the lower bound of rate fluctuations below which the temporal variation of the firing rates is undetectable from spike trains.
We also show that the relative entropy, as well as the lower bound, depends not only on the variability of spikes in terms of the coefficient of variation, but also significantly on the higher-order moments of interspike interval (ISI) distributions. This result suggests that it may be important to take into account the higher-order moments of ISIs for characterizing "irregularity" of cortical firing, in order to gain information on fluctuating firing rates.