Estimating Scaling Exponents in Auditory Nerve Spike Trains Using Fractal Models Incorporating Refractoriness

Fractal stochastic point processes FSPPs provide good mathematical models for long term correlations present in auditory nerve bers in a number of species Sim ulations and analytical results for FSPPs concur with experimental data over long times Refractoriness modi ed Poisson point processes in contrast model only the short term characteristics of these data FSPPs incorporating refractoriness yield superior results over all time scales and analytical results for the statistics of this model agree closely with both computer simulations and auditory data Furthermore these processes provide a superior method for estimating the fractal exponent which parametrizes the long term correlation found in these record ings