Wavelet analysis of fractal point processes

The response patterns of chinchilla auditory nerve fibers were analyzed using the wavelet transform. Previous studies have shown that the responses display a fractal characteristic when observed over relatively long time intervals (>100-200 ms). Both spontaneous (non-driven) and acoustically driven responses of neurons in the auditory nerve display this behavior. The wavelet transform was used to quantify this fractal behavior by estimating the slope of the variance of the wavelet coefficients across several levels of resolution. The estimates were similar to the ones obtained using the Fano factor (variance to mean ratio of events). A wavelet analysis done on a doubly stochastic point process model of the responses of the auditory nerve gave similar results