Measurement of temporal regularity of spike train responses in auditory nerve fibers of the green treefrog.

This paper describes a method which we have developed for quantifying the temporal regularity of neural spike trains in the sensory nervous system. Our method relies on the use of a modified correlation approach for identifying response firing patterns. We apply the concept of an ambiguity function and related coefficients to measure the tonic/phase character and statistical variability of spike patterns. We tested this method in recordings from auditory nerve fibers of the green treefrog (Hyla cinerea) in response to pure tone, multi-tone, and gaussian white noise. Our results indicate that there is a great deal of variability in the trains of spike times generated by any given fiber in response to identically repeated stimulus presentations. Nevertheless, despite this statistical jitter of the pattern to repetitive stimulation, the spike response trains from a single fiber maintain a high degree of individual detectability in signal metric space. The procedures in our method can be implemented in a relatively simple way on a Macintosh computer and the speed is fast enough for real-time spike analysis. This kind of quantification may especially be useful in studying habituation and plasticity in neural spike train data, as well as in judging the selectivity within the neural code of an individual fiber for particular stimulus features.