A Model for Predicting the Reliability of EEG Power Spectral Density Estimates of Subcortical Spiking Frequency

A model for predicting the reliability of power spectral density (PSD) indicators of spike-to-spike intervals taken from EEG time series is proposed. The analysis shows that the PSD is corrupted by irregularities in spike spacing and that conditions are improved if a larger number of spikes are included in the EEG epoch.