Spectral analysis vs. period-amplitude analysis of narrowband EEG activity: a comparison based on the sleep delta-frequency band.

This paper presents a comparison of spectral analysis with period-amplitude analysis when applied to the quantification of narrowband electroencephalographic (EEG) activity. In particular, it examines their respective usefulness in quantifying on the average the electrographic content within the delta-frequency band of EEG epochs during human stage 4 sleep. It is shown that while the power spectrum efficiently quantifies the overall power trends in the EEG data, period-amplitude analysis seems to offer more resolution than the power spectrum in detecting electrographic details in amplitude and incidence within relatively narrow frequency bands. Examples are given of the sensitivity of spectral analysis ot both wave amplitude and incidence, and of the fact that--due to the inherent averaging process in the power spectrum generation--spectral analysis cannot differentiate between low-amplitude, high-incidence EEG activity and high-amplitude, low-incidence EEG activity, in contradistinction to period-amplitude analysis. It is also shown that although two EEG epochs may exhibit similar power spectrum plots, their corresponding period-amplitude plots may not be similar. It is emphasized that discrepancies may exist when comparing spectral to period-amplitude analysis due to differences in the definition of "frequency" in the two techniques.