Differential spectrum approach to uncovering the electroencephalographic signatures of the opponent driving forces for sleep and wake underlying alternations of sleep and wake states

•The opponent model of sleep-wake regulation postulated the drives for sleep and wake.•The drives can be separated by scoring two principal components of the EEG spectrum.•Such possibility was confirmed and extended by using two new single EEG measures.•These measures score spectral EEG differences between distinct sleep-wake sub-states.•Analysis of these differences uncovered typical spectral EEG signatures of two drives.

The opponent model of sleep-wake regulation postulated two opposing drives for sleep and wake. Simple measurement of slow wave activity does not allow their separation in the electroencephalographic (EEG) signal. However, we previously showed that scores on the 1st and 2nd principal components of variation in the EEG power spectrum can serve as markers of the opposing sleep and wake drives, respectively. The major purpose of the present report was to confirm and extend methodology for measurement of these drives by applying a new approach aimed on uncovering differences in their EEG signatures. A set of new single EEG measures was calculated in analysis of the waking and sleep EEG signals recorded in experimental studies of night sleep, multiple naps and sleep deprivation with, in total, 62 participants. Most measures summarized differences between a pair of the EEG spectra representing two distinct sleep-wake sub-states. Analysis of these differences between spectra revealed only two typical patterns that were interpreted as the spectral EEG signatures of the sleep and wake drives. The calculated single measures were subjected to principal component analysis. It yielded two largest principal components representing these opposing drives. Time courses of scores on these two principal components of variation in the calculated single measures closely resembled time courses of scores on two principal components of variation in the EEG power spectrum. It was concluded that such methodology can facilitate quantitative evaluations and model-based simulations of the opponent regulatory processes underlying normal and abnormal alternations of sleep and wake states.