A model of human sleep homeostasis based on EEG slow-wave activity: quantitative comparison of data and simulations

Abstract

EEG slow-wave activity (SWA; spectral power in the 0.75-4.5 Hz band) is a function of the duration of prior waking and, thereby, an indicator of sleep homeostasis. We present a model that accounts for both the declining trend of SWA during sleep and for its variation within the successive nonrapid eye movement (non-REM) sleep episodes. The values of the model parameters were estimated by an optimization procedure in which empirical SWA of baseline nights (16 subjects, 26 nights) served as a reference. A sensitivity analysis revealed the model to be quite robust to small changes (+/- 5%) of the parameter values. The estimated parameter values were used to simulate data sets from three different experimental protocols (sleep in the evening or sleep in the morning after prolonged waking, or extended sleep initiated at the habitual bedtime; n = 8 or 9). The timing of the REM trigger parameter was derived from the empirical data. A close fit was obtained between the simulated and empirical SWA data, and even the occasional late SWA peaks during extended sleep could be reproduced. Minor discrepancies suggest indirect or direct circadian influences on SWA. The simulations demonstrate that the concept of sleep homeostasis as proposed in the two-process model of sleep regulation can be refined to account in quantitative terms for empirical data and to predict the changes induced by the prolongation of waking or sleep.