Simulation of an ultradian sleep homeostasis through fitting time courses of its EEG indicators obtained during baseline recordings of night sleep

Abstract

The elaborated version of the two-process model of sleep–wake regulation proposes more complex mathematical formulation for an ultradian process than for a homeostatic process. Numerical simulation of these two processes includes fitting time courses of electroencephalographic (EEG) indexes obtained for a baseline sleep and for the following episode challenged by shortening/prolongation of wakefulness. In the present paper, the possibilities of simplification of this approach were evaluated. Similar simple formulae were applied for description of both processes, and their parameters were derived by fitting time course of the EEG indexes obtained for only baseline sleep of 14 women. The derived parameters were then used to predict time courses representing 24-h sequences of very short (20 min) naps obtained for nine sleep-deprived and nine sleep-restricted men. The results suggest a possibility to apply such model-based analysis to any of numerous polysomnographic recordings of sleep routinely collected in clinical sleep laboratories.

Keywords:

• Sleep–wake regulation
• two-process model
• sleep homeostasis
• ultradian sleep cycle
• EEG spectrum
• principal component analysis

Acknowledgements

The experimental and theoretical studies were supported by the Russian Foundation for Basic Research, and the Russian Foundation for Humanities (grants 07-06-00263-а, 10-06-00114-а, 13-06-00042-a, 06-06-00375-a, and 12-06-18001-e). I have benefited greatly from the help of Olga Donskaya, Dr Vladislav Palchikov, Dr Konstantin Danilenko, and Dmitriy Zolotarev (Heffele) in the EEG data collection and analyses, and from help of Dr Evgeniy Verevkin in simulation of these data.