No abstract

Baseline sleep of 13 men (mean age of 23.5 years) and 15 women (21.9 years) was analyzed. Visual scoring of the electroencephalograms (EEGs) revealed no significant differences between the sexes in the amounts of slow-wave sleep and rapid-eye-movement (REM) sleep. Spectral analysis, however, detected significantly higher power densities during non-REM sleep over a wide frequency range (0.25-11.0 Hz) in the female versus male subjects. Also, during REM sleep, power densities were higher in the females. Analysis of the time course of EEG power density during sleep revealed that the differences between males and females persisted throughout the sleep episode. Comparison of these differences with published data on the effects of sleep deprivation on EEG power spectra did not suggest a common mechanism underlying sleep deprivation effects and the sex difference in sleep EEGs. It is concluded that sex differences in EEG power spectra are not likely to be caused by sex differences in sleep regulatory mechanisms but may, for instance, be caused by sex differences in skull characteristics.

Ke) wordy Sleep, Light, Two-process model, Clrcadmn rhythm In 8 subjects the spontaneous termination of sleep was determined after repetltwe exposure to either bright or dtm hght, between 6 00 and 9 00 h, on 3 days preceding sleep assessment Sleep duration was s~gmficantly shorter following bright hght than following &m hght During sleep the time course of EEG energy was not affected by the hght treatment Analysis of the time course of body temperature during sleep indicated an earher rise of body temperature following the bright hght treatment In terms of the two-process model of sleep regulation this can be interpreted as a &rect effect of hght on the clrcadmn phase of the wake up thresholdThe duration of human sleep is determined by both homeostatic and c~rcadtan factors. A homeostatic component has been demonstrated m expertments xn whtch sleep debt at a fixed sleep onset t~me was manipulated by varying the length of the preceding sleep episode. Wtth increasing sleep debt an increase in sleep duration was observed [2,17]. The changes m sleep duration are small and not proportional to the variations m sleep debt [10]. This may be explained by postulatmg that sleep has an intensity dimension. Electroencephalogram (EEG) studies revealed that the amount of slow-wave sleep mcreases with mcreasmg duration of prior wakefulness [16] Furthermore, spectral analysis of the sleep EEG showed that after sleep deprivation EEG power density mcreases within all sleep stages [4]. So, the homeostatic aspects of sleep regulatxon are not hm~ted to sleep duration but also encompass changes within sleepThe clrcadmn mfluence on sleep duration has been inferred from experiments m which the circadian phase of sleep onset was varied. If this was achieved by extending the duration of prior wakefulness, contrary to the predictions from a simple homeostatic model, sleep duration decreased with mcreasmg duration of prtor wakefulness up to about 32 h [1]. The circa&an influence on sleep duration is also present under

No abstract

The relation between EEG power density during slow wave sleep (SWS) deprivation and power density during subsequent sleep was investigated. Nine young male adults slept in the laboratory for 3 consecutive nights. Spectral analysis of the EEG on the 2nd (baseline) night revealed an exponential decline in mean EEG power density (0.25-15.0 Hz) over successive nonrapid eye movement--rapid eye movement sleep cycles. During the first 3 h of the 3rd night the subjects were deprived of SWS by means of acoustic stimuli, which did not induce wakefulness. During SWS deprivation an attenuation of EEG power densities was observed in the delta frequencies, as well as in the theta band. In the hours of sleep following SWS deprivation both the power densities in the frequency range from 1 to 7 Hz and the amount of SWS were enhanced, relative to the same period of the baseline night. Both the amount of EEG energy accumulating subsequent to SWS deprivation and its time course could be predicted accurately from the EEG energy deficit caused by SWS deprivation. The data show that the level of integral EEG power density during a certain period after sleep onset depends on the amount of EEG energy accumulated during the preceding sleep rather than on the time elapsed since sleep onset. In terms of the two-process model of sleep regulation (Borbély 1982; Daan et al. 1984) this finding indicates that EEG power density reflects the rate of decay of the regulating variable, S, rather than S itself, as was originally postulated.