Differential effects of prior wakefulness and circadian phase on nap sleep.

Studies of experimentally altered human sleep-wake cycles have shown that rapid eye movement (REM) sleep propensity exhibits a circadian periodicity, while slow wave sleep (SWS) is primarily responsive to the duration of prior wakefulness. What is not known is the extent to which REM sleep continues to show a circadian pattern under intense sleep pressure, and the extent to which SWS remains responsive to prior wakefulness at opposite phases of the circadian cycle. These questions were addressed by permitting healthy young adults a 2 h nap opportunity at opposite phases of the circadian cycle and with varying amounts of prior wakefulness, during a 54 h trial in a laboratory environment free of time cues. Three groups slept near the circadian peak (15.00 h) in the diurnal activity cycle, preceded by either 6, 30, or 54 h of prior wakefulness. Two other groups had naps near the circadian trough (03.00 h), midway between the peak naps and preceded by either 18 or 42 h of wakefulness. Comparisons both between and within groups revealed that latencies to sleep onset and to SWS decreased, while stage 4 sleep increased markedly in response to prior wakefulness up to 30 h, without any effect from the circadian placement of the nap. REM sleep propensity, as measured by the number of naps with REM and the amount of REM sleep among those naps that contained REM, was affected only by the circadian phase of the nap, with trough naps containing significantly less REM. Thus, no amount of sleep pressure changed the circadian phase-dependent expression of REM, and SWS remained wake-responsive at both phases of the diurnal cycle.