Ultrashort sleep-waking schedule. II. Relationship between ultradian rhythms in sleepability and the REM-non-REM cycles and effects of the circadian phase.

Eight subjects aged 20-30 years spent two 24 h periods in the sleep laboratory after having an adaptation night. At 16.00 h subjects began a strict 15 min waking-5 min sleeping schedule until 24.00 h. At 24.00 subjects retired for an uninterrupted monitored nocturnal sleep. Subjects were awakened after 6-7 h of sleep, either from REM sleep (in one experimental period) or 25 min after the end of a REM period (in the other experimental period) in a counterbalanced order, and a second 8 h 15 min waking-5 min sleeping schedule was initiated. There were no significant differences between the percentages of sleep stages 1 and 2 in the afternoon, evening and morning experiments. In each, stage 1 occurred in about 10 of the 24 'sleep attempts' and accounted for 15-19% of the total recording time; sleep stage 2 occurred in 2-5 sleep attempts and accounted for 3-8% of total recording time. Four of the 8 subjects showed REM sleep in 8 sleep 'attempts;' only one appeared during an evening period. Orthogonal spectral analysis revealed a dominant ultradian frequency of about 7.2 c/day during both experimental schedules. However, synchronizing the individual morning time series with the last nocturnal REM period resulted in the appearance of a single spectral peak at 14.4 c/day, which is the dominant ultradian frequency of the nocturnal REM-non-REM cycles.(ABSTRACT TRUNCATED AT 250 WORDS)