Ultrashort sleep-walking schedule. I. Evidence of ultradian rhythmicity in "sleepability'.

After a 7-8 h sleep at home, 9 young adults were placed on an ultrashort 15 min waking-5 min sleeping schedule for 12 consecutive hours, followed by a monitored recovery night in the laboratory. Six of 4 of the 9 subjects were also investigated in the same way after one night of selective REM deprivation and after one night of total sleep deprivation, respectively. In the first experiment, the amount of stage 1 in each of the sleep attempts varied rhythmically with a frequency of about 14.4 c/day. The average variance at the peak spectral frequency significantly exceeded the mean variance at the rest of the spectral frequencies and the mean variance at the adjacent frequencies. Stage 2 occurred more often around 15-16 h, with no evidence of ultradian rhythmicity. Both REM deprivation and total sleep deprivation disrupted the 100 min periodicity in stage 1 and modified the distribution of stage 2 toward a bimodal rather than a unimodal distribution. The result of the first experiment were interpreted in the light of Kleitman's BRAC model. The ultradian rhythmicity in sleepiness is suggested to play a role in the adaptability and flexibility of the circadian sleep-waking cycle.