Time of day effects in, and the relationship between, sleep quality and movement

The study aimed to measure the effects of a 27-h 'day' sleep-wake regime on actigraphic and subjective sleep variables, and to examine the relationshi ps between these variables. Nine subjects spent 30 days and nights in the l aboratory. After sleeping 8 h for each of 8 nights, the subjects had an imp osed 27-h 'day', for 18 'days', remaining in bed for 9 h on each sleep peri od. Sleep periods therefore started 3 h later each day, although subjects' circadian rhythms stayed entrained to 24 h, because subjects were not isola ted from the natural light-dark cycle. Time asleep, subjective sleep effici ency and subjective sleep quality, but not movement during sleep, were foun d to be significantly affected by time of going to bed. There were signific ant decreases in movement during recovery sleeps following each of two epis odes of 26 h sleep deprivation. Over the study there were significant withi n-subject correlations between subjective sleep quality and subjective slee p efficiency (r(av) = 0.65), movement during sleep and subjective sleep eff iciency (r(av) = -0.48), and movement during sleep and subjective sleep qua lity (r(av) = -0.26). We conclude that sleep movement, despite its low with in-and between-subjects variability, is nevertheless a statistically reliab le, but weak, indicator of subjective sleep efficiency and quality.