The effect of partial REM sleep deprivation and delayed recovery.

Four young adult subjects each slept ten consecutive nights in the laboratory without awakenings. The mean nightly REM sleep time was calculated for each subject and used as the baseline value for further procedures. Two subjects were allowed 75 per cent of their baseline REM sleep time for nineteen consecutive nights and slept without disturbance for the next five (recovery) nights. The first recovery nights were 54 per cent and 13 per cent respectively above the baseline with a more or less exponential decline to the basal level on subsequent nights. The compensatory rises in REM time were compatible with what might have been expected following five nights of complete deprivation and suggest that partial REM deprivation has a cumulative effect roughly proportional to the degree of partial deprivation. The other two subjects underwent complete REM sleep deprivation for five consecutive nights followed by five nights upon which REM sleep time was held to the baseline level. The next five nights were undisturbed and the REM sleep fraction was elevated 100 per cent and 66 per cent respectively above the baseline. These results suggest that the REM sleep deprivation effect can be reversed only by extra amounts of REM sleep and will persist if such compensation does not take place.