Abstract
16 Ss were deprived of rapid eye movement (REM) sleep for 2 nights by awakening them at the first sign of each REM period. 16 control Ss, initially matched with REM-deprived Ss on a paired-associate learning task, were awakened a similar number of times, but only from sleep without REMs. There were no significant differences between REM-deprived and control Ss on post-deprivation recall of paired-associate adjectives learned prior to deprivation or on the post-deprivation serial learning of trigrams. The results fail to support an important role for REM sleep in human memory or learning. Neither did REM-deprived and control Ss differ significantly on two mood check lists or on a number cancellation performance test.