Slow wave sleep during a daytime nap is necessary for protection from subsequent interference and long-term retention.

While it is now generally accepted that sleep facilitates the processing of newly acquired declarative information, questions still remain as to the type and length of sleep necessary to best benefit declarative memories. A better understanding could lend support in one direction or another as to the much-debated role of sleep, be it passive, permissive, or active, in memory processing. The present study employed a napping paradigm and compared performance on a bimodal paired-associates task of those who obtained a 10-min nap, containing only Stages 1 and 2 sleep, to those whose nap contained slow-wave sleep (SWS) and rapid eye movement (REM) sleep (60-min nap), as well as to subjects who remained awake. Measurements were obtained for baseline performance at training, after a sleep/no sleep interval for short-term retention, after a subsequent stimulus-related interference task, and again after a weeklong retention period. While all groups learned the information similarly, both nap groups performed better than the Wake group when examining short-term retention, approximately 1.5h after training (10-min p=.052, 60-min p=.002). However, performance benefits seen in the 10-min nap group proved to be temporary. Performance after a stimulus-related interference task revealed significantly better memory retention in the 60-min nap group, with interference disrupting the memory trace far less than both the Wake and 10-min nap groups (p<.001, p=.006, respectively). After a weeklong retention period, sleep's benefit to memory persisted in the 60-min nap group, with performance significantly greater than both the Wake and 10-min nap groups (p<.001, p=.004, respectively). It is our conclusion that SWS, obtained only by those in the 60-min nap group, served to actively facilitate the consolidation of learned bimodal paired-associates, supported by theories such as the Standard Theory of Consolidation as well as the Synaptic Homeostasis Hypothesis.