Since the 1920s, when the study of sleep and memory began, researchers have sought to uncover the role of sleep in memory.¹ Decades of experimental data show a strong link between rapid eye movement (REM) sleep, which is when dreams occur, and long-term memory formation.²

The study of sleep and memory "has always had to face skepticism.from people working in the field of sleep" and "from people working in the field of learning and memory," Hennevin et al comment in a 1995 article.¹ People in the sleep field, they claim, see sleep as serving basic biological functions while learning and memory researchers "do not easily accept the idea that information processing can take place in a nonconcious state."

Whether slow wave sleep (SWS) and REM sleep play different roles in memory trace processing is unknown, Pierre Maquet, PhD of the Department of Cognitive Neurology, University College London tells ADVANCE, the new perspective is that both SWS and REM sleep are involved.

Memory Consolidation
Sleep, especially the first night after learning a new task, is essential for memory consolidation, according to researchers at Harvard Medical School. Their study, published in the December issue of Nature Neuroscience, involved training volunteers on a low-level visual perception task-identifying diagonal bars against a background of horizontal bars in the peripheral vision field. Testing without an intervening period of sleep showed no improvement while testing the next day led to large improvements, explains Robert Stickgold, PhD, assistant professor of psychiatry and principle author of the study.

Dr. Stickgold's team then investigated what it is about sleep that may account for the improvement. Two factors that affected the amount of improvement shown, Dr. Stickgold says, were how much deep SWS a subject had in the first hours of the night and how much REM sleep he or she had in the last hours. These two parameters accounted for 80 percent of the variance in the findings. "If they don't get both of those things then they don't show improvement," he adds.

Building on this, the researchers trained another group and tested them three days after the perception task instead of the next day. They kept a subgroup awake for the first night, while the controls slept. All subjects slept on subsequent nights. They found that the group that slept the first night improved on the test while the sleep-deprived group showed no improvement.

"What the research tells us," Stickgold says, is that, "for some kinds of learning sleep plays a critical role."