Why Is Sleep Necessary?

Two studies in this issue of Nature Neuroscience (Vol. 3, Issue 12, 01 Dec 2000) further support this hypothesis by demonstrating that sleep is absolutely required for a particular type of memory consolidation. Sleep deprivation has long been used to test the importance of sleep for memory and learning. However, the results have been difficult to interpret because of complicating side effects, such as fatigue and reduced alertness. Stickgold and colleagues (Brief Communication, pages 1237-1238) circumvent these complications by allowing two nights of recovery sleep before performance is assessed: subjects are trained, allowed to sleep or sleep deprived on the first night, then allowed two nights of normal sleep before being tested on the third day. The discrimination task, which resembles a video game, requires subjects to identify targets within a mask of visual distractions. Subjects who were sleep deprived on the first night after training showed no significant improvement in task performance on the third day, while control subjects who were allowed sleep on the first night showed a substantial improvement. This demonstrates that there is an absolute requirement for sleep within the first 30 hours after training for consolidation of learning.

Born and colleagues (Article, pages 1335-1236) use a similar visual discrimination task to address what type of sleep is required for memory consolidation. There are two basic types of sleep, slow wave sleep, which predominates early in the night, and REM or rapid eye movement sleep, which occurs late in the night. The authors found that slow wave sleep alone was sufficient for task learning, but that both types of sleep were required for the maximal increase in performance. This suggests that there is a sequential mechanism for memory consolidation, which requires an initial period of slow wave sleep followed by periods of REM sleep. Does this invalidate the typical college student’s all night study session approach? It remains to be seen how implicit (unconscious) memory -- addressed in these studies -- relates to the explicit (conscious) memory needed for History 101.

The work is discussed by Pierre Maquet in an accompanying News and Views article (page 1235), and by John Spiro in the Editorial (page 1225).

CONTACT:

Dr. Robert Stickgold
Harvard Medical School
Massachusetts Mental Health Center
Laboratory of Neurophysiology
74 Fernwood Road
02115 Boston, Massachusetts
USA
tel: +1 617 626 9476
fax: +1 617 734 7851
e-mail: rstickgold@hms.harvard.edu

Dr. Jan Born
Department of Clinical Neuroendocrinology
Medical University of Lubeck
Ratzeburger Allee 160/ Hs 23a
D-23538 Lubeck
Germany
tel: +49 (0)451 500 3639
fax: +49 (0)451 500 3640
e-mail: born@kfg.mu-luebeck.de

Dr. Pierre Maquet
The Functional Imaging Laboratory
Wellcome Department of Cognitive Neurology
St. Johns House, 12 Queens Square
London WC1N 3BG
UK
tel: +44 (0)207 833 7472
fax: +44 (0)207 813 1420
e-mail: p.maquet@fil.ion.ucl.ac.uk

(C) Nature Neuroscience press release.

Message posted by: Trevor M. D'Souza