tory. One of these is a drop in attentiveness and a decreased ability to stay awake (Dinges et al., 1997). Impaired by sleep loss, individuals start a task feeling fine. Minutes later, however, heads begin to nod, and the rate of deterioration accelerates. Instead of being able to sustain attention for a 45-minute lecture in a classroom, for example, a student might be able to manage only 3 to 5 minutes. Wakefulness also becomes unstable, and young people experience rapid and involuntary microsleeps and increasing difficulty in staying awake. Reaction times get longer. This may not be serious in some situations, but a 1-second lapse in reaction time while driving a car at 60 miles per hour translates into 88 feet, Dinges said.

Laboratory tests also measured the ability of subjects to pay attention to a routine task. After one night without sleep, wakefulness is unstable, and with every lapse in attention subjects fail to notice input on a simple object identification test. Such unstable wakefulness not only undermines performance—for example, a student missing an important piece of information from a teacher—but it can also be incredibly dangerous—for example, a sleepy driver missing a stop sign, Dinges said.


Daily loss of sleep accumulates in a linear fashion. For each hour of nightly sleep that is lost, the price is paid in daytime sleepiness. With 6 hours of sleep a night, evidence of poor performance is clear; with 4 hours a night, the lapses increase day after day, Dinges said. Although conventional wisdom holds that individuals can “train” themselves to adapt to less sleep, laboratory tests belie this. Dinges described his research on young adults (ages 21 to 30) in a laboratory setting for 5 days (where they were carefully monitored so that they didn 't nap or do something that affected the amount of sleep they got). He said this study showed very similar results regarding lapse rates as those found by earlier research with the Multiple Sleep Latency Test (Carskadon and Dement, 1981). The correlation between those two laboratory datasets collected 16 years apart is very high, which Dinges said suggested the reliability of the finding that chronic sleep restriction has cumulative effects. Subjects also underwent longer tests, up to 20 days, with measurements taken for 14 days at various levels of sleep restriction—8, 6, and 4 hours. They were monitored by EEGs, and their sleep was recorded. By day 5 those with 6 hours of nightly sleep functioned at an equivalent of one night without any sleep. At 4 hours a night, this happened by day 3. Subjects who continue to get 4 and 6 hours of sleep a

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement