Skip Navigation

Sleep and Human Performance

Author(s): David F. Dinges, PhD

"Sleep debt" results from sleeping less than needed to be fully alert and at your best performance.

Slide Notes
Large scale, controlled laboratory experiments on healthy young adults have shown that the adverse effects of chronically sleeping less than needed to recover brain functions fully can accumulate across days of sleep restriction, reaching levels of impairment typically seen after 1 or 2 nights of acute total sleep deprivation. Interestingly, the participants in these studies were convinced they had “adapted” to the reduced sleep time (either 4 hours per night or 6 hours per night), when in fact, they were very experiencing both lapses of attention and cognitive slowing every few seconds across the entire day. Thus, their performance quickly degraded to levels well below their optimal performance capability.

Transcript of Videotaped Presentation (
Most people become impaired by sleepiness through the way they live their lives, as I mentioned earlier, and in adolescents and young adults and in millions of other people in the world, that can mean not sleeping enough at night. There were recently completed and published major scientific studies in which healthy adults were kept in a laboratory for 20 days. This graph shows 14 of those days, plus an original baseline day. For these 14 days, a group of people were given eight hours time in bed a night for sleep. So this is the control condition. Another group was randomized, or randomly assigned, to a condition in which they got six hours time in bed at night for sleep every night for 14 consecutive nights. A final group was randomized to four hours time in bed for sleep at night. Each point is the average of all the tests’ bounce, the performance test bounce, on an attention task that they completed that day from seven-thirty in the morning until midnight, each one of the groups. So thousands and thousands of points went into this experiment.

I want to remind you that people stayed in the laboratory in this study so we could control all the factors that might influence or introduce error into these assessments. We then looked at how well the subjects were able to sustain attention on simple tasks, remember things, process information. And this slide shows on the y ordinate the lapses of attention during a vigilance or sustained attention task. These are basically the number of times the signal was on, and they were holding the response button, but they failed to respond in a timely manner. And what you see is that the eight hour condition stays very near zero in the green area. This is all normal functioning. Only after about 12 days, the thirteenth to fourteenth day, does the average drift slightly above normal and that’s really due to two subjects who needed more time in bed than the eight hours allowed. But remarkably, the six hour and four hour condition showed an immediate and progressive impairment developing day after day. Around the fifth or sixth day, they separated so that the four hour was even worse than the six hour, and most importantly, in both of these conditions, the people developed performance impairments equivalent to people who had been awake in another experiment without any sleep for 24 to 40 hours. That’s a full night without sleep and the next day, and they actually progressed to an area, shown here by the upper arrow, to a group that had been awake 40 to 64 hours. That’s nearly three days without any sleep.

These progressive impairments indicate that the system keeps building up what is called debt for sleep, and that the system manifests this debt in the continuing inability, ever worsening inability, to sustain attention, to remember things, to process information quickly, make good decisions, drive, etc. So it’s a very sensitive homeostatic system and the only way to stop this is to get sleep. Once sleep is allowed, that’s being studied now in laboratories around the world, you then see the recovery function of how much sleep nets you how much recovery. But as long as sleep is reduced, as long as you’re only getting four hours or six hours, you get impairment. This tells us that most people need much more than six hours sleep and in fact they need something closer to eight hours or seven and a half hours of physiologic sleep a night.

One of the remarkable things not showing on this slide is that when people went through this experiment and we asked them every two hours, “how are you feeling, how are you doing, have you adapted, are you alert, are you fatigued,” etc., they kept telling us as the experiment went on that they were fine. In other words, the first day or two they said they were a little more tired, but as the experiment went on they expressed no increase in their fatigue. They just kept saying they’ve adapted. Yes, they were tired, but they’d adapted and were not feeling any worse. So at the time when they were maximally impaired, around the thirteenth day, they said that they were okay, that they were normal and adapted. So this makes chronic sleep restriction, which is widespread in the world and occurs certainly in young adults and adolescents, something that’s of grave concern, because people cannot introspect their impairment. They’re not aware of how dangerous they are if they drive in this condition or the extent to which they’re going to have trouble paying attention in class, remembering material they’ve read, or engaging in other activities, even physical activities like sports where reaction time matters. These, after all, are measures of slowed reaction time.

Funded by the following grant(s)

National Space Biomedical Research Institute

National Space Biomedical Research Institute

This work was supported by National Space Biomedical Research Institute through NASA cooperative agreement NCC 9-58.