The Science of Sleep

We spend roughly a third of our lives asleep, and yet the question of why we sleep at all remains, to a surprising degree, unanswered. From an evolutionary point of view, the behavior seems almost reckless: an animal that lies unconscious for hours is an animal that cannot eat, look after its young or escape from predators. The fact that virtually every animal studied so far shows some form of sleep suggests that whatever it does, it is too important to give up.

What scientists do understand reasonably well is the timing of sleep. Our bodies run on a circadian rhythm, an internal cycle lasting a little over twenty-four hours, which is controlled by a small cluster of cells deep in the brain. This internal clock is reset each day mainly by exposure to light. As evening falls and light levels drop, the brain increases its production of melatonin, a hormone that signals to the rest of the body that night has arrived. Exposure to bright light late in the evening, including the light from screens, can delay this release.

Alongside the clock works a second mechanism, often described as sleep pressure. The longer we stay awake, the more a chemical called adenosine builds up in the brain, and the sleepier we feel. Caffeine works largely by blocking the receptors that detect adenosine, which explains why a cup of coffee can hide our tiredness without actually reducing our need for sleep.

One of the most actively researched functions of sleep is its role in memory. A large body of evidence suggests that during deep sleep the brain replays and reorganizes what it has learned during the day, strengthening some connections and weakening others. Students who sleep after studying generally remember more than those who stay up all night, a finding that has been replicated in many laboratories. Another hypothesis, based largely on studies in mice, proposes that sleep helps the brain clear away waste products that accumulate during waking hours, although how far this applies to humans is still being investigated.

The science has particularly striking implications for teenagers. During adolescence, the timing of melatonin release shifts later, sometimes by two hours or more. A sixteen-year-old who cannot fall asleep before midnight is therefore not necessarily lazy or undisciplined; their internal clock is simply running on a different schedule from that of their parents. When school begins at eight in the morning, the result can be chronic sleep deprivation, with consequences for concentration, mood and even safety on the roads.

For this reason, a number of medical organizations have recommended later start times for secondary schools. In the United States, for example, the American Academy of Pediatrics has advised that classes for this age group should not begin before half past eight. Where schools have made the change, studies have often reported improvements in attendance and alertness, although the size of the effect varies and the research is not unanimous.

Perhaps the most honest conclusion is that sleep is less a single process than a collection of processes, some of which we are only beginning to understand. What is clear is that the common assumption that sleep is wasted time, an inconvenience to be reduced wherever possible, is very difficult to reconcile with the evidence.