Tool Module: The Possible Functions of Sleep

We spend one-third of our lives sleeping, so there is no doubt that we need to sleep. And if we are deprived of sleep for one night, we tend to “catch up” on our lost sleep the next, which further indicates the fundamental importance of sleep for the human organism. Likewise, from an evolutionary standpoint, the function of sleep has been universally preserved in birds as well as mammals. Yet astonishingly, scientists still don’t know exactly why we sleep.

Despite all these indications of the great functional importance of sleep, there is no consensus on what precisely makes it so important. Scientists have proposed many theories, and not all of them are mutually exclusive. But all of them have shown their limitations or been contradicted by other scientists’ experimental findings. The most commonly accepted theories of the function of sleep can be divided into two groups: restoration theories and adaptation theories.

Restoration theories make common sense: at the end of the day, you’re tired. You go to sleep, and when you make up the next morning, you feel restored. In other words, according to this theory, just like hunger and thirst, mental and physical fatigue induce a homeostatic response–in this case, sleep–that restores an equilibrium in the central nervous system.

In one example of experiments that support this theory, the more exercise animals got during the daytime, the longer their periods of non-REM sleep at night. But scientists still have not determined at what level of the central nervous system this restoration may occur, and whether it involves synthesizing molecules or breaking them down.

As far as the possibility of synthesis is concerned, there does not seem to be any evidence that any particular body tissues are rebuilt during sleep. But the observation that the neurotransmitter adenosine builds up in the brain during the daytime and declines during sleep has inspired a different kind of restorative theory: the “protective theory” of sleep, according to which we sleep mainly to protect our bodies from the negative effects of prolonged wakefulness.

Yet another restoration theory that some authors have proposed is that we sleep to save energy. The reasoning is as follows: during non-REM sleep, animals consume less glucose and oxygen both in their brains and in their bodies as a whole. Hence, if animals that were always awake had ever existed, they would have needed to eat more food than animals that could sleep (or hibernate), and over the course of evolution, natural selection would therefore have favoured those animals that could sleep.

But this theory leaves the following contradiction unresolved. During hibernation, animals expend almost no energy at all. Yet animals that hibernate periodically emerge from hibernation in order to sleep. In the process, they raise their body temperatures by a few degrees Celsius, to about 38°C. Hence it is hard to see how the role of their sleep could be to save energy.

Another counterargument to all these theories is that sleep presumably serves some function beyond mere resting. (Try resting in bed all night without going to sleep, and you’ll have no trouble agreeing that you don’t feel as refreshed as if you had slept!)

The other major group of theories of the role of sleep instead focuses on its adaptive value. According to these theories, for many animals, the absence of light and the lower temperatures at night made life more dangerous, so that it was more adaptive for them to remain hidden and motionless at night. Sleep was simply the most adaptive way to implement this protective strategy, and also let them save energy. Hence they evolved so that they slept at night. But that still doesn’t explain why nocturnal animals, which function so well at night, also need to sleep.

Another widely held adaptive theory of sleep is that REM sleep in particular may help the nervous system to mature. This theory is well supported by the strong correlation between how immature the individuals of a given species are at birth and how much time they spend in REM sleep. The intense neural activity of REM sleep may thus play a decisive role in the maturation of the central nervous system during a period when the brain receives fewer external stimuli. But this theory still doesn’t explain why REM sleep persists in adults.

The connection between sleep and learning and memorizing also has been studied extensively but still remains unclear. On the one hand, sleep definitely plays an important role in the optimal learning of new tasks. For example, several studies have suggested that when the REM sleep of experimental subjects is intentionally interrupted, they take longer to accomplish certain kinds of learning. REM sleep seems especially important for acquiring new visual and motor skills. When athletes are learning a new sequence of movements, such as a tennis serve, during the daytime, they spend appreciably more time in REM sleep the following night. In experiments where such athletes are deliberately awakened, some from REM sleep and others from non-REM sleep, the ones who are awakened from REM sleep have considerably more trouble in storing their new skill in memory.

On the other hand, it is also clear that people can successfully learn and memorize things without getting any REM sleep. Here is one substantial piece of evidence against a predominant role for REM sleep in the formation of memories: certain medications suppress REM sleep, but people who have taken them for years have not experienced any impairment of their ability to memorize.

Another, related debate is about which type of sleep, non-REM or REM, plays a role in learning. In humans, learning of some tasks seems to be facilitated more by REM sleep, while learning of others, such as those involving spatial orientation, seems to be facilitated more by non-REM sleep. Hence many scientists believe that both types of sleep may contribute to memory, in complementary ways, and that it is their alternating phases over a night’s sleep that enable the brain to correctly sort the information that it has accumulated during the day. For example, the long periods of non-REM sleep at the start of the night might amplify the memory trace, whereas the briefer but intense episodes of REM sleep might trigger the expression of genes that are necessary for storing the information processed during non-REM sleep.

The abundance of theories on the role of sleep suggests that it may in fact have a number of functions that are not mutually exclusive. Sleep may also serve different functions at different stages of life. In children, who dream a lot, sleep may plausibly play a role in building neuronal circuits. In adults, sleep may assist in selecting the relevant information learned during the day and placing it in long-term memory. Lastly, sleep may play a role in some as yet unidentified cellular function that is the foundation not only for maturational processes in children, but also for regulation of body temperature and for higher cognitive functions such as memory in adults.