Why do we sleep?

The function of sleep remains one of the fundamental unanswered questions in biology.

Although a mundane part of our lives, sleep is fundamentally bizarre. Unlike other inactive conditions such as anesthesia or hibernation, sleep is an intrinsic behavioural state that can be rapidly and spontaneously reversed. Most take it for granted and some even take pride in avoiding sleep, yet those suffering from insomnia become promptly aware of how crucial sleep is to full mental and physical function. Curiously, scientists have figured out how to edit DNA and fly to the moon, but not why we spend approximately a third of our lives practically unconscious. 

The field of sleep biology began in the 1900s, with the discovery of the neuron and the formulation of the first sleeping pill: barbital, a dangerous barbiturate. However, it wasn’t until the 1920s and 30s, when the electroencephalogram (EEG) was invented by Hans Berger and popularized by Lord Adrian, that the study of sleep began to blossom.  

The EEG is a non-invasive imaging technique that records summed electrical activity from surface neurons through electrodes placed on the scalp. The invention of the EEG, in combination with the electrooculogram (EOG) and electromyogram (EMG), which record electrical activity created by movements of the eye and muscles, respectively, allowed researchers to categorize sleep into two  stages: rapid eye movement (REM) sleep, and non-rapid eye movement (NREM) sleep, named after the characteristic eye activity exhibited during each stage. NREM is further subdivided into three stages of progressively deeper sleep: N1, N2, and N3, or slow wave sleep.

When we are awake and thinking, neurons are selectively activated and thus fire in an unsynchronized way, resulting in destructive wave interference which produces a low-amplitude, high-frequency EEG signal (also known as beta waves). However, during NREM, neurons are coordinated by the thalamus to fire in a synchronous, pulsative manner, and thus the EEG pattern consists of high-amplitude, low-frequency waves (delta waves).

REM sleep is also known as “dream sleep” or “paradoxical sleep” because although you are unconscious and your muscles are paralyzed, dreaming causes brain activity in some regions to resemble the beta waves of the waking state. Dreaming also causes physiological parameters such as heart and respiratory rate to vary greatly, depending on what you are experiencing in your dream. Interestingly, though the brain doesn’t activate muscles during NREM, it must actively inhibit motor neurons through the ventromedial medulla during REM sleep (atonia). Failure to induce muscle atonia results in a condition known as “REM sleep behavioural disorder,” characterized by patients who physically act out their dreams, often in action-packed ways that result in injury. This phenomenon is distinct from sleepwalking, which occurs during slow wave sleep and doesn’t involve acting out dreams. Conversely, sleep paralysis is a state wherein REM atonia occurs even after the brain has woken up, a condition often exhibited by sufferers of narcolepsy.

Humans experience an ultradian sleep rhythm comprising of approximately five 90-minute sleep cycles per night. Each cycle consists of a progression from the awake state through the three increasingly deeper stages of NREM down to slow wave sleep, then back up these three NREM stages, until finally transitioning to REM. A direct progression from the awake state to REM doesn’t occur in healthy individuals, but is common in patients with narcolepsy, and explains why they commonly experience a sudden loss of muscle tone known as cataplexy. It has been shown that REM and NREM serve distinct functions and are individually necessary, as deprivation of only one of these stages will result in its respective rebound or overcompensation in the future. For example, NREM sleep (especially slow wave sleep) is believed to facilitate the consolidation of declarative (explicit) memories, which are conscious, intentional recollections of facts and experiences. On the other hand, REM is necessary for the consolidation of nondeclarative memories, or unconscious knowledge and procedures.

As far as we know, all vertebrates sleep. Though multiple theories about the function of sleep have enjoyed scientific popularity at different times, the most famous debate endured in the 1970s between the instinct theorists and recovery theorists. Instinct theorists including Meddis and Webb argued that sleep is an adaptive behaviour; it provides an advantage in so far as conserving energy during temperature drops and avoiding danger by keeping us immobilized during the night—a time of limited sensory capacity. In other words, although sleep is beneficial, it’s not physiologically necessary. Conversely, recovery theorists like Morruzi argued that sleep is physiologically necessary because it allows the nervous system to recover from daily use; facilitates learning and the consolidation of memories; and removes toxins such as metabolic by-products. It would therefore follow that total sleep deprivation results in illness and eventually death.

For decades, scientists have worked to develop experiments to test the function of sleep. In biology, a common method to try to determine the function of something is to see what happens when it’s taken away. One of the pioneers of sleep biology is Dr. Allen Rechtschaffen of the University of Chicago, who developed the famous “disk over water” experiments to test the effects of sleep deprivation in rats. This setup involved two rats hooked up to EEGs, one control and one experimental, separated by a divider and standing on a rotating circular platform above water. When the EEG recording showed the beginnings of sleep in the experimental rat, the circular platform would rotate. Rats generally don’t like to get wet; therefore, this rotation would prevent the rat from falling asleep as they would be forced to walk on the platform like a treadmill to avoid falling off into the water.

In this way, Dr. Rechtschaffen determined that rats deprived of sleep would die in 3-4 weeks, as well as lose weight despite increased food consumption; develop a scrawny, debilitated appearance; and experience difficulties with thermoregulation. These results were strikingly similar to the course observed in patients with fatal familial insomnia, a genetic disease characterized by severe insomnia. Although the exact cause of death in the rats could not be determined, these experiments provided compelling evidence in support of the recovery theory.

Another line of reasoning is that if sleep isn’t physiologically necessary, we should be able to find an animal that doesn’t sleep. For some animals—especially prey—sleep is an extremely vulnerable state, and in fact prey animals are known to sleep less than predators. However, no matter how extreme the environment, all vertebrates examined to date exhibit some form of sleep-like state. One of the most interesting examples of the lengths to which some vertebrates go in order to sleep is that of the Bottlenose dolphin. Dolphins are mammals and can only hold their breath for a few minutes at a time, meaning they must regularly surface for air. They are also prey animals that live in tight social groups, and thus must remain vigilant to avoid predation or swimming into each other. Given all this, if it were possible to do without sleep, dolphins would be optimal candidates to have eliminated it throughout evolution.

Instead, dolphins exhibit a sleep state called unihemispheric slow wave sleep (USWS). During USWS, simultaneous EEG recordings from both cerebral hemispheres show that one half of the brain exhibits slow waves (N3 NREM) whilst the other exhibits normal waking patterns, with these states flipping sides every hour or two for one third of the daytime. During USWS, dolphins are known to swim in circles (using both sides of the body) with only one eye open—that which is controlled by the awake hemisphere found on the opposite side of the body. The awake hemisphere allows the dolphin to maintain respiration, as well as to periodically surface for air so it doesn’t drown, all while the other half of the brain can sleep.

Dolphins don’t appear to exhibit REM, but the evolution of such a complex system is evidence that sleep is physiologically vital. However, although we continue to discover new evidence underscoring the importance of sleep, science has yet to determine what exactly sleep does that makes it so necessary, and why total sleep deprivation invariably results in death.