THE DIFFERENT TYPES OF SLEEP

From a behavioural standpoint, sleep is defined by four criteria: reduced motor activity, diminished responses to external stimuli, stereotyped posture (in humans, lying down with eyes closed), and relatively ready reversibility. These criteria distinguish sleep from coma and from hibernation (see sidebar).

Compared with wakefulness and with REM sleep, non-REM sleep is characterized by an electroencephalogram (EEG) in which the waves have a greater amplitude and a lower frequency. From the time you fall asleep to the time you reach the deepest non-REM sleep, about 1½ hours later, the amplitude of these waves increases continuously, while their frequency diminishes correspondingly.

Scientists have somewhat arbitrarily assigned names to four frequency ranges of waves that can be distinguished in an EEG trace. From the highest to the lowest frequency, these waves are as follows.

• Beta waves: have a frequency range from 13-15 to 30 Hertz (symbol: Hz; 1 Hz equals 1 oscillation per second) and an amplitude of about 30 microvolts (µV). Beta waves are the ones registered on an EEG when the subject is awake, alert, and actively processing information. Some scientists distinguish the range above 30-35 Hz as gamma waves, which may be related to consciousness–that is, the making of connections among various parts of the brain in order to form coherent concepts.

Alpha waves: have a frequency range from 8 to 12 Hz and an amplitude of 30 to 50 µV. Alpha waves are typically found in people who are awake but have their eyes closed and are relaxing or meditating. Theta waves: have a frequency range from 3-4 to 7- 8 Hz and an amplitude of 50 to 100 µV. Theta waves are associated with memory, emotions, and activity in the limbic system. Delta waves: range from 0.5 to 3 or 4 Hz in frequency and 100 to 200 µV in amplitude. Delta waves are observed when individuals are in deep sleep or in a coma. Lastly, when there are no brain waves present, the EEG shows a flat-line trace, which is a clinical sign of brain death.

These four types of brain waves, and others discussed below, are important criteria that have been used to define four distinct stages of non-REM sleep. Obviously, falling into a deeper and deeper sleep as the night progresses is actually a gradual, continuous process, but these four stages still provide a convenient means of describing the relative depth of non-REM sleep.

	Stage 1 non-REM sleep begins when you first lie down and close your eyes. After a few sudden, sharp muscle contractions in the legs, the muscles relax. Then, as you continue falling asleep, the rapid beta waves of wakefulness are replaced by the slower alpha waves of someone who is relaxed with their eyes closed. Soon, the even slower theta waves begin to emerge. Though your reactions to stimuli from the outside world diminish, Stage 1 is still the phase of sleep from which it is easiest to wake someone up. In experiments where people are awakened from Stage 1 sleep and asked about their state of consciousness, they usually report that they had just fallen asleep or had been in the process of doing so. They also often report having had stray thoughts and short dreams. Each period of Stage 1 sleep generally lasts 3 to 12 minutes,
	Stage 2 non-REM sleep is a stage of light sleep in which the frequency of the EEG trace decreases further while its amplitude increases. The theta waves characteristic of Stage 2 sleep are interrupted by occasional series of high-frequency waves known as sleep spindles. These bursts of activity have a frequency of 8 to 14 Hz and an amplitude of 50 to 150 µV. Sleep spindles generally last 1 to 2 seconds. They are generated by interactions between thalamic and cortical neurons. During Stage 2 sleep, the EEG trace may also show a fast, high-amplitude wave form called a K-complex. The K-complex seems to be associated with brief awakenings, often in response to external stimuli. People in Stage 2 sleep are unlikely to react to a light or a noise, unless it is extremely bright or loud. It is still possible to awaken them, even if they then report that they were really sleeping during the 10 to 20 minutes that this stage lasts during the earliest of the night’s sleep cycles. But because people go through Stage 2 sleep several times during the cycles in a night, this is the stage in which adults spend the greatest proportion of their sleep–nearly 50% of the total time that they sleep each night.
	Stage 3 non-REM sleep marks the passage from moderately to truly deep sleep. Delta waves appear and soon account for nearly half of the waves in the EEG trace. Sleep spindles and K-complexes still occur, but less often than in Stage 2. The greater activity observed in the electro-oculogram (EOG) trace during stages 3 and 4 reflects the greater amplitude of EEG activity in the prefrontal areas, rather than movements of the eyes. Stage 3 lasts about 10 minutes during the first sleep cycle of the night but accounts for only about 7% of a total night’s sleep. During Stage 3, the muscles still have some tonus, and sleepers show very little response to external stimuli unless they are very strong or have a special personal meaning (for example, when someone calls your name, or when a baby cries within earshot of its mother).
	Stage 4 non-REM sleep is the deepest, the one in which we sleep the most soundly. The EEG trace is dominated by delta waves, and overall neuronal activity is at its lowest. The brain’s temperature is also at its lowest, and breathing, heart rate, and blood pressure are all reduced under the influence of the parasympathetic nervous system. In adults, Stage 4 lasts about 35 to 40 minutes during the first sleep cycle of the night; it accounts for 15 to 20% of total sleep time in young adults. The muscles still have their tonus, and some movements of the arms, legs, and trunk are possible. This is the stage of sleep that accomplishes most of the body’s repair work and from which it is most difficult to wake someone up. This is also the stage of sleep in which children may have episodes of somnambulism (sleepwalking) and night terrors.

Yawning is a stereotyped behavior with very ancient origins, for it is found in fish, reptiles, and birds, as well as in humans. Described in ancient times by Hippocrates (who thought it served to evacuate fever), yawning did not become a subject of serious interest until the advances achieved in neuroscience in the 1980s.  Generally speaking, yawning consists of three phases: first, a long intake of air, then a climax, and finally a rapid exhalation, which may or may not be accompanied by stretching. After yawning, you generally experience a sense of well being and relaxation and feel much more present in and aware of your body than you did before you yawned. Contrary to what was believed for centuries, yawning does not serve to improve oxygenation in the brain. This myth was first laid to rest when it was discovered that the human fetus can yawn as early as the age of 12 weeks, even though it is surrounded by amniotic fluid in its mother’s belly and so is scarcely likely to get any more oxygen to its brain from this effort. Second, if yawning really helped to raise the oxygen concentration in the blood, then inhaling pure oxygen would cause yawns to become less frequent, while raising the concentration of carbon dioxide in the blood would make them more frequent. But several studies have shown that neither of these things occurs. Also, yawning is no more common in people with acute or chronic respiratory problems than it is in the general population. The role of yawning has yet to be fully determined. But because we yawn more often when we first awaken, when we are bored, and when we are trying not to fall asleep, its primary function would appear to be to help make us more alert. Yawning also seems to play a role in non-verbal communication, especially among primates.   Which leads us to something truly singular about yawning: its contagiousness. That is, when we see someone yawn, it makes us yawn. Sometimes simply thinking about a yawn can be enough to trigger one! Obviously, the term “contagiousness” should not be taken literally here, because no germs are being transmitted. More precisely, yawning is a form of involuntary imitation. Some scientists believe that this characteristic of yawning may have developed as a mechanism for promoting social cohesion, for example, by enabling all the people present in a group to have the same level of alertness at the same time. In the rest of the animal kingdom, yawning is observed among predator and prey species alike. Among predators, its purpose might be to encourage the group to take a restorative nap so that all of its members can be well rested for an attack on their prey later on. Among prey, by encouraging all members of the group to fall asleep at the same time, yawning might reduce the risk that any one individual might be sleeping alone and hence highly vulnerable to attack by a predator.  There is no nerve centre strictly associated with the yawn reflex, but certain brain structures, such as the hypothalamus, the pituitary gland, and the brainstem are essential for its expression. Some scientists have even hypothesized that the strong contractions of the jaw muscles during yawning may stimulate the reticular formation and thereby encourage wakefulness. Lastly, one interesting linguistic note: the French verb bâiller (to yawn) has a circumflex accent on the “a” and not on the “i” because in Old French, when people pronounced this word, they stretched out the “a” to imitate the sound of someone yawning.

THE SLEEP CYCLES IN A NIGHT

How much time people sleep at night varies greatly with their age. Broadly speaking, from birth to death, the amount of sleep we get each night decreases steadily.

Newborns sleep an average of 16 hours per day, but even at this age, some babies sleep a lot more (20 hours) while others sleep a lot less (12). Newborns’ sleep is not affected by the alternation of day and night. Instead, it is broken up into periods of 3 or 4 hours, and the main thing that wakes newborns up is the need to nurse. Infants spend about half of their sleeping lives in REM sleep–double the proportion for adults.

This large amount of REM sleep in infants is believed to assist the development of their central nervous systems. Scientists know that neural activity helps developing synapses to find their targets. This neural activity is very intense during REM sleep, so the frequent episodes of REM sleep in babies may facilitate the activation of neural pathways and the establishment of the appropriate synaptic connections between them.

In older people,  REM sleep decreases to about 15% of total sleeping time. The deepest form of sleep (Stage 4 non-REM sleep) also diminishes gradually with age, so that older people’s sleep is more susceptible to disturbances of all kinds. Given the importance of non-REM sleep for the immune system, it may well be that this reduction in non-REM sleep also makes older people more vulnerable to illness.

DREAMS

What purpose do dreams serve? And does it even make sense to begin with to ask whether dreams serve a biological function just like eating and breathing, for example?

Some neurobiologists say not, seeing dreams as mere epiphenomena associated with brain activity. But others think that dreams contribute to epigenetic development or to the processing of recently acquired information. Still others have taken neurobiological data from brain-imaging studies done in the 1990s and used these data to support a theory holding, as Freud did, that dreams are psychological manifestations that can convey meaning.

Gérard Larguier, Father Freud’s Dream http://www.gerard-larguier.com/index_fr.php

One such theorist is neuropsychologist and psychoanalyst Mark Solms. Solms first observed that a number of his patients who had suffered damage to the neurons of the pons and therefore no longer had any periods of REM sleep nevertheless continued to dream regularly. He then identified two areas of the cortex that had nothing to do with REM sleep but that, when damaged, caused the loss of the subjective experience of dreaming. The first of these areas is located where the occipital, temporal, and parietal cortexes meet. This area is involved in spatial imagery, among other things, so Solms’s finding makes intuitive sense: it is hard to imagine being able to dream if your ability to form mental images were impaired. The other area of the brain that seems to be necessary for dreaming, according to Solms’s research, is located in the frontal cortex. The neural pathways that project to this area use dopamine as a neurotransmitter and are known as the mesolimbic system. The area itself is involved in positive reinforcement and motivation.

Why then should dreams disappear when this part of the brain is damaged? Probably because dopaminergic transmission has been disrupted. In any case, that is what is seen in people who take medications known to decrease their dopamine levels: they dream far less. And the opposite is also true: patients who take medications that increase dopaminergic activity along this pathway (for example, Parkinson’s patients who take L-dopa) dream more intensely than they used to, even though the frequency and duration of their periods of REM sleep are unchanged.

For Solms, it therefore seems clear that if REM sleep is generated in the most ancestral parts of the brainstem, dreams, in contrast, may arise in the cortex. The involvement of the frontal and the occipito-temporo-parietal cortexes, which regulate memory, feelings, and motivation, supports the idea that dreams in some way serve to reprocess subjective events that the individual has experienced previously. In short, Solms’s theory allows for the possibility that dreams may have meaning and thus preserves the foundations of psychoanalysis, in contrast to Hobson and McCarley’s model, in which dreams are simply the result of the random bombardment of the cortex by meaningless signals from the pons.

This theory of the cortical origin of dreams raises several issues. One in particular is the difficulty of reconciling the very fleeting nature of our memories of our dreams with the very fundamental role that this theory implies dreams play in our psychic equilibrium.

The strange and fragmentary nature of our dreams as we recollect them is central to another daring theory of their origin: we may dream not when we are sleeping, but only as we are awakening. This theory, developed by French neuroscientist Jean-Pol Tassin, is based on the paradox that consciousness vanishes during sleep, yet dreams cannot exist unless we are conscious of them. According to Tassin and his collaborators, during REM sleep, the brain is active, but its activity allows neither consciousness nor dreams.

There is a neurobiological correlate that supports this interpretation: some noradrenergic and serotonergic neuromodulatory neurons that are necessary for neural information to be stored in the brain for more than a few milliseconds–in other words, necessary for consciousness–cease to function when you are asleep, but become active again while you are waking up.

Thus, according to Tassin’s theory, as you awaken, these reactivated neurons enable you to become aware of the subliminal images generated during your sleep, and you then actually construct your dreams during the few hundredths of a second that it takes you to wake up. This brief interval might also be the time when, as sometimes happens, you incorporate into your dreams the light or the words that have woken you up.

But how then to explain the subjective impression that we dream during the night? Researchers who have analyzed EEG traces for entire nights of sleep have found that even sound sleepers may awaken as many as 10 times per night, then fall back to sleep again rapidly, even if the next morning they report that they slept straight through the night. During these “micro-awakenings” that last only a few seconds or fractions of a second, the brain finds itself in a state identical to wakefulness, but for such a short time that we very rarely remember it the next day. It would be during these micro-awakenings that we might dream, that is, organize our often bizarre mental images into coherent stories. And as the generator of these bizarre mental images, REM sleep seems the ideal candidate, though non-REM sleep can generate some strange images too. What makes this theory even more plausible is that REM sleep is the phase of sleep in which spontaneous awakenings are the most frequent.

This model thus provides an explanation for the illogical, impossible or unreal nature of the story lines of most of our dreams: because the return to consciousness that gives rise to dreams occurs in a very short time span, often following a period of REM sleep, the images we remember are too disparate to be integrated into a coherent story, and our conscious brain may therefore have to “force” reality a bit to assign a meaning to them. This would not be the only instance in which the brain plays tricks on us in an attempt to give a meaning to confusing stimuli; certain optical illusions and split-brain experiments offer other examples of this same phenomenon (follow the Experiment Module link to the left).

For Tassin, dreams would thus represent the conscious expression, during awakening, of the unconscious brain activity that occurs while we are asleep. Dreams would thus remain dependent on sleep, because they would arise from the sudden reactivation, at the moment of awakening, of the serotonergic and noradrenergic neurons whose activity is indispensable for consciousness.

If this theory proves correct, many observations could be interpreted differently. For example, when you awaken someone who is sleeping, you aren’t interrupting her dreams, but rather making them happen! And Jouvet’s sleeping but “disinhibited”cats were simply reproducing movements that they also made during the daytime, without consciously perceiving images associated with these movements–in other words, without dreaming.

This view of dreams has the further advantage of leaving open the possibility that dreams may have a meaning for the people who dream them. Because if their dreams occur in the space of a few hundred milliseconds, then the mental censor that may be active when they are awake is not in place, thus allowing bizarre dream content that might be worth interpreting.

SLEEP DISORDERS

Since the 1970s, laboratories that do research about sleep have been established in many parts of the world. Thanks to their discoveries, we now know that the health problems caused by lack of sleep are far more numerous than we once imagined. These laboratories have also identified over 100 different disorders that can affect our sleep. Besides insomnias and disturbances in circadian rhythms, hypersomnias and parasomnias represent the two other main categories of sleep pathologies.

Narcolepsy, formerly called “sleeping sickness”, is a hypersomnia that is characterized by excessive sleepiness during the day and, in extreme cases, by sudden irresistible bouts of sleep that occur several times per day. People with narcolepsy can thus literally fall asleep at any time. In addition, during these attacks, they pass directly from a state of wakefulness to a state of REM sleep, unlike healthy people, who always go through a period of non-REM sleep first. In fact, many of the symptoms of narcolepsy can be seen as the intrusion of a phase of REM sleep into a person’s waking life.

More and more studies in animals and humans (see sidebar) tend to suggest that hypocretins (also known as orexins), a class of neuropeptides produced solely by the neurons of the hypothalamus, play a role in narcolepsy. Several post mortem analyses have found far fewer of these neurons in the brains of people with narcolepsy than in those of healthy persons.

In its most complete form, narcolepsy is also accompanied by a condition that is startling, to say the least, to those who witness it: cataplexy, a sudden decrease in muscle tonus, varying in intensity and lasting less than a minute. The signs of cataplexy range from a simple weakness in the neck, knees, or facial muscles to total paralysis that causes the individual to fall to the ground.

An attack of cataplexy is usually caused by a strong emotional trigger such as laughter, anger, surprise, or sexual arousal. People having a cataplectic attack are often still conscious but unable to move, which makes this condition fairly terrifying. Once again, the connection with REM sleep is quite apparent: muscle atonia in all respects similar to that which occurs during REM sleep to prevent our bodies from acting out our dreams.

Sleep paralysis and sleep hallucinations are other symptoms of narcolepsy. Sleep paralysis is a temporary inability to speak or to move while falling asleep or waking up–a highly disconcerting experience, especially when the person having it doesn’t know its cause. Sleep hallucinations are strange, unpleasant experiences that resemble waking dreams. They occur during the transition from waking to sleeping, as well as during periods of reduced alertness in the course of the day.

Parasomnias is an umbrella term for a variety of abnormal phenomena that occur during sleep. Several types of parasomnias affect children in particular. One example is night terrors, a phenomenon completely different from simple nightmares.

The Nightmare, by Heinrich Füssli (1792). Private collection.

Nightmares are dreams involving visual images that are frightening enough or negative emotions that are strong enough to cause the dreamer to wake up scared and anxious. This feature differentiates a nightmare from a simple bad dream that doesn’t cause the dreamer to wake up. In children, nightmares are associated with normal aspects of psychological development, such as separation anxiety or sibling rivalry. In adults, nightmares tend to be precipitated by stress or by physical factors such as fever. Some violent, recurring nightmares may also be related to post-traumatic stress.

Night terrors are events that are biologically and psychologically distinct from nightmares. They begin when children are 3 to 6 years old and generally disappear during adolescence. Children in the throes of a night terror scream and cry. Their eyes are open, and they may say incoherent things while gesturing emphatically. Unlike nightmares, of which people can clearly recall some details once they awake, night terrors are characterized by confusion upon awakening, the lack of any recall of elaborate dream imagery, and intense activation of the autonomic nervous system, causing symptoms such as sweating, and elevated heart rate and blood pressure. Also, nightmares occur mainly during periods of REM sleep in the second half of the night, whereas night terrors typically occur during deep (Stage 3 and Stage 4) non-REM sleep, during the first part of the night. An entire night-terror episode can last 1 to 20 minutes. The next morning, the child usually wakes up in a good mood, having forgotten the entire incident.

Enuresis (involuntary bed wetting during the night) of course does leave obvious traces the next morning. Children are diagnosed as enuretic if they wet the bed more than twice per week after age 5 or 6–in other words, long after they are toilet-trained. The best approach to this problem is not to punish or humiliate the child, but rather to be supportive to help maintain the child’s self-esteem. This problem generally disappears on its own by adolescence.

Somnambulism is another form of parasomnia that is especially common in children. It involves sleepwalking during non-REM sleep. About one-third of all children display this behaviour at some time or other, and about 3% do so at least once per month. As with enuresis, episodes of somnambulism generally disappear gradually as the child grows older, so that only 1 to 4% of adults still have them occasionally.

Contrary to popular belief, it is not dangerous to wake up someone who is sleepwalking. But it can be fairly hard to do so, because episodes of somnambulism, which generally last about 10 minutes, typically occur during the deepest stage of non-REM sleep, Stage 4, and hence during the first sleep cycles of the night. Thus somnambulism is neither caused nor accompanied by dreams.

Episodes of somnambulism are believed to be triggered when something such as a noise, or the need to urinate, wakes up the body without waking up the brain. The sleepwalker may then get up, walk to the kitchen, open the fridge, eat a snack, pick up the telephone, or play some music, with no conscious awareness of any of these actions. Because this state of very partial cognitive functioning obviously entails some dangers, the best thing to do with a sleepwalker is gently guide him or her back into bed.

Somniloquy–talking in one’s sleep–can happen during either REM or non-REM sleep. The words are generally so poorly articulated and the sentences so meaningless that anyone who hears them will be at a loss to interpret them. Those utterances that occur during REM sleep do, however, tend to be somewhat more intelligible.

Bruxism is a strange parasomnia. It consists in repetitive, involuntary grinding of the teeth that causes them to suffer abnormal wear and tear and also causes discomfort in the jaw muscles. Though about half of all people move their jaws in their sleep, only about 6% display the tooth-grinding during the light stages of non-REM sleep that characterizes bruxism. The mechanisms of this disorder are not yet fully understood, though it is now agreed that they do originate in the central nervous system. People who suffer from bruxism will generally benefit from reducing their stress and from wearing a special device in their mouth to prevent tooth damage.

REM sleep behaviour disorder is a rare but fascinating pathology sometimes seen in older people. It consists in a form of sleepwalking that may superficially resemble somnambulism, but is significantly different, because the people engaged in this behaviour are in REM sleep rather than non-REM sleep. Normally, during REM sleep, people’s muscles are completely paralyzed, except for those involved in respiration and in moving their eyes. But individuals who suffer from REM sleep behaviour disorder do not experience this characteristic paralysis. Instead, they literally jump out of bed and mime their dreams while continuing to sleep! This disorder is very dangerous, because people who have it often injure themselves while externalizing their dreams, attempting to fight or flee some non-existent assailant. Sometimes the dreamers may cast their bedmates in the role of the assailant, who may then find his or her own dreams rudely interrupted! Luckily, this condition does respond to some medications, such as the benzodiazepine clonazepam.

Many people with this disorder have been shown to have damage in the areas of the brainstem normally responsible for the muscle atonia of REM sleep–the same areas where the production of lesions in cats enabled them to “externalize their dreams”. These areas that allow muscle atonia during REM sleep likely developed during the evolution of our species precisely to prevent what happens to people who have REM sleep behaviour disorder.

Sleep paralysis, which is very common in people with narcolepsy, can also occur in isolation, with no other associated pathology. This parasomnia is manifested when the individual is falling asleep or waking up, and it typically lasts just a few minutes. During this period, the person can neither move nor speak. This paralysis of course causes significant anxiety. It may also be accompanied by visual, auditory, and even tactile hallucinations, known as hypnagogic hallucinations.