s described in Part A, mammalian sleep consists of two very different states-NREM sleep and REM sleep, which are very different from wakefulness. The cortical EEG of REM sleep closely resembles the EEG of active wakefulness, and in some species they are virtually indistinguishable. This is a surprising finding, considering that these two states could almost not be more different behaviorally. REM sleep is usually not subdivided into stages, however, "tonic" and "phasic" aspects of REM sleep are often distinguished.

Phasic REM sleep events are intermittent (e.g., rapid eye movements and muscle twitches). Tonic REM sleep events are persistent (e.g., desynchronized [activated EEG] and striated [voluntary] muscle inhibition). As described below, these tonic versus phasic distinctions may be relevant to physiological changes that accompany sleep.

Thus far we have only considered the sleep-related changes in the brain that can be recorded from outside the brain. However, a more detailed picture of brain activity emerges as changes deeper within the brain are considered. PGO waves, which are one of the most distinctive of these changes, are spiky waves that herald the onset of REM sleep. They begin while the cortical EEG still shows the signs of NREM sleep and they occur most frequently during REM sleep, typically appearing in clusters. PGO waves are named for the sites where they can be easily recorded-the pons (where they originate), the lateral geniculate nucleus, and the Occipital (visual) cortex. They are important for two major reasons. First, they indicate that, prior to the cortical EEG signs of REM, profound changes in the neural activity are taking place within the brain and, secondly, they represent a powerful example of how other brain regions are influenced by activity emanating from the pontine brainstem.

Eye movement patterns are also used to distinguish sleep states. Involuntary, slow, rolling, pendular eye movements occur during drowsy wakefulness and during the transition from drowsy wakefulness to NREM sleep. This eye movement pattern can be replicated only by following a slowly moving object. Bursts of rapid eye movements occur during REM sleep; these bursts are interspersed with periods of no eye movements.

There are patterns of REM sleep eye movements that vary in a fairly consistent manner across the night. REM episodes occurring late in the night have more eye movement bursts than REM episodes occurring early in the night. It has been suggested that the bursts of eye movements represent "scanning" of the hallucinated dream scene. The scanning hypothesis predicts that during REM sleep the sequence of rapid eye movement directions will be that required to view the sequence of dream scenes. However, most of the evidence does not substantiate the scanning hypothesis. Nevertheless, striking correspondences between eye movement patterns and dream reports have also been noted, for example, in one report a series of horizontal eye movements were recorded; when awakened the subject reported that he had been watching a ping pong game. Regardless of such isolated reports, there is no substantial evidence which indicates that how our eyes move during a dream is directly related to what we "see" in the dream.

There is a vast difference in somatic motor activity during NREM and REM sleep (Figure 1). Few motor events occur during NREM sleep; however, body repositioning and some motor events, such as sleepwalking, are possible during NREM sleep. In general, the motor activity that occurs during NREM sleep is comparable to that present during very relaxed wakefulness. However, it is very different from that of REM sleep.



Intracellular recording from a trigeminal jaw-closer motoneuron: correlation of membrane potential and state changes. The membrane potential hyperpolarized rather abruptly at 3.5 min in conjunction with the decrease in neck muscle tone and transition from quiet (NREM) to active sleep (REM). At 12.5 min the membrane depolarized and the animal awakened. After the animal passed into quiet sleep again, a brief episode of active sleep occurred at 25.5 min accompanied by a phasic period of hyperpolarization. A minute later the animal once again entered active sleep and the membrane potential again hyperpolarized. Hyperpolarization of the membrane potential during active sleep is accompanied by postsynaptic inhibition.

There is a nonreciprocal flaccid paralysis of major muscle groups during REM sleep. Motor activity is actively inhibited by a process of postsynaptic inhibition of motoneurons; tdheir activity is also decreased by a process of membrane hyperpolarization (Figure 1). However, at intervals that often coincide with rapid eye movements, there are brief muscle twitches (Figure 2) that principally involve the peripheral muscles (e.g., the muscles of the fingers, toes, and face).


Action potentials from a tibial motoneuron during wakefulness (A) and a rapid eye movement period of active sleep (REM) (B). Depolarization (bar in A') preceded action potentials during wakefulness, whereas hyperpolarization occurred during a comparable period of active sleep (bar in B'). Action potentials in A' and B' are truncated owing to the high gain of the recording.

There is some evidence to indicate that the motor paralysis of REM sleep may be protective against the acting out of one's dreams; if the brainstem mechanisms subserving this state-dependent paralysis are damaged, truly bizarre sleep behaviors can occur. Individuals have been described who suffer from a clinically defined syndrome called REM Behavior Disorder. These patients exhibit normal behavior during wakefulness and NREM sleep. But during REM sleep they are not paralyzed as are normal individuals; rather, they appear to act out their dreams. One man dreamt that he was a linebacker for a successful football team and leapt out of bed and crashed into a wall. His behavior would have been appropriate had the wall been an opposing player. In yet another abnormality of motor functioning during REM sleep, the mechanisms that produce REM sleep paralysis can erupt during wakefulness, as in cataplexy. These patients, although they are awake, experience a sudden withdrawal of motor activity and can literally collapse to the floor.

Behavioral Responses at Sleep Onset

One of the hallmarks of sleep is a behavioral disengagement from the environment; that is, the organism is far less responsive to sensory input during sleep than when awake. This relative insensitivity has been demonstrated in a number of sensory domains. For example, subjects asked to respond to a flash of light in front of their eyes do so during wakefulness, but fail to do so at the moment of sleep. This failure to respond is not an inability to make a response, but a failure to see the stimulus, indicating that humans are functionally blind during sleep (Figure 3).

The failure to perform a simple behavioral task at the onset of sleep. The volunteer was required to tap two switches alternately, shown as pen deflections of opposite polarity on the channel labeled SAT. When the EEG pattern changes to Stage 1 sleep (arrow), the behavior stops for the period indicated by the word "gap," returning when the EEG pattern reverts to wakefulness. (SEMs=slow eye movement)

During the sleep state, there is also a failure to respond to an auditory stimulus that is clearly audible to the waking subject. When one is asleep, the highest auditory arousal thresholds occurs in NREM Stage 4 sleep, with somewhat lower and approximately equivalent auditory arousal thresholds in REM and Stage 2 sleep. Auditory arousal thresholds during sleep decline with increasing age. Young children at times seem to defy the notion that sleep is a reversible state, because they can be so difficult to arouse by any stimulus modality. In one study, a maximal auditory stimulus of nearly 120 db (equivalent to the noise of a large jet engine 500 feet overhead) during the early part of the night failed to produce any signs of arousal in a group of youngsters. In spite of a general decreased responsiveness to auditory stimuli during sleep, there is quite a bit of evidence that sleeping humans are able to discriminate significant auditory stimuli. For example, a person can be aroused more easily by the sound of his or her own name than to someone else's name. Similarly, a sleeping mother is more likely to hear her own baby cry than to arouse to the cry of an unrelated infant. Thus, it is possible to enhance auditory responses during sleep by linking stimuli to either reward or punishment. "That will be all for today's lecture" appears to be able to awaken one from any stage of sleep (or wakefulness). Clearly, sensory processing continues during sleep, though at a level far reduced from that of wakefulness.

A number of behavioral correlates accompany the transition from wakefulness to sleep. When the behavior ceases, it generally does not resume until the EEG reverts to a waking pattern. Thus simple, automatic behaviors may occur during NREM sleep. A very sleepy individual experiencing waxing and waning sleep onsets may nevertheless be able to carry out a fairly complex behavior, such as driving an automobile although their competence is diminished which can result in tragic consequences. When examining the motor system an infrequent though not rare behavioral concomitant of sleep onset has been observed, and that is generalized muscle contractions that accompany vivid visual imagery. These muscular contractions, called sleep related myoclonias are not pathological events, although they tend to occur more frequently with stress or unusual or irregular sleep schedules. Sleep paralysis-in which one consciously feels an inability to move at sleep onset-is another unusual motor phenomenon that can occur in normal individuals. In this case, the REM motor paralysis is present, although one is cognitively awake. As with hypnic myoclonia, the prevalence of sleep paralysis is highest in individuals whose sleep patterns are also very disrupted (e.g., graduate students, medical students, interns, residents, and shift workers).

Although learning during sleep has long been touted on matchbook covers and in comic books as an effective instructional tool, there is no scientific evidence that it has any validity whatsoever. On the other hand, there is evidence that material learned during the day may be "consolidated" across a night of sleep and that it may be recalled better the next morning; however, it is not clear whether this is a sleep-related phenomenon or simply represents the absence of interference from new memories.

Individuals have also attempted to learn by playing recorded information as they fall asleep. But information presented shortly before or within sleep is not consolidated into long-term memory (Figure 4). In one study, subjects had fairly good recall of all words presented in the 10 minutes before sleep onset when they were only allowed to sleep for 30 seconds. When they were allowed to sleep for 10 minutes, however, they were unable to recall the words. Thus, only a few moments of sleep are sufficient to inactivate the transfer of memory from short-term storage to long-term storage. This process explains a number of common sleep-related memory failures, such as forgetting a telephone conversation in the middle of the night, not remembering the ringing of an alarm clock, fleeting dream recall, and sleep-related loving promises of an ever-lasting nature.



Changes in Physiological Processes

Specific physiological changes accompany sleep. Because nearly every bodily function changes to one degree or another during sleep, only a small proportion can be presented in this overview of the physiology of sleep. We will concentrate on those systems that take on a greater degree of clinical relevance, because they can produce a sleep disorder in someone who is perfectly normal when awake.

Central Nervous System. Neurons in many different regions of the brain exhibit changes in activity related to the states of sleep and wakefulness. During NREM sleep, many neurons in the brain exhibit a reduced mean discharge rate. During tonic REM sleep, the mean firing rate in many regions increases compared with NREM or wakefulness. The high neuronal firing rates during REM sleep refute the old hypothesis that the brain rests during sleep. A number of regions, particularly the visual system, exhibit very great phasic increases in neuronal discharge during REM sleep. In a corresponding manner, the overall metabolism of brain tissue is generally decreased in most brain areas during NREM sleep.

Autonomic Nervous System. Because the autonomic system operates without conscious control, it is understandable that this system is very relevant to the sleeping organism. A few generalizations serve to highlight the function of the autonomic nervous system during sleep (Figure 5).



During NREM sleep, sympathetic activity stays at about the same level as during relaxed wakefulness, but parasympathetic activity increases resulting in a slight predominance of parasympathetic over sympathetic drives. During tonic REM sleep, the imbalance swings even more in favor of parasympathetic activation, not because parasympathetic input increases above the NREM level, but because the level of sympathetic input declines from that present during NREM sleep. During phasic REM sleep, both sympathetic and parasympathetic input increase (like putting your feet down on the accelerator and the brake at the same time). Transient imbalances in these autonomic drives generally favor a slight preponderance of sympathetic activation.

Cardiovascular System. With the transition from relaxed wakefulness to NREM sleep, systemic blood pressure tends to decrease slightly and to exhibit reduced variability. During tonic REM sleep, blood pressure stays at about the NREM level. During phasic REM sleep, on the other hand, blood pressure is highly variable and may occasionally increase as much as 40 mmHg, which is about 30% over the resting level. Cardiac output is reduced during NREM sleep versus relaxed wakefulness, and continues at about the same level during REM sleep. During NREM sleep there is active vasodilation of the vessels that supply resistance to the circulatory system. Most vessels remain dilated during tonic REM sleep, except for those of skeletal (striated) muscles, which undergo vasoconstriction. Vasoconstriction is generalized during phasic REM sleep, which may be the mechanism that results in greatly increased blood pressure during phasic REM sleep.

Studies of regional cerebral blood flow that have been carried out in cats have shown that only a few brain areas receive greater blood flow during NREM sleep than during relaxed wakefulness, although in general the increase is not very great. During tonic REM sleep, on the other hand, most brain areas show greatly increased blood flow, almost uniformly greater than 50% above the waking level, and as great as nearly 200%. During phasic REM sleep, there are transient further increases in blood flow to most brain regions, although precise quantification is difficult because the phasic episodes are so short. Brain temperature is generally decreased during NREM sleep versus wakefulness and there are occasionally increases above the waking level during REM sleep. These changes in brain temperature are correlated with changes in the temperature of the blood supplying the brain, even though the brain is slightly warmer than the blood (by 0.2-0.6C).

Respiratory System. With the onset of NREM sleep, there are immediate and remarkable changes in the respiratory system (Figures 6 and 7). Sleep removes not only the behavioral activity subserved by the breathing apparatus (e.g., its use for talking), but also a cortical state-related, nonmetabolic drive that tonically stimulates breathing during wakefulness. In addition, a decrease in skeletal muscle tone accompanying sleep relaxes upper airway dilator muscles so that resistance to inspiratory airflow increases. These two sleep related phenomena-the loss of the wakefulness drive to breathe, and the decrease of the activity of the muscles responsible for holding the pharyngeal passageway open during inspiration-combine to cause a decrease in ventilation during sleep. The net result is that during NREM sleep there is 13%-15% fall in minute ventilation. There is a corresponding fall in alveolar ventilation so that arterial PCO2 increases and arterial PO2 decreases.


Typical breathing pattern during NREM and REM sleep. V, ventilatory volumes (ml) obtained from a pneumotachograph attached to a face mask; insp, inspiration; exp, expiration; thor mvt, abd mvt, thoracic and abdominal movements; Ceco2, CO2 concentration (per cent) in the face mask; Sao2, oxygen saturation (per cent).

During NREM sleep, breathing is automatic and normally very regular, being exclusively under chemical and mechanical feedback control. Breathing appears to be most dependent upon the levels of CO2 in the arterial blood, so much so that below a certain level of CO2-the apneic threshold-breathing efforts cease. During the transition between wakefulness and sleep, due to fluctuations in the level of "alertness" and the changes between waking and sleep CO2 set point, breathing can be periodic with regularly occurring oscillations in amplitude. These oscillations dampen out, i.e., they become less pronounced, as sleep becomes established.

During REM sleep, breathing appears to be relatively free of chemical feedback control. Indeed, it appears to depend on some higher cortical drive, so that breathing during REM sleep can be quite irregular. The greatest degree of irregular breathing occurs during phasic periods of REM sleep.

Sleep alters not only breathing patterns but also the respiratory response to both external stimuli and internal stimuli. The hypoxic ventilatory response during sleep in men is lower during NREM sleep than wakefulness and shows a slight further fall in REM. In women, there is not much change from wakefulness to NREM sleep, perhaps reflecting a lower awake ventilatory hypoxic response. The hypercapnic ventilatory response also falls during sleep in adult humans, so there is about a 50% decrease during NREM sleep compared to wakefulness.

With the loss of muscle tone occurring during REM sleep there is an increase in upper airway resistance to about twice the normal waking values. In snorers, presumably due to altered upper airway anatomy, the airway resistance can increase to 10 times waking values. In certain individuals, the resistance during sleep becomes infinite and ventilation ceases in spite of continued respiratory efforts; this results in the Obstructive Sleep Apnea Syndrome.

There are a number of other ventilatory responses which are altered by sleep. For example, sleep suppresses the arousal response to bronchial irritation. Coughing in patients with lung disease is suppressed by sleep, and they must wake up simply in order to cough. The arousal response to hypoxia alone is poor, although hypercapnia is a strong arousal stimulus that results in the waking of most individuals before a 15mm rise in PCO2 occurs.

In summary, the changes in breathing which are linked to sleep are immediate and important. They can permit sleep related hypoxemia to occur and, in some individuals, can cause sleep apnea and even death. Remember, you have to breathe to live!

Endocrine Activity. In a very general sense, endocrines serve to maintain the homeostatic balance of the organism. They promote tissue growth, sexual development, the absorption of sodium, the response to stress and so forth. Secretions of several endocrine systems appear to be directly linked to sleep processes, whereas the secretion patterns of other endocrine systems are linked to circadian factors and not necessarily to sleep, per se (See Part G., Temporal Regulation of Sleep and Wakefulness).

Growth hormone secretion in humans is directly tied to sleep (Figure 8). If sleep is either advanced or delayed from its normal time, the peak episode of growth hormone secretion is advanced or delayed to coincide with the early part of sleep. The sleep-related release of growth hormone is not present in infants younger than three months of age. In older children just approaching puberty, virtually all growth hormone secretion occurs during sleep. With the advent of puberty, there is a great increase in the sleep-related secretion of growth hormone, and there may also be several minor episodes of growth hormone secretion during wakefulness. In the aged, the relationship between sleep and growth hormone release declines or ceases altogether. Thus, the relatively tight correlation in those events that occurs early in life is lost or diminished in old age.



The secretion of prolactin rises during sleep, beginning about 30 to 90 minutes after the onset of sleep (Figure 8). Maximal levels are reached in the early morning. As with growth hormone, prolactin release is linked to sleep (although its secretion is independent of growth hormone secretion). For example, prolactin is released during afternoon naps; if sleep is delayed, prolactin release is delayed until sleep occurs. The level of prolactin secretion increases greatly during pregnancy, and the sleep-related release is maintained. The link between prolactin release and sleep is present from puberty until old age.

Thyroid stimulating hormone (thyrotropin) release has a much different relationship to sleep. This hormone reaches a peak each day in the evening and then declines across the sleep period.

In pubertal children and those approaching puberty, the release of gonadotropins, luteinizing hormone and follicle-stimulating hormone occurs during sleep. Luteinizing hormone release occurs at sleep onset and is inhibited during wakefulness. This sleep-related release of luteinizing hormone is the first sign of puberty and is thought by some to be the event that initiates puberty. The pubertal release of follicle-stimulating hormone also occurs during sleep and is inhibited during wakefulness. Studies in which the sleep period has been reversed have shown that the pubertal release of gonadotropins follows sleep and therefore is directly linked to sleep. In adults, the sleep-related nature of luteinizing hormone release is no longer present, although testosterone levels in males continue to be highest during sleep. In elderly males, this sleep-related pattern of testosterone release is no longer present.

Normally, the highest levels of plasma cortisol occur toward the end of sleep or just after waking up. When cortisol levels have been examined under various conditions in which the timing of sleep has been varied, it is also clear that sleep onset, regardless of its time, has an inhibitory effect on cortisol release. Thus, sleep can modulate the ongoing pattern of cortisol release.

Melatonin release from the pineal gland follows a circadian rhythm that is influenced by light and not by sleep (See Part VII, Temporal Regulation of Sleep and Wakefulness). The release of melatonin occurs during the night in humans, as well as in mammals that are active at night. Sleep neither potentiates melatonin release in the daytime nor does wakefulness inhibit melatonin release at night. Nocturnal melatonin release is inhibited by light. Melatonin release patterns adjust slowly, taking 10-12 days to reverse to a change in a light/dark schedule.

Receptors for melatonin have been localized to the suprachiasmatic nucleus of the hypothalamus, which is strongly implicated as the mammalian circadian pacemaker, and melatonin is thought to function in part as an hormonal transducer of the light/dark signal (see Part VII, Temporal Regulation of Sleep and Wakefulness). Seasonal information provided by this system has apparently significant interactions with reproductive activity in many mammals, although no similar function for melatonin has been found in humans. Scientists have recently shown that melatonin administration may produce drowsiness and has minor effects on human sleep, though there is no evidence that endogenous melatonin release is involved in sleep processes.

Renal Activity. Many sleep-related changes affect kidney function. These include a sleep-related increase in plasma aldosterone levels; an increase in prolactin secretion (described above), which some consider to potentiate the action of aldosterone. There is increased parathyroid hormone release during sleep, which may affect calcium excretion. In general, the following are reduced during sleep: glomerular filtration rate, renal plasma flow, filtration fraction, and the excretion of sodium, chloride, potassium, and calcium. Smaller quantities of more concentrated urine are excreted during NREM sleep than during wakefulness; during REM sleep urine excretion is reduced and concentrated to a greater extent than during NREM sleep.

Alimentary Activity. A number of interesting relationships between sleep and the digestive system have been reported. In individuals with a normal digestive system, gastric acid secretion decreases during sleep; but for those with active duodenal ulcers, there is an increase in gastric acid secretion - three to twenty times normal - during sleep. This elevated secretion is not related to a specific sleep stage or sleep state. Swallowing occurs with lower frequency during sleep than when awake, and the motility of the esophagus is also reduced.

Sexual Activity. Activation of the sexual organs during sleep is a commonplace occurrence in both males and females. In women, vaginal blood flow has been measured with a thermoconductance flow meter. In men, the most common measurement technique involves placing two small elastic circular mercury-filled capillary strain gauges around the penis, one at the base and one at the tip. The strain gauges measure the changes in penile diameter that occur with tumescence. Both types of measures are recorded on a polysomnograph along with EEG, EOG, and EMG activity. By this means changes in vaginal blood flow or in penile circumference can be correlated with the different stages of sleep and wakefulness.

Studies in normal males from 3 to 79 years of age have shown that REM sleep-related penile tumescence occurs in all normal, healthy males. Episodes of tumescence are clearly associated with REM sleep periods, although tumescence is not exclusively confined to REM sleep, particularly in adolescents (Figure 9). The peak period of tumescence time as a percentage of REM sleep time occurs in the mid-teens. The few studies that have been performed in females have found a similar relationship between sexual arousal and REM sleep, although the link between sexual arousal and REM sleep does not appear to be as strong as in males.


Nocturnal penile tumescence cycle in a normal young man, showing five REM periods and associated maximum erections. Top graph shows EEG stages plotted against time (nine hours). Bottom graph shows penile circumference measured by strain gauge (SG). Full erection, 3 cm. Total REM time, 2-1/4 hours or 25% of total sleep time.

Among the earliest studies of penile tumescence during sleep were those attempting to relate tumescence to dream content. However, tumescence was found to occur in conjunction with REM sleep whether or not the dream content was sexual in nature.

There is some evidence that the occurrence of ejaculation during REM sleep (wet dreams) is more likely to occur if there is sexual content in the accompanying dream. In dreams in which anxiety, aggression, or rejection are prominent emotional phenomena, the accompanying tumescence shows very small, short-lived reductions in circumference.

The most common use of tests for nocturnal penile tumescence is to differentiate between organic impotence and impotence that is of psychogenic origin. For example, a male who is unable to have an erection during wakefulness for psychological reasons will have normal periods of tumescence during REM sleep. If tumescent episodes are not present during REM sleep, one would conclude that the impotence was of an organic nature. It should also be pointed out that few diagnostic tools are entirely accurate, for there are some recent data that nocturnal penile tumescence may also decrease in depression.

Thermoregulation Body temperature is regulated during NREM sleep at a lower set point. Therefore, temperature is kept at a lower level than during wakefulness. Shivering is also initiated at a lower temperature during NREM sleep than during wakefulness. Further, sweating will occur during NREM sleep when the ambient temperature is high or even in the waking thermoneutral range. Thus, although the ambient temperature may feel very comfortable when you go to bed, you may wake up later sweating.

Body temperature is not regulated during REM sleep. Therefore, shivering in response to a cold temperature stops during REM sleep as does sweating in response to a hot temperature. As a consequence, for as long as REM sleep persists, one's body temperature will drift toward the environmental temperature. In extremes of environmental temperatures sleep becomes disrupted; REM sleep is then reduced much more than NREM sleep, so that body temperature usually continues to be actively regulated. Newborn human infants may be at particular risk for catastrophic thermal events during sleep because they have such a large amount of REM sleep and because the drive to maintain REM sleep is so very great in infants.

Mammals and birds are ambient endotherms , and the cost in energy expenditure to maintain body temperature is great. It has been estimated that the metabolic rate of endotherms is 8 to 10 times that of reptiles of a similar size when they are passively heated to the same body temperature. Although there are clear state-related changes in thermoregulation, one must also keep in mind that one of the basic underlying body rhythms is the daily rhythm of core body temperature (Figure 10). This circadian rhythm will be described more fully in Part G., Temporal Regulation of Sleep and Wakefulness, but it is relevant to note that although a daily temperature pattern persists even in the absence of sleep, it does correlate with cycles of alertness. Thus, the state-dependent regulation of temperature involves control mechanisms independent from those responsible for the circadian pattern of sleep.


Average subjective alertness and body temperature of 15 subjects experiencing 72 hours of sleep deprivation under temporal isolation.

Infection. During systemic infections, people often experience increased lassitude or sleepiness. Unfortunately, studies of the relationship of infection to sleep have not been performed in humans. The sleepiness associated with hepatic failure may be relieved in seconds by the administration of benzodiazepine receptor antagonists. In experimental animals, initial work in this area indicates that: 1) very large changes in sleep patterns occur during infection; 2) sleep changes are a major sign of infectious disease; and 3) sleep changes are adaptive and possibly play a role in nonspecific host defenses. (A contrary note is that prolonged sleep deprivation does not reduce any of several measures of immune response). The general pattern after bacterial or fungal infection is an initial period of enhanced NREM sleep, followed 1-2 days later by a period of suppressed NREM. REM sleep is inhibited throughout the course of infection. It is clear that during infection and illness there is an increased tendency to sleep; it is suspected, but not proven, that sleep facilitates the healing process.

Proceed to Part E.

Return to Syllabus Home.