In 1926, during the Russian Academy of Sciences congress, the pediatricians Denisova and Figurin presented the results of their first formal pediatric sleep research showing that, several times during sleep, infants presented episodes, lasting for 10–15 min every half an hour, during which respiration and pulse became irregular and fast and small muscles presented numerous twitches. This periodic instability of physiological functions was present in healthy children, and the authors concluded that “normal sleep is not a state of rest” [8].

As reported by William Dement, this research inspired Kleitman and ultimately led to his decision to observe eye motility during sleep [9]:

Kleitman had become very interested in what he termed the “basic rest-activity cycle”. Being able to read Russian, he was aware of the report of Denisova and Figurin (1926) which described an impressively regular respiratory cycle in infants with a period of 50 minutes . He hypothesized that this short term periodicity ensured that a newborn infant would have frequent opportunities to respond to the stimulus of hunger pangs by waking up and crying, and would therefore get adequate nutrition…

…All of this suggested to Kleitman that eye motility could be the most sensitive measure of the basic rest-activity cycle and also be more representative of changing brain activity, i.e. changing depth of sleep. He then assigned graduate student, Eugene Aserinsky, to observe eye and body motility in infants.

However, the first description that eye movements occur in sleep was reported by de Toni (1933) describing slow rolling eye movements at the onset of sleep which appeared to decrease as sleep continued and presumably deepened [10]. This observation precedes the landmark study that suggested that rapid eye movements represented a “lightening” of sleep and might indicate dreaming, due to the close association with irregular respiration and an increase in heart rate [11].

Before the discovery of REM sleep in 1953, between 1949 and 1952, Aserinsky observed that sleeping infants exhibited a recurring “motility cycle manifested by ocular and gross bodily activity” paralleling the observation of Denisova and Figurin in 1926.

Aserinsky described “periods of motility” (writhing or twitching of the eyelids) and “periods of no motility.” The average duration of the periods of quiescence was about 23 min and of the entire motility cycle was approximately 50–60 min. This observation led Aserinsky and Kleitman to look for a similar phenomenon in adults and they discovered REM sleep.

After the description of REM sleep, the French school of Dreyfus-Brisac and Monod [12, 13] begins to study neonates and infants’ sleep. Since infants can be easily studied during daytime, Dreyfus-Brisac and Monod attempted to define the specific sleep electroencephalographic (EEG) patterns of infants. N. Monod, invited to the USA by Parmelee, introduced neonatal polygraphic recording to the latter’s laboratory and highlighted the need of a full polygraphic investigation including the recording of eye movements, respiration rate, and the electromyogram in addition to the ECG.

At the same time, sleep researchers in Prague described the development of sleep in infancy showing that “quiet” sleep (QS) (regular breathing with frequency of 30/min, closed eyes without movements, disappearance of body movements, spindles, and slow waves in EEG) alternated with “active” sleep (AS) (irregular respiration, eyes alternatively closed, half-open, or there were movements of bulbus oculi, increased frequency of body movements) in about 50–60 min intervals. These authors stated that the most striking changes took place in the first 12 weeks of life [14].

In the following years, Parmelee [15, 16] first showed two distinctive EEG patterns of sleep in infants called “active” sleep (AS) and “quiet” sleep (QS). QS is characterized by preserved chin EMG, few body movements, regular respiration and heart rate, and no eye movements; AS is characterized by rapid eye movements, frequent small face and limb movements, irregular respiration and heart rate, and the absence of or minimal chin EMG activity [17].

The same authors subsequently reported the changes of EEG in infants according to maturation related to conceptional age [18, 19] showing that QS in newborns at term is characterized by one of two EEG patterns: tracé alternant or high-voltage slow (HVS) activity:

Tracé alternant is an EEG pattern in which 3–8 second bursts of moderate to high voltage 0.5–3.0 Hz slow waves intermixed with 2–4 Hz sharply contoured waveforms alternate with 4- to 8-second intervals of attenuated mixed frequency EEG activity; because this pattern alternates between activity and much less activity it is considered to be “discontinuous.” In contrast, HVS consists of continuous moderately rhythmic 50–150 μV 0.5–4 Hz slow activity, without the bursting activity of the tracé alternant. HVS represents the more mature pattern of quiet sleep in infants.

Soon after the publication of the standards recommended by Rechtschaffen and Kales for the scoring of sleep stages in adults, it was clear that they were inappropriate for the scoring sleep stages in newborn infants. Therefore, a committee cochaired by Anders, Emde, and Parmelee worked on the definition of criteria for sleep scoring in infants that led to the publication of A Manual for Standardized Techniques and Criteria for Scoring of States of Sleep and Wakefulness in Newborn Infants in 1971 [20]. Afterward, Guilleminault and Souquet published a manual on the scoring of sleep and respiration during infancy [21].

In 1970, Dreyfus-Brisac [19] observed that active (REM) sleep could be identified in polygraphic tracings by 32 weeks of gestation because of the presence of frequent body movements, irregular respiration, and rapid eye movements while the eyes were closed.

In 1966 Roffwarg, Muzio, and Dement (1966) firstly described the ontogenesis of sleep states. They also tried to answer the question at what age do humans start having dreams. By observing infants, they confirmed the richness of their rapid eye movements; they therefore supposed that REM sleep was fundamental for the optimal development of the CNS. Roffwarg and colleagues found infants spent half of their total sleep time in REM sleep, leading to the theory that REM sleep must play an important role in the development and maturation of the immature brain [22].

Petre-Quadens in 1970 [23] described for the first time a decrease of REM sleep time and of rapid eye movements in mentally retarded subjects vs. normal children, supporting the hypothesis of the importance of REM sleep for CNS development and learning. This and other observations led the researchers to investigate the relationships between REM sleep and cognition and memory for the next two decades.

Later on, a better definition of the evolution of different physiological parameters during sleep in infants was achieved by Curzi-Dascalova leading to the publication of a manual of methods for recordings and analyzing sleep-wakefulness states in preterm and full-term infants [24].

The difficulties of the definition of scoring rules for infants, children, and adolescents are mainly related to rapid and dynamic changes that occur during the first two decades of life and to the extreme interindividual variability. The comparison of polysomnographic variables needs serial longitudinal assessments linked to the normal progression of maturation, rather than a single polygraphic study at a single point in time [1].

Besides these difficulties, standards for evaluating sleep in older infants, toddlers, children, and preadolescents have been published in the new American Academy of Sleep Medicine (AASM) manual in 2007 [25] in which a specific pediatric task force was appointed. Not clearly in the manual, but in the associated papers published in the Journal of Clinical Sleep Medicine, a critical review and collection of data defined better the features of the sleep structure during development [26]. Finally, a German group headed by Dr. Sabine Scholle published three papers attempting to define the normative polysomnographic data during development [27–29].

There are very few reports on the infant and child’s sleep in the antiquity. Aristotle’s treatises on sleep and dreaming reported only that “Children sleep more than other people” and that “Very young children do not dream at all” or “Children begin to dream from ages 4 or 5” [30].

In the Roman era, the children were not allowed to get much sleep since it was believed that too much sleep decreased intelligence and stunted growth [31].

According to medieval beliefs about beds and sleeping, between 7 and 9 h of sleep were recommended, but this depended upon individual body types; with all people categorized according to the Galenic four humors, too much or too little sleep could cause dangerous imbalances and lead to illness. Nor did children require more sleep: one late fifteenth-century manual suggested 7 h was sufficient. This would roughly equate to summertime daylight hours, with an extra hour in the winter. In the mid-sixteenth century, physician Andrew Boorde was recommending two periods of sleep at night, with people rising briefly between them. Sleepers should lie first on one side and then on the other, in dry rooms to which snails, spiders, rats, and mice had no access [32].

During the Renaissance, children went to bed early, often before sunset. In boarding school, they slept two in a bed until the age of 14 when they were adults and slept alone. Poor children slept at home in the same bed with their siblings or parents. Children’s beds were more like a hay pillow in a frame called a crib or they slept on hay mattresses on the floor. After the age of 7, children only slept with siblings of the same sex, a dog or two on cold nights. Even the aristocrat’s children shared their bedrooms with their siblings and their servants. Sleeping alone was considered odd, lonely, and sad. Until the industrial era, sharing the bed with infants and children was the norm: families in the lower ranks routinely slept two, three, or more to a mattress, with overnight visitors included to generate welcome warmth and even brought farm animals within sleeping quarters at night. Besides protecting cows, sheep, and other livestock from predators and thieves, boarding with beasts allowed greater warmth, notwithstanding the “nastiness of their excrement” [33].

In the nineteenth century, there was an increasing interest for pediatric medicine, but the first books published devoted no chapters or even paragraphs to sleep. Child-rearing manuals did not deal with sleep as a problem, despite or perhaps because of extensive health advice in other categories. Sleep was not considered as a problem at that time probably because most activities went on during the night and there were much more possibilities to recover sleep during daytime than actually in modern societies. Surely, individual parents faced children with unusual sleep difficulties, but a sense of a larger category of issues did not emerge.

The reasons why people and doctors did not pay much attention to children’s sleep can be different: (a) naps were common; (b) sleep patterns were less rigid; (c) many parents undoubtedly used opiates or alcohol to help the child sleep; and (d) the absence of much artificial light reduced nighttime stimulation and facilitated getting children off to bed.

From the late nineteenth century onward, there have been specific changes in sleeping arrangements with babies increasingly placed in cribs at a fairly young age, rather than rocked in cradles as their parents worked or relaxed. The infants and children had to learn to sleep alone as soon as possible.

Recurrent advices in health columns in popular journals dealt primarily with health precautions during sleep, rather than with sleep itself. There was a discussion of how much covering to place on the child, with concern both about overheating and underprotection; it is interesting to read that cold feet were to be avoided: “neglect of this has often resulted in a dangerous attack of croup, diphtheria, or fatal sore throat.”

The first generic recommendations and guidelines about infant sleep and expectations with regard to “normal sleep” can be found in the medical books of the nineteenth century (e.g., “newborns don’t sleep for more than 2 hours at a time,” “children won’t sleep through regularly until about 17 months,” “by about six months of age, babies could get used to sleeping at specific times of the day and that mothers should not rush to comfort the baby immediately but should instead see if it resettles on its own).

Contrary to what is expected, the recourse to drug treatment was frequent: for children troubled in sleep, the easy access to and wide use of opiates surely reduced the need for extensive expert comment on what to do to fight insomnia. At that time, however, there were recurrent warnings focused on the danger of opiates administered to children with fatal events [34].

Looking at the great debate on the adequate amount of sleep need for children, we could be really surprised in reading that in the nineteenth century, attitudes toward sleep involved surprisingly modest requirements of amount: (a) authorities urging early rising recommended going to bed by 10 pm, but then getting up as soon as the infant was compatible with not feeling sleepy or lethargic the next day; (b) infants, having slept uninterruptedly for 9 months in the womb, should sleep at least 12 h; (c) afternoon naps could be abandoned around 2 years of age; (d) by age 3, children should sleep no more than 12 h, and after this, sleep time should be shortened by 1 h per year; thus a 7-year-old should sleep 8 h and certainly no more than 9; (e) adolescents required less sleep still, and authorities explicitly discussing sleep sometimes advocated no more than 6–7 h for adults [35, 36].

With the advent of the industrial revolution, the artificial light, and the regulation of working and school hours, a disrupted night became a highly disturbing event. Social habits have dramatically changed, and obtaining a healthy night sleep was mandatory for optimal social and work functioning. At the same time, sleep has become more and more consolidated into one single bout per night, and also the possibility of ad hoc naps in children was limited by the new social and school rules.

At the end of nineteenth century, doctors noted the importance of sleep in building up “nerve force” in neurasthenic patients. Hypnotics were particularly recommended for sleepless patients, though drugs were often prescribed in the period as well, coming under more critical scrutiny by physicians only in the 1890s.

Problems of children’s insomnia began to receive explicit attention, with recommendations of special feeding. The frequency and fervor of advice against using opiates for children increased; parents who assimilated this warning become more concerned about what other remedies to employ. The first tables with indications of sleep timing and duration begun to be published. The 1910 table, backed by the Bureau of Education, insisted on 13 h of sleep for children 5–6, 12 for those 6–8, 11 for those 10–12, 10 1/2 for those 12–14, 10 for those 14–16, and still 9 1/2 for those up to 18. These recommendations were strikingly different from the approach of the nineteenth century. A 1931 table called for 14–16 h for infants, 13–14 for toddlers, 12–13 still until 8 years of age, and on clown to 9 for 16-year-olds.

It was not until the 1920s that child-rearing manuals picked up the question of children’s sleep and doctors dispensed sleep advice and recommend increasing amounts of sleep. Establishing a nighttime routine became important with rituals like daily bathing, story reading, toys, or night lights.

The importance of sleep and naps routine was greatly emphasized in the 1920s and 1930s. The American Medical Association highlighted the significance of a regular sleep schedule and even claimed that a “half-hour variation from this schedule … may induce masturbation, surreptitious reading in bed, restlessness, and inability to concentrate in school.”

Children’s sleep became a new kind of issue from around the 1920s. Nineteenth-century parents had undoubtedly worried about their charges’ sleep, at least in particularly difficult cases.

After 1920, specific advice increased the amount of sleep held to be essential and the explicit scheduling required. Children had been sleeping for hundreds of thousands of years, with considerable apparent success. Why the new fuss, and new directives, early in this century?

An interesting paper tried to answer to these questions [34].

An important contributing factor was the increase of specialists in children that delivered the guidelines for the “correct behavior of infants and children” and were eager to export the findings of science to a parental audience. Further, the major improvements in infant health with the decrease of deaths in childbirth as well as the possibility of a novel arrangement for children’s beds determined that infants were increasingly isolated from adults for sleep, placed in their own bedrooms, and early separated not only from parents but also from the nurse. This leads to a decrease of parental controls on infant’s sleep behavior with the difficulty to interpret the nighttime behavior (crying, awakenings).

Due to the decrease of the use of opiates, parental concerns about children’s sleep increased, and the opiates have been substituted by over-the-counter soporifics that had become the most widely prescribed of all drugs, as of today.

A specific search in PubMed looking at the first scientific publications on pediatric sleep found some interesting papers that could give us a picture of how sleep in infants and children was considered in the first decades of the last century. In one of these papers, Sleep Requirements of Children published in the California State Journal of Medicine in 1921, there were recommendations for the amount of sleep for each age and several statements of common sense that would have been demonstrated scientifically several years later by the literature [37]:

The Service commends the following precepts just issued by the London County Council: School children aged four years need twelve hours’ sleep a day; aged five to seven, eleven to twelve hours; eight to eleven, ten to eleven hours; and twelve to fourteen, nine to ten hours.

Children grow mainly while sleeping or resting; do you want yours to grow up stunted?

Tired children learn badly and often drift to the bottom of the class; do you want yours to grow up stupid?

When children go to bed late their sleep is often disturbed by dreams and they do not get complete rest; do you want yours to sleep badly and become nervous?

Sufficient sleep draws a child onward and upward in school and in home life; insufficient sleep drags it backward and downward. Which way do you want your child to go?

Tiresome children are often only tired children; test the truth of this.

That a neighbor’s child is sent to bed late is not a good reason for sending your child to bed late; two wrongs do not make a right. Going to bed late is a bad habit which may be difficult to cure; persevere till you succeed in curing it.

In a meeting of the British and Canadian Medical Associations, in 1931, Dr. Cameron categorized sleep disturbances as follows: (1) sleeplessness and continuous crying in young infants, (2) sleeplessness in older children, (3) night terrors, and (4) enuresis. He identified three causative factors for sleeplessness in infants: (a) pain (mainly colic or dyspepsia or aerophagy treated with chloral hydrate 10 min before each feed) or discomfort (nasal obstruction treated with few drops of adrenaline solution in the nostrils before the child is put to the breast), (b) inherited or constitutional neuropathy (which resembles the description of neonatal hyperexcitability), and (c) faulty management (which resembles the description of behavioral insomnia of childhood) [38].