The siesta culture shows that there is some flexibility in how and when we attend to our individual sleep needs. We can obviously split our sleep into two major episodes, one at night and another at midday. Under special conditions, we can even consolidate our sleep into one coherent episode of more than twelve hours, as some subjects in the Andechs bunker did. But even under these strange conditions, the average sleep duration amounts to somewhere between seven and eight hours per twenty-four hours. Our flexibility in how and when we catch up on our sleep need is, however, limited by our internal timing system, as Sergeant Stein discovered during his sleep study. At some times we can fall asleep easily whereas at others we find it almost impossible—no matter how exhausted we are. The lunchtime dip is an excellent window to position a second episode of sleep during the day. But remember, the local time at which an individual experiences “lunchtime” depends on internal time—for extreme late types it might be as late as 6 P.M.
What we haven’t addressed yet is the issue of how sleep patterns change with age. The older our fictional Mrk got, the longer he found himself sitting around the fire. Do these changes concern merely how we distribute our sleep within the day? Do our sleep needs change with age? Or are we even different chronotypes at different stages of our lives?
12
The End of Adolescence
Urf had been lying on the forest floor for a very long time but nothing had come along the trail so far. He had lost any sense for how long he had been lying there in almost complete darkness, surrounded only by the noises of the nocturnal forest. He and the other three hunters had left their dwellings at dusk and had walked for half of a summer’s night before reaching the part of the woods where they hoped to find the deer. It was a good place—the deer had to pass through a narrow opening between two mountains to get from the dense and safe parts of the forest to the rich grasslands. They knew that they had reached their hunting ground too early—the deer wouldn’t pass this spot until sometime before sunrise. It had taken them much longer to reach their destination than it would take them to go home to their dwellings. The shortest way between here and home led through the dense part of the woods, which was an excellent choice for their homeward journey once they were carrying their heavy prey. The shortcut was, however, a bad choice for the outward journey when they didn’t want to announce their presence to all of the forest’s animals—especially not to those they were trying to hunt. Today, the wind was blowing in the direction of the grassland. That was the best condition for hunting in these grounds: if the hunters took the precaution of making a wide detour around the big forest, the deer could not pick up their scent.
It was essential for the hunters not to fall asleep while lying in the dark for many hours on end, without moving or communicating. The forest—their potential prey and the potential predators for whom they might be an easy bounty—was to take no notice of them. They had walked fast, almost running, on their journey in the night, circumnavigating the forest for three-quarters of the way. But then they stopped and washed off their sweat in a small stream and rubbed themselves with moss and forest soil. They walked the last lap to their destination very slowly and then positioned themselves strategically around the narrow exit of the forest. Then they covered themselves with leaves, moss, and soil and embarked on their endless wait. They had learned all the tricks of night-hunting from their elders, whom they now had succeeded in the job. The long walks, the long waits, how to stay awake during the odd hours of the night, how to carry the heavy prey over long distances on as few shoulders as possible—all these were activities endurable only by men of a relatively narrow age range.
One way to stay alert was to keep the mind occupied. Urf thought of his woman and his children, who were safe back in the caves. He was one of the most successful members of the younger generation, and because of his success, he had been chosen by the chief ’s daughter. They already had two children and his woman was pregnant with a third. Urf and his family didn’t sleep in the big cave, where the majority of the clan members had their resting places, but resided in one of the small side caves of the labyrinth. He remembered the many challenges he had had to face before he was recognized as one of their future leaders. When he was chosen to accompany the night-hunters as their apprentice, his teachers made him stay up several nights in a row and even made sure that he couldn’t catch up on sleep on the day before he was allowed to accompany them on a real hunting trip. Now that he had been the leader of the night-hunters for several winters and summers, he and his fellows would have to choose a new young man to be taught all the tricks of the trade. After another couple of seasons, Urf would have to step down because he would not be able to lie quiet and motionless for most of the night without either falling asleep or becoming restless.
Different ages are associated with different sleep patterns. Newborn babies are governed by a feeding rhythm that is much shorter than twenty-four hours. Over the course of several months, their behavioral rhythms become longer until they gradually develop a daily pattern. This doesn’t mean that the circadian clock doesn’t function properly at that early age; all it means is that consolidating sleep and wakefulness into long stretches of time would be counterproductive since sleeping, feeding, growing, sleeping, feeding, and growing are the most important aims at that stage—in addition to shaping the development of the brain by sensory inputs during the short wake episodes. Human life is often depicted as a circle: we end up just as we arrived—with no teeth, wrapped in diapers, unable to sleep through the night. Although this may be true in many respects, the similarity of babyhood and old age is often only superficial. The reasons for not being able to sleep through the night and for falling asleep frequently during the day are very different at the two ends of our existence. Eus Van Someren, a Dutch sleep researcher, has shown that lack of light is a main cause for the ultradian sleep pattern in the elderly and especially in the mentally ill.1 Most elderly people hardly ever get the chance to go outside, and the television is often their main light source. As a result, their body clock is not synchronized properly. Van Someren recorded the activity of subjects living in nursing homes and could thus quantify how much their nighttime sleep was disrupted by activity and, conversely, how much their daytime activity was disrupted by naps. He displayed these data as an activity profile, recorded with an actimeter.2
Although we can see clear differences in the amount of activity between night and day, a lot of activity was still recorded during the night (between 8 P.M. and 8 A.M.). These disruptions became much less frequent when his team installed bright light sources in the common rooms of these homes.
There is no doubt that our body and our brain change with age and that many functions work less efficiently. The centers in our brain that control rest and activity, sleep and wakefulness, are not exempt from these age-related changes. The experiments performed by Eus Van Someren show, however, that the age-related changes in our behavior are not necessarily a direct result of degeneration in the brain but are caused indirectly by other changes in our lives, such as lack of light and activity.
But what about chronotype and age? Does our internal time change across our lifespan despite its genetic determination? Genetic control does not preclude change over time: our body height is, for example, genetically determined, but we don’t reach that goal for many years, and we shrink when we grow old.3 Similarly, chronotype changes with age—most noticeably in teenagers. Most of us experience twice in our lives (once when we are in our teens and once again when our children reach that age) the fact that teenagers can stay up easily until the early morning hours and possess an unchallenged ability to sleep through the day—almost the entire day. This may seem obvious, but it has to be proven with quantifiable data, and we want to understand the underlying causes and mechanisms. Epidemiology provides a good starting place for quantifying and eventually understanding a phenomenon—an approach that needs really big numbers.
The nighttime sleep and daytime ac
tivity of patients in nursing homes are heavily disrupted. Redrawn from E. J. Van Someren, A. Kessler, M. Mirmiran, and D. F. Swaab (1997). Indirect bright light improves circadian rest–activity rhythm disturbances in demented patients. Biological Psychiatry 41:955–963, with permission from Elsevier.
The introduction of bright light in the nursing-home common room alters activity patterns among patients. Redrawn from E. J. Van Someren, A. Kessler, M. Mirmiran, and D. F. Swaab (1997). Indirect bright light improves circadian rest–activity rhythm disturbances in demented patients. Biological Psychiatry 41:955–963, with permission from Elsevier.
A colleague once asked me whether it was scientifically ethical and sound to continue to add data to our database while the art of statistics had taught us to extrapolate from smaller samples to the entire population. She had a point, because it would be overkill to collect data beyond the point at which we can make good predictions with the help of statistics. This is certainly true if you only collect data to answer one specific question. The true value of a growing database is that you can constantly come up with new questions that go into more and more detail. If we wanted to know, for example, the chronotype distribution for every age group between ten and seventy (amounting to sixty-one age groups), separately for men and for women, then we would need 122 categories. If we wanted to have some statistical security for the results in each of these categories, we would like to have at least as many individuals within each group as we have groups. So in this case, we multiply 122 by itself, and the database should contain the information of at least 14,884 people. If we wanted to go into even finer detail, for example, by comparing those who live in the countryside with those who live in towns, the database ideally would have to be quadrupled (59,536). Our Munich ChronoType Questionnaire database is huge and steadily growing (at present more than 120,000 individuals, adding approximately 300 new entries per month), so that we are able to do what we just described. We looked at the changes in chronotype during development and across a human lifetime.
Young children are relatively early chronotypes (to the distress of many young parents), and then gradually become later. During puberty and adolescence humans become true night owls, and then around twenty years of age reach a turning point and become earlier again for the rest of their lives.4 On average, women reach this turning point at nineteen and a half while men start to become earlier again at twenty-one. This one-and-a-half-year sex difference is typical for many aspects of human development. Although adolescence is often used to describe the stage between puberty and adulthood, its end hasn’t been clearly defined. When we saw this clear turning point in the developmental changes of chronotype, we realized that this was the first biological marker for the end of adolescence.5
Chronotype depends on age. Teenagers belong to the latest chronotype in our populations. Women are generally slightly earlier chronotypes than men.
Since teenage boys continue to delay their clocks for one-and-a-half years longer than teenage girls, men are on average later chronotypes than women during adulthood. This difference becomes smaller with age and vanishes at around the age of fifty-two. This coincides statistically with the age when women enter menopause. Of course, entering menopause isn’t the reason for the disappearance of the sex difference in chronotype. The reasons for menopause are changes in the hormonal cocktail being produced at different ages. Men also go through substantial hormonal changes with increasing age. The decrease of testosterone levels, for example, is one of the reasons that men lose their six-pack abdomens and develop a belly. In a way, the graph of changing chronotype shows changes of hormones that start at the age of twenty and then gradually progress throughout our life until they result in the obvious features that initiate our old age, or, to put it in more biological terms, our post-reproductive stage.6
In my view, these developmental changes have clear biological roots. I often have had to defend this view against purely cultural explanations, which frequently come in the form of the disco hypothesis: if teenagers would only go to bed earlier (in other words, not party until the early morning hours), they would be able to wake up fresh as daisies to attend school. We have meanwhile collected statistics from many different parts of the world and found that the phenomenon of age-dependent chronotype is not exclusively a characteristic of modern and urban societies. It is found also in purely rural areas, from secluded valleys in the Italian Alps to rural Estonia, India, and New Zealand. The strongest argument against the disco hypothesis came, however, from experiments I learned about in two recent conferences. These experiments showed that similar changes in chronotype occur during the life cycle of rodents—lab rats that definitely never visited a disco.
If we are correct that a later bedtime during puberty or adolescence is a biological program and not simply a lifestyle choice, one could even turn the disco hypothesis around. If teenagers are so late in their internal time that they don’t find sleep before the small hours of the night, then where can they hang out without disturbing the earlier chronotypes of society? Maybe discos are necessary niches or institutions for adolescents.
But why do adolescents have late chronotypes? In modern industrial societies, young adults around the age of twenty are (statistically) still involved in some kind of education and will not start a new generation (their own family) for several years. This is a relatively new state of affairs. In my story of the night-hunters, Urf is in his mid-twenties and has already been confronted with crucial challenges, the outcome of which have severely determined his future, including his place in the social structure of his clan. Urf obviously has “made it.” The night-hunters, who provide the clan with high-quality protein by risking their lives, are regarded with great respect, as the leaders in their age group. Their qualities might have been similar to those of the heroes of our time, as for example the winners of Olympic medals. Qualities such as endurance, imperturbable focus and attention span, or physical excellence are limited to a relatively narrow age range, which roughly falls into the peak of lateness in chronotype. The average age of Olympic competitors at present is twenty-five, and it has been even younger in the past.
But why would being late be better than being early at that age? Is there more to being a late or early chronotype than just the timing of sleep? A possible answer to that question comes from highly artificial (and quite gruesome) experiments routinely performed by clock researchers. To investigate the control of the circadian clock over bodily functions, cognitive performance, behavior, or biochemical parameters without the interference of other factors, such as sleeping or running around, clock researchers invented “constant routines.” Well-paid and highly motivated subjects stay up thirty to forty-eight hours performing tests on an hourly basis. Unlike Sergeant Stein and his fellow soldiers, constant routine subjects are not allowed to sleep at all. Many physiological parameters, like blood pressure or body temperature, are controlled by the circadian clock, but also respond directly to changes in the environment or in our behavior. A change in posture, a big meal, moving around, even exposure to bright light may change the level of a physiological parameter so that its control by the circadian clock becomes masked. That is why constant routine subjects spend the entire time lying down in a dimly lit room, frequently eating small and isocaloric snacks while performing many different tests, giving blood or urine samples, or spitting saliva into a tube—all at hourly intervals.7
We and others have performed constant routines in which we assess the subject’s chronotype with the Munich ChronoType Questionnaire before the experiment starts. There are several reasons for doing that. One is determine how much a short questionnaire can reveal about internal time: how well can chronotype, based on subjectively assessed sleep times, predict the circadian rhythm of other clock-controlled parameters? The other reason lies in the possibility of evaluating the data collected in a constant routine (or other circadian experiments), not only on the basis of local time but also on the more biological basis of individual
, internal time. The results obtained for the validity of the MCTQ have been extremely encouraging. The daily ups and downs of biochemical factors, such as cortisol or melatonin, highly correlate with the midsleep (chronotype) assessed by the simple questionnaire. We have performed other validations of the MCTQ data, such as asking subjects to keep six-week-long sleep diaries and to wear actimeters. All these validations showed that people are remarkably precise in estimating their usual sleep times, and that an individual’s chronotype relates to far more than merely his or her timing of sleep.
But how can a constant routine answer the central question of this chapter: what is the advantage of teenagers becoming late chronotypes? Constant routines are extremely challenging and, despite good will, certain subjects don’t want to continue participating after the first eighteen hours or so. Some can be persuaded to continue and others plainly insist on stopping. Those subjects are almost always early chronotypes, who claim that they just cannot stay awake anymore. Logically the same thing should happen to late chronotypes a couple of hours later, but astonishingly it doesn’t. It seems that an additional quality of being a late or an early type is related to sleep pressure. Early types apparently build up sleep pressure more rapidly than late types. In addition, early types don’t have the ability, as late chronotypes do, to “sleep in” after having gone through sleep deprivation.
These insights into chronotype are relatively new, and therefore have to be treated with caution. Sleep timing and the speed with which an individual builds up sleep pressure may not always be tied to one another. There are certainly early types that can go on for a long time without sleeping and still perform with admirable focus and quality, but this ability seems to be much more prevalent in late than in early chronotypes. But if there is something to the relationship of being a late type and being able to endure and “sleep in,” then late types are better night-hunters than early chronotypes. The lateness of teenagers in our modern industrial age may be a remnant of a skill that accompanies the age of peak physical condition.
Internal Time: Chronotypes, Social Jet Lag, and Why You’re So Tired Page 10