Sleep habits

Jump to: navigation, search

This text is part of: "Science of sleep" by Piotr Wozniak (2017)

Constellation of sleep habits

In this section, I would like to demonstrate that people can differ vastly in their sleeping habits, and some of the differences have an important underlying biological cause. Scientists use the term chronotype to differentiate between different sleeping time and duration preferences that characterize different individuals. One person's chronotype might make him a short sleeper. Another's chronotype will make him an owl. Yet another's chronotype will make his doctor diagnose a sleep phase disorder. Despite a seeming variety, a small set of underlying variables should make it rather easy for you to figure out your own chronotype. Your chronotype may determine your suitability for certain professions. Luckily, you do not need to determine your chronotype before you choose your major or your job. Many people naturally gravitate towards activities and professions that match their natural sleep habits. A physician or a fireman needs to tolerate shift work and interrupted sleep. Milkmen get up early, while gym or disco owners need to stay up late, while a writer may be of any chronotype as he/she can adapt his/her writing hours to his/her sleep patterns. To illustrate individual sleep patterns I use a freeware application called SleepChart that you can download here to visually chart your own sleep (Wozniak et al. 2003[1]). If you collect a few months of data, I would be very happy to receive your data file for analysis and future research. Sending SleepChart data requires a single click in the program.

Four examples of sleep logs that illustrate that modern human sleep patterns are as varied as snowflakes

Four examples of sleep logs that illustrate that modern human sleep patterns are as varied as snowflakes.

Body clock

The cycle of sleep and waking is regulated by the body clock. Body clock is located in the brain and is primarily based in the suprachiasmatic nucleus (see the chapter devoted to the SCN). The clock has a period of about 24 hours. During a single 24 hour day we have a period of 5-10 hours when we are very sleepy. This is the time when we normally sleep. During the remaining 14-19 hours we are usually awake or take a nap at siesta time. As mentioned earlier, only a small portion of the waking time is suitable for top-quality intellectual effort (see: Optimizing the timing of brainwork). The period of maximum alertness may last as little as 2-4 hours. We should plan our day in such a way so that sleep comes at the time of maximum sleepiness, while activities that demand maximum focus or creativity fall into the hours of maximum alertness. It is very difficult and usually very unhealthy to force the body and the body clock to change the timing of waking activities and sleep. It is far easier to do the opposite: adapt one's life to the natural cycle governed by the body clock. That adaptation will depend on the unique properties of one's own body clock. In the following sections I will try to show different types of sleep habits determined by the properties of the body clock that characterize a given individual.

Components of sleep in phase disorders

There are two main mechanisms that regulate sleepiness (see: Two components of sleep). One is the body clock, and the other is the "wake-meter". Body clock produces increased sleepiness every 24 hours. The wake-meter increases sleepiness with prolonged wakefulness (i.e. the longer we do not sleep, the sleepier we are). In sleep literature, these two mechanisms are called the circadian and homeostatic components of sleep propensity.

Sleep control components:

  • circadian clock - circadian clock produces sleepiness in 24 hour cycles
  • homeostatic control - wake-meter measures the period in which we stay awake and triggers sleepiness after we stay up for long enough

In people affected by DSPS or ASPS, there may exist a combination of several factors that make it harder to get good sleep in normal hours:

  • Circadian clock runs in periods far different from 24 hours. For example, in DSPS people, the circadian clock may be set to 25-26 hours.
  • Circadian clock is not sensitive to time resetting factors (termed zeitgebers). Normal people reset their clock in the morning by light and activity. In addition, darkness and inactivity in the evening provide further clues for the clock. Normal people with normal lifestyles can easily synchronize their sleep with the day-night cycle.
  • Homeostatic wake-meter has an unusual time constant. Sensitive wake-meters will make people get tired very quickly after awakening. Insensitive wake-meter may make people tend to stay up for long. Caffeine abuse could contribute to a fast decline in alertness via adenosine receptor downregulation.
  • Lifestyle has a dramatic effect on the behavior of the circadian and homeostatic sleep regulation mechanisms. The same individual will show a different sleep pattern depending on such factors as: using artificial lighting, exercise, level of stress, timing of exciting activities, napping, diet, climate, changes in ambient temperature, health status, etc.

Lark-owl misconception

Research shows that 15% of people would classify themselves as "morning type" or lark. Another 20% would call themselves "evening type" or owl. The remaining 65% are indifferent or "mid-range". What is your type? You can find many lark-or-owl tests on the net. However, I have not yet seen even one that would be well-designed to truly answer the question of your genetic predispositions. In particular, the same person on a work-week schedule may be classified as a different chronotype than when he or she is on a free running schedule.

Few people know that they can easily adapt to a completely different schedule by means of chronotherapy (e.g. by shifting their sleeping hours by 30-45 minutes per day). If you ask a typical owl to go to sleep 30-45 minutes later each day, the owl will keep shifting its bedtime to later hours. Initially, it will sleep during the day. That sleep will shift gradually to even later hours until the owl finds itself going to sleep in the very early evening just to get up before the larks! Surprisingly, even the most committed owl can then comfortably stick to the early waking hours for quite long! There is little natural preference as to the sleeping time of the day!

However, there is a factor that drives people into believing they are of a given sleep-time preference type. This is the length of the circadian cycle and their ability to entrain it to 24 hours. As mentioned earlier, typical circadian clock period lasts longer than 24 hours. Those people whose cycle is particularly long tend to go to sleep later each day. They push the limit of morning hours up to the point when their compulsory wake-up time results in unbearable sleepiness. In other words, people with long cycles will tend to work during the night and sleep in the morning as long as it is only possible.

Larks and owls do not differ in their preferred timing of sleep in reference to daytime! The apparent difference comes from the length of the circadian cycle, sensitivity to zeitgebers, and lifestyle. You can easily make a lark work comfortably late into the night and make an owl get up at 3 am. This can be done by chronotherapy (cycle adjustment)! Moreover, owls can keep getting up at dawn if they adopt an ancient farmer's lifestyle (e.g. by giving up electricity)

A smaller proportion of people will experience short circadian periods and experience extreme sleepiness in the early evening. This is the lark type. Life forces larks to go to sleep slightly later than their natural preference (family, work, light, etc.). This keeps larks in line with time and they will often claim that the quiet of the morning, the singing of birds or the beauty of the sunrise keeps them getting up early. Yet it is still possible to forcibly push a lark to gradually shift sleeping hours and behave like an owl!

In a modern society, only a small fraction of people can boast a perfectly steady and regular natural sleep pattern. Not only are these the healthiest people around, they are also creatures of habit in reference to their sleep and waking rituals. They obey their rituals religiously, avoid alarm clocks, avoid evening entertainment, avoid medication that affects sleep, etc. Unlike those well-regulated individuals, owls shifted to a morning schedule will gradually tend to advance to their standard late-night rhythm. Similarly, larks will quickly shift back to getting up with the birds.

Some correlation studies showed that owls (as defined by the timing of melatonin release) exhibit slightly higher IQs than larks (Roberts and Kyllonen 1999[2]).

Charting sleep

Understanding the control mechanisms that produce sleep and wakefulness is extremely helpful in understanding sleep habits. It is particularly useful in individuals suffering from a number of sleep disorders, esp. insomnia and phase-shift disorders. Simple measurements of circadian variables and simple tools of chronotherapy may bring sound sleep to those who often struggled for years with insomnia, unsatisfying sleep, or sleep in wrong hours. Better understanding of chronobiology could also help extinguish dangerous practices such as poorly planned shift-work, disrespect for health consequences of the jet lag, cumulative sleep deprivation and the Internet fad of Uberman sleep.

To illustrate various sleep habits I use charts from a freeware program SleepChart. You can download SleepChart here and begin your own analyses today. All you need to do in the program is to click the beginning and the end of the sleep block in the graph. See the bottom of the SleepChart window for exact time corresponding with the position of the mouse pointer. If you set a wrong block, select it with a click and press Del.

Using SleepChart data, I will try to explain the main reason for which healthy people may not be getting refreshing sleep: sleep phase problems.

SleepChart attempts to approximate the circadian acrophase that correlates with maximum sleepiness, low body temperature, low ACTH, high melatonin, etc. The underlying assumption is that when you log your sleep with SleepChart, you do not attempt to artificially play with the sleep hours. Each intervention in the sleep schedule makes the tools used in SleepChart work with lesser accuracy. Here are the most important interventions that should be avoided:

  1. waking up with an alarm clock,
  2. combating sleepiness in the evening (e.g. in order to delay sleep), and
  3. controlling sleep with substances (e.g. alcohol, sleeping pills, etc.).

On those rare occasions when you delay sleep or use an alarm clock, you can disqualify the sleep episode with the appropriate markings. However, all attempts to modify the sleep schedule will partly fool the algorithm and your reading will be inaccurate or plain wrong. It is also very important that you do not attempt to follow the circadian approximation when determining your optimum sleeping hours! You should always give priority to your natural body signals, i.e. sleepiness. Following SleepChart approximations can result in a positive feedback of error. In other words, errors in the graph may be amplified by your attempts to follow the graph. This can disrupt the sleep cycle. At worst, you could even self-diagnose yourself with DSPS without actually suffering from the disorder! Your only and sole "go to sleep" criterion should be rapidly increasing sleepiness. You may use the graph to approximate the moment in which the readiness for sleep will occur so that you could "cool down" in time. You can also find SleepChart helpful in chronotherapy for ASPS or DSPS to make it easier to schedule your appointments without conflicting with your natural sleep rhythm.

Courtesy of the numerous contributors who sent in their SleepChart data, we can draw a number of interesting conclusions. The most compelling one is probably the confirmation of the hypothesis that we might be facing an epidemic of Delayed Sleep Phase Syndrome (DSPS) in younger generations, esp. among students and people employed in high-tech jobs. The epidemic is a result of an ever-growing discrepancy between the environment in which humans and their primate ancestors evolved over the last several million years, and the environment in which we live today with electric lighting, Internet, computers, TV, rat race, and 24-hour society. The increasing gap between lifestyles and biology leads many to seek radical solutions and take on drastic measures. A quick survey of those who attempted to adapt to an Uberman sleep schedule reveals an interesting truth. Although the idea to squeeze in more waking hours into a day is very appealing, most of the "experimenters" began their interest in polyphasic sleep as a result of troubles with achieving refreshing sleep!

Some people reacted with skepticism to the concept of using SleepChart as a sleeping prop: "it is just far too complicated and Ockham's razor needs to do a bit of shaving! Sleep is as natural as breathing air or drinking water and if you have to set up complicated charts and experiments, and utterly eccentric sleep-activity patterns just so as to get some decent shut-eye, then you must have a problem - but one more of a psychological than a physiological nature". It is true that sleep will occur naturally in a natural setting. The trouble begins when we interfere with nature using caffeine, alcohol, nicotine, artificial lighting, 24/7 society, night-time entertainment, etc. SleepChart may seem complex, but it might still be the easiest way to predict the optimum timing of sleep in free-running conditions for people who may have problems with sleeping. SleepChart will only ask you when you go to sleep and when you wake up (naturally). All computational complexity is hidden in the background. The approximation procedure needs no further input from the user and it predicts the circadian acrophase as well as the optimum bedtime. SleepChart can even disentangle homeostatic and circadian components of sleep. Understanding these can also be helpful in planning healthy sleep.

I agree that the need to resort to tools such as SleepChart is a sign of troubled times. However, SleepChart has a proven record of helping people understand their seemingly irregular sleep patterns and organizing their sleep. Falling asleep might be natural, but there are many factors that mask sleepiness or magnify it. For people on very irregular sleep schedules this can pose an insurmountable obstacle!

People with sleep problems are often little understood by the naturals: "If you work solidly 8 hours a day, have 3 decent meals, have a proper family life, and treat other people as human beings, then in the evening you go to bed happily knocked out and wake up next morning happily refreshed. Surely this is as it always has been for most people throughout history and surely this is how it will always remain". This attitude towards sleep problems is not much different from telling a clinically depressed person: "Pull yourself together", or expect a heroin addict to go cold turkey and instantly return to normal life. A tortured insomniac will only get more upset with himself or herself if (s)he is told that sleepless nights come from "unsolid work", "indecent meals", "improper family life" or treating others "inhumanely". The trouble stems from the clash of biology with modern lifestyle. With the arrival of artificial lighting sleep disorder statistics skyrocketed. These were only made worse by television, computer games and the Internet. With the advent of mobile telephony and instant messaging, insomnia and sleep phase disorders seem to reach epidemic proportions. Fewer people are able to leave work behind, cope with stress, or give up evening activities. Without a major change in lifestyle or a breakthrough in circadian control methods, people affected with lifestyle-related sleep disorders are faced with a choice between a daily sleep deprivation misery and radical solutions such as throwing away the alarm clock. Certainly, we can expect science to come up with answers to the problem. Until that happens though, waking up "happily refreshed" remains a privilege of a shrinking subset of the population in industrialized nations.

SleepChart in SuperMemo

To make it possible to analyze the connection between sleep and learning, SleepChart has been integrated with SuperMemo speed-learning software. Instead of explaining SleepChart itself, I will shortly describe its functionality in SuperMemo. Keep in mind that some of the functions related to memory are not included in the freeware version due to the fact that it does not have access to your learning data.

SleepChart was included in SuperMemo a few years ago upon the understanding that sleep is vital for learning. To sleep well and to learn well, one needs to understand his or her own circadian rhythm. SleepChart in SuperMemo was designed with the view to assisting in that task. It can help you optimize the timing of sleep as well as to optimize the timing of your learning. Moreover, you can submit your sleep and learning data for analysis and have your own contribution in our research over the impact of sleep on memory. You can access SleepChart in SuperMemo with: (1) Tools : Sleep Chart on the Main menu, (2) SuperMemo commander, or (3) by just pressing F12.

SuperMemo: Tools: Sleep & Learning Timeline displaying repetitions blocks of the current collection (in red) and sleep blocks (in blue)

Sleep blocks are marked in blue. Learning blocks are marked in red. Total learning time on individual days is displayed on the right. Selected sleep block is displayed in yellow. The length of that block is displayed at the bottom.

In SuperMemo, the learning timeline is generated automatically. Each time your make repetitions with SuperMemo, the learning block is added to the timeline (displayed in red on the graph). On the other hand, your sleep data must be logged in manually (displayed in blue). At minimum skill level, you can use SleepChart for a basic visual inspection of your favorite learning and sleep hours. However, more advanced functions such as optimizing the time for learning or the time for sleep require advanced analysis and understanding of circadian rhythms. If you start logging your sleep data today, you will be able to use future, more advanced versions of SuperMemo to study and understand your sleep and learning.

Sleep log

The timeline of sleep in SleepChart must be logged manually. To log a block of sleep, click the beginning of the block (sleep start) and then click the end of the block (sleep end). You can also start from clicking the end of sleep first. Sleep blocks above 22 hours are disallowed. Sleep blocks cannot overlap with repetitions timeline (you cannot learn with SuperMemo and be asleep at the same time). If you have already collected your sleep data in SleepChart Freeware, you can import this data to SuperMemo with File : Import : SleepChart file (you can also import data from a spreadsheet). If you import files from SleepChart Freeware, you can test for sleep and learning overlaps with File : Verify : Block overlaps. Protection from block overlaps is an important advantage of using SleepChart in SuperMemo as opposed to a standalone SleepChart, in which it was very easy to fall out of phase in logging data (e.g. by failing to fill out a single day and noticing that only a month later). You can mark blocks of forcefully delayed sleep, as well as mark blocks cut short with an alarm clock or other factors. Please note that you can get best analytical results if you do not artificially regulate sleep (e.g. with an alarm clock, sleeping pills, etc.). Applied models will not fully account for artificial intervention. Last but not least, natural sleep is what you should aim for in learning as well as for the sake of maximum health and well-being.

Sleep and learning timeline

Combining sleep timeline with repetition data taken from SuperMemo opens an array of new research and optimization options.

Various sleep statistics pertaining to individual days can be displayed on the right. Sleep blocks can be consolidated with the Consolidate button on the toolbar. For example, if you woke up for 5-10 min. in the night, consolidation will make SuperMemo treat the entire night sleep episode as a single entity. Short nocturnal awakenings are a norm, even if we are not aware of them, and have little impact on learning. Sleep block consolidation often unmasks important properties of sleep (e.g. see Preference for night sleep). It helps treat successive sleep episodes as an expression of a single period of high sleep propensity.

In addition to sleep statistics, optimum bedtime can also be estimated in SleepChart. Two independent models are used to predict middle-of-the-night points as well as the expected optimum retirement and awakening times. Those approximations may be helpful in optimizing sleep in people who work shifts or sleep in irregular hours for various reasons. For example, after a week of irregular sleep, it may be difficult to determine the optimum retirement hour that is likely to produce best quality sleep. Going to sleep too early might result in premature awakening (which may often ruin the night sleep entirely). Going to sleep too late may result in short night sleep, sleep deprivation, and reduced alertness on the following day. Predicting optimum sleep time on the basis of sleep history is inexact science, and the two models used may produce different outcomes. Important! Your natural instinct should always take precedence over mathematical models. Moreover, best results in sleep optimization are accomplished in free-running sleep. If you use an alarm clock, or force yourself awake through the night, or take sleeping pills, the models may not adequately account for the chaotic change that is occurring in your sleep control systems.

SuperMemo: Sleep and repetitions timeline displaying repetitions blocks of the current collection (in red) and sleep blocks (in blue) with recomputed circadian approximations on the current data

Blue and red continuous lines are predictions of optimum sleep time using the SleepChart model (based on sleep statistics). Yellow continuous line shows the prediction of the maximum of circadian sleepiness (circadian middle-of-the-night peak) using a phase response curve model. Note that theoretically, yellow line should roughly fall into the middle between blue and red lines. However, when a disruption of the sleep pattern is severe, those lines might diverge testifying to the fact that it is very hard to build models that fully match the chaotic behavior of the sleep control system subjected to a major perturbation. Aqua dots point to the predicted daytime dip in alertness (i.e. the time when a nap might be most productive).

Circadian graph

The circadian graph in SleepChart can help you better understand your sleep patterns, as well as to visualize the degree of cycle instability (i.e. how difficult it is for your sleep-wake cycle to fit into 24 hours). You will need a few months of data before the graph becomes meaningful. In addition, subjective night approximation lines in the sleep log are subject to substantial hysteresis. If your lifestyle changes dramatically (e.g. as a result of a therapy), you may need a few weeks for the approximation lines to align properly with data. The circadian graph may then be more difficult to interpret. In such cases, you can use From the first day and To the last day options to demarcate the period of interest. This will limit the analysis to a selected period characterized by a selected lifestyle.

Exemplary circadian graph of a monophasic sleeper

Exemplary circadian graph of a monophasic sleeper

Sleep initiation

Blue line shows the preferred time to fall asleep. It corresponds with sleep propensity derived from the number of sleep blocks falling into a given hour of the waking day, where zero on the horizontal axis refers to the hour of waking up. Percentage of sleep episodes initiated at any given time is displayed on the right vertical axis. The blue line roughly expresses your "tiredness of wakefulness". It also expresses your ability to fall asleep. Your own optimum bedtime hour is your personal characteristic as it differs between people. For most people the optimum bedtime falls into the range of 16-20 hours from waking. In the example, the most favored bed time occurs in the 18th hour of waking.

Sleep maintenance

Red line shows the average length of sleep. This line is a rough reflection of the ability to maintain sleep, i.e. the longest sleep episodes occur during the subjective night. The average length of sleep is displayed on the left vertical axis. The graph will tell you that even if you are able to initiate sleep during the day, it will never last long. In most cases of regular sleepers, only after 11-14 hours of waking does the length of initiated sleep start increasing. Note that the sleep length graph is slightly phase shifted in reference to the preferred sleep initiation time due to the fact that long sleep is mostly achieved by initiating sleep early.

Optimizing bedtime

If you are trying to determine your optimum bedtime, find the evening peak in the blue curve and choose nearby points that produce sufficiently long sleep (red curve high enough). In addition, pay attention to the fact that your wake and sleep time should add up to 24 hours, otherwise you will experience phase shifts.

Some people take naps during the day. Others don't. In nappers, the blue curve should also point to the maximum mid-day alertness dip. Short nap time may actually be a sign of good nap timing as long as the nap is not taken too early in reference to the blue curve (see: Best nap timing). Non-nappers will also experience a peak of sleepiness around the 7th hour even though their blue curve will not show as a prominent bulge.

If the graph shows that your optimum nap time falls into the 8th hour, and you wake up at 6 am, you should take a break at around 14:00 (2 pm) and look for a secluded place for a few minutes of rest. You could also plan your lunch at around 13:00-13:30 to complete a perfect setting for a siesta.

Example 1: Unstable circadian cycle

In the exemplary circadian graph below, on average, the best night time sleep is obtained when it is initiated after 18 hours from the morning awakening (assuming the graph was created without any artificial form of sleep control such as an alarm clock, delaying sleep, etc.). The blue line shows that the 18th hour is the preferred time to initiate sleep, while the length of sleep (red line) is long enough to add up to a 24 hour sleep-wake cycle.

Unstable circadian cycle

As blue peaks are of the same height, we can conclude that the graph represents a religious napper, whose optimum siesta time occurs 7 hours from awakening. In this case, for an awakening at 8 am, the siesta should begin at 3 pm, and the night sleep around 2 am. For both blue peaks, 7.4% of all sleep episodes being at the optimum hour, while the remaining 85% are suboptimum.

Maximum length of sleep can be achieved at the 16th hour, however, this does not indicate this is the optimum hour of going to sleep. If sleep is initiated too early, it may or may not catch on the full circadian low of the subjective night. In other words, there is a risk of a premature awakening after just a couple of minutes of sleep. Such an awakening makes it harder to fall asleep again. This is one of chief causes of insomnia. The difficulty in re-initiating sleep is due to a very rapid loss of homeostatic sleep propensity during sleep. In addition, sleep initiated before the full circadian low does not seem to be of more value than slightly shorter sleep initiated a bit later (e.g. as reflected by the subjective feeling of being refreshed in the morning, or as measured polysomnographically). The blue homeostatic line indicates that the sleep is more likely to be initiated effectively at the 18th hour, while its average length is then 6 hours. If your graph is generated without attempts to artificially regulate sleep, the second peak in the homeostatic curve will often indicate the optimum bedtime. The graph also indicates that if the sleep is delayed by an hour, it will be shortened by 10-30 minutes. It is possible, that even this little shortening will affect the performance during the day. If the sleep is advanced by an hour, it may be 10-30 minutes longer but its quality is not likely to increase proportionally.

The graph can also show how the length of the circadian period can be determined by the bedtime hour. The green line shows the set of breakeven points for a stable 24 hours sleep-wake cycle where the sleep and wake times add up to 24 hours. All the circadian graph points that lie to the right of the green line cause a phase delay, while points on the other side will cause a phase advance. Aqua blue line shows where the 24-hour-cycle green line crosses the red sleep length line. Due to the fact that the angle between green and red lines is large, this sleep pattern is pretty unstable. This means that going to sleep before the 18th hour will result in a cycle that is less than 24 hours long, while going to sleep after the 18th hour may lengthen the cycle and result in phase shift delays. For example, early bedtime (around the 15th hour) will result in a day that lasts 21 hours (15 hours on the horizontal axis corresponds with the average sleep length of 6 hours read from the red curve). Later bedtime (around the 18th hour) will result in a perfect 24 hours day, while a very long waking day (e.g. 20 hours) will produce a day lasting 25.5 hours. Naturally, all manipulations in the length of the day would better be avoided as early bedtime increases the chances of insomnia, while a very late bedtime increases the chances of sleep deprivation, and REM sleep deficit. Understanding one's sleep preferences can be very helpful for planning shift-work or combating jetlag in long-haul flights.

Example 2: Stable circadian cycle

The second graph shows a sleep pattern that is much more stable that the one from the first example above.

Stable circadian cycle

The graph shows a habitual napper who shows a preference for a waking day of 19 hours. As opposed to the graph shown earlier, the zone of stable sleep-wake cycle, demarcated by vertical aqua lines is much wider due to the fact that red and green lines are nearly parallel. This means that if the sleep is initiated after the 20th hour of waking, the night sleep will be shortened to fit the 24h cycle. Naturally, even if delayed sleep does not cause a phase shift, it will always result in lesser sleep quality due to stage compression. Such sleep will result in sleep deficits. Days lasting less than 20 hours may result in a phase advance. Despite running free, the longest average sleep period (initiated at around the 16th hour) isn't even 6 hours long. This illustrates that excessive sleeping is not a problem in free running sleep. In the graph, the optimum siesta time again falls in the 7th hour and is executed religiously (over 14% of sleep episodes executed "on the dot").

Phase shift disorders

As shown in both graphs above, with sufficient discipline, people with phase disorders should be able to accomplish 24 hour free running rhythm independent of the desired waking hour. In practice, due to various perturbations in lifestyle (exams, stress, socializing, etc.) as well as due to the stress of the need to wake up early, adherence to the optimum 24h sleep schedule may be very hard to achieve for people with severe phase shift problems. For those who need to wake up at a specific early hour, free running sleep may become unobtainable without the use of an alarm clock, melatonin, or other unwelcome measures.

24-hour sleep cycle

Perfect 24-hour cycle

Let us now consider an ideally synchronized 24-hour cycle. In the picture below, an octogenarian female wakes up naturally everyday around 3:00-3:30 am. She sleeps 5.4-5.5 hours per day, wakes up refreshed and is active throughout the day.

The sleep timeline of the perfect 24 hour cycle of an octogenarian female.

There is no synchronization with daylight as the waking hour falls into the period of darkness. The cycle is synchronized by evening activities, not daylight. The subject keeps in her mind a "must go to sleep" hour estimation that helps synchronize body clock with the time of day. This "psychological imprint" is illustrated by a smooth change in the sleeping rhythm after the end of the daylight saving time on Sunday October 27, 2002 (the graph disregards DST so that the waking hour before the change is set at 2:00-2:30 am).

Even though aging is said to increase nocturnal awakening, perhaps due to the cell loss in sleep control centers, this subject reported no awakening in the study period.

Circadian graph shows a single favored bedtime in the 19th waking hour. As the average nighttime sleep episode is 5 hours long, the sleep-wake cycle lasts exactly 24 hours, and daily fluctuations in bedtime are minimal. As the green breakeven line and the red circadian line are nearly parallel in the span of 3 hours, this sleep pattern is very stable, and all delays in bedtime occur at the cost of sleep time without causing a phase delay.

The circadian graph of the perfect 24 hour cycle of an octogenarian female.

Sleep and stress

Stress can ruin the fabric of sleep. The following SleepChart graph demonstrates the impact of stress on a well-balanced 24 hour sleep pattern:

The impact of stress on a well-balanced 24 hour sleep pattern

In the presented example, a middle-aged self-employed male wakes up naturally everyday around 6:00-6:20. However, on Jun 3, 2003, a severe family problem threw the rhythm into chaos as evidenced by frequent nocturnal awakenings. The rhythm returned to the norm one month later as soon as the family conflict was resolved.

Monophasic sleep

Monophasic sleep graphs will often show a small siesta-time sleep propensity peak due to the fact that even the purest monophasic sleeper hits crisis days in which a postprandial nap brings a welcome relief. Due to their "crisis nature", such naps may last longer than in a habitual napper. The mid-day peak is particularly visible in irregular sleepers who show less discipline in sheltering their natural regular sleeping hours.

Circadian graph of monophasic sleep showing a mid-day siesta sleep preference

Preference for night sleep

Independent of the innate circadian cycle, light has a powerful impact on sleep. In particular, its phase-shifting capacity will always ensure that humans naturally gravitate towards sleeping at nighttime. Only the advent of lifestyle that involves electricity and 24h work cycles triggered the present epidemic of sleep disorders, which indirectly contributed to the appeal of concepts like "Uberman sleep".

The preference for sleeping in the night can best be seen in irregular sleepers, esp. those who suffer from phase shift disorders and run their sleep free, or those who are on a free running schedule and phase shift "by choice" (i.e. by not trying to fit any particular sleeping hours). In those cases, using the circadian graph in SleepChart, we can see the impact of nighttime on the ability to initiate and maintain sleep.

In the presented circadian graph, we see a clear preference for night sleep in free running sleep. The graph shows that sleep initiation (blue line) is easier at nighttime between 7 pm and 4 am, while the length of sleep (red line) is greatest if the sleep is initiated between 10 pm and 5 am.

Absolute time circadian graph showing a preference for night-time sleep in free running sleep

The graph can also be interpreted as a phase space. It shows how difficult it is to achieve "wasteful" 8 hours of sleep in an efficient free running sleep pattern. It can also be used to demonstrate that no trajectory in the phase space will lead to an entrained polyphasic sleep. When alarm clock and/or sleep delay are introduced into the system, sleep control may become chaotic. However, in free running mode, it quickly stabilizes around a roughly biphasic rhythm, often with a degree of phase-shift dependent on the lifestyle. The timing of phase-shifting, excitatory and inhibitory stimuli, even if they are repetitive and regular, may still lead to a degree of chaos in the system. This occurs if the period of the stimulus cycle is different from the period of the entrained circadian rhythm.

In contrast to the first graph, the second example can be used to argue that artificial lighting can virtually eliminate the impact of natural light on the cycle in a well-disciplined sleeper with a more regular cycle and better adherence to free running sleep rules.

Absolute time circadian graph showing a preference for night-time sleep maintenance with no preference for night-time sleep initiation

The question remains open to whether the nighttime sleep preference isn't to a large degree caused by social entrainment. Despite the fact that we live in a 24/7 society, there is still more fun to be had during the day or in the evening than during the night when still the larger portion of the population is asleep. A big clue comes from the fact that despite little difference in sleep initiation preference throughout the day, sleep initiated in the evening or in the night (8 pm - 6 am) is still likely to last up to twice as long as sleep initiated at 3 pm.


  1. Wozniak P.A., Gorzelanczyk E.J., "Formula for healthy sleep" (2003)
  2. Roberts R.D. and Kyllonen P.C., "Morningness–eveningness and intelligence: early to bed, early to rise will likely make you anything but wise!," Personality and Individual Differences / Volume 27 / Issue 6 (December 1999): 1123-1133