Polyphasic Sleep & Adaptation

Sleep Pressure and achieving SOREM

REM and SWS sleep are both necessary for the body to function properly. While abstaining from them, sleep pressures build, until the body tries to get the rest it needs. REM and SWS have different speeds of building pressure, and feel different. Severe sleep deprivation results in change in the structure of sleep and proportion of time spent in the individual sleep stages.  SWS and REM are prioritized over light sleep, which leads to Sleep Onset REM (SOREM) or Sleep Onset SWS (SOSWS) depending on which Homeostatic pressure is highest.

SWS debt is more likely to unpredictably override REM and light sleep, since its pressure is strongly dependent on homeostatic pressure rather than circadian pressure. REM pressure, however, stays more limited to its circadian rhythm1 i.e. REM pressure is particularly likely to take over before dawn till soon after the end of midday slump, especially at the 06:00-09:00 REM peak.

If a person is sleeping a reduced amount of REM or SWS from their daily needs, sleep pressure builds up over time. REM deprivation typically happens on any reduced-sleep schedule, while SWS deprivation is typically only problematic with schedules that have less than 3 core cycles (nominally 4.5h of core time).

Achieving SOREM plays an important factor in naps: Normally, the first bit of a sleep phase is NREM2. Due to the increased need, the brain can learn to put REM in this timeframe instead, leading to a restructuring of the sleep cycle. Once sleep pressures normalizes, this ability often remains, leading to REM in naps.

NREM2 never gets prioritized: during a study it was observed that 53% of missed REM was regained, 68% of missed SWS, yet only 7% of missed NREM2 was caught up on2. These observations serve to support the belief that NREM1 and 2 are less essential parts of sleep.

After a sufficient amount of recovery sleep, the brain switches back to the usual sleep cycles. With SOREM, this still happens roughly 25 minutes in; after this, SWS will begin. This is the reason why naps are recommended to last for 20 minutes.

If more sleep is scheduled than needed, you will wake up once you’re done sleeping. Mainly underaged people often claim to need less sleep on mono than they would get on biphasic schedules like segmented. Scheduling in the standard amounts of sleep according to the schedules is still recommended. More info about age restrictions can be found in “Lifestyle considerations”.


Before repartitioning the naps will not contain REM, and one will only enter SWS if sleep deprived. The core will also have a significant portion of light sleep. During adaptation the body will slowly start trading off light sleep for quality sleep in the core (refer to “adaptation” for more information). Once SOREM is achieved often enough, the body will also learn to rely on it in order to get the needed quality sleep regardless of one’s level of sleep deprivation, which means REM filled naps happen even after adapting, when all sleep debt is recovered.

Once adaptation is complete, the sleep schedule is stabilized and predictable amounts of each sleep type should be present in every session. Each cycle in the core(s) will have a percentually larger amount of quality sleep compared to monophasic sleep. The earliest cycles will be the densest quality-wise.

Naps can have different amounts of quality sleep depending on the time of the day. It is common to have a “long” feeling nap around dawn, suggesting that the nap is REM dominant. The body still needs to go through NREM1 and NREM2 in order to have REM, but it is a much shorter amount compared to the beginning of adaptation, and NREM3 is bypassed to enter REM. If the nap is taken late in the day, close to evening one might enter REM very fast after lying down, or it could be an SWS dominant nap, in which you recalling dreams is rare and it feels short and deep. The nap can also contain both SWS and REM.

Conclusively, after repartitioning is complete, NREM1 and NREM2 are reduced in the sleeps, saving space for SWS and REM.

Minimal schedule length

Regardless of our wishes we still have to sleep for a certain amount every day. When choosing a schedule, it is wise to consider the “Minimum Sleep Threshold”. This number is found when you add up your daily requirements for both REM and SWS; typically this is between 3 and 4 hours. NREM2 is also needed to some extent as a transitioning phase; however on extreme schedules this can be as low as 20 minutes total.
Some people may have (significantly) lower sleep requirements, making adaptation to more extreme schedules easier compared to people that need an average amount of sleep.

Sleep schedules that are shorter than the time needed to fit all needed sleep once repartitioned, are extremely difficult to adapt to. These have an extraordinarily low success rate, since both a high cycle depth and a high level of compression must be reached. The long-term sustainability and health risks of these schedules are questionable. Any and all oversleeping on these schedules immediately destabilizes them.

Interrupted Sleep

If you wake during the night and go back to sleep, the sleep cycle continues in the same place where it left off, replacing quality sleep with light sleep as the cycle resumes, leading to an overall lowered amount of quality sleep3.

If you schedule two sleep segments too close together, the same occurs, leading to unpredictable cycles which leads to bad wakes, lower sleep depth, and lower compression.
It is recommended to leave at least 1 BRAC (approx.2 hours) between sleep segments because of this; however some people may find that they prefer a slightly longer amount. There is anecdotal evidence suggesting that instead of 1 BRAC, 2 or even 3 BRAC lengths would be preferable (3-4 hours), as some people have reported a feeling of great uncomfort and a very hard time adapting to anything less than 2 BRACs apart.

Wake Time Programming

If you wake up at the same times of day consistently, the brain will learn to insert light sleep at that exact time, leading to an easier and more pleasant wake. In time, it even becomes likely that you’ll wake up entirely of your own accord at this point without needing an alarm.

Smart alarms should be avoided, because they do not maintain a consistent waketime. Since they rely on very limited information from the phone sensors to track your sleep, they are not effective at this task either4–7.

Wake Time Reduction

By reducing the length of a sleep segment, the ability to stay awake afterwards is also reduced. While a monophasic sleeper can stay awake the rest of the day after 8 hours of sleep, a polyphasic sleeper can not afford to do this. This effect is likely attributable to NREM2’s functions as a wakefulness sustainer.
However, an effective 20-minute nap has a similar effect on wakefulness restoration as a full 90-minute cycle. For optimal results, it is a good idea to keep the circadian and ultradian rhythm in mind when scheduling naps.

A common pitfall when adjusting a schedule is to make the gap between sleeps too large. While this might seem like a good idea on paper, the pre-made schedules are the way they are for a reason. If a gap is too big, you’ll feel tiredness during the later part of said gap, regardless of other adaptation processes. The schedule then becomes impossible to adapt to fully.

Second Wind

If you go beyond your normal sleep times, you may encounter a sudden boost of energy a little while after (caused by cortisol, which activates adrenalin8). These boosts are known as Second Wind, or “the wake maintenance zone”9. During these boosts, it may be difficult to fall asleep, regardless of tiredness levels throughout the day. Usually Second Wind lasts 2-3 hours.

This boost of wakefulness is caused by your circadian rhythm cycling back into a phase when you’d normally be awake; it is most commonly experienced in the (early) morning.
After second wind, you may hit third wind etc. in 24-hour intervals.


When starting a polyphasic schedule, it is important to keep in mind that most processes remain the same as when sleeping monophasically. Processes like the circadian rhythm can be used to determine optimal sleep placement, and sleep cycle length can be used to determine the optimal sleep length for you to avoid sleep inertia. While most of the scheduling articles offer good guidance on the placement and length of the individual sleep blocks and naps, sleep is very individual and therefore some slight experimentation may be needed to figure out what works best for you, depending on the mentioned factors and your individual sleep need.

Main author: Jelte1234
Scientific sources: Crimson


Dinges DF. Differential effects of prior wakefulness and circadian phase on nap sleep. E. 1986;64(3):224-227. doi:10.1016/0013-4694(86)90170-7
GULEVICH G. Psychiatric and EEG Observations on a Case of Prolonged (264 Hours) Wakefulness. A. 1966;15(1):29. doi:10.1001/archpsyc.1966.01730130031005
Aaron J, Carlisle C, Carskadon M, Meyer T, Hill N, Millman R. Environmental noise as a cause of sleep disruption in an intermediate respiratory care unit. Sleep. 1996;19(9):707-710. [PubMed]
Policies and procedures of accreditation for programs in nursing education. No. 14-1437. NLN Publ. January 1972:1-21. [PubMed]
Kolla BP, Mansukhani S, Mansukhani MP. Consumer sleep tracking devices: a review of mechanisms, validity and utility. E. 2016;13(5):497-506. doi:10.1586/17434440.2016.1171708
Lee J, Finkelstein J. Consumer sleep tracking devices: a critical review. Stud Health Technol Inform. 2015;210:458-460. [PubMed]
Lee-Tobin P, Ogeil R, Savic M, Lubman D. Rate My Sleep: Examining the Information, Function, and Basis in Empirical Evidence Within Sleep Applications for Mobile Devices. J Clin Sleep Med. 2017;13(11):1349-1354. [PubMed]
Wilson JL. Clinical perspective on stress, cortisol and adrenal fatigue. A. 2014;1(2):93-96. doi:10.1016/j.aimed.2014.05.002
Shekleton J, Rajaratnam S, Gooley J, Van R, Czeisler C, Lockley S. Improved Neurobehavioral Performance during the Wake Maintenance Zone. J Clin Sleep Med. 2013;9(4):353-362. [PMC]