Accounting for the time to fall asleep
When choosing a schedule a question that might arise is “how much time should I add to these sleep blocks to account for the time it takes for me to fall asleep”. This is a reasonable question, however the answer to it is: none at all. The existing schedules, which can be found in the Polyphasic Schedules-section of this site, are all created to display the total bed time (TBT). TBT is a scientific way to express the total duration of attempting to sleep and actually sleeping, or in other words, the time you spend laying down when you wish to sleep. After a successful adaptation, the time it takes to fall sleep should regardless not be any longer than a few minutes, so the TBT will be very close to the total sleep time (TST).
Polyphasic sleep adaptations utilize several different defence mechanisms to allow for the needed amounts of SWS and REM to be gained. One of these defence mechanisms is very short sleep onsets, and is the main reason why polyphasic sleep is a good method to fight insomnia. So while it might take a very long time to fall asleep at the beginning of adaptations, this issue will fix itself later. Simply adding more sleep time to account for the time it takes to fall asleep is a bad idea, since after sleep onsets shorten the alarms may go off during SWS, which increases the risk of oversleeping.
One suggested method to deal with long sleep onsets versus sleeping at all is to allow for additional time to fall asleep, and then gradually decrease the duration of the sleep times as sleep onsets shorten. There are a few reasons why decreasing the time it takes to fall asleep gradually is a bad idea. The main reason is that fast sleep onsets can be triggered very drastically and unpredictably, which means that a normal nap or core can turn into ending in a mid-cycle wake. Another reason is that the sleep onset times might not actually decrease at all, since there is no pressure for them to do so. This results in unnecessary time being spent in bed. A third reason is that either moving the alarm times or the start-of-sleep times can disorient the body, which can cause feelings of uneasiness and discomfort, possibly even affecting the adaptation negatively.
If falling asleep proves to be a major issue increasing the sleep time of the naps by 2 minutes should be tolerable. Still, the optimal nap duration is 20 minutes. Cores are still recommended to be kept at whole cycle lengths. However, if there are issues with severe insomnia persisting after a month the total sleep time of cores can be increased to allow for sufficient time to fall asleep and sleep for whole cycle lengths. Having anything of this sort happening is extremely rare.
Falling asleep faster
There are ways to shorten the time to fall asleep, and these are particularly useful to utilize at the beginning of adaptations, since it is normal for the homeostatic pressure to not be alleviated at as large degrees as further into adaptations from only NREM1 naps. It should also be pointed out that you might become dependant on these to fall asleep if you use them throughout adaptation, so pick wisely which you go for. It is also possible that some don’t work for you, while others do. This short (non-exhaustive) list of sleep onset aids contains the following suggestions:
- Visualize a pendulum swinging to occupy your mind with something calming.
- Create a sleep-inducing environment like the dark.
- Try focusing on your breath, inhale and exhale at regular intervals. The Wim Hof breathing method is good further reading on this subject.
- Keep away from thoughts; perhaps try reading something entertaining before sleeping.
- Have a hot shower around 1h before sleep time (this especially works for some people, but not for others, so it is best to assess the alertness effects from hot showers while monophasic or at the early stages of adaptation).
- Listen to lulling music, brown/white noise or ASMR.
- Stay away from situations that increase your alertness levels, like looking at electronic screens, at least 15m before bedtime.
Main author: Crimson
Page last updated: 12 March 2020