The last "Polyphasic Sleep Experiment" blog entry
I still love to hear from people who have suceeded or are thinking of trying it, so feel free to drop me an email if you want.
This is a web log of our sleep experiments.
Christphe Saout writes:
In chapter 12 he discusses the "Leonardo da Vinci Ultrashort Sleep Strategy" which is actually exactly the 6x 30min napping strategy.
Stampi had the luck to find a subject that was willing to try living on the Uberman schedule. The subject was a graphic artist, so he appearantly didn't have any trouble to sleep when he wanted to.
The first test lasted only 19 days. He tried it with 15 minute naps, but these were appearantly much too short and he had a lot of napping incidents, overslept, etc...
One year later he tried it again. This time with 30 minute naps. But he didn't try to follow the all-or-nothing strategy described on kuro5hin, he gradually reduced the sleep amount over 10 days. Starting with 8 hours of monophasic sleep, he reduced the core sleep to 3 hours over 5 days while adding 80 minute naps over day (80 minute naps). After these 5 days the nap duration was gradually reduced to 30 minutes. This was done to minimize sleep deprivation in the beginning. At the beginning it's easier to get asleep if have 80 minutes to get asleep, while later, when your brain gets the hang of the rhythm, your ability to get asleep for the 30 minutes naps increases. This is also the result of other studies (there a very many described in the book!).
At the beginning everything went the usual way. The subject was very tired, had some oversleeping accidents, etc... but it was motivated enough to continue and also had enough work. No real crashes.
The interesting thing now is that Stampi told the subject to do some test on a regular basis. Usually 30 minutes after waking up or so to minimize the effects of sleep inertia. These test were 2 kinds of performance analysis tests. The first was the so called "Memory and Search Test" (MAST), the second one the "Descending Subtraction Test" (DST). Okay, the MAST focuses on memory while the DST focuses on "thinking".
Both performance levels rapidly degraded after being on the schedule and remained on this level for two weeks.
But somewherey day 21 something strange happened: The MAST performance suddenly raised over the baseline levels! (baseline = the levels before the schedule with normal 8hour night sleep). And it stayed on this high level for the rest of the test. The DST performance however didn't change much. On day 34 the subject was told sleep for several consecutive hours (how much he could) and then immediately return to the Uberman schedule. Appearantly this worked fine. After that the DST performance also jumped over the baseline levels and stayed there.
After 48 days the test was aborted (not because of any problems, I suppose the subject simply didn't know what to do with all the time or something).
The subject had to wear a special gadget like a wrist watch that recorded when he was asleep or awake. Also an EEG,EOG, EMG and EKG were recorded when possible (to distinguish the types of sleep, REM, NREM, SWS).
What was found: At the beginning the naps mostly containted stage 2 NREM sleep and a reduced amount of stage 3/4 SWS sleep but nearly no stage 5 REM sleep.
After adapting (around day 21) suddenly the relative distribution of the sleep stages (stage 2 NREM, SWS, REM) returned to nearly the same distribution a normal night sleep has (30% REM, 70% NREM). This is very unsual because normally REM sleep is only possible after at least 60 minutes of NREM sleep and only lasts for a few minutes, except in the last half of an 8 hour sleep, where you find a lot of stage 2 and REM sleep. But after having adapted to the Uberman schedule some naps contain a lot of SWS, other contain "only" stage 2 NREM (light sleep) and other contain REM sleep. That's what is being described at kuro5hin, being able to jump directly to REM sleep. But where he was wrong: You won't get only REM sleep, the sleep distribution returns to normal, that's all (but under very unusual and surprising conditions).
Conclusions (from the book):
1. Adult humans appear to have a natural ability to adapt to polyphasic sleep schedules
2. The 4-hr ultradian cycle of sleep-wake pressure previosly described (note: from other studies before) may be an important factor in allowing adaptation to polyphasic patterns
3. The sleep-wake system appears to show a high level of flexibility in terms of sleep timing and duration
4. Polyphasic sleep may be a feasible, and perhaps the only, strategy allowing remarkable levels of sleep reduction during prolonged quasi-continuous work situations, without unduly compromising performance effectiveness.
5. THis may be analogous to what is observed in a considerable number of mammalian species, particularly in thos living in dangerous enviroments.
6. Further studies extended to a larger sample of subjects may provide powerfull tools for developing sleep-wake schedules for individuals involved in irregular or quasi-continuous work situations.
7. These findings and hypothesis raise challenging questions concerning what is known about the regulatory mechanisms of sleep functions.
I find especially the last one very interesting. A lot of scientific research is done without even noticing that other cycles than the monophasic sleep patterns! A lot of assumptions that are right on a normal schedule can't apply to polyphasic sleep. As seen, after 2-3 weeks there is some sort of switch.