I first tried polyphasic sleep almost 5 years ago, in summer 2011. About 6 days into my uberman adaptation, I gave up. Two years later, I tried adapting to everyman 3, which I persisted with for several months with some success, but ultimately it didn’t quite work for me. Towards the end of that summer (2013) I tried uberman again, because a bunch of people were trying it all at once and I still aspired to greatness.
The results of that experiment are pretty telling: out of a dozen people, none of them successfully adapted to uberman or everyman. This, despite doing nearly everything right, and being all in a house together where they could ensure each other stayed awake. But within a month or two, all had reverted, and I hear that there were some negative effects in the form of narcolepsy and one or two other issues.
So if you’re planning to adapt to one of these schedules, your odds of success are low.
That being said, I maintain that my polyphasic sleep experiments ended up having one of the most positive effects on my life. Why? I learned to nap and became biphasic, fixing a sleep issue I’d had for as long as I can remember.
When I was a kid, my parents used to insist I had my lights out by a certain time, but I was almost never able to actually sleep then, so I would sneakily read with a flashlight, or othertimes follow the letter of the law by doing things in my room with the lights out (mostly pushups and sittups).
I went to boarding school starting in grade 9, and fairly shortly no longer had trouble falling asleep. I assumed I’d grown out of this issue, or something.
Years later I realized that this hadn’t happened at all. The reason that I was able to fall asleep when I tried is that I’d gotten a laptop when I went to boarding school, and now I was staying up late and had become chronically very sleep deprived.
I had kind of known for awhile that my sleep habits weren’t ideal, but it hadn’t occurred to me that chronic sleep deprivation might be the only stable point. I learned this in May 2011, when I decided that I would actually go to bed on time, like for real, I mean it this time. I slept from 11-7 Sunday night, then again Monday night. Tuesday night, I went to bed at 11pm, and promptly laid awake for over an hour, unable to sleep.
It felt remarkably similar to when I was 12.
The effect, in this case, was that I was very tired the next day, and then I was able to fall asleep on time, but at that point I’d sort of given up on the idea that sleeping at the same time every night was going to work. It was a few months later that I tried my first uberman adaptation. As I said, this failed, though it gave me a sense of how REM-naps work. I did have several successful naps that included dreams, during my adaptation.
It gave me an affordance for napping too: over the following couple of years, if I noticed myself flagging midday, I would sometimes stop and take a nap.
But for the most part I was back to my old routine, chronically somewhat sleep deprived but relatively stable and able to fall asleep on time.
Two years after this first uberman attempt, I tried adapting to the everyman sleep schedule. I lasted much longer on this, closer to a couple of months, although by the end it was pretty janky, with me oversleeping every other day. But I got better yet at napping, both in terms of getting good sleep when I did, and in terms of being good at finding places to nap and remembering to do so.
I was using a Zeo EEG sleep tracker this time and recorded many REM naps and a fairly substantial amount of deep sleep many nights. It does actually seem plausible that this could have been nearly sustainable if I hadn’t messed up some other things.
By the end of summer 2013, my sleep schedule was pretty obviously broken, so I reset. I think I did maybe a week of monophasic, a week of trying to do uberman again, but when that didn’t work I said “how about something much less extreme?”
So I tried biphasic sleep: I slept for 6h every night (I think around 3am to 9am originally) and then took a 20-minute nap midday (around noon). This worked SO WELL.
I haven’t gone back since. It seems that 2am-8am is better as a core time, but other than that this sleep schedule is just clearly way better for me. I’m no longer sleep-deprived during the day (as measured by e.g. me not falling asleep during class when I was in school) but I’m also able to fall asleep almost immediately each night.
I fall asleep during virtually all of my naps (and even when don’t, it’s a very relaxing pseudomeditation) and have dreams at least a quarter of the time (probably other times, without me remembering). It’s amazing to get a midday recharge and reset, while also generally having a pretty similar schedule to everyone else’s.
Based on Zeo data, I get 5-15 minutes of REM sleep each nap, and my core is (roughly) 2h each of REM, deep, and light sleep. Contrast this with many healthy adults, who get 2h REM, 2h deep, and 4h light sleep.
Light sleep, as far as we know, isn’t good for much of anything, so this reduction is fine. Don’t believe me? Seriously, there’s so little to say about light sleep that it doesn’t even have its own wikipedia page, even though it’s one of the most popular human activities (we spend like 15% of our time on it). Light sleep just gets a brief mention on the page non-REM sleep page.
One of the premises behind polyphasic adaptation is that your body sort of knows to prioritize REM and deep sleep, so if you get yourself sleep deprived, it will learn to break the normal deep-REM-light cycles and do mostly deep and REM sleep at night, and REM instead of light sleep during power naps. This seems to be roughly true–most people only get REM sleep during naps when they’re exhausted, but after having done this adaptation, I can get it fairly regularly when I’m not.
Extreme polyphasic sleep seems to not work nearly as well as we hoped, based on the 2013 experiment where nobody succeeded. So you probably shouldn’t expect that you can adapt to those schedules. Also, this sort of sleep hacking can have negative side-effects (e.g. narcolepsy) so even if you don’t mind throwing a week away (although that’s not necessarily accurate) it’s not quite a free thing to try.
So if your existing sleep experience is already pretty good, or if you know that you can get it to be pretty good if you make sure to go to bed at the same time every night, then I say: don’t fix what isn’t broken.
If your existing sleep patterns are already kind of messed up, then it might be worth experimenting with weird schedules/blocks of sleep! There’s lots of precedent for naps being natural. The REM-nap hack is a bit of an innovation on top of that, but evidence suggests it does work for a decent number of people.
There’s an interesting question that I have, which is: can you go straight to biphasic sleep or do you need to do more intense schedules first in order to learn to REM-nap? I lean towards the latter for two reasons:
So yeah, that’s my latest update of what I’m up to, sleepwise. To be honest, it’s so fixed that I mostly don’t think about it unless I get out of whack due to a late-night party or jet lag or similar. That’s a pretty fantastic result, especially since I get an hour more of being conscious than most people (or the same number of waking hours, but I’m more rested).
(If you want to read my polyphasic adaptation advice from a few years ago, go here.)
I’m really curious to find out if this works for other people, so let me know in the comments (particularly if you used to have the same issue I had and then you fix it with biphasic too).
Have you heard about the online mattress wars? Anyway, I had just been reading about them when I got an email offering to pay me money to link to some sites that are trying to sell mattresses, and I figured “well, I might as well get a slice of that absurd pie.” 😂 So here’s a link to one of their articles, also about biphasic sleep.
Constantly consciously expanding the boundaries of thoughtspace and actionspace. Creator of Complice, a system for improvisationally & creatively staying in touch with what's most important to you, and taking action towards it.
Malcolm » 12 Apr 2016 »
I use an alarm for getting up in the morning (currently really really digging the sound “Dione” on my android, which isn’t obnoxious to listen to but is very not-conducive to being asleep) and during my nap I use Matt Fallshaw’s audiotrack, which gently takes you down into sleep then lifts you back out. (mp3 & context.
When things are going well though, it’s not uncommon for me to wake up midmorning, after maybe 5 or 5.5h and feel reasonably rested, but I’m not in the habit of actually getting out of bed then, so the alarm ensures that I actually do. During my nap, it’s rarely hard for me to get up after 20 minutes (unless my nap is like 5pm or later) but I don’t actually have a good internal gauge of how long 20 minutes is. My naps sometimes feel like they’re over an hour long, other time 2 minutes long, even though they’re all exactly the same length.
Regarding shortening my core, there *was* one time when I was stably biphasic and then again I couldn’t fall asleep at 2am as planned, so I stayed up an extra 1.5h and then took two naps (2nd at ~4pm) the following day, and again it felt like I could stay up til 3:30am, but then the following day I only got one nap, and was quite tired, so that was the end of that. But yeah, it feels like it would be pretty possible for me to switch to that kind of 4.5h+2×20min triphasic, if I made sure to get 2 naps/day.
My sense is that ‘taking a nap midday’ is pretty common in some circles/cultures. Would you expect people that do that are getting most of the benefits of ‘biphasic’? If not, what would you recommend those people do, set alarms to sleep less in main cycle?
Dunno! I don’t know much about what their habits already are, or what works well for large groups of people.
It also seems possible that for many people, biphasic / napping doesn’t make sense. But it’s important to know that 5 years ago we would have thought I was one of those people, so it’s not like you can easily tell if the person hasn’t tried.
Oh, I guess the main thing is like, do people in Spain dream during their siestas? If not, they’re not getting REM. Although maybe they’re taking more like a 90min nap, which would probably contain REM but also light sleep. According to wikipedia, Spain is mostly not napping anymore:
According to a 2009 survey, 16.2 percent of Spaniards polled claimed to take a nap “daily”, whereas 22 percent did so “sometimes”, 3.2 percent “weekends only” and the remainder, 58.6 percent, “never”
More than the rest of the world, no doubt, but not that common.
Guy » 13 Apr 2016 »
How do you know whether you had heavy sleep phase or light sleep phase?
What do I need to read to understand all this stuff about REM sleep, sleep trackers and other things you mention?
Malcolm » 13 Apr 2016 »
Originally with my Zeo sleep tracker, a device that unfortunately isn’t for sale anymore. I think some smartwatches can sort of track this stuff, although they’re probably not going to be as good as something that’s actually measuring your brainwaves, especially for naps, which they’re probably not trained on.
Now I just kind of know what it feels like to wake out of each different kind of sleep.
Crystal Gillis » 6 Dec 2016 »
I read an interesting article (or part of a book) about natural sleep patterns prior to electric lighting. People would commonly wake up in the middle of the night and go out and chat with neighbors and such for quite a while, and then go back to bed!
Harry Taussig » 10 Nov 2022 »
Curious if this is still going well for you?
Malcolm » 28 Nov 2022 »
Mixed. I probably take 3-4 naps/week these days. I’ve been enjoying waking up with the sun, but I’m still going to bed fairly late, which means naps help a lot.
Tim Telleen-Lawton » 12 Apr 2016 »
Thanks for the suggestion Malcolm!
Apologies if you’ve already answered any of these questions. I’m curious about:
– Do you use alarms to enforce either phase of your system? Do you expect others to need to?
– Have you tried shortening your main sleep cycle even more to see if what you lose on the margin is still light sleep instead of REM/Deep?
– My sense is that ‘taking a nap midday’ is pretty common in some circles/cultures. Would you expect people that do that are getting most of the benefits of ‘biphasic’? If not, what would you recommend those people do, set alarms to sleep less in main cycle?