Alternate Sleep Schedules

by [anonymous]1 min read11th Jan 201145 comments


Personal Blog

My friend and I are starting the Uberman sleep schedule (six 20-minute naps spread evenly throughout each day) tonight. Have other lesswrongians experimented with alternate sleep schedules? Are any of you qualified medical experts who can give input or advice? Success stories and failure stories would both be appreciated, and I'll keep you guys posted on our progress.

45 comments, sorted by Highlighting new comments since Today at 11:20 AM
New Comment

I started attempting uberman at 12/13/2010; that is, two days short of a month ago. I was dreaming during naps pretty quickly, but was routinely oversleeping. I would go about 10 cycles (i.e. 40 hours) of 20 minute naps (not managing to fall asleep for the first one, most of the time) then would sleep through any alarms I set for 8 hours. I was able to push back the 8 hour crash for a day through concerted effort, but never went more than 72 hours without sleeping for more than an hour.

On the 6th, I declared the experiment a failure and switched to attempting Everyman. That has been going much better- I've still noticed some oversleeping, but it'll be an extra hour or two tacked on to my core rather than blowing the schedule like it was with the Uberman.

Advice: Get this alarm clock. The vibrator isn't as strong as I would like, but the physical sensation is a pretty strong incentive to get up. Condition yourself to get up with your alarm- with the caveat that my conditioning stopped working after enough cycles. Have dance music set up as your alarm / have a dance party with your friend at the start of every cycle. Hang out with your friend continuously- I noticed that human interaction was good at making me more awake (but obviously has limits).

I expected a failure point to be my insomnia, but that wasn't a big issue. Staying awake also wasn't that terrible (but, again, I was regularly crashing, so I probably didn't reach the lowest point). Waking up was the problem for me, and you should take it seriously.

My plan is to attempt Uberman again in the summer with two helpers (day shift / night shift), whose job is to make sure I get up when I need to get up and stay awake when I need to stay awake. Trying this with a friend, you'll have some help, since both of you will need to fail at the same time for the two of you to fail. But it strikes me as strongly likely that that will happen.

Overall, my impression is mixed. I certainly had a number of interesting experiences attempting to transition, and I may stick with Everyman long-term, but it's not clear to me that polyphasic works much better than monophasic. I had a lot of dead time where I was too fatigued to do anything besides watch television (something I stopped doing years ago), and so despite spending more time awake I got less done this break. If I had succeeded at transitioning, it would have been worth it, so I'm not complaining too much.

The thing that worries me the most is that I didn't seem to be able to nap unless really tired. That may be particular to my sedentary lifestyle; when sleeping monophasically, it takes me 30-40 minutes to fall asleep. I'm not sure how to interpret this, because it makes me think that I was in terrible sleep debt all the time but failed to notice after good naps for a few hours, before the demon reared its head again; that doesn't compare favorably to monophasic, where I had the feeling that I had a giant sleep surplus all the time, such that I had to bore myself to sleep.

But I also have evidence that my mind was working well the periods where I thought it was, rather than that just being an illusion / too tired to notice that I was tired. I was reading quite a bit (when I was awake enough to focus my eyes on the page) and incorporating / remembering what I read. I was coding quite a bit and it was working well (again, when I had the energy).

I'm single, live alone, and have a weekly social outlet- so I didn't have much social life for this to disrupt. My friends were perfectly understanding about my need to nap, and so I didn't have any issues on that end.

I've made several stabs at uberman. I once made it 9 days, and thought I was on the downslope, but got cocky and screwed it up.

After setting up a really loud alarm on the computer that I had to get up to turn off, the main problem was lack of willpower. At a certain point I would just stop caring enough to continue.

The changes in the way you sleep are quite real and tangible, though. I had 20 minute naps where I would dream bizarre states of mind, where I would feel like I was going insane, only to wake up completely sane and refreshed, as though I had slept eight hours.

In the end I came to the same conclusion as you, that to do this I needed two helpers working in shifts. I tried to recruit some people, but nothing panned out. Do you have specific people lined up? Where do you live?

I'm also single, live alone, and have a weekly social outlet. Very similar experience overall.

Oh, another thing I found helpful was a little device that goes on your ear, which beeps at you if your head tilts forward.

I had 20 minute naps where I would dream bizarre states of mind, where I would feel like I was going insane, only to wake up completely sane and refreshed, as though I had slept eight hours.

My dreams so far have been odd/boring. A number of times I got the impression I was having some sort of REM daydream. The first one that was unambiguously a dream that I could recall was, oddly enough, a nightmare about falling asleep. I also had a dream with no visual input, which I think has been repeated a couple of times.

The naps I wake up from where I was definitely dreaming tend to take 7-10 minutes from lying down to getting up, which is sort of worrisome, since 6 of those means at most an hour of REM. I'm not sure whether the right response is more naps or trying to lengthen the naps.

I tried to recruit some people, but nothing panned out. Do you have specific people lined up? Where do you live?

I have friends I expect I can get to help. Worst comes to worst, I'm considering the possibility of hiring day laborers, but I anticipate quality control issues that make me reluctant to take that approach. I'm living in Austin.

Oh, another thing I found helpful was a little device that goes on your ear, which beeps at you if your head tilts forward.

Did you make it or buy it somewhere?

I believe I bought it from amazon. It's for people driving long distances.

it takes me 30-40 minutes to fall asleep

Do you get enough exercise? (not that 30-40 minutes is so long, but it's a big high)

Enough to get the heart pumping but not enough to fatigue me. I do expect that if I had a more physically demanding life it would be easier to fall asleep. I'm not sure I get any time savings by moving in that direction, though, and doing something like listening to binaural beats or taking melatonin seems more efficient.

Make sure to do some kind of objective test of cognitive performance like dual n back to determine if you're losing (or gaining!) ground.

With the proviso that n-back performance often improves over (at least) the first few weeks of daily training. So if Mark_Gomer & his friend start n-backing at about the same time they change sleep schedule, they might get a spurious improvement in n-back scores.

Seth Roberts uses simple arithmetic problems, and you can do those online at the Genomera website. Just watch out for practice effects, especially during the first few weeks.

[-][anonymous]11y 2

Good call; we'd figured that we would just notice if our cognitive performance declined, since we perform lots of cognitive tasks daily, but maybe we should be suspect of our ability to judge our performance.

Does anyone have any suggestions for test we could use? Would, for example, my mean Tetris score (or distribution of Tetris scores, since I could get better or worse with each passing game as my brain got more used to Tetris or as it got more tired) per waking period (not that I plan to play Tetris every waking period), be a good reference?

EDIT: Tetris sounds like a really stupid way to measure cognitive skills, now that I re-read this. I guess it was just the first thing I thought of that I could easily keep quantitative track of.

EDIT 2: I play a lot of Tetris (~1 hour a day, but not usually for more than ~20 minutes at a time), and it is the only video game I play.

[This comment is no longer endorsed by its author]Reply

This post features a comment (search the world "cultish" -- that's the one) in which someone reports to have tried polyphasic sleep and noticed a decline in memorization ability, as found through using Anki. Perhaps browse through their shared decks (install Anki and browse the list through the application or Ctrl+F via this method) and find something that is vaguely familiar or unfamiliar and try and learn it.

Honestly, you might be better off postponing until you get some good data before to make sure you have a non-affected control sample. Taking tests while conducting your experiment might make it difficult to gauge when any potential negative effects have worn off. At least you have a "default" opportunity now -- afterward, perhaps you should just pick an arbitrary time to recoup (like 3mos) and take the tests again (or pick an equally unfamiliar new topic to learn). Then compare your performance.

Lastly, what are your target criteria? Awareness? Learning abilities? Ability to follow conversation? Deep thought development? You might want to establish some clear goals beforehand -- what if you can memorize like a champ... but you develop twitches or feel like you're about to collapse all the time? Just saying -- think through your experiment before you do it... even if it means postponing your start. Better to do this on the front end.

Honestly, you might be better off postponing until you get some good data before to make sure you have a non-affected control sample.

This is my greatest regret with my experiment. It's clear to me when I'm not functioning at all, but reduced function is a lot more subtle. It's also hard to establish a baseline- driving to the store to pick up some alarm stuff before I transitioned, I noticed I was consistently making navigation errors, leading to it taking significantly longer to get there. I found it amusing because of how reliable it was that I would make the wrong choice, but it made me wonder how I would be able to compare cognitive function.

[-][anonymous]11y 0

My target criterion is having more time per day to do stuff, and that's it. A small decline in some cognitive abilities might be worth the significant increase in available time (if it's not permanent!).

[This comment is no longer endorsed by its author]Reply

Reaction times can do amazing things when you get clever.

Relatively simple reaction times are reported to correlate g somewhere around 0.3 to 0.5 depending on details, with faster times associated with higher IQ. Brutally simple procedure: hold a ruler up vertically and have it dropped by an unpredictable process while holding fingers in a position to pinch it and stop the fall after you detect that it has begun to fall. Record how many centimeters it falls in a spreadsheet. If the number gets big, something is probably going wrong with your brain.

Having a friend on hand to drop the ruler is a big help. Also, for any drinkers out there, try this while sober and then after having some drinks (tested one at a time, because you'll be surprised how quickly results show up).

You probably want a bunch of quick, dissimilar tests. One interesting one which hasn't received much attention is the Embedded Figures Test.

If he's on Linux, he could try a suite of puzzlegames like gbrainy.

[-][anonymous]11y 0

I can't believe I didn't think of that! I'm on Ubuntu, so that's perfect!

[This comment is no longer endorsed by its author]Reply

I'm going to point out that since the writing of this post, a tool has been created for exactly this purpose: Quantified Mind

I wouldn't suggest playing Tetris as a data gathering strategy, but if you're playing it anyways, you might as wel collect the data (scores), as long as setting up the logging can be made into a one-time task.

You could try sudoku or a crossword puzzle.

I've been trying to track my cognitive performance, and it turns out simple arithmetic isn't so good at detecting sleep deprivation.

I've been deprived of sleep to the point of starting hallucinations, and my math times went up by about 10%, which is within the noise. However, I was so impaired that I not only soldered the wires on before running them through the cap, but when I undid it, I soldered it right back on without passing them through the cap first.

It seems like different parts of the brain are at work. If you're trying to make sure you're not sleep deprived, I'd look for tasks that lobotomy patients are not good at. I just haven't found any that are easily quantifiable yet.

One sort of alternative sleep schedule I've practiced on occasions that enables me to get more active hours for a few days in a row without being too sleepy is going to bed later and later each day and sleeping fewer hours than usual. Normally, if I go to bed at my usual time and wake up after, say, five hours, I'll be sleepy, irritable, and unproductive all day, and after a few days of sleeping less than 7 hours or so, I'm good for nothing. But if I go to bed very late and wake up sometime after my regular waking time, I feel much better, so that constant pushing ahead of both bedtime and waking up time enables me to sleep much less for several days while feeling tolerable.

Of course, this is practical only when working on projects where you don't need to coordinate with others. Also, it probably wouldn't work for many other people, especially those who become extremely sleepy late at night, and who wake up easily and full of enthusiasm in the morning. (Whom I envy -- I always find it hard to fall asleep in the evening, getting up after less than ~8 hours of sleep is for me sheer torture, and after waking up I'm slow and sluggish for at least an hour.)

My sleep 'schedule' is something like this, in practice. I go to sleep when I'm tired and get up when I wake up and am no longer tired, without worrying about the clock at all, and I've been doing so for 2 years now. My days usually last between 23 and 28 hours, and seem to run longer, not shorter, when I'm stressed or busy, both in terms of the length of time from the start of one sleep to the start of the next and in terms of the awake:asleep ratio.

I find the exact opposite works for me: if I normally go to sleep at midnight and wake up to an alarm at seven (which is not an unusual pattern for me), if I go to bed at 11pm, I tend to wake up between 5am and 6am, well before my alarm. To a point, the earlier I go to bed, the stronger the effect is, so going to bed at 9pm will usually have me up and refreshed at 1am or 2am, and I feel like I've slept "for the night".

Thats interesting. Do you have to keep pushing later and later, or does a fixed offset give you this benefit?

I ended up in this routine fairly recently and was quite surprised how functional I was on so much sleep debt- given how worthless I feel when waking up a couple hours early. There were other factors that might be at play, but I hadn't considered this one.


Thats interesting. Do you have to keep pushing later and later, or does a fixed offset give you this benefit?

For the same number of reduced sleeping hours, I feel much better if I'm pushing both bedtime and waking up time later and later each day. But even if I wake up at my regular time, the loss of the first few hours of sleep hurts far less than if I were waking up that much earlier, at least for the first two days or so.

I tried Uberman once before. I simply couldn't adapt and kept blowing naps (or not sleeping at all).

I'd like to try it again in the future, but this time I would probably try Everyman since my Zeo sleep-tracking seems to indicate I need at least 2 hours of REM (and preferably another hour of deep sleep) to feel normal. I would also definitely be tracking dual n-back performance and seeing what the Zeo has to say like some others.

Edit: Wow; I've essentially changed my mind completely about this thanks to the comment by Nornagest and the link provided by icebrand. See my more recent comment for my current take.


My understanding is that this is not a good idea. You may want to read some of Piotr Wozniak's work. He's the author of SuperMemo which had a long article in Wired HERE. This program's algorithms are at the root of those used in Anki and Mnemosyne

In any case, while Wozniak isn't a neurologist or sleep expert, per se, my reading of his site seems to present quite a compelling case for his wide familiarity with sleep models, it's function, and practical applications at the very least.

Here is his main page for sleep.

Note the link at the above to myths and facts. Here is one of them verbatim (#13):

Myth: We can adapt to polyphasic sleep. Looking at the life of sailors, many people believe they can adopt polyphasic sleep and save many hours per day. In polyphasic sleep you take only 4-5 short naps during the day totaling less than 4 hours. There are many "systems" differing in the arrangement of naps. There are also many young people ready to suffer the pains to see it work. Although a vast majority will drop out, a small circle of the most stubborn ones who survive a few months and will perpetuate the myth with a detriment to public health. Fact: We are basically biphasic and all attempts to change the inbuilt rhythm will result in loss of health, time, and mental capacity. A simple rule is: when sleepy, go to sleep; while asleep, continue uninterrupted. See: The myth of polyphasic sleep.

At the link about the myth of polyphasic sleep, the Uberman sleep cycle is discussed at length. Wozniak does not seem to concede any validity to the various write ups about this method that have led to people wanting to try it.

Even glancing at the wiki article on this does not present convincing evidence that this is a desirable experiment. Stampi seems to have adopted a sleep pattern like you suggest, but he along with the other two military groups seem to prefer monophasic sleep to polyphasic and only advocate this pattern to relieve deprivation effects under extreme conditions.

Perhaps we should back up: what are you using to support this as a good idea? Are you just interested in the results (positive or negative) for experimentation's sake, or are you trying to implement something you have been led to think is a method of self-improvement (efficiency, learning capacity/speed, etc.)?

Wozniak seems to suggest that there are at best misconceptions about this method's supposed ability to give you more time (and therefore learning/productivity) and at worst serious detriments to sleep and life quality as well as health.

Then again, like I said, I'm not sure about Wozniak's position as any kind of expert... I just know he's written quite a bit on sleep. I could not tell you if any studies had been directly carried out about this, what they found, or what the collective body of those who have attempted it report.

I'd probably be more inclined to believe that Uberman and other polyphasic sleep patterns have serious drawbacks than that they don't, but Wozniak's page is distinctly long on rhetoric and anecdote and short on data.

The section about the biological basis of sleep timing is about the only thing that I'm not inclined to throw out as obviously biased, although even that isn't good news for Uberman.

Point taken, though it seems like that's a common issue for the whole subject. It's been hard for me to find (in a short time) any studies about polyphasic sleep. The primary source of any information seems to be anecdotes. It seems that the entire concept arose from someone who appears to have invented this schedule here. "Uberman" was his naming convention since he was reading Nietzsche at the time.

I did find it quite interesting to read that post as well as this one which reports that he's been on the "Everyman schedule" for 3 years as of 2009 and this one where he compares Everyman to Uberman.

While I'm still skeptical, such positive reviews from someone who's tried it does entice me...

Edit: He... is a she!

Here is PureDoxyK's response to the Wozniak criticism. There's a funny bit...

  • There are no women doing polyphasic sleep. Which makes myself, my friend who first did Uberman with me, and my hero Heidi who’s gone more than a year-and-a-half on Uberman by now the most attractive, clean guys in HISTORY! Mind you, he doesn’t just state this silliness, but uses it as EVIDENCE for the fact that polyphasic sleep can’t work (because women’s “hormones” don’t allow it somehow). Yeah, ROFLcopter.

Wow -- thanks for providing that link -- it was great to read her whole response and the comments. This is a fun reply, as I can say that:

  • I've spent the evening reading tons of posts from PureDoxyk and Steve Pavlina
  • I essentially retract everything I wrote above; I agree that Wozniak is heavy on rhetoric and short on data
  • I think PureDoxyk's posts, blog, and book show that even if she is only one data point... at least it's an empirical data point vs. Wozniak's speculation.
  • Given this... I'm going to try this myself! How's that for a rapid mind changing?

I believe I was wrong and that even if Wozniak has a reasonable understanding of one method of proper sleep habits... this does nothing to rule out the testimony given by those who have adapted to polyphasic and state that they are energetic, alert, etc. on less sleep.

I still have one lingering concern -- long term effects. PureDoxyk states that she is not aware of any long term data on polyphasic sleep.

Lastly, PureDoxyk has a series of vlogs documenting her most recent adaptation to the Everyman schedule (3hr core + 3 naps of 20ish minutes). I watched the first and a couple others. The first has some great tips. There's also a google group with some helpful tips on getting ready for something like this.

Be sure to read the most recent Wozniak where he talks about PureDoxyk and provides more data:

This link (which looks reputable to me) states that REM usually occurs around 90 minutes after you doze off. I think that before trying the experiment, you should consider that you won't be getting any REM sleep at all if you actually sleep for the prescribed amount of time, which strikes me as very unhealthy.

Personally, this feels like a batch of psuedoscience, but I could be wrong because I've not actually researched it -- the link above and some similar things said in a sleep book I own are my only sources of knowledge.

Edit: This comment has been retracted. See my comment below.

The point of these sleep schedules is to get rid of everything other than REM. Numerous people, including several on this thread, find that after a couple of days of this, they go directly into dreams. That much works. The question is whether getting rid of everything else is sustainable and useful. Incidentally, I have known one person who naturally only had REM. He had problems and was diagnosed with narcolepsy. I have known several other people I suspected were similar, but who hadn't been through sleep studies. ETA: also, he found short naps quite useful.

After reading your reply, my impulse was to shrug and think "that objection is wrong, but I still don't think it's a very good idea." I since realized that what I had written really was my true rejection (besides a blind appeal to lack-of-popularity) and I didn't know otherwise about the success or failure of polyphasic sleep, so I'd better go look it up or risk missing out on some possibly-low-hanging-fruit.

Anyways, I'm still reading through PureDoxyk's blog, but I've seen that it definitely seems to work for some people. I plan to continue to look into it and decide whether I want to try Everyman once I go back to college.

I retract my previous comment. Thanks for the correction.

The 90 minute thing for monophasic sleep seems to be true. I was a little surprised how regular my sleep turned out to be:

At the start of the night, down into light sleep, then deep sleep, then up to light sleep, and finally some REM an hour or so later. Regular as clockwork, practically every night.

Incidentally, an interesting link (well, it would be if people would post more of their data) is the Zeo sub-forum dedicated to polyphasic sleeping:

Any updates? How is the experiment progressing?

At this point, I'd be VERY careful about the Uberman sleep schedule, until we learn more about the association between sleep deprivation and Amyloid-Beta buildups (and how we can increase the clearance of amyloid-beta).

Amyloid-β (Aβ) accumulation in the brain extracellular space is a hallmark of Alzheimer’s disease. The factors regulating this process are only partly understood. Aβ aggregation is a concentration-dependent process that is likely responsive to changes in brain interstitial fluid (ISF) levels of Aβ. Using in vivo microdialysis in mice, we found that the amount of ISF Aβ correlated with wakefulness. The amount of ISF Aβ also significantly increased during acute sleep deprivation and during orexin infusion, but decreased with infusion of a dual orexin receptor antagonist. Chronic sleep restriction significantly increased, and a dual orexin receptor antagonist decreased, Aβ plaque formation in amyloid precursor protein transgenic mice. Thus, the sleep-wake cycle and orexin may play a role in the pathogenesis of Alzheimer’s disease.

Mouse studies are not worth the paper they are printed on; compared to other, human, issues with polyphasic sleep, those studies are of epsilon interest.

This looks interesting; I regret that my life is too scheduled, and is entwined too strongly with other people's schedules, to try it myself. On another note: if polyphasic cuts out everything but REM, does that make people more likely or less likely to remember their dreams? I enjoy dreaming and can remember it some mornings, but other mornings I know I dreamed but can't recall it. Still other mornings, I have no idea whether or not I dreamed at all.

On another note: if polyphasic cuts out everything but REM, does that make people more likely or less likely to remember their dreams?

Steve Pavlina's experience was positive- he would lucid dream and remember them better, I believe.

My experience has been negative. My first dream was a nightmare about falling asleep. A later dream had no visual input (but I recognized where things were and what they were regardless). Recently, at least half of the dreams I recall are just blackness and possibly an ache somewhere (though that has become less recent), where I notice it's a dream and attempt to move my hands until I wake up.

My ability to remember what happened in my dreams appears to have deteriorated, but I have no baseline measurement. I remember the content of 0 dreams from this month, and 0 from last month, and the last one which had details I still recall was in August (luckily, at the time I was keeping a dream journal, so I know the night it happened). I typically write in the diary that I'm keeping of the experience right after waking up from naps, but I don't think I've had much to say about dream content after the minute it takes to get from bed to computer.

I tried the uberman schedule about 18 months ago and only made it about 10 days, including several oversleeps, then went back to everyman, and am now back to a normal monophasic schedule. My ability to nap robustly within 20 minutes has persisted to this day, which I still make use of when tired. I was not able to do that before experimenting with polyphasic sleep.