Just don’t. Read on for why.

—

Back in 2015 I was working actively as a software freelancer, on a startup (which later became Corti) and running a small incubator at the Technical University of Denmark. At that time I had no kids, just myself and my girlfriend to answer to so after reading a couple of blogposts about it I decided to give polyphasic sleeping a try. 

In a nutshell (you can google this yourself for more info but please finish this post first), the basic idea behind polyphasic sleeping is to adjust your sleep to be more efficient by timing when you sleep to your circadian rhythm. At the same time you limit your sleep a lot and hereby you “repartition” (according to the polyphasic “literature”) your sleep meaning that you condense all your sleep phases into just the “perceived good stuff”: Deep sleep. Even one of the objective metrics you are encouraged to used by polyphasic sleep sites is how little you remember your dreams (which is so silly in so many ways).

The classic polyphasic sleep schedule is called the Uberman in which a person only sleeps 6x30mins naps (or even shorter) carefully spaced out across a full 24 hour period hereby leading to a total sleep time of 3h, hereby effectively releasing 4-5h of time where you can — in my case — code or do other stuff. The schedule I tried was called the Everyman 3 in which I slept for 3h “core” when I would normally go to bed and then 3x 20min naps spread out through the day. This lead to a total of 4hours of sleep and thus freeing up lots of time for coding and other things. 

I managed to keep this going from May and until September. And the only thing, at the time, that killed it for me was that I would so often fall asleep at meetings. Now the Michael version 2023 would say that if you can fall asleep during a meeting you’re probably not supposed to be in it (but that’s another matter for another post). Still it was a bit inconvenient to fall asleep at investor and customer meetings etc so I decided to stop because of that. 

At all times I at least claimed that I was feeling great. That I had no trouble thinking whatsoever and I really was quite productive. I got at least 4 more hours out of the day that I was spending working. I actually missed it at times over the years after I stopped. I missed the early completely dead silent mornings, where it was only me, a cup of coffee and my mac watching the sunrise. I felt like I had found some way to cheat life for more time. In addition, I felt that one thing I had gained from this experiment was the ability to fall asleep in a few seconds any time all the time something I still to this day believes does come from this experiment (but I’m not a doctor so I don’t know of course).

Then in 2018 I became a dad and as all new parents know sleep stops. Or at least there are days where you feel like you might as well just not get down to bed again because you know your baby is going to wake again in the next 20 mins so why bother. I actually at that time started doing Everyman again because I was so desparate to just try something and to feel in control of my sleep again (ultimately scrapped it again pretty quickly because it’s impossible to sleep at stable hours with kids also).

My hypothesis (and I should note of course that I’m no neuroscientist) of what I think happens with polyphasic sleeping is simply that the brain and body adapts to sleeping very little. Just as the brain and body seem to adapt to new situations and new environments over time so does it adapt to little sleep - is my hypothesis. Indeed among polyphasers the first period where you’re starting on your chosen schedule is known as “adaptation” and in general the period that will feel the most shit. But after that you come out “triumphant” or so you think. Likely your brain just doused you with chemicals of some sort so you stay alive and don’t lie drooling on the floor.

So “adapting” in this case doesn’t necessarily mean that the sleep deprived brain solves the problem in some other way - and you can now magically do without it. Like when you take an ice bath your blood vessels constrict to stop blood flow to non-vital organs and that does indeed make it possible for you to stay in the ice bath for some time. But it doesn’t solve the problem of the new cold environment you’re in. Stay too long and you’re bound for hypothermia anyway. Biology always wins.

As a new parent you adapt to the little sleep too. The first time you managed to scrape 5 hours together during a 10hour sleep period you feel horrible. The umpteenth time, less horrible and you actually feel like you’re “functioning”. But really you probably just adapted so you can actually live - and who knows maybe make a new baby and hereby evolution wins again.

But don’t take my word for it.

The past 3-4 months I’ve listened to just about everything that Andrew Huberman has put out through his Huberman Lab podcast and I recently learned in this episode with the great Matt Walker (whom I’ve also followed quite a bit) that in 2021 a study was published that concluded very bluntly:

1. The claims of benefits from polyphasic sleep schedules are not supported by scientific evidence.
2. Polyphasic sleep schedules and the sleep deficiency inherent in the schedules that are most highly promoted in popular culture have been associated with adverse physical and mental health, as well as with decreased performance.
3. Striving to adopt a schedule that significantly reduces the amount of sleep per 24 hours and/or fragments sleep into multiple episodes throughout the 24-hour day can have significant adverse consequences for daytime performance, mood and health; and is clearly not recommended.

And here’s a link to the study.

While considering whether or not to do this polyphasic sleep experiment back in 2015 I spent a lot of time trying to get to the bottom of it from a scientific standpoint and I couldn’t find any clear indication that the benefits claimed by the polyphasic sleep community weren’t to be had. On the contrary once I started doing it I felt great, I felt productive and using a Fitbit I could see that evidence my sleep “repartitioned” and that I could go into deep sleep for a much longer period of time during the core and much faster than before.

Later on through Dr. Huberman’s podcast I learned that there’s a linear relationship between REM sleep and mortality. The less you get the sooner you die. Plus listening through this, this and this of Huberman’s podcasts on sleep (along with the Matt Walker episode linked to above) makes it absolutely clear to me that there’s absolutely no reason in the world to try a polyphasic sleep schedule. Not unless you specifically want to die earlier, get Alzheimers, be less productive etc.

I hope this post will reach somebody who are considering to start a polyphasic sleep schedule. If you still want to do it after listening to Dr. Huberman then seriously, you should seek help. 

I tried it - it “works” - in the sense that I didn’t die — at least I didn’t in the 5 month period I did it. And I adapted, at least for my sleep deprived brain it looked like I adapted. 

Maybe if you’re in your 20s reading this you might think: “Sure but I have all this important stuff I need to do. I can risk a bit of Alzheimers. I wanna die when I’m young anyway so who cares. This guy clearly just didn’t drink enough coffee” (ahem you should also check out Huberman Lab’s episode on coffee btw).

So final argument in this way too long blogpost is that I tried to setup a way for the logical side of my brain to stop the experiment in case I experienced some stuff out of the ordinary. And one of the things I did find in my research of polyphasic sleep schedules prior to starting it was that people apparently are poor at assessing how sleepy they are. So I needed some objective way to judge whether I was too sleepy and should stop my experiment. For that I found an app that through some reaction-type tests determined how mentally alert I was. I tried to use those scores as a way to say, ok if I get below this score then I should stop. The test is simple. A light will flash on the screen and you need to press a button as fast as you can. So the test measures your reaction time.

So I did those tests diligently after my 3h core and through out the day. And some of them I “aced” others I had multiple of what’s called a lapse where I totally mised that the light was flashing. As I “only” had maybe a couple of those out of 20ish flashes my sleep deprived brain just pushed those aside. Had I owned a car at the time a lapse would mean hitting a cyclist. Or pushing my cup of coffee off a table instead of grabbing it. Or accidentally wiping a production database. The fact that I just brushed those aside is mindblowing to me now.

Another big alarm bell that rang but I didn’t hear was that I would fall asleep every time I just sat still in a chair for more than 5-10 seconds. Those were also just brushed aside as nothing to be particularly concerned about.

Then there were those “zombie awakenings” where you wake up to your real alarm bell after the 3h core and shutting off the alarm clock and go to sleep again and then when you wake in the morning have no clue why you slept past your alarm clock. Seriously - I thought my alarm clock was broken so later on I had two different alarms - one where I had to get out of bed to turn off, but I still managed to turn it off in this zombie state and I was not a sleep walker before that.

But really all of these appear to be my brain’s way of yelling STOP THIS SHIT YOU’RE DOING RIGHT NOW! But I wasn’t really paying attention to all of that yelling, because, you guessed it, I was sleepy.

It’s getting late - I’m going to bed.

Just don’t. Please?

Feel free to reach out if you have any questions about the above (reibel@hey.com).