Thursday, June 25, 2009

Want to be brilliant? SLEEP MORE!

Paper we’re looking at: Sleep inspires insight Ullrich Wagner

If you’re like me, the concept of “Uberman” or polyphasic sleep is like a holy grail. The usual patter goes something like this:

“Hey, did you know that DaVinci only slept for a total of about 2 hours per day? Yeah, apparently he napped for 20 minutes every 4 hours.”

It was popularized on Seinfeld, and it’s made the rounds among bloggers & others who are looking for weird things to do. Mens health got into the game, with this article.

At first blush, it’s such a nice fantasy – Cut out all that ‘useless’ time you spend sleeping, and do more productive things! Finally get all that work done! Learn for the love of learning! Become the person you’ve always wanted to be, but never had time to become!

Sadly, it’s a pipe dream. The first clue, of course, is that NOBODY does this for any period of time. Some try it, and some can even manage to do it for weeks (sometimes a couple of months), but nobody sticks with it. That, in and of itself, should tell us something.

But this paper’s not a slam on the polyphasic sleep pipedream. In fact, it doesn’t even mention polyphasic sleep. This paper is about the link between sleep and learning, and more importantly, the link between sleep and INSIGHT.

Because this short paper destroys ANY illusion that polyphasic sleep is a good idea for any sort of growth, and shows us why getting sleep is so critical – ESPECIALLY if we want to generate insights or understanding on things we’ve learned, or problems we’re trying to solve.

Here’s the “big idea” from this paper: Learning something (ie a rule set), then sleeping, then getting a chance to test out that rule set makes it MUCH more likely that you’ll be able to extrapolate & find “hidden patterns” in said rule-set. In other words, you can go beyond what’s explicitly taught a LOT more reliably if you get sleep between initial learning & next application.

The experiment performed as the backbone of this paper is incredible – Not just in the findings, but in the authors ability to conceive of an experimental framework that could be used to test “insight.”

Think about that for a second – How would YOU go about creating a test to see if people had some insight about something they learned? How could you do it in a way that you could be sure that the insight was ONLY related to sleep, and not other factors?

Here’s how this study did it. I’m going to give you the exact instructions the study used.

Paper Excerpt:
"On each trial, a different string of eight digits was presented. Each string was composed of the digits ‘1’, ‘4’, and ‘9’. For each string, subjects had to determine a digit defined as the ‘final solution’ of the task trial (Fin). This could be achieved by sequentially processing the digits pairwise from left to right according to two simple rules.

"One, the ‘same rule’, states that the result of two identical digits is just this digit (for example, ‘1’ and ‘1’ results in ‘1’, as in response 1 here).

"The other rule, the ‘different rule’, states that the result of two non-identical digits is the remaining third digit of this three-digit system (for example, ‘1’ and ‘4’ results in ‘9’ as in response 2 here). After the first response, comparisons are made between the preceding result and the next digit.

"The seventh response indicates the final solution, to be confirmed by pressing a separate key. Instructions stated that only this final solution was to be determined and this could be done at any time. Not mentioned to the subjects, the strings were generated in such a way that the last three responses always mirrored the previous three responses. This implies that in each trial the second response coincided with the final solution (arrow). Subjects who gain insight into this hidden rule abruptly cut short sequential responding by pressing the solution key immediately after the second response."

End Excerpt

The subjects were broken up into 3 groups. One group did the task in the morning, then stayed up all day, then slept the night through. The second group did the task at night, then stayed up all night. The 3rd group did the task at night, then had 8 hours sleep.

So who did better? It was the group who did the task before sleep. And they did better by a WHOPPING margin.

Often we’ve been told that we should review notes & journal entries & such before sleep. We’ve been told that we should think about problems we want to solve before sleep.

This research suggests that that’s a pretty darned good idea. It suggests that sleep is where we take sets of rules & experiences, and turn them into genuine insights.

Chalk up another win for team sleep!


  1. You're misinterpreting research. You cite an interesting study, but it does not support your conclusions about polyphasic sleep.

    The study WOULD have supported your position, if they had a group of people who took a short nap AND that group would have done significantly worse than the full-sleep group.

    I think polyphasic sleep is worth researching. Even if "Uberman" sleep turns out ineffective, at least it would provide some insight into how sleep affects thinking... Sleep on it...

  2. Polyphasic sleep is a big pile of shite - If the fact that NOBODY does it wasn't evidence enough, the emerging links between sleep and learning (and more importantly, sleep and INSIGHT) is more than enough to poke 8 million holes into that ridiculous pipe dream.