Sunday, July 5, 2009

Tough problem? Sleep on it.

Research shows you're twice as likely to solve a difficult problem by sleeping on it. And it's not necessarily because you are rested.

It's because you solve problems differently while sleeping, engaging other parts of the brain. After sleep, you're twice as likely to arrive at a "non-linear", creative answer. These findings are consistent with other research I discuss in my post: the muse comes quietly.

When I'm struggling with a really tough problem, I deliberately utilize my overnight problem solving machine (also known as sleep). It works. And I feel more rested, because I can remove the difficult problem from my to do list.

Often the good ideas like to be invited, not chased.


  1. I totally agree with this.

    As kids we were asked to solve the difficult MATH problem before sleep. Wake up in the morning and try again. It worked, atleast I got closer to the solution if not 100% complete.

  2. I've solved many of my programming dilemmas while sleeping, and answers have come to me in dreams..

    So I agree completely

  3. While it sounds like a good idea and the anecdotal evidence is certainly inspiring, I'm not sure the science behind it is sound. The Boston Globe piece may be misrepresenting/oversimplifying for its audience. I confess I do not have a copy of the 2004 issue of Nature to which the article refers, but 66 participants is a small sample size, and I suspect the distribution of mathematical skills across the 3 different groups may not have been entirely random. To accomodate margins of error, I'm pretty sure you need ~1000 samples to predict for a general population the size of, say, the US citizenry.

    That being said, I have my own anecdotal evidence against dreams solving my problems; the dreams are usually better than the reality! :)

    But don't get me wrong: I do think the benefits of rest and sleep should be more widely publicized. Though I may prefer to describe it in terms of reducing brain /body temperature and improving nervous conductivity, there's no doubt I feel more clear-headed perform better at most tasks when I'm well rested.

  4. Paul, I'd be skeptical of 66 data points as well, except the results seem to match the direct observations of the brain scans. But I admit I'm relying mostly on a single data point (mine). :-)