About the SA Blog Network

The Scicurious Brain

The Scicurious Brain

The Good, Bad, and Weird in Physiology and Neuroscience
The Scicurious Brain Home

Exercise to sleep? Or sleep to exercise?

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

It’s something we feel we’ve always known: if you can’t sleep, you need to exercise more. Wear yourself out, make yourself good and tired, you’ll sleep like a baby!

So when I started having trouble sleeping, I just figured I needed to work out more. Of course, it kind of figures that often, you have trouble sleeping because of life stress, which often means you’re really busy, which in turn means it probably puts MORE stress in your life just trying to find the time to work out. But that’s just details.

So sometimes, when I catch myself constantly waking up in a panic over several days, I’ll fit in some hard exercise. Maybe I’ll go for a long run, or try a really hard new class or something. By the time I go to bed I am WIPED. Physically and mentally. My body is so exhausted that the feeling of lying down is one of total bliss.

…so why can’t I SLEEP?!?!

Turns out I was suffering under expectations that were a little too high for reality.


Glazer Baron, et al. “Exercise to Improve Sleep in Insomnia: Exploration of the Bidirectional Effects” Journal of Clinical Sleep Medicine, 2013.

(Side note: before we get into this I feel obligated to point out that I am writing this because I couldn’t sleep.)

First off, we’re not wrong. Exercise DOES improve sleep. It does. But not necessarily immediately. And perhaps, instead, we should ask a different question. Instead of asking how exercise impacts sleep, perhaps we should ask how sleep impacts exercise.

The authors of this study were looking at exercise and sleep, especially in the elderly. We all sleep less as we get older, but chronic insomnia is a different beast entirely. When we don’t get enough sleep, we get snappish, have trouble concentrating, suffer from daytime sleepiness, and are more susceptible to things like getting sick, or getting in to accidents.

There are of course various pills you can take to help increase your sleep (remember the glowing Lunesta moth?). But they have side effects, you can become addicted to them (in the sense that you can’t sleep without them and require increasing doses), and in general, people tend to prefer a more “natural” (to them) cure, such as a lifestyle change (though of course, then you have to keep it up, so).

Exercise is one of the ones that people prefer. It’s got many benefits in addition to helping you sleep. But how much does it really help you to sleep?

To find this out, the authors took 11 women (small sample, which doesn’t thrill me), and asked them to exercise 3 times a week at a specific intensity (75% max heartrate) for 16 weeks, but allowed them to exercise more or less, as they chose. They wore activity wristbands so that the authors could be sure they were sticking to the treatment, and also so that they could assess their sleep both before and after. They looked at how much they exercised, how well they slept, etc.

And what they found was not that exercise predicted sleep. Instead, sleep predicted exercise.

The authors found that working out did NOT immediately affect your next night’s sleep, though after 16 weeks of the study, people slept about an hour more per night than they had before. But on any given night, whether you worked out didn’t affect how well you slept. But instead, how much you slept the night before predicted how much exercise you got the next day. What you can see above is trajectory, how much exercise someone got as dependent on the amount of sleep they got the night before. You can see the trajectories go down. The less sleep people got, the less they worked out. This makes major sense to me, as since I did not get a lot of sleep last night, I was both too tired and unproductive to make a go at it today.

Now, you might think: well, then we kind of KNEW that, so what difference does it make? It might, actually, make some big differences in how people approach treatment of insomnia. For example, long term exercise does help people sleep more, but it requires the development of the HABIT of constant exercise. And if someone’s not getting a lot of sleep the night before, they are less likely to exercise, and so less likely to get in the habit. So it might be a good idea to pair the development of exercise with another type of treatment when you start out treating insomnia.

And of course, it also puts paid to the idea that you can work yourself into a good exhausted sleep in a single day. If you’re going to make it work, you’ve got to keep up the habit. Me, I’m off to exercise.

Scicurious About the Author: Scicurious is a PhD in Physiology, and is currently a postdoc in biomedical research. She loves the brain. And so should you. Follow on Twitter @Scicurious.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 9 Comments

Add Comment
  1. 1. David Cummings 6:38 am 08/19/2013

    Interesting. Thanks.

    Link to this
  2. 2. rouksaar 10:49 am 08/19/2013

    It’s been more than a year, i’m suffering from sleep problem. thanks a lot for this awesome article.
    From now on, ultimate objective is to do maximise my exercise daily in order to see the difference .

    Link to this
  3. 3. deathby2 9:01 am 08/20/2013

    As one who exercises 10 hours a week, sleep has more effect on exercise than vice versa. The more sleep I get the better I perform. How much I exercise has little effect on how much I sleep. But the one thing that does effect my sleep is calories. The more I eat the less I sleep. The worst culprit is sugar. So when they say don’t eat before bed, don’t.

    Link to this
  4. 4. wmterhaar 3:06 pm 08/20/2013

    This is purely anecdotal, but maybe the type of exercise has something to do with it. Exercise at 75% heart rate to me is both boring and not very tiring. Those modern, short high intensity training methods are; usually when I come home afterwards I have to take a short nap. The other thing that makes me really tired is long days – like several hours – outside walking and cycling.

    Link to this
  5. 5. 5:29 pm 08/20/2013

    Again anecdotal, but perhaps the type of exercise is impt. Try dancing, which gets the brain working as well as the body. Beats boring gym work.

    Link to this
  6. 6. denisosu 5:53 pm 08/20/2013

    I find exercise dramatically improves the quality of sleep – when I exercise, I sleep fewer hours but wake up feeling more refreshed.

    To get the maximum benefit of this (which I usually do not manage) I should go to bed at the same time every night. I am not at all one of those lucky people who survives and thrives on only a few hours sleep, but in a period where I biked 10 hilly miles each way two and from work and kept regular sleeping hours, I used to crash each night around 1.00 or 1.30, wake up at 7.00 fully reinvigorated and be wide awake and alert until 1.00 the next night, when a “wave” of sleepiness would overwhelm me.
    I know that sometimes insomnia is a medical condition requiring treatment, but I’d strongly encourage anyone to try first combining a lot of exercise and very regular sleeping times for a few weeks.

    Link to this
  7. 7. basudeba 9:59 pm 08/21/2013

    Sleep is related to the built-in governors in the brain to prevent a break down due to load (excess or impoverishment). This load excess can be caused in two ways: draining out of body due to physical work or exercise, and draining out of oxygen from the brain due to mental work. During stress, most of our mental exercise is reduced, because we are confused about what to do. Hence, we require less sleep. But the confusion causes us various trouble. Exercise does help in the long run by creating a load excess.

    Link to this
  8. 8. bucketofsquid 12:56 pm 08/26/2013

    That does kind of explain why people with sleep disorders tend to gain weight. They probably eat the same but exercise less.

    Link to this
  9. 9. Fitsteve 10:39 pm 02/14/2015

    But it’s a sample of 11 people with so many unknown variables it’s hard to draw any conclusions. Fair enough conclusions which may be true, but would to see a comprehensive study of more than 11 people

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article