Field note
The 20-minute rule: get out of bed, and where the number comes from
If you have been lying awake a while, get up, leave the bedroom, do something dull in dim light, and go back only when you feel sleepy. It sounds like folk advice and it is not: it comes from stimulus control therapy, formalised by the American psychologist Richard Bootzin in 1972, which remains the only nonpharmacological insomnia treatment holding the highest standard-of-practice rating from the American Academy of Sleep Medicine. The point is not to pass the time. It is to stop your brain learning that bed is a place where you lie awake. As for the twenty: across the studies using this instruction, the threshold ranged from 10 to 25 minutes. Twenty is a convention, not a measurement, and you are not supposed to be timing it anyway.
- It is a clinical technique, not a tip.
- The mechanism is conditioning, not relaxation.
- The number varies 10 to 25 across studies.
- Do not watch the clock. Watching the clock is the problem.
The rule itself
Lie down to sleep. If sleep has not arrived after a while, get up, leave the bedroom, and do something undemanding in low light. Go back to bed when you feel sleepy rather than when you feel it has been long enough. Repeat as often as needed, including four or five times in a night if that is what happens.
And get up at the same time in the morning regardless of how the night went. That last instruction is not an afterthought; it is what stops the whole thing unravelling.
Why it works, which is not what you think
The intuitive explanation is that getting up helps you relax, or breaks the frustration. That is not the mechanism, and understanding the real one changes how you do it.
The mechanism is conditioning. Your brain forms associations between places and states. If you consistently sleep in bed, bed becomes a cue for sleepiness. If you spend hours in bed awake, frustrated, checking the time and rehearsing tomorrow, bed becomes a cue for exactly that instead. The bed stops being a sleep signal and starts being an anxiety signal, and the harder you try in it the stronger the wrong association gets.
Leaving the room protects the association. You are not going somewhere else to calm down. You are refusing to let your bed accumulate any more evidence that it is a place for being awake.
If the point were relaxation, reading in bed would be fine. Because the point is conditioning, reading in bed is the thing to avoid, and the same logic covers your phone. That is why getting the phone out of the bedroom belongs to the same protocol rather than being a separate tip about blue light.
Where it came from
Richard Bootzin, an American psychologist, published the stimulus control treatment for insomnia in 1972. The origin is unusually human: a student approached him after a class asking for help sleeping, and Bootzin developed and tested the idea with him.
In that first case the student had to get out of bed four or five times a night at the start. By the end of a two-week trial, about half his nights passed without getting up at all.
More than fifty years later, stimulus control therapy is the only nonpharmacological insomnia treatment to hold the highest standard-of-practice rating from the American Academy of Sleep Medicine. It is a core component of CBT-I, the first-line treatment for chronic insomnia.
Which makes this an unusual entry on this site. Most of the numbers I trace turn out to be softer than advertised. This one turns out to be considerably more serious than the listicles it appears in.
Where the 20 came from
Now the part I actually came here to check, and the answer is the familiar one.
Across studies that use this instruction, the threshold for getting up ranged from 10 to 25 minutes. Twenty sits in the middle of that range. It is a reasonable convention, and it is not a measured optimum, and I could find nothing establishing that twenty outperforms fifteen or twenty-five.
What is consistent across the studies is the instruction: get out of bed after a period of being awake, and repeat as needed. Twenty studies used a specific figure. The figure varied. The instruction did not.
This is the same shape as the 20-minute nap, the 10-3-2-1-0 rule and most of the source index: a well-supported direction with a precise-sounding number bolted on, where the number is a convention and the direction is the finding.
There is an irony built into the popular version. Knowing whether twenty minutes have passed requires watching the clock, and clock-watching in bed is exactly the arousal-raising, deadline-creating behaviour the technique exists to stop. The clinically sensible version is not "twenty minutes" but "when it is clear this is not happening, and before you get frustrated." Turn the clock away.
Doing it properly
- Leave the bedroom if you reasonably can. Somewhere else is the point.
- Keep the light low. Bright light at 3am is a signal to your circadian clock that it is morning, which is not the message you want to send.
- Do something genuinely dull. A boring book, folding laundry. Not your phone, not email, not anything with an ending you want to reach.
- Go back when sleepy, not when it feels fair. Returning on a timer puts you back in bed awake, which is the thing you got up to avoid.
- Expect to repeat it. Bootzin's first patient got up four or five times a night initially. That is the technique working, not failing.
- Keep the wake time fixed anyway. A rough night is not a reason to sleep in. Sleeping in is what carries the bad night into the next one.
- Do not time it. See above.
The hard part is the last-but-one. After a broken night, every instinct says recover the sleep in the morning, and doing so pushes tonight later and makes the following night worse. The regularity evidence is on the side of holding the wake time.
The rest of the protocol
The 20-minute rule is the famous instruction, but it is one of several, and the others are what make it work.
| Instruction | What it is protecting |
|---|---|
| Go to bed only when sleepy | Stops you lying awake in the first place |
| Get out of bed if not asleep | The bed-sleep association |
| Use the bedroom for sleep and sex only | The same association, during the day |
| Get up at the same time every morning | Sleep pressure and circadian timing |
| Do not nap | Sleep pressure, so that bedtime works |
Note the last row, which sits awkwardly next to a nap calculator. If you sleep well, a short afternoon nap is fine and the nap pages on this site apply. If you have insomnia, the protocol says no naps, and it is right. Those are different situations and it would be dishonest to blur them.
When this is not enough
Occasional bad nights are normal. Persistent difficulty falling or staying asleep, for months, affecting your days, is insomnia, and it has a first-line treatment that is not sleep hygiene: cognitive behavioural therapy for insomnia, of which stimulus control is one component.
CBT-I outperforms sleeping tablets over the long run and is what a clinician should be offering. If that describes you, the useful step is asking for CBT-I by name rather than assembling techniques from articles like this one. I am a medical student, not a licensed clinician, and this is where a page stops being the right tool.
Questions people actually ask
Should I get out of bed if I cannot sleep?
Yes. It is the central instruction of stimulus control therapy, formalised by Richard Bootzin in 1972 and the only nonpharmacological insomnia treatment holding the highest AASM standard-of-practice rating. Leave the bedroom, do something dull in low light, and return when you feel sleepy.
Why 20 minutes specifically?
It is a convention rather than a measured optimum. Across studies using this instruction the threshold ranged from 10 to 25 minutes, and nothing establishes that 20 outperforms 15 or 25. The instruction to get up is the well-supported part; the specific number is not.
Should I watch the clock to know when 20 minutes have passed?
No, and this is the contradiction in the popular version. Clock-watching raises arousal and creates a deadline, which is precisely what the technique exists to prevent. Turn the clock away and get up when it is clear sleep is not coming.
What should I do when I get out of bed?
Something dull, in low light, outside the bedroom. A boring book or a repetitive chore. Avoid your phone, anything work-related, and anything with a plot you want to finish. Bright light also signals your circadian clock that it is morning.
What is stimulus control therapy?
A behavioural insomnia treatment developed by Richard Bootzin in 1972. Its instructions are to go to bed only when sleepy, get out of bed when not asleep, use the bedroom only for sleep and sex, get up at the same time every morning, and avoid napping. It works by protecting the learned association between bed and sleep.
References
- Bootzin RR. Stimulus control treatment for insomnia. 1972. Origin of the technique, the instruction set, and the first case described here.
- Systematic reviews and meta-analyses of stimulus control for insomnia, including work published in the Journal of Sleep Research (2024). Source for the 10 to 25 minute range across studies and for the consistency of the instruction itself.
- American Academy of Sleep Medicine practice standards for behavioural treatment of insomnia. Source for the standard-of-practice rating.
- Windred DP, Burns AC, Lane JM, et al. Sleep regularity and mortality: a prospective analysis in the UK Biobank. eLife. 2023. Cited for holding the wake time.
Keep the wake time whatever happened
A broken night is not a reason to move the morning. The calculator works backwards from a fixed wake-up time and prints its assumptions.
Open the sleep calculator