The 20-20-20 rule: what the research says
Updated
TL;DR
- The 20-20-20 rule says: every 20 minutes, look at something 20 feet away for 20 seconds. The American Optometric Association and Harvard Health both recommend it.
- The evidence is mixed. Three small studies found fewer symptoms or less change in eye function with breaks, and one 30-person study found no effect from 20-second breaks.
- No study shows the rule prevents long-term eye damage. The studies are small and short.
- No study here reports harm from taking a break, so trying it costs little.
The evidence for the 20-20-20 rule is mixed and the studies are small. Some found fewer symptoms with regular breaks, one found no effect, and two newer studies lean positive. This is not medical advice. See an eye doctor if symptoms persist.
What the rule says
Every 20 minutes, look at something about 20 feet (6 m) away for at least 20 seconds. The American Optometric Association (AOA) and Harvard Health (Harvard Health) both give this advice.
Researchers note the rule is widely cited, with "relatively little peer-reviewed evidence to support" it (PubMed 36473088).
What the studies found
Reminders eased symptoms. In a 2-week study, 29 computer users with symptoms got 20-20-20 reminders. They took more breaks and had fewer eye strain and dry eye symptoms (PubMed 35963776). One week after the reminders stopped, the improvement was gone. There was no comparison group, so people may have felt better because they expected to.
One lab test found no effect. In a 30-person experiment, young adults read on a tablet for 40 minutes. Breaks of 20 seconds made no significant difference to their symptoms (PubMed 36473088). It was a single short session, so it says little about a full workday.
Two newer studies lean positive. In a 2025 trial of 24 young adults, people with no break reported more eye irritation and strain than people who took a break every 10 minutes or at their own pace (PubMed 40466853). A 2026 lab study of 100 healthy adults found that gamers who did not follow the rule showed more decline in eye function after 40 minutes than gamers who did (PubMed 42929271). Both measured a single session. Neither followed people for longer.
People who skip breaks report more strain. A November 2026 meta-analysis found that people who did not follow the rule had higher odds of eye strain (PubMed 43024191). That shows a link. It does not show a cause. People with symptoms may also start taking breaks after the symptoms appear.
Do people follow it?
Mostly not. In one online survey of 432 people, only 34% practised the rule, regularly or occasionally (PubMed 37203083). In a survey of 445 young adults, only 17.0% did (PubMed 40186443).
Whether the rule works and whether people do it are separate problems. Why you skip break reminders covers the second one.
What this adds up to
- No study here shows that the 20-20-20 rule prevents long-term eye damage, and nobody should claim it does.
- Three studies (29, 24 and 100 people) found fewer symptoms or less deterioration in eye function with breaks. One study (30 people) found no effect on symptoms from 20-second breaks.
- Break frequency may matter more than the exact 20-20-20 numbers. The 2025 trial found benefit from breaks every 10 minutes and from self-paced breaks.
- A break costs little. No study listed here reports harm from taking one.
The broader background on symptoms and other measures is in digital eye strain. For a short routine to use during a break, see desk eye exercises. EyeHack's own summary is at /science/20-20-20.
The research in detail
Talens-Estarelles and colleagues (PubMed 35963776). 29 computer users with symptoms. Software on each laptop used the webcam to watch breaks, eye gaze and blinking, and sent personalised 20-20-20 reminders. Symptoms and eye measurements were taken before, after 2 weeks of reminders, and 1 week after the reminders stopped. Results from the abstract:
- Participants took more breaks per day and worked for shorter stretches.
- Digital eye strain symptoms and dry eye symptoms decreased with the reminders.
- The symptom improvement was not maintained one week after the reminders stopped.
- Tear film and ocular surface measurements did not change. Binocular vision measures did not change either, apart from an increase in accommodative facility.
The authors concluded the rule is "an effective strategy for reducing" digital eye strain and dry eye symptoms, while 2 weeks was not enough to improve binocular vision or dry eye signs. Limits: one group of 29 people, no control group, self-reported symptom scores. A single-group design cannot rule out expectation effects.
Johnson and Rosenfield, 2023 (PubMed 36473088). 30 young adults did a 40-minute, mentally demanding reading task on a tablet, four times. Each time they got a 20-second break every 5, 10 or 20 minutes, or no break. Symptoms rose after the task in all four conditions. Scheduled breaks had no significant effect on symptoms (p = 0.70), reading speed (p = 0.93) or accuracy (p = 0.55). The authors wrote that the results do not support 20-second scheduled breaks as a treatment for digital eye strain. Limits: one 40-minute session per condition, a young sample, one task on a tablet. It tests symptoms within one session and does not cover a full workday.
2025 trial, 24 young adults (PubMed 40466853). Four schedules during a 40-minute reading task: no break, one break at 20 minutes, a break every 10 minutes, and self-paced breaks. Participants reported more eye irritation or burning, and more eye strain, with no break than with breaks every 10 minutes or self-paced breaks. The abstract does not report such a difference for the single break at 20 minutes.
2026 lab study, 100 healthy adults (PubMed 42929271). Two groups during 40 minutes of gaming. One followed the 20-20-20 rule and one did not. The group that did not follow it showed greater deterioration in near point of accommodation, near point of convergence and vergence facility. The authors concluded the rule appears effective. These are eye-function measures after one session.
Observational work. The November 2026 meta-analysis found that failing to follow the 20-20-20 rule was associated with an odds ratio of 3.04 for eye strain (PubMed 43024191). The 445-person survey found 17.0% adhered to the rule, and screen time correlated weakly with dry eye symptoms (PubMed 40186443). In the 432-person survey, people with burning sensation and headache were more likely to practise the rule, and the authors noted that little is known about its validity (PubMed 37203083).
How to read them. The trials have 24 to 100 people. Sessions run from 40 minutes to 2 weeks. Outcomes are mostly self-reported symptoms. Each study used a different break schedule, so results do not add up cleanly. The Talens-Estarelles study measured symptoms across two weeks of real use, Johnson and Rosenfield measured symptoms inside one reading session, and the 2026 study measured eye function after 40 minutes of play.
Try EyeHack
EyeHack is a macOS menubar app built around the 20-minute, 20-second cycle. After every 20 minutes of screen use it covers every connected screen for 20 seconds, with no skip button. It waits while you're on a call (any app using your microphone or camera) and while a common screen recorder is running. Its price is $14.99 once, with a 14-day free trial. See EyeHack.
Questions
The evidence is mixed. A 2-week study of 29 computer users found fewer symptoms with reminders, and a 30-person lab study found no effect of 20-second breaks on symptoms.
No. The studies are small, short and use different designs, and one of them found no benefit from 20-second breaks.
The American Optometric Association and Harvard Health both recommend it for screen use. The research on the exact numbers is thin.
In the 29-person study, the improvement was not maintained one week after the reminders stopped.
No study here reports harm from breaks. The authors who found no effect wrote only that 20-second breaks did not work as a treatment in their test.