Almost every parent asks this question, and almost nobody gives a straight answer to it.
So here is one, age by age. But before the numbers, the thing worth hearing: you are not looking for a figure to feel guilty about. Most parents reading this are doing considerably better than they think, and the families who need to change something usually already know which thing it is.
Two caveats. First, no country's guidance is a law of nature — the AAP has moved away from hour limits, Canada still gives numbers, and the World Health Organization grades its own evidence as very low quality and says so openly. Second, the number of hours matters far less than what the hours are replacing, and whether your child can stop.
Under 2
12–16 hours of sleep (4–12 months) · 11–14 hours (1–2 years)
What is appropriate: video calls with a relative they know, and nothing else. The American Academy of Child & Adolescent Psychiatry puts video chat only under 18 months, and educational programming with an adult sitting beside them from 18 to 24 months.
The main risk is that screens replace speech. This is the most concrete finding in the whole field. A JAMA Pediatrics study of 220 families, using 16-hour audio recordings, found that every additional minute of screen exposure at 36 months was associated with 6.6 fewer adult words, 4.9 fewer child vocalisations and 1.1 fewer conversational turns. At the average exposure the families actually recorded, that is roughly 1,139 adult words a day the child does not hear. Notably there was no such association at 12 months — it appears as language is being built.
What screens push out: language input, and the back-and-forth of being answered.
Ages 2 to 5
10–13 hours of sleep, including naps
What is appropriate: AACAP suggests about an hour of non-educational screen time on a weekday and up to three at weekends. The Canadian 24-Hour Movement Guidelines say a maximum of one hour, and that less is better.
The main risk is that screens replace playing with other children. The Upstate KIDS cohort of 3,894 mother–child pairs found screen time displaced peer play but not reading time, and that this was the route by which it raised the risk of developmental delay — motor, communication and social. The authors' own conclusion is the useful part: enough peer play appears to offset it.
Worth knowing: the AAP's own parent explainer says heavy solo screen use carries the risk. The word solo is doing real work there, and it is the argument for sitting down and watching with them.
Ages 6 to 9
9–12 hours of sleep
What is appropriate: AACAP gives no hour figure from age six and talks about habits instead. The Canadian guidelines give the only firm number anywhere: two hours of recreational screen time at most. Present that as one country's guidance rather than a consensus, because it is.
The main risk is displacement — of sleep, of running around, of reading. The AAP's June 2026 explainer lists for this band: poor sleep, reduced academic performance, weakened attention, vision problems and not enough exercise.
This is the band where the device enters the bedroom, which is the moment to not let it. See the sleep finding on every other page here.
Ages 10 to 12
9–12 hours of sleep
This is the tipping point, and it happens in one year. Ofcom's May 2026 figures: 56% of ten-year-olds own a smartphone, and 83% of eleven-year-olds do. Starting secondary school is the decision point, and it is the year to organise with other families rather than the year to decide alone.
The main risk is the phone itself arriving. The ABCD study of more than 10,000 US adolescents, published in Pediatrics in December 2025, linked smartphone ownership at twelve with higher risk of depression, obesity and insufficient sleep — and found that the earlier the phone arrived, the higher the obesity and sleep-insufficiency rates. It is observational, it makes no causal claim, and the lead author is explicit that they are not saying smartphones harm every adolescent. It is still the best-designed study we have on the question.
For girls, the sensitive window starts here. Orben and Przybylski's Nature Communications analysis found higher social media use predicted lower life satisfaction a year later specifically at ages 11 to 13 in girls. The authors call the effects small, and they are, but the timing is specific.
Ages 13 to 15
8–10 hours of sleep
The sensitive window for boys is 14 to 15, on the same Orben analysis.
This band is squarely the target for sextortion. The FBI's stated primary targets are boys aged 14 to 17. If your son has a phone and you have not had the conversation on this page, have it this week. It is one sentence and it works.
Night-time use, ranked. The ABCD study of 10,280 adolescents gives the most practically useful table in the whole literature — the specific behaviours, in order of how much they raise the risk of trouble sleeping: using a device after waking in the night (1.34), being woken by notifications (1.28), a TV or connected device in the bedroom (1.27), leaving the ringer on rather than the phone off (1.23). Every one of those is fixable tonight without a single argument about apps.
Ages 16 to 18
8–10 hours of sleep
On the life-satisfaction measure this band is comparatively lower risk, with risk rising again at around nineteen and the move to independence.
AI use peaks here: 92% of 16 to 17-year-olds use AI, 30% of them daily. Emotional reliance on an AI companion is the newest risk and the one least likely to be on your radar — see the page on it.
Sextortion continues through this band, and sleep debt is cumulative by now.
What screens push out at this age: sleep, and the practice of managing real social and romantic life in person, which is the thing they are about to need.
The one rule that applies at every age
No device in any bedroom overnight, for anybody in the house, including you. It is the best-evidenced thing a parent can do, it is free, and it takes an evening.
Michael has spent much of his working life online—and much of the rest of it trying to stop that technology taking over. He has thrown a phone into a lake, locked devices in boxes and experimented with long periods offline. He writes about what helped, what failed and how to keep the useful parts of digital life.
Last updated 9 August 2026. This page is checked and corrected as new research comes out. If you spot something out of date or wrong, tell me and I will fix it.