Screen time and play: what the evidence supports, and what it does not

What UK guidance actually says

The Royal College of Paediatrics and Child Health published guidance on screen time in 2019 and took a position that surprised people who expected a limit. Having reviewed the evidence, it declined to recommend a universal threshold, on the grounds that the evidence was not strong enough to support one.

Instead it directed families to a set of questions: whether screen time is controlled, whether it interferes with what the family wants to do, whether it interferes with sleep, and whether snacking during screen use is controlled. It also advised avoiding screens in the hour before bed.

The World Health Organization's guidance for children under 5 does set limits: no sedentary screen time for children under one, and no more than one hour for those aged two to four, with less being better. The two positions are not contradictory. They address different age groups and reflect different judgements about how to handle uncertainty.

Displacement is the mechanism worth attending to

The clearest concerns are about what screen time replaces rather than about the screen itself.

Sleep. This is the best supported association. Screen use, particularly in the hour before bed and particularly in a bedroom, is associated with later sleep onset and shorter sleep duration, and sleep in children affects mood, attention, behaviour and learning substantially. If one change is to be made, this is the one with the best evidence behind it.

Movement. Time on a screen is generally sedentary, and it competes with the activity described in the guidance for under fives.

Interaction. Time on a screen is time not spent in the back and forth exchange that language development runs on. This applies to adult screen use as well as to children's, and adult phone use during interaction is a less comfortable and less discussed part of the picture.

The video deficit, and its exception

Young children learn less from video than from the same content delivered by a person in the room, an effect documented consistently in the research literature for children under roughly two and a half. The most credible explanation is the absence of contingency: the screen does not respond to this particular child.

The exception supports the explanation. Live video calls, where the person on the other end responds, do support learning, because contingency is restored. This is why a video call with a grandparent belongs in a different category from a video for toddlers, and it is one of the more useful practical distinctions available.

Where the two hour rule came from

The widely repeated two hour limit has a documentary rather than an experimental origin. It derives from American Academy of Pediatrics recommendations dating back to the 1990s and early 2000s, formulated in the era of television, and it was a reasonable professional judgement rather than a threshold derived from dose-response data.

It has since been widely abandoned or qualified, including by the body that issued it, in favour of guidance about context and content. It nevertheless persists in popular discussion as though it were a scientific finding.

Content and context

Why the public debate outruns the evidence

Large studies linking screen time to wellbeing generally find associations that are statistically detectable and very small, and several widely publicised findings have shrunk substantially on reanalysis. Reviews have repeatedly noted the problems: self-reported screen time is inaccurate, cross-sectional designs cannot establish direction, and reverse causation is plausible, since children who are struggling may use screens more rather than the other way round.

None of this means screens are harmless or that concern is irrational. It means the confident claims in both directions are running ahead of the data, and that specific well documented issues, sleep, displacement of activity, and the way some products are designed to be difficult to stop using, are better places to put effort than a total.

What follows practically

Protect sleep first, because it has the best evidence: no screens in the hour before bed, and none in the bedroom overnight. Notice what is being displaced rather than counting hours. Watch with a young child where you can, because adult narration improves what they take from it. Treat video calls as a different category. Be sceptical of anything designed to autoplay, since the difficulty of stopping is a product design decision rather than a failure of the child.

And apply the same standard to adults. The interaction lost when an adult is on a phone during a child's play is the same interaction lost when the child is on a screen, and it is much less frequently discussed.