A study reported today finds that more screen time during infancy and around the start of school was associated with poorer academic performance and weaker working memory years later, suggesting early childhood may contain particularly sensitive windows. It is exactly the kind of finding that gets turned into advice it cannot support, so it is worth walking through how to read it.

First question: what was measured?

“Screen time” is not one thing. A video call with a grandparent, a maths game, a cartoon playing in the background and an autoplay video feed are entirely different activities that a stopwatch cannot distinguish.

Most studies in this field measure duration because duration is what parents can report. That is a real limitation rather than a quibble: if content and context matter more than minutes, then a study measuring minutes will find a weak, blurry effect no matter how large it is.

Second question: what is it being compared against?

This is the point most coverage skips. An hour spent on a screen is an hour not spent doing something else, and what that something else would have been is doing much of the work.

If screen time displaces being read to, conversation, sleep or play, the comparison is against activities with well-established developmental benefits. If it displaces a child waiting in a hospital corridor, it does not. Researchers call this the displacement problem, and it means the same quantity of screen time can carry different consequences in different households.

Third question: could something else explain both?

The most important question in any observational study, and the hardest to resolve.

Households where very young children have high screen exposure differ systematically from those where they do not — in parental working hours, income, housing space, stress, and access to childcare. Every one of those independently predicts later academic outcomes. A study can adjust statistically for the factors it measured, but not for those it did not, and residual confounding is the standard explanation for effects of this kind.

Reverse causation is also live: children who find sitting still and concentrating harder may be given screens more often, rather than the other way round.

Fourth question: how big is the effect?

Coverage reports whether an effect exists. Parents need to know how large it is.

With large samples, effects far too small to matter for any individual child reach statistical significance. A difference that is real across a population can be undetectable in a household. The number to look for is the effect size, and it is almost never in the headline.

The “sensitive windows” framing is genuinely interesting and worth testing — the claim that timing matters, not just total exposure, is more specific and therefore more falsifiable than most in this literature.

Why the caution runs both ways

None of this means the finding is wrong. Scepticism about a study that says screens are harmful is not an argument that screens are harmless — it is an argument for not overstating what one study can show.

We set out a related version of this in our fact check on the hundred-year flood: a genuine result, described in a way that invites the wrong inference. The questions above are the ones that separate the two.

This is a guide to reading research, not advice about any individual child. A health visitor or doctor is the right person for that.

Sources

• ScienceDaily — screen time at ages 1 and 6 and later academic outcomes, 24 August 2026