Screen Time

The best-documented casualty of screen-saturated childhood is vision: myopia has reached epidemic levels in East Asia — 80-90% of school-leavers in Singapore, Seoul, Shanghai and Taipei are short-sighted, with high myopia (which carries lifetime retinal-detachment and maculopathy risk) rising fastest — and prevalence is climbing in Europe and North America a generation behind. The mechanism evidence indicts near-work and, above all, lost outdoor daylight: randomised trials adding 40-80 minutes of outdoor time to school days measurably cut myopia onset, making daylight the rare screen-adjacent intervention with trial-grade proof. A projection widely used in ophthalmology expects half the world myopic by 2050. Almost none of the screen-time debate mentions any of this — a panic aimed at minds while the measurable damage accrues in eyes.

For cognition and mental health, total screen minutes predict remarkably little — associations in large datasets are small, often vanishing under proper controls — because screen is not an exposure but a container: video calls with grandparents, homework, algorithmic shorts and violent streams share a category. What carries measurable harm is displacement: screens consumed at the expense of sleep (the strongest, most replicated pathway — devices in bedrooms and evening use robustly predict shorter, worse sleep in children), physical activity, outdoor light (the myopia pathway) and, in young children, live language interaction. The practical translation: guard the things screens displace — bedtime, movement, daylight, conversation — and the minutes largely take care of themselves.

The under-5 evidence is the most consistent: heavy screen exposure in infancy and toddlerhood associates with language delay and attention problems, background television measurably suppresses parent-child talk (the raw material of language development), and no educational claim for under-2 screen products has survived testing — infants learn from live humans, poorly from video (the well-replicated video deficit). Hence the guidance convergence: WHO and paediatric bodies recommend no screen time under 1-2 years, an hour or less of quality content at 2-4, co-viewed where possible. The honest caveats: associations partly reflect family circumstances, video-calling is sensibly exempted everywhere, and the guilt industry around occasional cartoons vastly exceeds any measured effect of them. The dose that matters is the one displacing talk, play and sleep daily.

Past early childhood, the evidence fragments by what and how rather than how much: violent-content effects on aggression are real but small and shrink in better studies; educational and creative use associates neutrally or positively; gaming shows no reliable population harm and earned a narrow ICD-11 disorder category only for the small minority whose use escapes control; and the social-media-specific questions — comparison, algorithms, girls — have their own hub. Adults get one underrated finding and one overrated one: evening light and stimulation genuinely disrupt sleep (though blue light per se is a smaller factor than marketing claims — brightness, timing and arousal matter more than colour), while attention-span decline narratives run far ahead of their evidence. Moderate use with intact sleep, movement and relationships shows no detectable harm at any age above toddlerhood.

Screen time became the metric because it is measurable, not because it is the active ingredient — a classic streetlight effect that produces guideline hours with thin evidential foundations (the famous two-hour recreational limits trace to expert consensus, not dose-response data), fuels parental guilt disproportionate to measured effects, and lets platform-design questions hide behind family-responsibility framing. The cost of the misallocation is double: real mechanisms (sleep, daylight, content curation, algorithmic design) get less attention than the clock, and the one epidemic with trial-grade prevention — myopia, answerable with outdoor time — barely features in public debate. Counting minutes is easy; guarding what minutes displace is the actual health intervention.

By age: under 2 — screens are for video calls; the developmental work happens face to face. Ages 2-5 — an hour or so of decent content, ideally shared; guard talk and play. School age — two rules do most of the work: no screens in the bedroom overnight, and two hours outdoors in daylight daily, which protects eyesight with trial-level evidence while covering activity and sleep. Teenagers — see the social-media hub; sleep and content beat minute-counting. Adults — dim and wind down evenings, and audit whether use is displacing sleep, movement or people; if not, the hours are yours. Every age — the family rule that survives contact with reality is not a number but a structure: screens stay out of bedrooms and mealtimes, and daylight is non-negotiable.