More screen time tracked with better working memory in teens

Physical activity is widely assumed to help children's thinking, and screen time is widely assumed to harm it, but much of the supporting evidence captures a single moment rather than years of accumulated habit. What matters for a developing brain is presumably the total exposure built up over childhood, not a snapshot of one week. That requires following the same children repeatedly with both device-measured and self-reported behaviour. This study set out to link cumulative physical activity, sedentary time and screen time from childhood into adolescence with cognitive performance at ages 15 to 17.

The analysis drew on 260 adolescents, 136 of them boys, aged 15 to 17 at the 8-year follow-up of the Finnish PANIC study. Physical activity and sedentary time were measured with an Actiheart device, while different types of physical activity, non-screen sedentary time and screen time were captured by questionnaire at baseline and at the 2-year and 8-year examinations. Cognition was assessed at the 8-year visit using the CogState battery, which includes working memory tasks measuring both reaction time and accuracy. Exposures were expressed as cumulative measures across the whole follow-up and related to cognition using standardised regression coefficients.

Self-reported cumulative screen time was inversely associated with reaction time and accuracy scores in both working memory tasks and directly associated with overall cognition. Non-screen sedentary time showed a similar inverse pattern with working memory reaction time and accuracy. Cumulative unsupervised physical activity was inversely associated with working memory accuracy. All the coefficients were small, and the authors summarise the overall pattern as less unsupervised physical activity and more screen time being associated with better cognition in adolescence. The device-measured physical activity and sedentary time did not produce the headline associations.

This runs against the common expectation that screen time harms cognitive development and that more activity helps it, so it complicates rather than confirms the prevailing story. It is also a reminder that screen time is not one thing: the questionnaire cannot distinguish between different uses of a screen, some of which may overlap with schoolwork or reading. The study leaves open whether the associations reflect the behaviour itself, family and educational circumstances, or the imprecision of self-reported habits. Larger cohorts and studies that actually change children's behaviour would be needed to know.

Take this as a reason to treat blanket claims about screens and children's brains with caution, not as evidence that screen time builds cognitive skill. The relationships found here were small, self-reported and measured in one group of Finnish adolescents, and nobody was assigned to more or less of anything. It also does not say physical activity is bad for young people; the specific finding concerned cumulative unsupervised activity and one working memory accuracy measure. The honest position is that the link between these habits and adolescent cognition looks messier than the public conversation implies.