Just pointing out that that does stress that this is not detecting causality but correlation. I only have the abstract of the 1st one, but the second only uses sex, BMI, family income, parental education, and race and ethnicity as covariates, they could easily be missing some underlying factor that is causative to both. Not that it is a particularly radical hypothesis, getting kids to run about outside rather than sit in front of the TV all day is probably going to be good for them.I also met plenty of folks online. It's not that the medium is bad or has no use. And I wouldn't say it's 'unknown', the damages of screentime replacing RL interactions are known and will be as obvious as smoking damages over the coming decades.