Screen time for kids has been measured, argued over and legislated for two decades, almost always by counting the hours the child spends in front of a device. A cohort study published in JAMA Pediatrics on 17 August 2026 counted those hours too, found almost nothing, and then found something in a place most guidance does not look: the phone in the parent's hand.
The finding is small, it is an association rather than a cause, and it comes from a single country. It is still worth an hour of a parent's attention, because it points at a lever most families have not tried.
What did the new Norwegian study actually find?
Researchers followed 747 Norwegian children recruited through municipalities in 2023, testing them once a year for three years from roughly age 5 to age 7. Most indicators of the child's own screen use showed no association with how fast their language grew. Two things did show an association: the primary parent's daily smartphone time, and whether the child owned a digital device of their own at age 4 to 5.
The numbers, from Yang and colleagues in JAMA Pediatrics: each additional hour per day of the primary parent's smartphone screen time was associated with 0.42 points less annual growth in the child's receptive vocabulary (95% confidence interval, minus 0.66 to minus 0.18). Children who owned a device at 4 to 5 scored 5.13 points lower in receptive vocabulary and 4.56 points lower in receptive grammar at the start of the study than children who did not. Language was measured with the British Picture Vocabulary Scale and the Test for Reception of Grammar, and the models adjusted for parental education and household income.
Why did most screen time measures show nothing?
Because this study separated two questions that older research tended to blend together: where a child starts, and how fast a child moves. Many earlier studies compared children at a single point in time, so a difference in scores could reflect differences that were already there before any screen was switched on. Latent growth modelling lets you look at the slope instead of the snapshot.
That distinction matters for how you read the older literature too. A widely cited 2020 meta analysis, Madigan and colleagues on screen use and child language skills, found that greater quantity of screen use was associated with weaker language skills, but also that the quality of what was watched and whether an adult watched alongside changed the picture substantially. The Norwegian result is consistent with that: hours alone are a weak predictor. If you are trying to reduce those hours anyway, we have a list of screen-free alternatives that hold a child's attention long enough to matter.
What does a parent's phone have to do with a child's vocabulary?
The most likely explanation is not mysterious: a parent looking at a phone is a parent not talking. Vocabulary at this age is built almost entirely out of conversation with adults, and every interruption is a sentence the child did not hear and a question that went unanswered. Researchers call these small interruptions technoference, and this study is one of the first to link a parent's own screen time to the child's language trajectory rather than to the child's behaviour.
Two honest caveats. First, 0.42 points of annual growth per hour is a small effect, and the confidence interval runs from 0.18 to 0.66, so even the upper end is modest. Second, this is observational: parents who spend more time on a phone may differ in other ways the model did not capture, and the authors do not claim causation. What the finding does justify is a cheap experiment. Nobody has to buy anything to try it.
Does owning a tablet at four cause a smaller vocabulary?
No, and the study is careful about this. The device ownership effect was found at baseline, meaning those children already scored lower when they were first tested, and their subsequent growth was not slower. That pattern is more consistent with device ownership being a marker of something else in the household than with the tablet itself shrinking a vocabulary.
It is still a useful marker. A 5.13 point gap in receptive vocabulary is not trivial, and a parent deciding whether a four year old needs their own device now has one more reason to wait. The relevant question is what the device replaces. A tablet that replaces twenty minutes of nothing is different from a tablet that replaces twenty minutes of a parent reading aloud.
What do paediatricians still recommend on screen time by age?
Broadly, less than most families currently manage, with more emphasis on content and company than on a stopwatch. The American Academy of Pediatrics has moved away from a single hourly number toward a family media plan, and its guidance in its updated media recommendations and in its explanation of the underlying policy is worth reading before setting a rule you will not keep.
For the youngest bracket the AAP is more specific, and its answer on screen time limits for young children is the clearest short summary available. We have set out the numbers by age in our breakdown of screen time recommendations from 3 to 12, and the separate question of what screens do to sleep in what actually keeps kids awake.
| What the study measured | Result | What it suggests for a family |
|---|---|---|
| Child's daily screen hours | No association with language growth | Counting hours alone tells you little |
| Child owned a device at 4 to 5 | Lower scores at baseline, growth unaffected | A reason to delay a personal device, not a verdict |
| Primary parent's daily phone time | Slower annual vocabulary growth | The cheapest change available is your own |
| Mother's and father's social media time | No significant association | The mechanism is likely interruption, not screens as such |
What can a family change this week?
One thing, chosen because it is measurable. Pick a single recurring slot, the drive home, the meal, the twenty minutes before bed, and make it phone free for the adults. That is the variable this study actually implicates, and it costs nothing to test for a fortnight.
The second change is to make sure the replacement is something that asks the child to talk, not something that talks at them. Reading aloud is the strongest option and always will be. Beyond that, the useful test for any toy is simple: does the child end up producing more words, or fewer? It is the test we apply to Ted, our interactive plush toy for children aged 3 to 12. Ted is built to ask a question and then wait rather than to fill the silence. It has no camera and no screen, and the parent sets the language, the topics and the tone from the app. It is a single purchase at 129 euros with no subscription, and the full specification is on the Ted product page if you want to judge it against the checklist above rather than against our description of it.
Frequently asked questions
How much screen time is safe for a 5 year old?
There is no threshold that becomes unsafe at one minute past. The AAP's current position is to plan media use as a family rather than to enforce a single number, with attention to what is watched, whether an adult is present and what the time displaces.
Does my phone really affect my child's language?
One good study associates the primary parent's daily phone time with slightly slower vocabulary growth between ages 5 and 7. It is an association, not proof, and the effect is small. It is also the easiest thing on the list to change.
Should I take my child's tablet away?
The study gives no support for that. The children who owned devices did not grow more slowly; they simply started lower. Removing a device abruptly usually costs more than it gains. Restricting when and where it is used tends to work better.
Is educational content different from other screen time?
The evidence suggests yes, with a caveat. Content quality and adult co-viewing both moderate the relationship in the 2020 meta analysis. A child watching alone is a different situation from a child watching with someone who talks about it.
My child is 7 and behind on vocabulary. What should I do first?
Speak to a speech and language therapist or your paediatrician rather than adjusting screens on your own. Language delay has many causes and screen time is at most one of them, and a professional assessment will tell you far more than a fortnight of measurement at home.
