Developmental psychologist Dr. Sheri Madigan tracked thousands of toddlers over several years to answer the question that hangs over the whole screen-time debate: do screens actually come before the developmental differences, or after them?
Dr. Sheri Madigan holds the Canada Research Chair in Determinants of Child Development at the University of Calgary. A research chair of this kind is not a courtesy title; it is a national appointment given to researchers whose work is shaping a field, and Madigan's field is precisely the question parents most want answered: what, in the earliest years, actually determines how a child develops?
That framing matters. Madigan is not a campaigner against screens, and her research does not start from the assumption that technology is the villain. She studies determinants, the full set of early influences on a developing child, and screens earn their place in her work simply because they now occupy a large share of many toddlers' waking hours. When a researcher with that brief turns her attention to screen time, the result is worth more than a hundred opinion pieces, because the method is built to find out what is true rather than to confirm what anyone already believed.
The study at the centre of this article, published in JAMA Pediatrics in 2019, followed roughly 2,400 children over several years. Mothers reported how much screen time their children had at 24 months and 36 months, and the children were assessed at 36 months and 60 months on a developmental screening measure covering communication, motor skills, problem-solving and social skills: the everyday machinery of early childhood, from forming sentences to climbing stairs to working out how a toy fits together to playing alongside another child.
The headline finding was clear. Children with higher screen time at 24 and 36 months went on to perform less well on the developmental screening at 36 and 60 months. More screen time earlier predicted poorer scores later, across exactly the domains a parent of a toddler cares most about.
The source
Madigan et al. (2019), Association Between Screen Time and Children's Performance on a Developmental Screening Test, JAMA Pediatrics · Read the original
But the most valuable part of the study is not the headline, it is the direction. There has always been a perfectly reasonable objection to research in this area: perhaps children who are developing more slowly are simply given more screen time, because they are harder to occupy, and the screen is a symptom rather than a cause. Because Madigan's team measured the same children repeatedly over time, the 2019 study could test that objection directly, and it did not hold. The pattern ran from screen time to later developmental performance, not from earlier developmental performance to later screen time. Screens came first.
As for why, Madigan's proposed mechanism is displacement, and it is disarmingly simple. The early years run on enormous quantities of practice: thousands of exchanges of babble and reply, thousands of climbs and falls and grips and stacks, thousands of small problems solved with the hands. An hour in front of a screen is an hour in which a toddler is, for the most part, not talking with anyone, not moving their body, and not manipulating anything. The screen does not have to inject harm. It only has to crowd out the practice, and the practice is where development comes from.
First, the honest caveats. This is observational research, not a controlled experiment, and no study of this kind can rule out every alternative explanation. The differences it describes are patterns across a large group, not predictions about any individual child, and a toddler who watched television this morning has not been marked by it. Madigan's own framing is measured, and a parent's reading of it should be too.
What makes the study harder to wave away than most is the time structure. The screens-to-outcomes direction undercuts the most comforting counter-argument, the one that says the screen time is merely a marker for something else. And the mechanism it points to is one a parent can actually act on. If displacement is the engine, then the question to ask about any given hour is not "is this content educational?" but "what would this hour otherwise hold?" For a two-year-old, the otherwise is talking, moving and handling things, and those are exactly the domains the screening measured: communication, motor skills, problem-solving, social skills.
There is real comfort in that framing. Displacement is not a one-way street. An hour reclaimed for chatter on the kitchen floor, for stacking and posting and climbing, is an hour of precisely the practice the research says was missing. The same mechanism that explains the risk also describes the remedy, and the remedy is cheap, ordinary and available this afternoon.
Take Madigan's four screening domains and put a wooden shape sorter on the floor between you and a two-year-old, and you can watch all four run at once. The child grips the beech cylinder, a fist's worth of motor practice with a real weight in it. She tries it against the square hole and it refuses, which is problem-solving conducted at the speed of a small hand, not a swipe. You say "turn it round", she looks up at you, and that look, the word, the second try, the triumphant clunk as it drops through, is communication and shared attention folded into a single ten-second loop. Then she tips the whole thing out and hands you a block, because the game is better with you in it, and that is the social skill the screen could never have rehearsed.
This is the texture of what the study calls displacement, run in reverse. Nothing about it is rare or clever. Threading beads, hammering pegs, counting steps on a caterpillar puzzle, matching animal dominoes with the click of wood on wood: every one of these is an hour that holds talking, moving and handling all at the same time. At Jaques London we have been making toys to this recipe since 1795, longer than any other games company on earth, and our educational wooden toys built for small hands and shared play → are designed around exactly the loop described above: grip, try, fail, adjust, succeed, show somebody.
The practical move is unglamorous. Pick the one slot in the day where the screen most reliably appears, often the stretch before tea, and pre-empt it: basket of sorting, stacking and counting toys → already out on the floor, you nearby and occasionally drawn in. You do not need to narrate like a teacher or supervise like a lifeguard. The toys do the inviting; your job is to be the person the block gets handed to. One reclaimed hour, repeated daily, is a very large amount of practice by the time a child turns five.
The original paper, Association Between Screen Time and Children's Performance on a Developmental Screening Test, is clearly written and careful about its own limits, and worth reading in full. To place it in the wider evidence, start with our guide to the science of play and screens, then our practical guide to reducing toddler screen time for the how, and why real play matters for the bigger picture the displacement argument sits inside.
Proof is too strong a word for any observational study, and Madigan does not use it. What the study showed, in around 2,400 children followed over time, is that higher screen time at 24 and 36 months predicted poorer developmental screening scores at 36 and 60 months, and crucially that the pattern ran in that direction rather than the reverse. That rules out the most common alternative explanation, but it remains an association, describing group-level risk rather than individual destiny.
Worry is not the useful response; rebalancing is. The findings describe patterns across thousands of children, not a verdict on yours, and the proposed mechanism, displacement, points straight at the fix: restore some of the talking, moving and hands-on play that screen hours were standing in for. Early development is built on volume of practice over years, so the hour you reclaim this week counts, and the one after that counts more.
The displacement argument is largely indifferent to content. The concern is not that the programme teaches nothing, it is that while it plays, a toddler is mostly not conversing, not moving and not manipulating anything, and those are the domains where the study found later differences. Better content is a reasonable choice when a screen is going to be on anyway, but it does not return the practice the hour displaced.
The study's own screening domains are the best shopping list: communication, movement, problem-solving and social skill. In practice that means back-and-forth chat during ordinary tasks, free movement indoors and out, and toys that make hands work, sorters, stackers, puzzles, threading and simple matching games, ideally with you in reach so the play keeps turning social. Nothing exotic, just the ordinary ingredients of a toddler's day, present in quantity.
Britain's oldest toy maker. Every toy open-ended, wooden, and built for exactly the kind of play the research recommends.