What the research actually says about screens

By Susanne Rice · Grounded Parenting

*Not the headlines. Not the panic. Not the dismissal. What the evidence

actually shows, and what it means for your family.*

By Susanne Rice \| Grounded Parenting

Every few weeks, a new headline. 'Screen time linked to developmental

delays.' 'Children's tech use hits record levels.' 'Phones are

destroying childhood.' Then, a week later, a counter-headline:

'Educational apps improve literacy.' 'Screen bans don't work.' 'Stop

panicking about iPads.'

If you are a parent trying to make thoughtful decisions about technology

in your family, you are navigating a noise level that makes genuine

clarity almost impossible. The research exists. It is substantial and

growing. But it rarely reaches parents in a form that is actually

useful, specific enough to act on, nuanced enough to trust, and honest

about what we do not yet know.

This article tries to do that. It draws on the most current

peer-reviewed evidence to give you a clear picture of what screens

actually do and do not do in early childhood, so that your decisions are

grounded in something real.

What we know: the evidence that is now clear

Research on screens and young children has advanced significantly in the

last five years, and several findings are now sufficiently consistent

across studies to be considered settled.

Background television slows language development.

This is one of the most replicated findings in the field, and one of the

least well-known. Television playing in the background, even when no one

is watching it, even when it is simply on in the room, reduces the

number of words adults say to young children, reduces the number of

conversational turns between adult and child, and disrupts the quality

of play. The child does not need to be watching the screen for it to

have an effect. The adult's attention is partially captured by it, and

the conversational environment that is the single most important factor

in language development becomes thinner.

A large Danish study of 31,125 two and three-year-olds, published in

2024, found that mobile device screen time was associated with poorer

language comprehension and expressive language, even after controlling

for other factors. The effect was not catastrophic, but it was

consistent and present at levels of use that many families would

consider moderate.

**Parental screen use matters as much as, and possibly more than, child

screen use.**

This finding consistently surprises parents, and it is important enough

to say clearly: research on US families found that parental device use

during interactions with young children was associated with more child

behaviour problems, including both withdrawal and tantrums, with the

strongest association running through the interruption of parent-child

interactions that researchers have termed 'technoference'. The mechanism

is not the screen itself, but the moments when the adult's presence and

responsiveness is partially elsewhere.

This is not a criticism of parents who use their phones. It is a

description of a mechanism. The developing brain needs a responsive,

attentive human face and voice. When the adult who is present is also

partially absent, absorbed in a device, even briefly and benignly, that

responsiveness is reduced. The child notices. The interaction thins. A

parent reading a message at the kitchen table while their two-year-old

tries to show them something is not a negligent parent. They are a human

being living in 2026. But the two-year-old has already looked up,

checked the face, found it elsewhere, and turned back to what they were

doing, smaller, quieter, having learned something small about whether

this moment is worth sharing.

Content type matters more than duration for most outcomes.

The research does not support a simple 'more screen time is worse'

relationship. What it consistently shows is that the kind of screen use

matters enormously. Passive, fast-paced content, particularly

entertainment videos and short-form content, is most consistently

associated with negative outcomes for attention and language. Co-viewed

content, educational programmes watched with an engaged adult who talks

about what is on screen, is largely neutral or, in some cases, modestly

positive.

Under-twos cannot learn novel words from screens in the same way they

learn them from real human interaction. This is a well-established

finding sometimes called the 'video deficit.' A child who hears a new

word from a screen does not acquire it in the same way as a child who

hears that word from a person who is looking at them, responding to

them, and using it in a real exchange. The screen is not a substitute

for the human. It is a fundamentally different kind of input. This is

why the practical advice later in this article recommends video calls

with grandparents for very young children rather than passive viewing. A

live, responsive video call, where the person on screen reacts to the

child, uses their name, responds to what they do, is not subject to the

video deficit in the same way. It replicates the relational quality that

the brain is looking for.

Short-form video is a new and significant concern.

The 2025 Common Sense Census, the most comprehensive survey of

children's media use in the US, which tracked children aged zero to

eight, found that short-form video viewing among young children has

increased substantially since 2020, from a few minutes a day to an

average of around 14 minutes for under-eights. TikTok, YouTube Shorts,

Reels: content designed to be fast-paced, algorithmically optimised for

continued viewing, and structurally incompatible with the sustained

attention that is a prerequisite for learning.

The research on fast-paced content and attention regulation in young

children is consistent: exposure to it is associated with reduced

capacity for self-directed, sustained attention. This does not mean that

a child who watches YouTube Shorts is damaged. It means that the

attention architecture built in early childhood, which is the foundation

for reading, learning, and the kind of deep play that Chapter 1

described, is shaped by what it is asked to practise. A diet of very

fast, very reactive content practices a different kind of attention than

a diet of slow, child-directed play.

What we do not know: where the research is still developing

Honesty requires naming what is not yet settled, because a significant

proportion of screen time headlines, in both directions, overstate the

certainty of the evidence.

We do not have reliable long-term data on tablets and smartphones

specifically. Most of the longitudinal research was conducted with

television, and the extrapolation to interactive devices is reasonable

but not proven. The evidence is catching up, but it is not yet as robust

for tablets and phones as it is for background TV.

We do not know what moderate, intentional, co-viewed screen use does

over the long term. Research on harmful effects is largely focused on

passive, high-volume, or background use. The parent who watches one

episode of something with their three-year-old and talks about it

afterwards is operating in a space where the evidence raises no alarms.

And we do not have good data on use that is genuinely contextual and

parent-led, the phone used as a camera to document a child's play, the

tablet used for a video call with grandparents, the programme chosen

deliberately and watched together. These are meaningfully different from

the scroll, the background noise, the hour of YouTube before bed, and

the research has not yet fully distinguished between them.

What this means in practice

The research does not tell you to eliminate screens. It tells you

something more specific and more useful than that. It tells you where

the effects actually come from, which means you can address those

specifically rather than managing screen time as a blunt total.

Turn the television off when no one is watching it.

>

This is the single most evidence-supported thing you can do.

Background television affects language development even when your

child is in another room, and you are not aware of it. It is an easy

change, with consistent evidence to support it.

>

Notice your own phone use around your child.

>

Not to judge it, but to see it. The moments when your child is talking

to you, and your eyes are on your phone. The mealtimes when you're

both somewhere else. These are the moments the research points to, and

they are changeable without drama or self-flagellation.

>

Treat content type as more important than time.

>

An hour of slow, co-viewed content watched with an engaged adult who

talks about what is happening is developmentally different from twenty

minutes of fast-paced algorithmically served video watched alone.

Duration is a crude measure. Quality and context are what the research

actually cares about.

>

Under-twos: treat screens as social tools, not solitary ones.

>

Video calls with people who love your child. The occasional programme

watched together. Not the babysitter, not the background filler. The

research on very young children is most detailed here: the human voice

and face, responsive and attentive, is what the developing brain is

looking for. Screens are not a substitute for that. They are, at best,

a supplement to it.

>

Use the screen as a launchpad, not a destination.

>

If your child watches something about dinosaurs, the value is not in

the watching. It is in the conversation afterwards, in the books that

follow, in the plastic dinosaurs that come out, in the game that grows

from it. Screens are best understood as a door into something rather

than the thing itself.