Screen time: all the 2026 numbers (France, ages, attention)
Everyone quotes numbers about screen time, and almost nobody cites their sources. So we've gathered here the most solid data available — the kind that comes from public institutions, audience-measurement bodies and peer-reviewed scientific journals. Every figure comes with its source and its date. Feel free to reuse them, cite them, share them.
The number to remember: about 4 hours a day
According to the Baromètre du numérique 2025, carried out by Arcep, Arcom and the Conseil général de l'économie among 4,066 people, French people's daily screen time for personal use comes to about 4 hours. To put that in perspective: that's a quarter of your waking hours in a normal day. And this is personal use alone — screen time at work comes on top of it.
Source: Baromètre du numérique, 2025 edition, Arcep / Arcom / CGE.
How much time, and for whom?
| Metric | Figure |
|---|---|
| French people spending more than 2 hrs/day in front of a screen (personal use) | 72% |
| French people spending more than 5 hrs/day | 25% |
| 18-24-year-olds spending more than 5 hrs/day | 39% |
| Own a smartphone (aged 12 and over) | 91% |
| Own a computer | 89% |
The generational contrast is stark: one young adult in three tops five hours a day — nearly a full working day spent in front of a screen, every day, outside of any work or school obligation.
Video alone: 4 hrs 14 min a day
Médiamétrie measured that in 2025, French people watched an average of 4 hours and 14 minutes of video content a day, across all screens (television, computer, tablet, smartphone). This figure partly overlaps with the previous one: it doesn't add to it, it shows just how much video dominates screen time.
The most interesting number: 42% think it's too much
It's probably the most telling data point in the 2025 Baromètre: 42% of French people feel they spend too much time in front of screens, and 19% feel they spend far too much. So this isn't a problem that experts diagnose from the outside — it's a discomfort people feel for themselves.
And it grows with how heavily people use screens: among 18-24-year-olds who spend more than 3 hrs a day in front of a screen, 61% consider that time excessive, compared with 31% of those over 70. The heaviest users are also the most clear-eyed about their own habits. What's missing isn't awareness. It's a way to act.
What screen time does to your attention
47 seconds before switching screens
Gloria Mark, a researcher at the University of California, Irvine, has spent more than twenty years measuring how long we stay on a single window before switching elsewhere. In 2004, the average was 2 minutes 30 seconds. In her most recent data, it has dropped to 47 seconds. In twenty years, our ability to stay on one thing has been cut to less than a third.
23 minutes to get back to the task
The same team established that after an interruption, it takes an average of 23 minutes and 15 seconds to get back to your original level of concentration. This is the number that makes screen time so deceptive: "I'm just checking my phone for ten seconds" doesn't cost ten seconds. It costs the next quarter of an hour.
More than double the ADHD symptoms among heavy users
A study published in JAMA in 2018 followed 2,587 high-schoolers with no ADHD symptoms at the outset for two years. Among those who reported 14 high-frequency digital activities, 10.5% showed ADHD symptoms two years later, compared with 4.6% among those who reported none.
Source: Ra et al., JAMA, 2018. We break down these findings in our article ADHD and screens: what the science says.
Children: what France recommends
In April 2024, a panel of experts appointed by the French Presidency delivered a report of nearly 140 pages, "Enfants et écrans — À la recherche du temps perdu" (Children and Screens — In Search of Lost Time), after three months of expert hearings and consultations with 150 young people. Its age recommendations are now the benchmark in France:
| Age | The panel's recommendation |
|---|---|
| Before age 3 | No screens |
| Before age 11 | No mobile phone |
| Ages 11-13 | Phone with no internet access |
| Ages 13-15 | Connected phone, but no social media |
| Age 15 and up | "Ethically designed" social media only |
Source: report "Enfants et écrans — À la recherche du temps perdu", April 2024. Our detailed guide: screens and children, the age-by-age guide (0-11 years).
How to read these numbers without getting them wrong
Three precautions, because this topic is saturated with dubious statistics.
1. Be wary of numbers with no source or date. The famous "we touch our phones 2,617 times a day" is still circulating everywhere: it comes from a 2016 study run on a small American sample. It's not wrong, but it's neither French, nor recent, nor representative.
2. Self-reporting overestimates… or underestimates. Surveys like the Baromètre rely on what people report about themselves. Yet automatic measurements show that we generally underestimate our own screen time. The real figure is probably higher than 4 hours.
3. Duration doesn't tell the whole story. The CHU Sainte-Justine study (4,000 young people followed for 5 years) found an association between ADHD symptoms and social media, television or video games — but none with simply using a computer. Two hours of video editing and two hours of endless scrolling don't produce the same effects. The format matters more than the count.
So now what?
If these numbers hit a little too close to home, the most useful thing isn't to measure more — your phone already does that, and it hasn't changed anything. It's to add friction where there's none. The Lock-IN box (39 €, no subscription) blocks your distracting apps with a single tap: you set your phone on it, the apps lock, and you have to physically come back to the box to reopen them. Calls, texts and GPS keep working.
Read next: why Screen Time mode isn't enough, how to stop scrolling and digital detox: where to start.
Using these numbers for an article, a talk or a presentation? Feel free to cite them, as long as you credit the original source (listed under each data point) and, if you like, link back to this page.