In brief
Whether screen time has anything to do with autism is open. People on the autism spectrum on average use the internet problematically more often and social media less often; this shows most clearly in problematic gaming. This does not mean that screens cause autism, and autism does not explain every heavy use either. Tested rules for everyday life hardly exist.
Screen time and autism: what the meta-analysis says
Ophir and colleagues evaluated 46 observational studies, five longitudinal and 41 cross-sectional studies with 66 effect sizes. They came from 4,682 studies found. The researchers found a small, statistically significant association between screen time and autism. It was strongest in studies on general screen use by children under twelve. After accounting for publication bias the association was small and no longer significant. Publication bias means that not all results are published. The authors call the results inconclusive and the question far from settled. [2]
The studies are observational and show only that the two occur together. It could also be that children with autism find screens particularly attractive. Equally, parents could be more generous with children who have a diagnosis. The authors name such explanations expressly. [2]
The kind of use matters
In the meta-regression the association disappeared when television, video games, computers and smartphones were considered separately. For social media the authors even found a negative association with autism. They conclude that distinguishing by device and activity is important. This could explain why studies find different things.
Screen use could have positive as well as negative consequences. Background television did worse in related meta-analyses, learning programmes and watching together better. Some games were developed specifically for children with autism to offer emotional support and joy. The authors see their results as a preliminary indication that supports existing medical recommendations to limit screen time in young children. Parents should make sure screen time does not come at the expense of real experiences and relationships. This is the authors’ assessment and not a result of the evaluation. [2]
Problematic internet use and gaming
Muris and colleagues carried out two meta-analyses. The first covered 46 studies with 42,274 participants. 26 studies concerned general problematic use, 15 problematic gaming and 5 excessive smartphone use. Autism or a higher level of features of autism was linked to problematic internet use (mean effect size r = 0.26). For problematic gaming the association was stronger (r = 0.31) than for general problematic use (r = 0.24). For excessive smartphone use no significant association showed. [1]
The authors explain this as follows. Games have built-in rewards and are clearly defined, whereas “problematic internet use” covers a lot. In children and adolescents the association was weaker than in young adults. It is possible that parents steer use more strongly at a young age. The results vary very strongly between studies (heterogeneity I² = 95.5 %). [1]
Social media: less use, but no withdrawal from the net
In the second meta-analysis with 15 studies and 7,036 participants, people on the autism spectrum used social media less (r = −0.28). 13 of 15 studies found this. The authors see this as preliminary because there are few studies and publication bias cannot be ruled out. They suspect that people on the autism spectrum do seek contact but use other digital services such as games for it. Measurement of social media was mostly crude, for example only the frequency of visits. [1]
How robust are the studies?
The authors rate the quality of the studies as “moderate”. Autism was often recorded without standard procedures. In 17 of the 46 studies the diagnosis was unclear or self-reported. Problematic use was mostly recorded by self-report, which can also be under-reported. Most studies are cross-sectional. Many participants belong to the group with lower support needs, and girls and women were hardly considered separately. [1]
What this means for everyday life and family
The meta-analyses we read test no measures and give no time limits. If use burdens sleep, school or relationships, a conversation about what screen time means for the person is worthwhile. It can be withdrawal, relaxation, contact or a special interest. This is an assessment by the editorial team that is expressly not evidence-based. Gaming and online contacts can also have social meaning.
If someone hardly sleeps any more, refuses school, withdraws strongly or becomes violent when restricted, counselling or therapy can make sense. The meta-analyses test no treatments for this.
Further articles are on Leisure and special interests, on Routines, structure and predictability in everyday life, on conflicts at home in the article Family life with a child on the autism spectrum and on counselling in the article Psychotherapy for people with autism.
What we know and what is open
What we know. Autism and problematic internet use occur together more often than in people without autism, most clearly in problematic gaming. Social media is on average used less. [1]
Open.
- Whether and in which direction screen time and autism are linked. [2]
- How robust self-reports on internet use are. [1]
- Which measures work in everyday life. The meta-analyses test none.
- Positive sides such as contact, learning and special interests have hardly been studied in the sources we read. [2]
- For people with higher support needs, girls and women the results are hardly informative. [1]
Frequently asked questions
Does screen time cause autism? No, this cannot be derived from the studies we read. A meta-analysis of 46 studies found an association that was no longer significant after accounting for publication bias. The direction of the association is unclear. [2]
Do people with autism more often game problematically? A meta-analysis found higher scores for problematic internet use, most clearly for problematic gaming. The results vary greatly between studies. [1]
Do people with autism use less social media? On average yes, according to 15 studies. This is regarded as preliminary. [1]
How much screen time is okay? The meta-analyses we read give no time limits and test no measures. If use burdens sleep, school or relationships, a conversation about what it means for the person is worthwhile. This is an assessment by the editorial team. More in the article Routines, structure and predictability in everyday life.
You might also be interested in
- Leisure in autism: activities and interests
- Routines, structure and predictability in everyday life
- Family life with a child on the autism spectrum
- Psychotherapy in autism: anxiety and depression
Sources and literature
- Muris P, Otgaar H, Donkers F, Ollendick TH, Deckers A: Caught in the Web of the Net? Part I: Meta-analyses of Problematic Internet Use and Social Media Use in (Young) People with Autism Spectrum Disorder. Clinical Child and Family Psychology Review 2025;28(2):392-413. https://doi.org/10.1007/s10567-025-00524-8 (frei lesbar, CC BY; online 23.04.2025) – Meta-analyses (46 and 15 studies)
- Ophir Y, Rosenberg H, Tikochinski R, Dalyot S, Lipshits-Braziler Y: Screen Time and Autism Spectrum Disorder: A Systematic Review and Meta-Analysis. JAMA Network Open 2023;6(12):e2346775. https://doi.org/10.1001/jamanetworkopen.2023.46775 (frei lesbar, CC BY) – Systematic review with meta-analysis (46 studies)
You might also be interested in
Knowledge articleShaping everyday life
Out and about: public places, shopping and travel
Knowledge articleShaping everyday life
Leisure in autism: activities and interests
Knowledge articleShaping everyday life
Overload, meltdown and shutdown: understanding the differences
Knowledge articleShaping everyday life
What helps in a meltdown?
Found a mistake?
If something is factually wrong, out of date or unclear, let us know. You do not have to give any personal details.
