Research
Every source I quote, in one place. If I say it in a video, the evidence is here, linked to the original.
Young children and screens
70 Canadian toddlers (mostly middle-to-upper income families). Mothers reported screen time at ages 2 and 3; words were counted on a parent-completed checklist at 2. More weekday and weekend screen time at 2 went with fewer words at 2. Among toddlers with smaller volume in one language-related brain area, fewer words at 2 predicted more screen time at 3, so the link may run both ways. Observational: the authors say causal inferences cannot be drawn.
Sumeet Kumar and Madhu Viswanathan (Indian School of Business) and Ravi Bapna (Santa Clara University), in Information Systems Research (published online September 2026). Across more than 32,000 “made for kids” videos from 93 YouTube channels, 10% more visual motion went with about 1% more views. In a randomised field test, adding moving stars behind an unchanged lesson raised view rates by 1.6% (medium motion) and 2% (high motion). In an eye-tracking study of 53 children aged 6 to 10, motion raised on-screen attention by about 3.4 to 3.6 percentage points, but most of the extra attention went to the moving background, not the instructor. Learning (comprehension or retention) was not measured. Viswanathan: “More animation is certainly not necessarily better.”
The 2025 Common Sense Census of media use by US kids aged 0 to 8: 40% have their own tablet by age 2 and 58% by age 4; by age 8 about one in four has their own cellphone. Young children average about 2.5 hours of screen time a day.
NPR coverage of the Common Sense Census. Common Sense Media chief executive Jim Steyer's advice to parents: “Delay, delay, delay” on tablets and phones.
More than 150 clinicians, including GPs and paediatricians, wrote to the UK Education and Health Secretaries urging schools to “start rolling back screen use”. The group says nearly a million children use screens in most lessons, and cites an analysis of 276 studies (5.4 million people) in which short-sightedness among children and adolescents rose from about 24% in 1990 to almost 36% in 2023. The Department for Education said that, used well, technology can be a powerful tool for learners who need it, and that it is consulting on screen-time guidance for ages 5 to 18.
A New Zealand Public Health Expert Briefing (University of Auckland and University of Canterbury) says almost two-thirds of babies and toddlers in Aotearoa New Zealand exceed recommended screen time. It argues for looking beyond individual parenting choices, noting that digital products are designed to keep people engaged, and calls for screen-free early childhood and healthcare settings, better parental leave and childcare, and limits on marketing to young children. An opinion briefing, not peer-reviewed.
Western Australia will trial less classroom screen time for Kindergarten to Year 2 in 10 public schools for six months from Term 1, 2027, with $50,000 per school for hands-on materials. Students can still use devices for the technology curriculum and NAPLAN.
From the 2027 school year, NSW public schools limit classroom screen time to 15 minutes a day for Kindergarten to Year 2, 30 minutes for Years 3 and 4 and 45 minutes for Years 5 and 6, averaged across the school week, becoming mandatory in 2028. The government cited record-low PISA results in maths and reading. Students with disability, additional needs or English as an additional language can be exempt.
No screen time for under-1s or 1-year-olds; no more than 1 hour a day for 2 to 4-year-olds, and less is better. Not restrained (prams, high chairs) for more than 1 hour at a time; 11 to 14 hours' sleep at ages 1 to 2 and 10 to 13 hours at 3 to 4; reading and storytelling encouraged when sitting.
Screens, sleep and child development
67,710 students aged 11–19 from 154 Quebec schools, surveyed in 2025. Every type of recreational screen use was linked with later bedtimes. Compared with teens on 1 hour a day or less, those on more than 3 hours went to bed about 45 minutes later for social media, video gaming and messaging, and about 30 minutes later for web surfing and TV or streaming. The gap was larger for 11–12 year olds than for those 15 and older (except gaming) and among late sleepers. Cross-sectional: a link, not proof.
23 meta-analyses covering 2,817,252 children and teens from 64 countries. More screen use was associated with lower school achievement (r = −0.15) and language skills (r = −0.14), and with more internalising (r = 0.10) and externalising (r = 0.11) problems, emotional problems and peer problems. No clear association with executive function, prosocial behaviour, maths, science or reading. The authors say screen exposure may be a modifiable factor in children's well-being.
Independent experts note the associations are very small, that most included studies are cross-sectional (association, not cause), and that the review could not account for how or why screens were used.
Tailisha González, Griselda Chapa (Lehman College, CUNY) and Corey H. Basch (William Paterson University) studied 58,881 children (6–12) and teens (13–17) in the 2023–24 National Survey of Children's Health. Compared with 2 hours or less of recreational weekday screen time, 3 hours a day was linked to 30% lower odds of adequate sleep in children and 4+ hours to 42% lower (slightly smaller after full adjustment); in teens, 25% and 32% lower. 60.6% of children and about 70% of teens met sleep guidelines. Cross-sectional and parent-reported: a link, not proof.
“The AAP recommends keeping all screens—TVs, computers, laptops, tablets and phones out of children's bedrooms, especially at night.” And: “To prevent sleep disruption, turn off all screens at least 60 minutes/1 hour before bedtime.”
The American Academy of Pediatrics' 5 Cs: Child, Content, Calm, Crowding out and Communication. “Research shows that content quality shapes whether kids have positive or negative relationships with media.” Rather than only cutting minutes, look at what media may be crowding out.
NPR's Rhitu Chatterjee: set clear boundaries (new apps need permission); put phones away at meals and bedtime; have a family media plan the parents follow too; “be curious, not furious” (Sunny Xun Liu, Stanford Social Media Lab); and start talking early. Dr Jason Nagata (UCSF): “If you're not on your phone late at night, then you have less use.”
Report on the same review (published in Nature Human Behaviour, 2,451 studies, almost two million participants under 18): social media was consistently associated with depression, anxiety and risk-taking, while co-viewing and educational content were linked to better outcomes. Lead author Dr Taren Sanders: it matters more what kids are doing than how much.
An umbrella review led by Dr Taren Sanders of Australian Catholic University pooled 2,451 studies covering almost two million children and teens. Most harms were small and mainly tied to heavy use and social media; social media was consistently linked to depression and other mental health problems with no sign of benefit. Educational and co-viewed content can be neutral or beneficial, and watching TV with a parent reversed the link with weaker literacy.
UAE doctors and school leaders report more children with sleep disruption, eye strain and behaviour problems linked to heavy screen use. Paediatrician Dr Sudha Rao advises keeping devices out of bedrooms overnight and charging them in a shared space one to two hours before bed.
Survey of parents of 500 children at primary health care centres in Dammam, Saudi Arabia. Among children with more than four hours of screen time a day, 66.7% (40 of 60) had trouble falling asleep, compared with 9.5% of those with under one hour. Parent-reported and measured at one point in time, so it shows a link, not cause and effect.
A review in JAMA Pediatrics covering nearly three million children aged 18 and under in 64 countries linked more screen time to poorer outcomes in learning, speech and emotional regulation. Senior author Sheri Madigan says the findings are correlational and do not prove that screens cause these outcomes, and most of the links were small. Outside experts note it cannot draw firm conclusions about specific types of use such as social media or gaming. I read CNN's report as republished by CP24.
A systematic review and meta-analysis of 20 studies covering 125,198 children aged 6 to 19. Using a device at bedtime was linked to not enough sleep (odds ratio 2.17), poor sleep quality (1.46) and excessive daytime sleepiness (2.72). Having a device in the bedroom at night, even unused, was linked to not enough sleep (1.79). The studies were cross-sectional, so they show links, not cause.
Recommends keeping all screens (TVs, computers, laptops, tablets and phones) out of children's bedrooms, especially at night, and switching screens off at least an hour before bedtime.
Teens and social media
A qualitative study of 50 Australian families with 118 children aged 1–18 suggests using technology as a reward or punishment can have adverse effects. A 15-year-old: “I don't tell my parents much now about what happens to me because I don't want my phone taken off me.” Device consequences make sense when the problem involves technology, such as online bullying.
A nationally representative survey of 1,211 US 9- to 17-year-olds. A typical teen spends about 8 hours a day on screen media for entertainment (median 8 h 15 min) and a typical tween about 5 hours; 43% of teens and 23% of tweens spend 10 hours or more. Smartphone ownership: 24% at age 9, 47% at 10, 57% at 11, 94% at 17. Short-form video is a daily habit for 82% of teens and 58% of tweens. 77% of teens use AI in their free time, but only 11% of teen AI users enjoy it “a lot”; 48% of teens think they spend too much time on screens.
More than 21,000 Swedish twins were followed from age 9 into adulthood. Fifteen-year-olds who spent more than 6 hours a day on interactive screen activities were over twice as likely to be diagnosed later with depression or anxiety as those at 2 hours or less. Most of the links weakened when twins in the same family were compared, particularly identical twins, and the lead researcher says the association "may in some cases have been overestimated". Screen use was self-reported and the data predates today's smartphones and social media. Published in Nature Human Behaviour.
Teens spending more than 3 hours a day on social media face double the risk of depression and anxiety symptoms. Up to 95% of 13 to 17-year-olds use social media, and a third say they use it "almost constantly". Advice for parents includes tech-free zones, a family media plan and modelling healthy use.
A review of reviews covering nine million people under 25 across 121 reviews found consistent links between problematic digital media use and harm to health, including mental health, sleep and physical activity. Moderate use was linked to higher wellbeing. The authors say the links cannot establish cause and effect and that the certainty is low. They call for technology companies to be required to assess new products and features for harm before release.
The press summary of the Lancet series, with its headline numbers and recommendations. The figures I quote from the series were checked against this summary.
How apps are designed to hook us
Everyone on Instagram and Facebook was given the option to hide like counts on other people's posts and on their own. Instagram had tested hiding counts to see if it might “depressurize people's experience”; it said not seeing like counts was beneficial for some and annoying to others, so it became a choice.
James Lynden and Teis Rasmussen's mixed-methods study links read receipts to “an array of behaviours and emotional effects”, including avoidance strategies, more frequent checking of messaging apps and feelings of anxiety, alongside strongly negative attitudes to the feature.
Ying Xie (UT Dallas), Mina Ameri (University of Pittsburgh) and Elisabeth Honka (UCLA), Journal of Marketing Research, 2023, using data from an online fan community. Similarity mattered most when strangers became friends: common friends, common interests and shared demographics or location. Having more active friends made users more active (if friends watched two shows the day before, the chance a user watched the next day rose by 15%). On fast-growing (evolving) networks, recommending a friend of a friend was the most effective friend suggestion.
Florida State University researchers Cary Stothart, Ainsley Mitchum and Courtney Yehnert tested 166 students on an attention task. Receiving a call or text they did not answer still hurt performance, by an amount similar to studies where people actually used their phones (Journal of Experimental Psychology: Human Perception and Performance, 2015).
Explains variable reward schedules, introduced by psychologist B.F. Skinner in the 1930s: when rewards arrive unpredictably and checking is cheap, people check out of habit. Compares the effect to how slot machines keep players engaged.
Aza Raskin, credited with inventing infinite scroll in 2006, says he built it to help people before social media took off, that “incentives eat intentions”, and that he had to write software to break his own addiction.
New York's attorney general alleges, in a newly unsealed complaint, that in 2023 TikTok gave thousands of users, including teens and children, non-working versions of Algo Refresh, a feature meant to reset their recommendations, to compare them with users who got the working version. TikTok says the allegations mischaracterise the feature and that it regularly tests new products. These are allegations in a lawsuit, not findings by a court.
Aza Raskin, who designed infinite scroll, testified about its impact and said he regrets it.
When Autoplay is on, another related video plays automatically after a video ends. On Wi-Fi, Autoplay stops automatically after 4 hours; on a mobile network, after 30 minutes of inactivity. For users aged 13 to 17, Autoplay is off by default. To switch it off on a phone: profile picture, Settings, Autoplay.
Phone settings that help
From the parent's phone: open Family Link > select your child > Screen time > Time limits > App limits > choose the app > Set limit > pick the hours and minutes > Done.
Autoplay Next Episode is set per profile. In the Netflix app: My Netflix > tap the profile name > Manage Profiles > choose the profile > turn off “Autoplay Next Episode”. In a browser: Account > Profiles > choose a profile > Playback settings.
The same profile screen has an “Autoplay Previews” switch. iPads and some older TVs and streaming devices don't support preview autoplay.
Google's path: in Google Messages tap your profile picture > Messages settings > RCS chats (or Chat features), where “Send read receipts” and “Show typing indicators” are switched on or off.
For one conversation: open Messages, tap the chat, tap the contact's name, then turn off Send Read Receipts. For everyone: Settings > Messages > Send Read Receipts. Read receipts apply to iMessage, not SMS.
Apple's overview of the redesigned parental controls: open Settings, tap Family, tap your child's name, then Screen Time. The new controls need iOS 27, iPadOS 27 or macOS 27; Ask to Browse is on by default for children under 13 and Limit Adult Websites for under-18s.
Apple's steps: Settings > Focus, choose a Focus (for example Do Not Disturb), choose which people and apps may notify you, and tap Add Schedule to turn it on automatically at set times.
Google's steps: Settings > Sound & vibration > Do Not Disturb > Turn on automatically, then add a time rule (on newer phones, Settings > Modes). Menu names vary by phone.
Google's steps for a daily limit: in the Family Link app, select your child > Screen time > Time limits > turn on Daily limit, then set each day. Daily limits apply to each Android device or Chromebook the child uses, so a 2-hour limit gives 2 hours on each device.
Apple's iPhone User Guide for limiting time in apps with Screen Time (Settings > Screen Time).
Psychologist Alex Holte and colleagues found that switching a phone's display to grayscale cut smartphone use by an average of 39 minutes a day in a study of about 220 college students; a follow-up with about 135 participants also found less screen time and lower anxiety.
Apple's steps for colour filters, including Grayscale: Settings > Accessibility > Display & Text Size > Color Filters.
Google's steps for Bedtime mode, including a grayscale screen at bedtime (Android 13 and later): Settings > Digital Wellbeing & parental controls > Bedtime mode > Customize > Screen options at bedtime > Grayscale.
On Android 17 or later, with a parental controls PIN: Settings, Parental controls, Downtime, turn on Use downtime, then choose the days and the start and end times. The device locks during downtime; a parent can add more time with the PIN.
To remove the red badge from an app's icon: Settings, Notifications, choose the app, then turn off Badges (iOS 27).
To remove notification dots from app icons: Settings, Notifications, then turn off "Notification dot on app icon". Each app also has its own "Allow notification dot" switch.
On a child's iPhone (iOS 27 or later): Settings, Family, your child's name. Time Allowances limit how long apps can be used; Screen Time Schedules choose which apps work before, during and after school.
Time Allowances set limits by app category such as games and social media; schedules can differ on weekdays and weekends.
Bedtime mode can switch the screen to grayscale at night (Settings, Digital Wellbeing, Bedtime mode). App timers close an app when its daily limit is up; Focus mode pauses distracting apps.
Six steps: start a conversation about screen time, set up a Child Account, set safety settings for websites and apps, make a schedule for device use, manage your child's contacts, and keep the conversation going as your child grows. "Parental controls often work best when children understand why they're in place."
News report on the guide's release on 29 September 2026.
Laws around the world
Would bar anyone under 16 from using a social media platform, with no parental-permission exception. Platforms must verify age before accepting an account holder and delete the verification data immediately afterwards; a verified parent can ask for a known minor's account to be deleted, and the platform must do so within 10 days. Violations count as deceptive trade practices. Takes effect 1 September 2027; applies to access from 1 January 2028.
Rep. Jared Patterson filed HB 212 for the 2027 session, which begins in January. His 2025 bill (HB 186, under-18s) passed the Texas House 116–25 and cleared a Senate committee but never got a full Senate vote before a deadline.
After the Education Ministers Meeting, ministers agreed to introduce, trial or consider limits on screen use in schools, with each state, territory and sector deciding how. Student use of generative AI will be restricted in primary schools and in secondary assessments, Year 5 NAPLAN writing returns to paper from 2028, and an Australian Schools Digital Detox Taskforce led by an independent expert will advise on next steps.
AAP report on the ministers' decisions: generative AI scaled back in primary schools and high school assessments, Year 5 NAPLAN back to pen and paper, and a taskforce expected to report three times in 2027. Education Minister Jason Clare: “There's a reason we don't give kids a calculator in kindergarten.”
A survey of 9,750 US adults (26 May–1 June 2026): 56% support a ban that prevents anyone under 16 from using social media sites, 21% oppose and 23% are not sure. Parents of children under 18: 65% support. Support for requiring social media companies to limit how much time minors spend on their sites rose from 69% (2023) to 78%; opposition to each of the three policies asked about was no more than 10%.
Seven High Court judges heard two constitutional challenges to Australia's under-16 social media ban over two days from 3 November 2026. The Digital Freedom Project argues the law strips young Australians of the implied freedom of political communication; Reddit argues it raises privacy and political expression issues and should not be grouped with the other platforms. Communications Minister Anika Wells said the laws are under “sustained ideological attack”. Platforms face fines of up to A$49.5 million if they fail to take reasonable steps to remove under-16 accounts.
Australia's education ministers agreed to “set or consider screen time limits in schools across the country”, to restrict generative AI in primary schools and in secondary school assessments, and to return Year 5 NAPLAN writing tests to pen and paper. A new Australian Schools Digital Detox Taskforce will survey students, teachers and parents and advise ministers next year. The release says: “The evidence is showing us that the more time kids spend online the more likely it is that their results decline.”
ABC News on the Education Ministers Meeting: an Australian Schools Digital Detox Taskforce, Year 5 NAPLAN writing back on paper, and a survey of parents, students and teachers before further reforms. Details of any limits are not yet set.
UK House of Commons Library briefing. Australia's ban on social media for under-16s has been in force since 10 December 2025. The UK Government announced a ban for under-16s on 15 June 2026; it aims to lay regulations before Parliament by the end of 2026, and the ban is expected to take effect in spring 2027.
Communications Minister Anika Wells told Bloomberg she expects the eSafety Commissioner to take legal action against Facebook, Instagram, TikTok, Snapchat and YouTube over the under-16 social media ban. No case has been filed yet; the regulator is using new evidence-gathering powers. Meta says it removed more than 750,000 Australian accounts and argues for a single age check at the operating-system or app-store level.
eSafety found many children kept existing accounts, opened new ones or passed age checks, and some platforms allowed unlimited attempts at age checks. About 5 million accounts had been deactivated since the ban began on 10 December 2025. Snap said it had locked 450,000 accounts; Meta said it is committed to complying.
A three-judge panel of the US 11th Circuit Court of Appeals put on hold a 2025 injunction against Florida's 2024 law, so the state can enforce it while the case continues. The law bans social media accounts for children under 14 and requires parental consent for 14- and 15-year-olds, on platforms that use algorithms and certain addictive features. Industry groups NetChoice and the Computer & Communications Industry Association, which sued on free-speech grounds, say they are weighing their options.
Rules published in Turkey's Official Gazette on 10 October 2026 bar children under 15 from social media from 1 November, with age verification required. For 15 to 17-year-olds, accounts must be private, personalised advertising is not allowed, livestreaming is banned and infinite-scrolling feeds must be switched off. Platforms have until 31 March 2027 to verify existing accounts.
Ages are checked through the government e-Devlet portal with a token that only confirms whether a user is over 15 or 18. Teen accounts (15 to 17) are private by default, may not use autoplay or endless-feed designs, and cannot be shown personalised ads.
Confirms the 1 November start and the April 2026 vote in parliament. Rights groups say the ban comes amid wider restrictions on free expression in Turkey.
On 24 August 2026 a Fourth Circuit panel voted 2 to 1 to keep Virginia's law blocked while the state's appeal is heard. The law (SB 854) would limit under-16s to one hour a day on social media unless a parent consents to more.
Judge Patricia Tolliver Giles blocked the law on free speech grounds, writing that it cannot infringe on First Amendment rights. It had been due to apply from 1 January 2026. Virginia's attorney general has appealed.
Proposed 17 September 2026: no social media accounts under 13, 13 to 14-year-olds only through a parent's account, and bans on infinite scroll, reward tricks and night-time push notifications. Fines up to 6% of global revenue. Still a proposal.
The Constitutional Council struck down the ban on 14 August 2026, ruling it disproportionately infringed freedom of expression and communication.
Parliament passed the under-15 ban in July 2026, pending Constitutional Council review.
The UK plans a ban on social media for under-16s from spring 2027, using powers from the Children's Wellbeing and Schools Act 2026. Messaging apps such as WhatsApp are exempt.
Three months in, 81% of under-16s still had access to at least one restricted platform; about half kept accounts because their age was never checked. By 16 January about 4.7 million accounts belonging to Australian children had been revoked. Denmark, Malaysia, Greece and the UK are moving on similar bans.
Families describe workarounds, including apps re-downloaded through iCloud without permission.