Social Media Curfew Trial: What the UK's 1-Hour Cap Will Test

What happens to a 13-year-old’s anxiety, sleep and friendships if their social media is capped at one hour a day and switched off from 9pm to 7am? Nobody knows for sure, because nobody has properly tested it. A social media curfew trial now running in Bradford, England, is designed to find out. The IRL Trial is randomizing year groups of about 4,000 students aged 12 to 15 across ten schools, with a custom app enforcing the limits, and comparing them with students who have none. Alongside it, the UK government has run a smaller 300-home pilot of bans, caps and curfews. Here is what each test does, why a randomized design can answer questions that a decade of surveys couldn’t, and what parents can borrow from it before the results arrive in 2027.

Key Takeaways

• The IRL Trial randomizes school year groups of about 4,000 students aged 12 to 15 in ten Bradford schools to a 1-hour daily social media cap and a 9pm to 7am block (details provisional), versus no limits.
• It measures anxiety, sleep quality, time with friends and family, wellbeing, body image, social comparison, school absences and bullying, with extra attention to already-anxious teens.
• A separate UK government pilot in 300 homes compares a full ban, a 1-hour cap, a 9pm curfew and a control group over six weeks.
• Randomization breaks the link between who uses social media and why, which is the flaw in most survey research.
• The UK government announced an under-16 ban in June 2026, before either test reported. Analysis of the IRL Trial is targeted for mid-summer 2027.

The Two UK Social Media Curfew Trials, Explained

The IRL Trial in Bradford

The IRL Trial is a randomized trial of social media limits for adolescents, co-led by Professor Amy Orben of the University of Cambridge’s MRC Cognition and Brain Sciences Unit and Dr. Dan Lewer of the Bradford Institute for Health Research, working with the Born in Bradford study. It is funded by the Wellcome Trust. Cambridge describes it as the world’s first major scientific trial of reducing adolescent social media use to see whether mental health and wellbeing improve.

The design, as announced on January 20, 2026:

• Participants: about 4,000 students in years 8, 9 and 10 (ages 12 to 15) at ten Bradford secondary schools.
• Randomization: half of the school year groups are randomly assigned to restrictions; the rest have no limits and serve as the comparison group.
• Intervention: a bespoke app caps total daily social media use and blocks it at night. The planned limits are one hour a day and a 9pm to 7am block, subject to change.
• Apps covered: TikTok, Snapchat, Instagram, Facebook, X, YouTube, Reddit and LinkedIn.
• Apps excluded: messaging apps such as WhatsApp, because initial consultation found they are integral to family communication.
• Outcomes: anxiety, sleep quality, time with friends and family, wellbeing, body image, social comparison, school absences and bullying, measured by interviews, questionnaires and, with consent, app data on social media and browser use.
• Timeline: a feasibility study in spring and summer 2026; the full six-week study in autumn and winter 2026; data analysis targeted for mid-summer 2027.

Orben summarized the gap it fills: “What we currently lack is research on what interventions work.”

The government’s 300-home pilot

On March 25, 2026, the Department for Science, Innovation and Technology announced six-week pilots in 300 teenage households across all four UK nations. Families were assigned to one of four conditions:

1. Full ban: parents use parental controls to remove or disable selected social media apps.
2. One-hour cap: a one-hour daily limit on popular apps, including Instagram, TikTok and Snapchat.
3. Curfew: social media blocked from 9pm to 7am, with access before and after school.
4. Control: current access continues.

Government officials and a panel of academics were to assess the data alongside responses to the “Growing up in the online world” consultation, which closed May 26, 2026. I could not find published pilot results as of November 2026.

Why a Randomized Trial Can Settle What Surveys Can’t

Most of what we know about teens and social media comes from surveys and cohort studies. They have a built-in problem, and it’s worth spelling out.

The confounding problem. Teens who use a lot of social media differ from those who use little in many ways: sleep habits, family stress, friendships, existing anxiety. A survey that finds heavy users are more anxious can’t tell whether social media caused the anxiety, the anxiety caused the use, or a third factor caused both.

The reverse-causation problem. Several studies have found that unhappiness predicts later social media use. A 2022 Nature Communications study by Orben and colleagues found that lower life satisfaction predicted more use a year later at all ages. A 2026 Manchester study found that boys with more symptoms later gamed less. Mood shapes use, not just the other way around.

What randomization does. In a randomized trial, a coin flip, not the teen’s personality or circumstances, decides who gets the limits. On average, the two groups start out the same in every way, measured or not. So if anxiety falls in the restricted group compared with the control group, the limit is the most plausible cause.

The IRL Trial randomizes year groups rather than individuals. That’s a deliberate choice: if half the students in a class had limits and half didn’t, friends would message each other across the line, and the groups would blur. Randomizing whole year groups reduces that spillover, at the cost of needing more participants for the same statistical precision.

The closest earlier evidence is a 2020 randomized experiment by Allcott and colleagues in the American Economic Review, in which adults who deactivated Facebook for four weeks reported higher subjective wellbeing. That study was in adults, involved one platform and removed it entirely. The IRL Trial tests partial limits, across many platforms, in children.

What Each Kind of Evidence Can Tell Us

Evidence type	Example	Can it show cause?	Main weakness
Cross-sectional survey	Many national teen surveys	No	Can’t separate cause, reverse cause and confounders
Longitudinal cohort, within-person	Orben et al. 2022; #BeeWell (Cheng et al. 2026)	Partly, under strong assumptions	Unmeasured changes over time; self-reported use
Natural experiment	Australia’s under-16 ban; school phone bans	Sometimes	Enforcement gaps; other changes happening at the same time
Small home pilot	UK government 300-home pilot	Limited	Small, short, families may not stick to conditions
Large randomized trial	IRL Trial, Bradford (~4,000 students)	Yes, for the tested limits	Six weeks; one city; app workarounds; participants know their group

What the Trial Can’t Answer

Even a well-run trial has limits, and it’s worth knowing them before the headlines arrive in 2027.

• Six weeks is short. Effects on sleep may show up quickly. Effects on body image or long-term mental health may take longer, or fade.
• Participants know their condition. Teens and parents can’t be blinded to a phone limit. Expectations can shape self-reported outcomes.
• Workarounds exist. A second device, a friend’s phone or a browser could weaken the dose, though the app’s browser data should help measure that.
• Messaging is excluded. If much of teen social life runs through WhatsApp, the trial tests limits on feeds and video, not on group chats.
• One city. Bradford is diverse, but results may not generalize to every country or school system.
• A null result is not proof of harmlessness. If the trial finds no change in anxiety, it means this limit, for this period, didn’t move that outcome. It won’t settle every question about content or vulnerable teens.

Policy Isn’t Waiting

The UK government has already decided. Its “Growing up in the online world” response says it will ban social media companies from providing services to under-16s, apply default restrictions and curfews to 16- and 17-year-olds, restrict infinite scrolling for that age group, and require companies to detect and prevent VPN use by under-16s. A final response was published July 15, 2026. Those decisions came before either the pilot or the trial reported.

That isn’t necessarily wrong. Governments often act on precaution. But it means the IRL Trial will evaluate an idea the country has already moved past for under-16s, while still offering the best evidence yet on whether caps and curfews help, which matters for 16- and 17-year-olds and for families everywhere. For how a similar ban has gone in practice, see Australia’s under-16 ban, three months in.

What Parents Can Borrow Now

Try the curfew before the cap

Of the two limits being tested, a nighttime block is the simplest to enforce and the most directly tied to a measurable outcome: sleep. Many phones let you schedule downtime. A 9pm to 7am block, with the phone charging outside the bedroom, is a reasonable family experiment.

Run your own six-week test

The trial runs six weeks. You can too. Agree on a limit, note how your child sleeps and feels at the start, and check again at week six. Treat it as an experiment you’ll review together, not a punishment.

Exempt family messaging, as the trial does

The researchers excluded WhatsApp because families rely on it. If your limit blocks the channel your child uses to tell you they’ve missed the bus, rethink the design.

Pay attention if your child is already anxious

The trial gives particular attention to teens with heightened anxiety. If your child is anxious, changes to their online life deserve more care, and possibly a conversation with their doctor first, since social media may be part of how they cope.

What not to do

Don’t announce a cap as a cure. The evidence isn’t in yet, and a teen who feels punished will hunt for workarounds. For context on what the debate’s two sides would expect the trial to show, see Haidt vs. Odgers, explained fairly.

What to Watch For Over the Next 3 Months

• Week 4: Look for news from the IRL Trial’s full six-week phase, running autumn and winter 2026, and any published findings from the UK’s 300-home pilot.
• Month 2 red flags: If you’re running a home curfew, watch for a second device appearing, or a child who seems more anxious without access. Both are signals to adjust the plan together.
• Month 3 self-check: Did sleep improve under your curfew? That’s the outcome you can measure at home most easily, and the one most likely to move.

Frequently Asked Questions

What is the UK social media curfew trial?

The IRL Trial is a randomized study in Bradford, England, in which school year groups of about 4,000 students aged 12 to 15 are assigned to a one-hour daily social media cap and a 9pm to 7am block, or to no limits. It measures anxiety, sleep, bullying and other outcomes, with analysis targeted for mid-2027.

When will the IRL Trial results be published?

The full six-week study runs in autumn and winter 2026, and the team targets completing data analysis by mid-summer 2027. Results will then be submitted to academic journals, so peer-reviewed publication may come later in 2027 or after.

Why does a randomized trial matter more than surveys?

Surveys can’t tell whether social media causes problems or whether struggling teens use it more. Random assignment makes the restricted and unrestricted groups alike at the start, so differences afterward can be attributed to the limits themselves.

Does the UK trial block WhatsApp?

No. Messaging apps such as WhatsApp are excluded, because consultation found they are integral to family communication. The limits apply to platforms including TikTok, Snapchat, Instagram, Facebook, X, YouTube, Reddit and LinkedIn.

Is the UK banning social media for under-16s?

The government has said it will ban social media companies from providing services to under-16s, with default restrictions and curfews for 16- and 17-year-olds. It published its final consultation response in July 2026. Check gov.uk for implementation dates, which were not set out on the summary page.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. University of Cambridge. (2026). “Thousands of UK schoolchildren to take part in major study of social media use and teen mental health.” January 20, 2026. https://www.cam.ac.uk/stories/irl-trial-social-media-study-launch
2. UK Department for Science, Innovation and Technology. (2026). “Children and parents to pilot social media bans, time limits and curfews at home as government tests next steps to give UK kids their childhood back.” March 25, 2026. https://www.gov.uk/government/news/children-and-parents-to-pilot-social-media-bans-time-limits-and-curfews-at-home-as-government-tests-next-steps-to-give-uk-kids-their-childhood-back
3. UK Government. “Growing up in the online world.” https://www.gov.uk/growing-up-online
4. Orben, A., Przybylski, A. K., Blakemore, S.-J., & Kievit, R. A. (2022). “Windows of developmental sensitivity to social media.” Nature Communications. https://doi.org/10.1038/s41467-022-29296-3
5. Cheng, Q., Panayiotou, M., Finserås, T. R., Andersen, A. I. O., & Humphrey, N. (2026). “How do social media use, gaming frequency, and internalizing symptoms predict each other over time in early-to-middle adolescence?” Journal of Public Health. https://doi.org/10.1093/pubmed/fdaf150
6. Allcott, H., Braghieri, L., Eichmeyer, S., & Gentzkow, M. (2020). “The Welfare Effects of Social Media.” American Economic Review. https://doi.org/10.1257/aer.20190658
7. Al Jazeera. (2026). “Most Australian teens still online three months after under-16 ban.” August 1, 2026. https://www.aljazeera.com/news/2026/8/1/most-australian-teens-still-online-three-months-after-under-16-ban