Meta reaches more UK news users than all newspapers combined.

The problem

Most adults now meet news through platforms whose feeds are regulated for online harms rather than accuracy, trust in news is low, and there is no UK evidence that media-literacy teaching has lasting effects.

Who is affected, and where

UK news users, starting with younger adults who rely most on platform feeds. Territory: UK. Ofcom’s media-literacy duty is UK-wide; school curricula are devolved, so a schools route would need separate decisions in Scotland, Wales and Northern Ireland.

Policy status

Proposed by researchers or others, not government policy. Ofcom’s media-literacy duties already operate (the existing option on this page). A pre-registered UK prebunking pilot with a retention test is a research-led proposal, not government policy.

The evidence

What the research found:

After a 90-second video, participants were 1.95 times as good as a control group at identifying a false dichotomy and 2.14 times as good at spotting incoherence. On YouTube, recognition rose by 5% on average when tested about 18 hours later. A 2025 follow-up found text and video formats can remain effective for one month, game-based formats decay faster, and memory “boosters” slow the decline.

Evaluated on:

US adults: 6,464 in six controlled experiments; 22,632 YouTube users who answered a test question; 11,759 in five longitudinal experiments (2025); Online survey panels and YouTube advert slots, United States

Transfer to this case:

All the studies used US participants. The field test was run with Google’s Jigsaw unit, which co-funded and co-authored the research.

Our recommendation (separate from the research):

Run a UK pilot only with a pre-registered audience, comparison group and outcome; test recognition again after a month or more; and decide on expansion from the retention and discernment results, not from views.

Strongest counter-evidence:

A reanalysis of gamified inoculation studies argues they make people more sceptical of all headlines rather than better at telling true from false. Recognising a technique in a test is not proof of changed sharing, trust or voting.

Effects on record

• Recognition of common manipulation techniques: Direction plausible; size not established. evaluated, limited confidence; United States (evaluated); UK effect untested. On YouTube recognition rose by 5% on average about 18 hours after viewing; lab effects were larger. These are US results on a test question, not changes in sharing, trust or voting, so no UK size is given and confidence for a UK benefit is limited until a UK pilot is evaluated.
• Trust in accurate news: A mixed effect; size not established. plausible spillover, limited confidence; Not established. A reanalysis of gamified inoculation studies argues they raise scepticism towards all headlines. Any pilot should measure discernment, not just recognition.

Cost and funding

Cost completeness: Not yet costed. Funding: Not yet costed / funding not identified.

• Prebunking pilot: content, UK adaptation, distribution and independent evaluation (recurring spending; paid by regulator, central government): Not yet costed. The US$0.05 per meaningful view in the 2022 YouTube test is an advertising price in another country; it excludes content production, UK adaptation and evaluation, so it is not used as a cost. Payer depends on who commissions the pilot (Ofcom, an education department or a philanthropic funder).

Who pays:

Whoever commissions the pilot — Ofcom, an education department or a philanthropic funder — plus the time of teachers, librarians and participants.

Financing proposed:

A named media-literacy or philanthropic allocation. A funder with an interest in the result, including a platform, must not control the findings.

Gap remaining:

Unknown until a pilot is specified: content, adaptation, distribution and evaluation are all uncosted. The US$0.05 per meaningful view in the 2022 YouTube test was an advertising price, not the cost of a pilot.

What delivery needs

Skills:

Content designers and an independent evaluator.

Operational capacity:

A delivery channel (schools, libraries or advert slots) that can randomise who sees the material.

Infrastructure:

None beyond existing platforms and classrooms.

Permissions:

Research ethics approval; platform agreement for any in-feed test.

Law:

None: Ofcom’s media-literacy duty already exists.

Public consent:

Examples must not appear to take political sides; contested examples undermine trust in the whole exercise.

First step and stages

First step: Pre-register the intended audience, comparison group and outcome before producing any material. (owner: Commissioning body (for example Ofcom) with an independent evaluator)

1. Pilot with a comparison group — Delivery partner and independent evaluator (proposed stage, not an official timetable)
2. Measure immediate learning and retention after a month or more — Independent evaluator (proposed stage, not an official timetable)
3. Decide whether to expand, adapt or stop — Commissioning body (proposed stage, not an official timetable)

Depends on: An evaluator independent of the funder; A delivery channel that allows a comparison group

Trade-offs

Who gains and who loses:

The intended beneficiaries are people who rely most on feeds, mostly younger adults; people who are rarely online may be missed.

What else the money or effort could do:

Money and teaching time spent here are not available for other literacy work or for platform-side measures.

Strongest counterargument:

It puts the burden on individuals rather than on the platforms whose ranking systems decide what spreads.

Risks:

Indiscriminate distrust of accurate news, overconfidence, contested examples, and effects fading without boosters.

How we would judge it

Baseline:

No matched UK measure of recognising manipulation techniques was found for this review; Ofcom’s media-literacy surveys are the nearest baseline.

Leading indicator:

Recognition of manipulation techniques immediately after, against a comparison group.

Outcome:

Recognition and discernment (telling reliable from unreliable posts) after a month or more, and sharing behaviour where it can be measured, by age group.

Continue, adapt or stop:

Expand only if better discernment — not just general scepticism — persists at follow-up; redesign or stop if the effect disappears within a month.

Earlier option on this page

This card replaces the option “Media literacy for younger news users” (operating). Its evidence: Ofcom has statutory media literacy duties. This site has not found UK evaluations showing lasting effects of media literacy teaching on how people judge news sources. DSIT and DCMS, Online Safety Act: explainer

Sources

• University of Cambridge, Social media experiment reveals potential to “inoculate” millions of users against misinformation (University of Cambridge story on Roozenbeek et al., Science Advances 8(34), 2022), Story paragraphs on the six controlled experiments and the YouTube field study (2022-08-24; relevant passages checked)
• Nature Communications 16, Psychological booster shots targeting memory increase long-term resistance against misinformation, Abstract; Methods (tests at T0, T10 and T30) (2025-03-11; relevant passages checked)
• Journal of Experimental Psychology: General, Gamified inoculation interventions do not improve discrimination between true and fake news: Reanalyzing existing research with receiver operating characteristic analysis (2023-09; could not be accessed)

Source published 24 August 2022 (Cambridge); 11 March 2025 (Nature Communications); evidence reviewed 2026-10-04. Reconsider if: A UK evaluation of prebunking is published; Ofcom publishes an evaluation of its media-literacy work; A meta-analysis separates discernment from general scepticism.

The problem

Councils, courts and other public bodies need independent scrutiny, but news use has moved to platforms and no UK evaluation shows which support model sustains useful local reporting without government or platform influence.

Who is affected, and where

Residents of areas with little local public-interest reporting, and the local publishers who serve them. Territory: UK. Media policy is reserved, so a scheme could be UK-wide or set per nation. The Irish schemes are a comparison only; their law, market and funding source differ.

Policy status

Proposed by researchers or others, not government policy. The UK government does have a grant scheme for local news: a Local News Fund of up to £12m over two years, run by DCMS through a centrally managed competitive bidding process (announced 17 March 2026). An arm’s-length, contestable fund administered independently of government, as proposed here, is not government policy. The BBC already funds up to 165 Local Democracy Reporters employed by regional news organisations under its Charter commitment; Ireland’s media regulator announced more than €15m for several journalism schemes in January 2026.

The evidence

What the research found:

The Reuters Institute’s Ireland report records Coimisiún na Meán announcing more than €15m in January 2026 for schemes including Local Democracy and Courts Reporting. Trust in news in Ireland fell 9 points to 42% in the same report, so no trust benefit can be read from the schemes. The BBC says its scheme funds up to 165 reporters covering councils and public services.

Evaluated on:

None evaluated: descriptive accounts of the Irish schemes and of the BBC scheme; Ireland (2026); United Kingdom (BBC Local News Partnerships since 2017)

Transfer to this case:

Ireland’s scheme is state support through a media regulator in a market of 5.3 million people; no per-head figure is carried over to the UK.

Our recommendation (separate from the research):

Map coverage gaps first, then run time-limited awards with transparent eligibility and editorial-independence protections, and renew only on evidence of additional original reporting.

Strongest counter-evidence:

Public money can create real or perceived government influence, reward incumbent publishers and displace subscriptions; a count of stories is not the same as useful reporting.

Cost and funding

Cost completeness: Not yet costed. Funding: Not yet costed / funding not identified.

• Local-reporting pilot: reporter grants, administration and evaluation (recurring spending; paid by central government): Not yet costed. The Irish €15m+ (January 2026) covers several national schemes in a different market and is not scaled to the UK. Payer is a design choice (taxpayers, licence-fee payers, platforms or philanthropy); publishers also bear application and reporting costs.

Who pays:

Depends on the design: taxpayers, licence-fee payers, platforms or philanthropic funders; publishers bear application and reporting costs.

Financing proposed:

None specified. Any public, industry or philanthropic source must be named with its independence safeguards.

Gap remaining:

Unknown: a pilot needs a reporting-output, administration and evaluation budget. The Irish total is not a UK cost and is not scaled per head.

What delivery needs

Skills:

Local journalists, and an administrator able to judge public-interest reporting.

Operational capacity:

Publishers in under-covered areas able to host reporters.

Infrastructure:

Content-sharing arrangements such as the BBC Local News Partnerships.

Permissions:

None.

Law:

None for a pilot; a statutory scheme would need legislation.

Public consent:

Cross-party confidence that the administrator is independent of government.

First step and stages

First step: Map genuine coverage gaps and consult a range of providers before designing eligibility. (owner: DCMS or an independent body, with publishers)

1. Map coverage gaps and consult providers — Independent researcher with DCMS (proposed stage, not an official timetable)
2. Time-limited, contestable awards — Independent administrator (proposed stage, not an official timetable)
3. Evaluate and decide renewal — Administrator and funder (proposed stage, not an official timetable)

Depends on: A credible measure of local coverage gaps; An administrator independent of government

Trade-offs

Who gains and who loses:

Residents of under-covered areas gain; publishers that receive awards gain relative to competitors that do not.

What else the money or effort could do:

Public or philanthropic money used here is not available for other public-interest uses; licence-fee money already funds the BBC scheme.

Strongest counterargument:

State-linked funding may compromise the independence it is meant to protect, and markets or philanthropy may fill the gaps without it.

Risks:

Government influence, capture by incumbents, displacement of subscriptions, and counting stories rather than useful reporting.

How we would judge it

Baseline:

No UK measure of local public-interest coverage was found for this review.

Leading indicator:

Additional original reports on councils and courts in funded areas, against similar unfunded areas.

Outcome:

Public access to and use of local reporting; editorial independence, with no funder interference reported.

Continue, adapt or stop:

Renew only where funded areas show additional original reporting compared with similar unfunded areas and no breach of independence.

Sources

• Reuters Institute, University of Oxford / Jane Suiter, Digital News Report 2026: Ireland, Country page, paragraph on Coimisiún na Meán funding (January 2026) (2026-06-16; full text checked)
• BBC, Local Democracy Reporting Service (BBC Local News Partnerships), Service description: up to 165 reporters, funded by the BBC, employed by regional news organisations (date not stated; full text checked)
• Reuters Institute, University of Oxford / Jane Suiter, Digital News Report 2026: Ireland, Opening paragraphs: overall trust in news fell to 42%, down 9 points (2026-06-16; full text checked)

Source published 16 June 2026 (Reuters Institute); evidence reviewed 2026-10-04. Reconsider if: An evaluation of the Irish schemes is published; A decision on the BBC’s local news partnerships in Charter review; A UK government local-media strategy.