Prebunking: inoculation theory in the classroom

Evidence grade: PROMISING-to-STRONG for adults and adolescents; adaptation-stage for under-13s. Inoculation against manipulation has decades of theory (McGuire, 1960s) and a modern wave of real trials — including large randomized studies of games and short videos. Effects are real, moderate, and they decay, which turns out to be the most instructive finding of all. Primary-age versions exist but carry far less evidence.

Debunking arrives after the false belief has moved in and unpacks slowly — the misinformation literature's "continued influence effect" documents how corrections chase but rarely catch the original claim. Prebunking inverts the timeline: expose people to a weakened dose of manipulation — the technique, named and demonstrated, before they meet it live — and they build resistance, the way vaccines work on the same metaphor McGuire chose in the 1960s.

The modern evidence

The Cambridge program (van der Linden, Roozenbeek and colleagues) turned the theory into trials: Bad News (2019) — a game where players run a misinformation operation using the standard techniques (impersonation, emotional amplification, conspiracy framing) — measurably reduced susceptibility to those techniques afterward; short inoculation videos tested at YouTube scale (Roozenbeek et al., 2022, Science Advances) improved technique recognition in the wild. Two design features matter for educators:

Technique beats topic. Inoculating against a claim protects against that claim; inoculating against a technique (fake urgency, fake experts, us-vs-them framing) generalizes across every claim built on it. Teach the trick, not the instance — the same logic as phishing pattern-training, which is prebunking's older sibling in the safety curriculum.
The active version is stronger. Performing the manipulation in a sandbox (running the fake account, crafting the clickbait) produces more durable resistance than watching a warning — generation beats reception, as the learning science predicts.

The decay problem — and the familiar answer

The finding that should shape implementation: inoculation fades — follow-ups show effects decaying over weeks-to-months without reinforcement, and booster exposures restoring them (Maertens et al.). To a reader of this library the shape is instantly recognizable: it's the forgetting curve, and the answer is the same — spaced, retrieved reinforcement, not a one-off assembly. A single media-literacy day is a vaccine with no booster schedule; the trials say to expect exactly what you'd expect.

Scaling down to primary

Junior adaptations of the games exist and pilot promisingly, but the under-13 evidence is thin (the honest map) — so primary use is principled adaptation: fewer techniques, concrete ones first (fake urgency, too-good-to-be-true, pretending-to-be-someone — the ones that overlap with scams and dark patterns, where children already meet them), always in the empowering register — you're learning the magician's tricks — and always with the calibration guard: recognizing manipulation must not curdle into disbelieving everything, a measured failure mode of overdone skepticism training.

What the evidence doesn't say

Effect sizes are moderate — inoculation shifts recognition rates; it doesn't immunize. Layer it with lateral reading; they answer different questions (is this manipulative? / is this true?).
Durability without boosters is poor — decay is the design constraint, not a footnote.
Primary-age trials are the gap — everything below age ~12 here is adaptation, graded as such.

In the classroom

Name the techniques like specimens: fake urgency, fake experts, too-good-to-be-true — a small bestiary children learn to spot and call out by name.
Let them build one (in a sandbox): writing their own over-the-top clickbait headline for a fictional product teaches the anatomy from inside — the active-inoculation result.
Schedule boosters — revisit the techniques across the year in new guises; the decay data makes this non-optional.
Pair every "spot the trick" with a "how truth gets checked" — calibration, always.

How Wiz Kids applies this

Prebunking is threaded through the simulated apps rather than delivered as a unit: phishing drills in Owl Post, consent dark patterns as playable puzzles in the browser, too-good-to-be-true offers in the Oracle's web — each technique met as a weakened dose in a safe world, named on the spot, and returning later in spaced reviews because the decay literature says once is not enough.

References

McGuire, W. J. (1964). Inducing resistance to persuasion. Advances in Experimental Social Psychology, 1.
Roozenbeek, J., & van der Linden, S. (2019). Fake news game confers psychological resistance against online misinformation. Palgrave Communications, 5.
Roozenbeek, J., van der Linden, S., et al. (2022). Psychological inoculation improves resilience against misinformation on social media. Science Advances, 8(34).
Maertens, R., Roozenbeek, J., et al. (2021). Long-term effectiveness of inoculation against misinformation. Journal of Experimental Psychology: Applied, 27(1) — the decay and booster data.
Lewandowsky, S., et al. (2012). Misinformation and its correction. Psychological Science in the Public Interest, 13(3) — the continued-influence case for acting before exposure.

© Glu IO Pty. Ltd. — Wiz Kids (wiz.kids). Link freely; republication requires permission — see terms. Found an error in our reading of the research? We correct fast: tell any teacher piloting Wiz Kids.