Prebunking

From Liberpedia

Prebunking prepares people for misleading claims or persuasion techniques before they encounter them. Psychological inoculation combines a warning with examples and explanations of how the attempted manipulation works.[1]

Recognising techniques

Training can address recurring techniques such as scapegoating, personal attacks and false dichotomies. It aims to help people recognise manipulation in unfamiliar material.[1] For example, a claim that people must accept a particular policy or tolerate a problem can conceal other possible responses.

Recognising a rhetorical technique and establishing a claim’s truth are separate tasks. Emotional or polarised language can accompany accurate information. An intervention that makes people reject everything can reduce belief in false information while also reducing acceptance of true information.[2]

Experimental evidence

Roozenbeek and colleagues tested short videos in six randomised studies involving 6,464 participants and a YouTube field study with 22,632 responses. The videos improved recognition of the techniques taught. The pooled field effect was small: Cohen’s h was 0.09, slightly below the researchers’ predeclared threshold of 0.10. The field test used one question per respondent within 24 hours and did not measure actual sharing behaviour. The studies did not establish long-term persistence, transfer to other cultures, or protection against techniques that had not been taught.[1]

The Bad News game lets players practise techniques used to spread misinformation. In a separate study, Seabrooke and colleagues tested 150 English-speaking participants of Indian nationality before and after the game. They counterbalanced which headlines appeared in each test, reducing the risk that an easier second set would resemble a training benefit. They found no significant overall change in discrimination or response bias. The study used few headlines and lacked an inactive control group; it was an extension of an earlier experiment rather than an exact replication.[2]

In a 2026 meta-analysis, Almog Simchon and colleagues reported improved discrimination without a systematic increase in general scepticism across 33 experiments involving 37,025 participants. They used a hierarchical Bayesian signal-detection model and included both games and videos.[3]

Accuracy at first exposure

An initial accuracy judgment can encourage the use of existing knowledge. Brashier, Eliseev and Marsh found that this reduced susceptibility to repeated falsehoods when participants knew the correct facts, including after a two-day delay. The benefit did not extend to unknown facts. This intervention asks people to assess particular claims, whereas technique-based inoculation teaches patterns of manipulation.[4]


Measuring improvement

Discrimination concerns separating true from false information; response bias concerns the tendency to accept or reject information generally. Signal-detection methods distinguish these questions. Test materials, comparison groups, effect sizes and follow-up periods matter when evaluating an intervention. A fall in ratings of false headlines alone does not show improved discrimination.[2]

See also

References

  1. ↑ 1.0 1.1 1.2 Jon Roozenbeek et al., “Psychological inoculation improves resilience against misinformation on social media”, Science Advances 8(34), August 24, 2022, eabo6254; full text.
  2. ↑ 2.0 2.1 2.2 Tina Seabrooke, Ariana Modirrousta-Galian and Philip A. Higham, “Re-examining the bad news game: No evidence of improved discrimination of Indian true and fake news headlines”, Psychonomic Bulletin & Review 33, article 13, 2026, especially pp. 2–5 and 7–9; published paper.
  3. ↑ Almog Simchon et al., “A signal detection theory meta-analysis of psychological inoculation against misinformation”, Current Opinion in Psychology 67, February 2026, 102194; abstract and publication details.
  4. ↑ Nadia M. Brashier, Emmaline Drew Eliseev and Elizabeth J. Marsh, “An initial accuracy focus prevents illusory truth”, Cognition 194, 2020, article 104054, especially pp. 2–5; published paper.