Prebunking

Working Definition

Prebunking is the preemptive building of psychological resistance to manipulation: instead of correcting a falsehood after it has already gone viral (debunking), an intervention exposes people in advance to a weakened, controlled dose of a manipulation technique—emotional fearmongering, false dichotomies, scapegoating, incoherence, or impersonating experts—alongside an explicit demonstration of how the trick works. In technical social psychology, it is the contemporary application of William J. McGuire's 1960s inoculation theory: a forewarning of an impending persuasive attack combined with preemptive refutation generates cognitive “antibodies.” In popular and policy usage, the term has expanded to cover almost any advance warning about false claims, including intelligence disclosures, though its core methodological insight is that one inoculates against recurring techniques of manipulation rather than against every individual factual lie.

Genesis and Global Lineage

Inoculation theory originated in American social psychology at the height of the Cold War. Prompted by the question of why seemingly self-evident “cultural truisms” (such as “you should brush your teeth after every meal”) crumbled so easily when challenged, William J. McGuire proposed a biological analogy in 1961: a belief that has never been attacked is like a pathogen-free organism with no immune defenses, whereas exposure to a weakened counterargument stimulates active cognitive defense. For half a century the theory lived quietly in health-campaign, advertising, and courtroom research (systematized by scholars such as Josh Compton). Its pivot to digital disinformation began after 2016: Sander van der Linden (trained in Amsterdam, PhD at the London School of Economics in 2014) and Jon Roozenbeek at the University of Cambridge developed the browser game Bad News (2018), in which players step into the role of a fake-news tycoon and learn six recurring manipulation tropes by deploying them. The word prebunking—coined as the mirror twin of debunking—entered the title of peer-reviewed papers in 2020. In August 2022, the Cambridge and Bristol team, partnering with Google's Jigsaw unit, demonstrated in Science Advances that 90-second animated prebunking videos run as YouTube ads measurably improved users' ability to spot manipulation at platform scale. Van der Linden's book Foolproof (2023) brought the concept into mainstream policy vocabulary across the WHO, the European Commission, and national governments—while strategic communications simultaneously borrowed the word to describe Western intelligence agencies' preemptive declassification of Russian false-flag plots in early 2022.

Central European Lineage

Central Europe did not originate the theory of prebunking, but it served as the primary empirical proving ground for its transition from laboratory experiment to mass deployment. First, the February 2020 Harvard Kennedy School Misinformation Review paper by Roozenbeek, van der Linden, and Thomas Nygren that brought the term prebunking into a scholarly title did not rely on an Anglo-American sample: it tested localized adaptations of the Bad News game in German, Greek, and Polish (following a Swedish pilot). The Polish sample of 1,332 self-selected players (d = 0.33 reduction in perceived reliability of manipulative posts) was one of the three core datasets on which the authors grounded their claim that psychological inoculation works “across cultures.” Second, and more consequentially, on August 24, 2022—simultaneously with the publication of the landmark Science Advances study—the University of Cambridge and Google's Jigsaw unit announced the world's first population-scale prebunking video campaign across social platforms in Poland, Czechia, and Slovakia. Designed in partnership with local NGOs and fact-checkers, the campaign targeted emerging anti-refugee manipulation tropes in the three countries that had absorbed millions of Ukrainian refugees after Russia's full-scale invasion in February 2022. The region was chosen precisely because Kremlin-aligned information operations were actively trying to weaponize economic anxiety against Ukrainian refugees, turning Central Europe into the real-world testbed for scalable cognitive inoculation.

Why Now

After 2016, newsrooms and researchers realized that traditional post-hoc fact-checking suffers from a structural timing defect: corrections travel slower and narrower than sensational falsehoods, arrive after memory traces have already consolidated (the “continued influence effect”), and frequently trigger partisan backlash. The COVID-19 pandemic and the WHO's declaration of an “infodemic” in 2020 gave inoculation theory both urgent public funding and a resonant metaphor, spawning the Go Viral! game in partnership with the UK government and the UN. Then 2022 delivered a double watershed: Western intelligence agencies publicly “prebunked” staged Russian pretexts days before the invasion of Ukraine, and the Cambridge–Jigsaw partnership proved that short video inoculations could be delivered through standard ad-buying infrastructure to millions of YouTube viewers. Since 2023, the European Commission has integrated prebunking into Digital Services Act compliance discussions, while the rise of generative AI—which reduces the cost of fabricating bespoke falsehoods near zero—has made item-by-item fact-checking mathematically hopeless, shifting policy hope toward building upstream resilience in audiences.

Explanatory Power

The concept clarifies that the struggle over public deception has three temporal moments—before, during, and after exposure—and that almost all journalistic and regulatory infrastructure had been crowded into the third, when the damage is already done. Second, technique-based prebunking shifts the terrain of dispute from contested facts to formal rhetorical mechanics. One does not need to adjudicate complex macroeconomic statistics about migration to show an audience how a headline uses a false dichotomy or emotional scapegoating—which allows an intervention to bypass partisan tripwires and work across ideological divides. Third, the immunological metaphor forces researchers to ask precise questions about dosage, decay rates, and “booster shots”: longitudinal studies by Rakoen Maertens and colleagues have shown that the protective effect of a single inoculation session fades over several weeks unless reinforced, turning resilience from a one-off literacy certificate into an ongoing maintenance problem. Concrete applications now span pre-election video ads during the 2024 European Parliament elections, fraud-prevention warnings by banks against AI voice-cloning scams, and classroom exercises where students build mock propaganda campaigns to recognize their joints.

Axes of Debate

Underappreciated Angle

The biological vaccine metaphor conceals an evolutionary asymmetry: measles does not rewrite its own genome in response to a public-health poster, whereas human propagandists read the prebunking literature and adapt their styles overnight. Once audiences are trained to spot overt emotional hysteria or crude false dilemmas, sophisticated operators pivot to calm, “just-asking-questions” pseudo-academic framing or weaponized mal-information where no classic trope is visible. Yet the deeper, underappreciated consequence of prebunking is political: its success has quietly provided the strongest empirical indictment of traditional fact-checking, shifting the burden of epistemic defense from institutional gatekeepers onto the cognitive immune system of the individual user. For giant platforms facing regulatory pressure under the EU's Digital Services Act, funding 90-second educational videos is vastly cheaper and less disruptive to engagement-driven business models than redesigning the recommendation algorithms that amplify outrage in the first place.