Across five survey experiments covering 52 ideologically charged political misstatements and more than 10,100 respondents, Wood and Porter (2019) found no 'backfire effect' — no case in which a factual correction moved people further into a misperception — and concluded that citizens largely heed corrective information even when it challenges their ideological commitments. Yet the authors themselves cautioned that a complete rejection of the backfire hypothesis was not warranted: backfire might still emerge under heightened partisan or ideological salience (elections, crises, emotionally intense environments); only immediate belief acceptance was measured, leaving persistence over time and downstream effects on attitudes and behavior unknown; the samples excluded the undergraduates on whom the original backfire evidence rested, so subpopulations higher in need for cognition might still backfire; and aggregate nulls cannot rule out counterarguing by some individuals. The open problem is to establish the boundary conditions of the backfire effect and the individual-level dynamics of factual updating: under what conditions, in which subpopulations, and through what cognitive processes — if any — do corrections strengthen rather than weaken false beliefs, and how do corrected beliefs evolve over time after a correction is received.

Background

A political misperception is a false or unsupported factual belief about politics — for instance, the belief that US troops found weapons of mass destruction (WMD) in Iraq after the 2003 invasion. The remedy most studied in political communication is the corrective message or 'fact-check': a news-style statement that a claim is false, accompanied by accurate information. The foundational result was Nyhan and Reifler's (2010) finding that corrections can fail and even 'backfire': in two experiments with undergraduate samples, exposure to a correction increased belief in the targeted misperception among the ideologically threatened group. This so-called backfire effect (more precisely, a worldview backfire effect, since the correction allegedly clashes with the recipient's ideological worldview) resonated with the 'motivated skepticism' account of Taber and Lodge (2006): when people encounter counter-attitudinal arguments, they tend to counterargue them — generating defensive thoughts that refute or discount the evidence — more vigorously than congenial arguments, which can push attitudes further apart rather than together. A related construct is the 'belief echo' (Thorson 2016): even when a correction updates factual beliefs, exposure to the misinformation can leave a lingering negative impression on attitudes.

Wood and Porter (2019) mounted the most extensive challenge to the generality of the backfire effect. Across five survey experiments enrolling more than 10,100 subjects — recruited from Amazon Mechanical Turk, an online crowdworking platform, and from Lucid, an opt-in survey panel matched to US demographic margins — they tested 52 real political misstatements spanning both ideological sides (gun violence, immigration, abortion, taxes, and a direct replication of the WMD correction from Nyhan and Reifler 2010). No correction triggered backfire in the average respondent or in any ideological subgroup; the modal outcome was significant factual updating. Two auxiliary analyses sharpened the null: a re-run of the original WMD experiment suggested that the original backfire was partly attributable to the complexity of the original survey question (a simpler item produced no backfire); and the semantic proximity between each misstatement and its correction — rated on a 100-point scale by 261 independent raters — showed no relationship to the size of correction effects, contradicting both leading counterarguing variants (which predict backfire from, respectively, weak and tangential corrections or strong and directly contradictory ones).

The authors nevertheless declined to pronounce the backfire hypothesis dead. Their stated caveats define the open question: (1) the experiments measured beliefs immediately after treatment, in low-stakes survey settings that cannot reproduce the emotional intensity of elections, crises, or other moments of heightened ideological salience that might prime counterarguing; (2) nothing was learned about whether corrected beliefs persist over time or whether corrections spill over into related attitudes and behaviors; (3) the samples, though broader than prior work, did not include the college undergraduates on whom the original backfire evidence was largely based — and if backfire is a cognitively effortful process, samples high in need for cognition (a personality trait reflecting the enjoyment of effortful, deliberative thinking) may be precisely the ones to exhibit it; (4) the null proximity-effect relationship came from aggregate ratings and cannot exclude that some individuals counterargue even when the average respondent does not. Subsequent research has converged on the view that backfire, if real, is rare and confined to limited circumstances, and has proposed that many historical backfire observations are artifacts of low test-retest reliability (the stability of a measure when the same respondent answers the same item twice without any treatment) or of regression to the mean; but the delimited conditions under which genuine backfire might still occur — and the individual-level cognitive processes that govern who updates and who resists — remain unsettled and actively contested.

Problem Statement

Wood and Porter's (2019) null replication left the boundary of the backfire effect open: their five experiments found no correction capable of triggering backfire across 52 ideologically charged issues, yet a complete rejection of the backfire hypothesis was explicitly not warranted. The problem is to determine the boundary conditions of the backfire effect and the individual-level dynamics of belief updating after corrections. Concretely: (1) Does backfire — an increase in belief in a false claim after exposure to a corrective message — emerge in contexts of heightened ideological or partisan salience, such as real election campaigns, crises, or emotionally intense political environments, which the low-stakes survey experiments of Wood and Porter could not reproduce? (2) Do the factual-belief changes produced by corrections persist over time, and do corrections have downstream effects on related attitudes and behaviors? (3) Does genuine backfire occur in identifiable subpopulations, for example respondents high in need for cognition — such as the undergraduate samples on which the original backfire evidence was largely based — even though the average survey respondent updates factually? (4) Do some individuals counterargue corrections even when the aggregate response is factual updating, and what individual-level cognitive processes govern who updates and who resists? The problem is to identify the conditions, populations, and mechanisms (if any) under which corrections strengthen rather than weaken misperceptions, to distinguish genuine backfire from measurement artifacts such as regression to the mean and low test-retest reliability, and to characterize how corrected beliefs evolve after the correction is received. The evidence bearing on these questions must be adequately powered and ecologically valid enough to reach the conditions and populations at issue: the question is whether genuine backfire occurs in a delimited set of conditions or populations, or whether its absence is well supported within the specifiable populations and contexts that such evidence can reach. The evidence currently available may not extend to all of the conditions and populations at issue.

References

  1. Wood, Thomas, and Ethan Porter (2019). 'The Elusive Backfire Effect: Mass Attitudes’ Steadfast Factual Adherence.' Political Behavior 41(1): 135-163 (online 2018). DOI: 10.1007/s11109-018-9443-y. https://doi.org/10.1007/s11109-018-9443-y
  2. Nyhan, Brendan, and Jason Reifler (2010). 'When Corrections Fail: The Persistence of Political Misperceptions.' Political Behavior 32(2): 303-330. DOI: 10.1007/s11109-010-9112-2. https://doi.org/10.1007/s11109-010-9112-2
  3. Taber, Charles S., and Milton Lodge (2006). 'Motivated Skepticism in the Evaluation of Political Beliefs.' American Journal of Political Science 50(3): 755-769. DOI: 10.1111/j.1540-5907.2006.00214.x. https://doi.org/10.1111/j.1540-5907.2006.00214.x
  4. Thorson, Emily (2016). 'Belief Echoes: The Persistent Effects of Corrected Misinformation.' Political Communication 33(3): 460-480 (online 2015). DOI: 10.1080/10584609.2015.1102187. https://doi.org/10.1080/10584609.2015.1102187
  5. Nyhan, Brendan, Ethan Porter, Jason Reifler, and Thomas J. Wood (2020). 'Taking Fact-Checks Literally But Not Seriously? The Effects of Journalistic Fact-Checking on Factual Beliefs and Candidate Favorability.' Political Behavior 42(3): 939-960 (online 2019). DOI: 10.1007/s11109-019-09528-x. https://doi.org/10.1007/s11109-019-09528-x
  6. Porter, Ethan, and Thomas J. Wood (2019). 'False Alarm: The Truth About Political Mistruths in the Trump Era.' Cambridge University Press. ISBN 9781108688338. DOI: 10.1017/9781108688338. https://doi.org/10.1017/9781108688338
  7. Swire-Thompson, Briony, Joseph DeGutis, and David Lazer (2020). 'Searching for the backfire effect: Measurement and design considerations.' Journal of Applied Research in Memory and Cognition 9(3): 286-299. DOI: 10.1016/j.jarmac.2020.06.006. https://doi.org/10.1016/j.jarmac.2020.06.006
  8. Ecker, Ullrich K. H., Stephan Lewandowsky, and Matthew Chadwick (2020). 'Can corrections spread misinformation to new audiences? Testing for the elusive familiarity backfire effect.' Cognitive Research: Principles and Implications 5: 41. DOI: 10.1186/s41235-020-00241-6. https://doi.org/10.1186/s41235-020-00241-6
  9. Autry, Kevin S., and Shea E. Duarte (2021). 'Correcting the unknown: Negated corrections may increase belief in misinformation.' Applied Cognitive Psychology 35(4): 960-975. DOI: 10.1002/acp.3823. https://doi.org/10.1002/acp.3823
  10. Nyhan, Brendan (2021). 'Why the backfire effect does not explain the durability of political misperceptions.' Proceedings of the National Academy of Sciences 118(15): e1912440117. DOI: 10.1073/pnas.1912440117. https://doi.org/10.1073/pnas.1912440117
  11. Swire-Thompson, Briony, Nicholas Miklaucic, John P. Wihbey, David Lazer, and Joseph DeGutis (2022). 'The backfire effect after correcting misinformation is strongly associated with reliability.' Journal of Experimental Psychology: General 151(7): 1655-1665. DOI: 10.1037/xge0001131. https://doi.org/10.1037/xge0001131
  12. Ecker, Ullrich K. H., Stephan Lewandowsky, John Cook, Philipp Schmid, Lisa K. Fazio, Nadia Brashier, Panayiota Kendeou, Emily K. Vraga, and Michelle A. Amazeen (2022). 'The psychological drivers of misinformation belief and its resistance to correction.' Nature Reviews Psychology 1(1): 13-29. DOI: 10.1038/s44159-021-00006-y. https://doi.org/10.1038/s44159-021-00006-y
  13. Chan, Man-pui Sally, and Dolores Albarracín (2023). 'A meta-analysis of correction effects in science-relevant misinformation.' Nature Human Behaviour 7(9): 1514-1525. DOI: 10.1038/s41562-023-01623-8. https://doi.org/10.1038/s41562-023-01623-8 (figures cited are those of the dataset corrected by the 2025 Author Correction, DOI: 10.1038/s41562-025-02294-3)
  14. Porter, Ethan, Yamil Velez, and Thomas J. Wood (2023). 'Correcting COVID-19 vaccine misinformation in 10 countries.' Royal Society Open Science 10: 221097. DOI: 10.1098/rsos.221097. https://doi.org/10.1098/rsos.221097
  15. Porter, Ethan, and Thomas J. Wood (2023). 'Factual Corrections: Concerns and Current Evidence.' PsyArXiv preprint. DOI: 10.31234/osf.io/svbru. https://doi.org/10.31234/osf.io/svbru
  16. Prike, Toby, Phoebe Blackley, Briony Swire-Thompson, and Ullrich K. H. Ecker (2023). 'Examining the replicability of backfire effects after standalone corrections.' PsyArXiv preprint. DOI: 10.31234/osf.io/u7s2b. https://doi.org/10.31234/osf.io/u7s2b
  17. Carey, John M., Elizabeth Chun, Alice Cook, Brian J. Fogarty, Leyla Jacoby, Brendan Nyhan, Jason Reifler, and Lilian Sweeney (2025). 'The Narrow Reach of Targeted Corrections: No Impact on Broader Beliefs About Election Integrity.' Political Behavior 47(2): 737-750 (online 2024). DOI: 10.1007/s11109-024-09968-0. https://doi.org/10.1007/s11109-024-09968-0
  18. Chae, Je Hoon, Tim Groeling, and Hyunjin Song (2026). 'Time is the fire in which message effects burn: Decay and sustenance of correction effects over time.' Journal of Communication 76(1): 36-48 (online 2025). DOI: 10.1093/joc/jqaf030. https://doi.org/10.1093/joc/jqaf030
  19. Carey, John M., Brian Fogarty, Marília Gehrke, Brendan Nyhan, and Jason Reifler (2025). 'Prebunking and credible source corrections increase election credibility: Evidence from the US and Brazil.' Science Advances 11(35): eadv3758. DOI: 10.1126/sciadv.adv3758. https://doi.org/10.1126/sciadv.adv3758
  20. Borah, Porismita, Pablo González-González, and Homero Gil de Zúñiga (2025). '"Stop the steal": misinformation correction and misperceptions about election fraud.' Online Information Review 49(5): 913-932. DOI: 10.1108/oir-08-2024-0512. https://doi.org/10.1108/oir-08-2024-0512
  21. Mavropoulos, Georgios (2025). 'Tracing the Backfire Effect in the Brexit Referendum.' SSRN working paper. DOI: 10.2139/ssrn.5280269. https://doi.org/10.2139/ssrn.5280269