You have probably heard this one. Show a stubborn person the facts, and they dig in harder. Correct a lie, and people believe it even more. It even has a name: the backfire effect. * Here is the short answer. The backfire effect is real, but it is rare. Big studies with more than 10,000 people found that when you correct a false claim, most people move toward the truth — even when the truth hurts their own side. The real problem is different. The correction fades, and the lie keeps getting repeated. * ## Where the idea came from * In 2010, two political scientists, Brendan Nyhan and Jason Reifler, published a paper called "When Corrections Fail" in the journal Political Behavior. They showed people made-up news articles. Some articles had a false claim from a politician. Others had the same claim plus a correction. * One test was about Iraq. The article quoted President George W. Bush hinting that Iraq could pass weapons to terrorists. Then it added the facts from a CIA-backed report: Iraq did not have banned weapons when the US invaded in 2003. * For conservative readers, the correction did the opposite of its job. Those who saw it became more likely to believe Iraq had the weapons, not less. * That result spread fast. Journalists, bloggers and cartoonists loved it. It seemed to explain everything — why your uncle won't change his mind, why facts "don't matter" any more. * ## Then other researchers tried to repeat it * In science, one study is a start, not an answer. So other researchers ran the test again, much bigger. * Thomas Wood and Ethan Porter ran five experiments with more than 10,100 people. They tested 52 different false claims, many on hot-button political issues where you would expect people to dig in. Their paper, "The Elusive Backfire Effect", came out in Political Behavior in 2019. * They did not find a single correction that caused a backfire. Their conclusion: most people accept facts, even when the facts go against their politics. * Wood and Porter later teamed up with Nyhan himself to test corrections to Donald Trump's claims about crime. People who saw the correction got more accurate. That held for Trump supporters too, not only for his opponents. * ## The man who made it famous says it was overblown * In 2021, Nyhan wrote a paper for the science journal PNAS with a very plain title: "Why the backfire effect does not explain the durability of political misperceptions." In everyday words: backfire is not the reason false beliefs stick around. * He pointed out that his own 2010 paper found backfire in only two of its five experiments. The finding, he wrote, was often misread as "corrections always backfire." They don't. * The Debunking Handbook 2020 says the same. It was written by 22 researchers, led by psychologists Stephan Lewandowsky and John Cook. Backfire can happen in some conditions, it says, but it is unusual. And repeating a myth in order to knock it down is safe in most cases. * ## So why do lies stick? * If corrections work, why do so many people still believe false things? Nyhan's answer comes in three parts. * First, the correction wears off. People get a bit more accurate right after a fact-check. But that gain often fades over time. * Second, the lie keeps coming back. Politicians and TV hosts on your side repeat it again and again. One fact-check against a hundred repeats is not a fair fight. * Third, the fact-check never reaches the right people. Nyhan cites research on the 2016 US election. Of the people who read a story from an untrustworthy website that had been fact-checked, only 3% also read the fact-check. * ## What this means when you read the news * The backfire myth is a little too convenient for newsrooms. If facts "don't work", there is less reason to push back hard on a false claim. It becomes easier to print both sides and move on. * The research says the opposite. Clear corrections do work. They just need to be repeated, and put where people will actually see them — not buried in paragraph twelve. * The Debunking Handbook suggests a simple shape, often called a "truth sandwich". Start with the fact. Mention the false claim once and say why it is wrong. Then end with the fact again. Studies found this works better than a story that ends on the myth. * Watch for the opposite in headlines. "Politician repeats claim that X" puts the lie in the biggest type on the page. The correction, if there is one, sits lower down, where fewer people read. * And if you are arguing with a relative at dinner? The science is more hopeful than the meme. People do change their minds about facts. Just don't expect one conversation to beat a year of cable news. * ## Related guides * Want to go further? Read our guides on what fact-check ratings actually mean, the difference between a correction, a clarification and a retraction, and how one unchecked claim becomes "widely reported" through circular reporting.