What is the backfire effect?
The backfire effect refers to the idea that a factual correction would strengthen the false belief it targets instead of weakening it. The phenomenon was described in 2010 by Brendan Nyhan and Jason Reifler on a few American political topics, and outside academia it enjoyed a success far broader than the data supported. Replication work has largely disconfirmed it: Thomas Wood and Ethan Porter, across more than ten thousand participants and fifty-two subjects, found it nowhere, and corrections moved beliefs in the expected direction. Nyhan himself now considers the effect rare and unstable. The nuance matters for our subject: if correcting false information does not strengthen the belief, neither is it enough to change anyone’s mind. What limits fact-checking is not a backfire, it is the cost of refutation, the identity attachment to the belief, and the fact that a correction places the other person at fault. It is important to note that the backfire effect remains true when the factual correction is accompanied by ridicule, judgement and condescension. The way you "correct" someone can indeed produce the opposite of the desired effect. See also Brandolini’s law, identity-protective cognition and "Does fact-checking work in a debate?".