Unmasking the Backfire Effect: An Exploration of Resistance to Fact-Checking in the Digital Age

Social media has become a critical battleground in the fight against misinformation, where fact-checks play an essential role in addressing falsehoods. Despite these efforts, growing polarization has led to widespread hostility toward fact-checks, with responses often characterized by toxicity or direct confrontation. We investigate the backfire effect, where fact-checks paradoxically strengthen false beliefs, as manifested through toxic responses on social media. This study explores the drivers of toxic reactions to fact-checks on social media. Using a dataset of fact-checks spanning three months, we analyzed user comments on Twitter (now X) to understand the dynamics of these reactions. Our results reveal that toxic responses are not a universal reaction but a selective one, concentrated on specific triggers. We find that fact-checks with definitive refutations (Pants on Fire or False) attract significantly more negative responses than ambiguous ratings, with True ratings showing minimal impact. Furthermore, toxicity is heavily driven by high-salience topics and the frequency of fact-check tweets. Our findings offer actionable strategies for fact-checkers to mitigate pushback and enhance truth dissemination in a landscape fraught with skepticism and resistance.