Testing the resilience of fact checks against rhetorical challenge

Fact-checks correcting political misperceptions promise to set the record straight—but anyone who has glanced at the comments section underneath a fact-check knows they rarely have the last word. Corrections are quickly met with suspicion: accusations of bias, appeals to personal experience, or claims that instincts matter more than statistics. So, how resilient are fact-checks against this rhetorical challenge? We tested this question in a preregistered survey experiment with a representative sample of British adults. Participants first rated common false claims about immigration. They then saw an expert fact-check of whichever claim they believed most. Next, treated respondents read a second message urging them to take the statistics “with a big pinch of salt.” In a 2×3 design, we varied both the source (professor vs. blogger) and the content of the challenge (questioning the expert’s neutrality, appealing to personal experience, or encouraging respondents to “trust their instincts”). The fact-checks worked: they significantly reduced belief in false claims, even among skeptics. Crucially, most of this corrective effect endured: the undermining comments only cancelled out about a quarter of the effect of the fact-check on belief in false claims. Our findings support prior research on the corrective power of fact-checks and add evidence of their resilience. Fact-checks may not flip beliefs overnight, but they loosen conviction in false claims—even when they do not have the last word.