The use of confirmation and refutation frames in fact-checking war-related misinformation

We implement a survey experiment to measure the effects of confirmation versus refutation frames in fact-checking wartime corrections. Respondents were presented with semantically equivalent statements that either confirmed accurate information (e.g., “It is TRUE that President Zelensky remained in Ukraine”) or refuted its inaccurate version (e.g., “It is FALSE that President Zelensky left Ukraine”). We evaluate whether confirmation frames increase sharing behavior compared with refutation frames. We also test whether refutation frames elicit more negative sentiments—such as anger and disgust—while confirmation frames generate more positive emotional responses. The experimental design mimics a Facebook post and employs four randomized treatments that vary in framing (confirmation vs. refutation) and news source (The New York Times vs Fox News). The survey was administered to 2091 U.S. adults in May 2022.