Separate Effects of Correction Formats and News Source Trustworthiness on Beliefs in False Headlines in a U.S. Sample of Democrats and Republicans
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Restating false information before correcting it reduces beliefs more than correcting it with only true information. However, correction effectiveness may also depend on the trustworthiness of sources, especially in the polarized political environment of the United States. Across two experiments, we examined the roles of correction format and source credibility on beliefs in false news headlines. Samples of Republicans and Democrats tested online read both true and false social media headlines as well as corrections of false headlines attributed to media sources associated with different political and ideological leanings (e.g., Fox News, CNN). The first experiment found that reminder-based corrections were more effective than corrections without reminders at reducing false beliefs across partisan affiliation and news sources, so the second experiment only used a reminder-based correction format. In both experiments, Democrats initially believed false headlines more when it was attributed to an ideologically congruent (liberal) source than an ideologically incongruent (conservative) source, but Republicans did not show this difference. Corrections reduced false beliefs to the same levels within party regardless of sources. For Democrats, the magnitude of belief change was driven by the initial headline source and not the correction source. Republicans’ magnitude of belief change, in contrast, was relatively consistent regardless of the sources. Recognized corrections were consistently associated with reduced false beliefs. Collectively, these findings show how the memorability of corrections of false headlines is a key determinant of their effectiveness in improving belief accuracy, across the dimensions of partisanship and news source trustworthiness.