Study: Misinformation Spreaders Reframe News From Reputable Sources to Support False Claims

A growing body of research tracking the spread of misinformation online has largely focused on presumably false stories from unreliable sources—websites that fabricate content or intentionally mislead readers. But a recent study co-authored by researchers affiliated with the University of Maryland indicates that some social media users deliberately repurpose credible journalism to advance misleading claims.

The multi-institutional team of researchers examined sharing patterns on Twitter/X, finding that accurate, mainstream stories shared by misinformation spreaders often echoed themes common in false content, including vaccine doubt, crime exaggeration and political distrust, according to the results published last month in Nature Human Behaviour.

“Our analysis suggests that users promoting misleading narratives find mainstream sources particularly attractive—especially when those sources publish information that can be spun to support their views,” said lead author Pranav Goel Ph.D. ’23, a postdoctoral research associate at Northeastern University who conducted the research while at UMD. “They are leveraging the credibility of these sources to give more weight to their own narrative.”

Goel’s co-authors include Jon Green, an assistant professor of political science at Duke University; David Lazer, a University Distinguished Professor of Political Science and Computer Sciences at Northeastern University; and Philip Resnik, an MPower Professor of linguistics at UMD with an appointment in the University of Maryland Institute for Advanced Computer Studies (UMIACS).

Resnik was Goel’s adviser when he was at UMD, where they were both active in UMIACS’ Computational Linguistics and Information Processing Lab.

The team used a dataset of tweets sent by about 1.6 million U.S. residents from May 2018 to November 2021 that shared articles from sources ranging from fake news outlets to reliable ones such as The Chicago Tribune and The Washington Post.

The researchers analyzed those tweets using specialized clustering and graph-pruning algorithms that are used to simplify, or reduce, the complexity of informational graphs constructed from large volumes of text.

The resulting data determined that during the COVID-19 pandemic, for example, headlines such as “Vaccinated people now make up a majority of COVID deaths” or “A ‘healthy’ doctor died two weeks after getting a COVID-19 vaccine” were widely shared among anti-vaccine groups on Twitter/X. Although factually accurate, Goel notes these stories were often stripped of context and reframed to undermine public trust in vaccines.

“These headlines became ammunition for people pushing anti-vaccine narratives,” Goel said. “And because the origin of the content came from credible outlets, they were harder to dismiss.”

The researchers argue that classifying content solely by its source’s reliability misses how information operates socially. A single article can inform or mislead, depending on how it is framed and interpreted. Goel added that the study moves beyond binary labels of true or false to examine how people use information—and how that shapes its impact.

The study also raises important questions for newsrooms. While factual accuracy remains essential, the findings suggest story framing—especially headlines—can influence how information is received and potentially misused online.

For both journalists and scientists, Goel emphasized the need for awareness that credible reporting can be co-opted in harmful ways, particularly on social media.