People who frequently share fake news on social media tend to use words related to anger, power, religion, death, and money while using fewer words associated with family, friends, and ‘low arousal’ emotions. To counter extremism, a new era of media literacy is encouraged.
Recent research from Columbia Business School analysed 3,200 existing posts by X (Twitter) users, identifying linguistic patterns that predict which social media users are likely to share fake news.
These users tend to be more open and neurotic but less agreeable, extroverted, and conscientious. Notably, these traits overlap with users who share fact-checks, suggesting both groups are driven by anger.
The study suggests that platforms could improve misinformation detection by focusing on users’ language patterns rather than just flagging content. By incorporating textual cues into predictive models, platforms could better prioritise fact-checking efforts.
In experiments to curb fake news sharing, researchers found that directly reducing anger in the moment had little effect.
However, messaging that made users feel empowered, such as advertisements emphasising their control over misinformation, successfully encouraged the adoption of fact-checking tools.
This insight suggests that fostering a sense of personal agency might help reduce the spread of misinformation on social media. “But of course, that’s something you don’t always control,” concludes Gita Johar, Meyer Feldberg Professor of Business who authored the research.
A new generation of media literacy
In an article published by the Reuters Institute, Bellingcat’s founder Eliot Higgins and Dr. Natalie Martin of the Centre for Media, Politics and Communication Research discuss the growing threat of weaponised information, using the UK news media landscape as a case study.
While misinformation and disinformation have always existed, advancements in technology – particularly smartphones and artificial intelligence – have accelerated their spread. This has made it easier for state and non-state actors to manipulate narratives, destabilise democracies, and erode trust in liberal institutions.
Weaponised information takes multiple forms, including misinformation (false but unintended), disinformation (false and deliberately harmful), and malinformation (true but shared with harmful intent). The authors introduce a fourth category: non-scrutinising, bland information that fills the space where critical journalism should be.
Various actors, from state-controlled media to radicalised “true believers” in online echo chambers, contribute to the dissemination of disordered information. The result is an erosion of trust in mainstream media and institutions, creating a vacuum filled by alternative, often extreme, narratives.
Fact-checking and regulation
Fact-checking and regulation efforts have proven insufficient due to public scepticism towards mainstream news and the challenges of regulating free speech in democratic societies. Instead, the authors argue for a proactive educational approach: integrating open-source investigation (OSI) training in schools, colleges, and universities.
Inspired by Bellingcat’s methodology, this training would equip individuals with skills to critically assess and verify information, fostering a new generation of media-literate citizens capable of identifying and countering disinformation.
The proposal envisions embedding OSI skills into curricula, encouraging student-led investigations, and creating pathways into journalism and civic engagement.
Ultimately, the article argues that OSI education should be a policy priority for the UK government to mitigate the impact of disordered information, strengthen democratic resilience, and counteract the manipulation of news ecosystems by illiberal actors.
[Edited By Brian Maguire | Euractiv’s Advocacy Lab ]