To help explain the increasing spread of misinformation, a collaboration involving The University of Western Australia has, for the first time, quantified a fundamental shift in the way American politicians communicate on social media.
Our analysis identified clear linguistic signals associated with the sharing of low-quality information. It follows that the public might learn to recognise these linguistic signals which would enable them to avoid being misled by that information.
Professor Stephan Lewandowsky, UWA School of Psychological Science
The international study, published today in Nature Human Behaviour, analysed 3.8 million tweets by Democratic and Republican members of Congress over the past decade.
It found that, since Donald Trump's 2016 election victory, both Republican and Democratic politicians increasingly shared their beliefs and opinions as well as evidence-based information – but the honestly held beliefs and opinions stated among Republicans were strongly linked to less trustworthy information sources.
Researchers developed a unique method to identify and measure the speech patterns of "belief-speaking," which relies on authentic expression of a conviction regardless of evidence or fact, and "fact-speaking," which examines evidence and substantiates opinion with facts.
The two groups of language patterns were used as a blueprint to produce algorithms which rated politicians' tweets, each tweet receiving a belief-speaking score and a fact-speaking score.
Using statistical models, the findings demonstrated a clear correlation between belief-speaking and the linking of poorly rated sources for Republican members of Congress.
Co-author Professor Stephan Lewandowsky, from UWA's School of Psychological Science and the University of Bristol, said the distinction between fact-speaking and belief-speaking might explain why three-quarters of Republican voters considered Donald Trump to be honest, despite an extensive record of false or misleading statements.
"The key insight is that one aspect of honesty is sincere expression of one's beliefs, no matter whether or not they are accurate," Professor Lewandowsky said.
"This is where Donald Trump scored highly because he always seemed to speak his mind and reported how he felt in the moment."
Lead author Dr Jana Lasser, a postdoctoral research fellow in computational social science at Graz University of Technology in Austria said the sharing of poor-quality sources followed a particular trend.
"In spreading their opinions and beliefs on Twitter, the Republicans are moving more and more in the direction of right-wing populists," she said.
But Professor Lewandowsky said the research could also provide potential solutions to combat misinformation.
"Our analysis identified clear linguistic signals associated with the sharing of low-quality information," he said.
"It follows that the public might learn to recognise these linguistic signals which would enable them to avoid being misled by that information."
