- Article published by Mit sloan
The MIT Sloan study finds that similar solutions to slow the spread of online misinformation could be effective in countries around the world.
Since the 2016 US presidential election and the British Brexit referendum – and then COVID-19 – opened the floodgates to fake news, academic research has delved into the psychology behind online disinformation and suggested interventions that could help curb the phenomenon.
Most studies have focused on disinformation in the West, even though misinformation is a very serious global problem. In 2017, false information on Facebook was implicated in the genocide against the Rohingya minority group in Myanmar, and in 2020 at least two dozen people were killed in mob lynchings following rumors spread on WhatsApp in India.
In a new paper published today in Nature Human Behavior, “Understanding and Combating Misinformation in 16 Countries on Six Continents,” MIT Sloan School of Management professor David Rand found that the traits shared by disinformation spreaders are surprisingly similar around the world. With a dozen colleagues from universities around the globe—including lead author Professor Antonio A. Arechar of the Center for Research and Teaching in Economics in Aguascalientes, Mexico, and co-senior author Professor Gordon Pennycook of Cornell University—the researchers also found common strategies and solutions to slow the spread of fake news that could be effective across countries.
"There is a general psychology of misinformation that transcends cultural differences. And broadly across many countries, we found that getting people to think about accuracy when deciding what to share, as well as showing them tips on media literacy, made them less likely to share false claims," Rand said.
The study, conducted in early 2021, found wide variations across countries in how likely participants were to believe false statements about COVID-19 — for example, those in India were twice as likely to believe false claims as those in the United Kingdom. Overall, people from individualistic countries with open political systems were better at distinguishing true statements from lies, the researchers found. However, in every country, they found that reflective analytical thinkers were less likely to believe false statements than people who relied on their gut.
"Almost everywhere we looked, people who were better critical thinkers and cared more about accuracy were less likely to believe false claims," said Rand. In the study, people who got all three questions wrong on a critical thinking quiz were twice as likely to believe false claims compared to people who got all three questions right. Valuing democracy was also consistently associated with being better at identifying truth.
On the other hand, endorsement of individual responsibility over government support and belief in God were associated with greater difficulty distinguishing true from false statements. People who said they would not definitely get vaccinated against COVID-19 when a vaccine became available were 52.9% more likely to believe false claims than those who said they would.
The researchers also found that at the end of the study, while 79% of participants said it was very or extremely important to share only accurate news, 77% of them had nevertheless shared at least one false statement as part of the experiment.
What explains this surprising disconnect? "A lot of them are people who just don't pay attention to accuracy," Rand said. "In a social media environment, there's so much to focus on around the social aspects of sharing – how many likes will I get, who else shared this – and we have limited cognitive latitude. So people often just forget to think about whether claims are true before they share them," he added.
This observation suggests ways to combat the spread of misinformation. Participants across countries who were prompted to focus on accuracy by rating the truth of an initial non-COVID-related question were nearly 10% less likely to share a false statement. Very simple media literacy tips also help to focus attention on accuracy, Rand said, and those who read them were 8.4% less likely to share a false statement. The researchers also found that when averaging up to 20 participant ratings, it was possible to identify false headlines with very high accuracy, leveraging the “wisdom of crowds” to help platforms detect misinformation at scale.
METHODOLOGY
In their study, Rand and colleagues recruited social media users from 16 countries across South America, Asia, the Middle East, and Africa. Australia and the United States were also among the countries, and nine languages were represented. Of the final sample size of 34,286, 45% were female, and the average age was 38.7 years.
The researchers randomly assigned study participants to one of four groups: Accuracy, Dissemination, Encouragement, and Advice. All participants were shown 10 false and 10 true statements related to COVID-19, taken from the websites of health and news organizations, including the World Health Organization and the BBC. For example, one false statement read: "Masks, gloves, vaccines and synthetic hand soaps suppress your immune system," while a true statement said: "The chance of shoes spreading COVID-19 is very low."
Participants in the Accuracy group were asked to rate the accuracy of the statements on a six-point scale, while the Sharing group assessed how likely they would be to share the statements. Study participants in the Encouragement group were first asked to rate the accuracy of a non-COVID-related statement. "Stewardess slaps crying baby during flight" – before deciding whether or not to share the COVID statements. Finally, the Advice group read four simple digital literacy tips before deciding whether or not to share them.
Implications for policy
The research has implications for policymakers and organizations like the United Nations seeking to combat disinformation around the globe, as well as for social media companies trying to reduce the sharing of fake news. "Really for anyone who works against disinformation," Rand said. "This research helps us better understand who is vulnerable to misinformation and shows us that there are important global interventions to help."
Social media companies could regularly ask users to rate the accuracy of headlines as a way to encourage them to focus on accuracy and share more opinions, while also collecting useful data to help identify misinformation. Future research should test these types of interventions in field experiments on social media platforms around the world, Rand said. "If you focus people back on accuracy, they are less likely to share bad content," he finished.