Many Americans seem to worry that their parents or grandparents will fall for fake news online. But as it turns out, we may be collectively concerned about the wrong generation.
Contrary to popular belief, Gen Zers and millennials could be more susceptible to online misinformation than older adults, according to a poll published online on June 29 by the research agency YouGov. What’s more, people who spend more time online had more difficulty distinguishing between real and fake news headlines. “We saw some results that are different from the ad hoc kinds of tests that [previous] researchers have done,” says Rakoen Maertens, a research psychologist at the University of Cambridge and lead author of a study on the development of the test used in the poll, which was published on June 29 in Behavior Research Methods.
Maertens’s team worked with YouGov to administer a quick online quiz based on the test that the researchers developed, dubbed the “misinformation susceptibility test” (MIST). It represents the first standardized test in psychology for misinformation and was set up in a way that allows researchers to administer it broadly and collect huge amounts of data. To create their test, Maertens and his colleagues carefully selected 10 actual headlines and 10 artificial-intelligence-generated false ones—similar to those you might encounter online—that they then categorized as “real” or “fake.” Test takers were asked to sort the real headlines from the fake news and received a percentage score at the end for each category. Here are a couple of examples of headlines from the test so you can try out your “fake news detector”: “US Support for Legal Marijuana Steady in Past Year,” “Certain Vaccines Are Loaded with Dangerous Chemicals and Toxins” and “Morocco’s King Appoints Committee Chief to Fight Poverty and Inequality.” The answers are at the bottom of this article.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Maertens and his team gave the test to thousands of people across the U.S. and the U.K. in their study, but the YouGov poll was given to 1,516 adults who were all U.S. citizens. On average, in the YouGov poll, U.S. adults correctly categorized about 65 percent of the headlines. However, age seemed to impact accuracy. Only 11 percent of Americans ages 18 to 29 correctly classified 17 or more headlines, and 36 percent got no more than 10 correct. That’s compared with 36 percent of the 65-and-older crowd who accurately assessed at least 16 headlines. And only 9 percent in the latter age group got 10 or fewer correct. On average, Americans below age 45 scored 12 out of 20, while their older counterparts scored 15.
Additionally, people who reported spending three or more leisure hours a day online were more likely to fall for misinformation (false headlines), compared with those who spent less time online. And where people got their news made a difference: folks who read legacy publications such as the Associated Press and Politico had better misinformation detection, while those who primarily got their news from social media sites such as TikTok, Instagram and Snapchat generally scored lower. (“I didn’t even know that [getting news from Snapchat] was an option,” Maertens says.) This could be part of the reason that younger people scored lower overall, Maertens’s team hypothesized. People who spend a lot of time on social media are exposed to a firehose of information, both real and fake, with little context to help distinguish the two.
Personality traits also impacted a person’s susceptibility to fake news. Conscientiousness, for instance, was associated with higher scores in the study conducted by Maertens and his team, while neuroticism and narcissism were associated with lower scores.
“They’ve done a good job in terms of conducting the research,” says Magda Osman, head of research and analysis at the Center for Science and Policy at the University of Cambridge, who was not involved in the study. She worries, however, that some of the test’s AI-generated headlines were less clear-cut than a simple real/fake classification could capture.
Take, for example, the headline “Democrats More Supportive than Republicans of Federal Spending for Scientific Research.” In the study, this claim was labeled as unambiguously true based on data from the Pew Research Center. But just by looking at the headline, Osman says, “you don’t know whether this means Democrats versus Republicans in the population or Democrats versus Republicans in Congress.”
This distinction matters because it changes the veracity of the statement. While it’s accurate to say that Democrats generally tend to support increased science funding, Republican politicians have a history of hiking up the defense budget, which means that over the past few decades, they have actually outspent their Democratic colleagues in funding certain types of research and development.
What’s more, Osman points out, the study does not differentiate which topics of misinformation different groups are more susceptible to. Younger people might be more likely than their parents to believe misinformation about sexual health or COVID but less likely to fall for fake news about climate change, she suggests.
“The test shouldn’t be taken as a 100% reliable individual-level test. Small differences can occur,” Maertens wrote in an e-mail to Scientific American. “Someone who has 18/20 could in practice be equally resilient as someone scoring 20/20. However, it is more likely that a 20/20 scorer is effectively better than let’s say a 14/20 scorer.”
Ultimately both Osman and Maertens agree that media literacy is a crucial skill for navigating today’s information-saturated world. “If you get flooded with information, you can’t really analyze every single piece,” Maertens says. He recommends taking a skeptical approach to everything you read online, fact-checking when possible (though that was not an option for MIST participants) and keeping in mind that you may be more susceptible to misinformation than you think.
In the example in the third paragraph, the headlines are, in order, real, fake, real.