How news producers can drive consumers toward misinformation - even when they want the truth

New MIT Sloan research reveals flaws in common assumptions about how misinformation spreads and shows how news producers can manipulate consumer behavior.

Cambridge, MA, March 05, 2024 — In a new paper published in the Proceedings of the National Academy of Sciences (PNAS), MIT Sloan School of Management professor David Rand and three coauthors use game theory to show that news producer behavior — rather than consumer preferences — may explain why misinformation gets engagement online. 

In the marketplace of ideas, the truth will win out — or so the old theory holds. But does that assumption hold true in an online marketplace where content producers can strategically promote misinformation?

The misinformation game

Until now, most research on responses to misinformation has been observational and consumer-focused, asking questions like who falls for fake news and why. Rand and his coauthors saw that this approach left out a fundamental piece of the equation — misinformation producers. 

“A key feature of news consumption is that it is interactive,” Rand explained. “Consumers don't just make unilateral choices about what they consume, but instead, their choices depend on the content that producers produce. Game theory provides a formal language for understanding these kinds of multi-party interactions.”

To explore this interactive process, Rand and his coauthors developed a “misinformation game” model that considers how news outlets decide to publish true or false information and how readers and viewers choose whether or not to engage with stories from a given outlet. 

The model shows how news producers can adjust their strategies over time and manipulate consumers — even those actively seeking the truth — into engaging with false information. This is possible when two conditions are met: news producers can present different stories to different consumers using microtargeting and consumers are somewhat inattentive. Both are central features of the social media environment. 

“Facilitating the microtargeting of news makes it easier for misinformation producers to ensnare users who prefer accurate information,” Rand explains. 

The model also illustrates how news producers can spread misinformation using subtler tactics than consumers might expect. It shows that the most effective strategy isn’t publishing only blatantly false (but potentially attention-grabbing) headlines. Instead, it’s posting a mix of true and false information over time and increasing the rate of falsehoods as consumer engagement grows and readers let down their guards. 

Consumers’ behavior may not predict their preferences

Rand and his coauthors paired their game theory model with several empirical studies to find out whether real-life news consumers prefer true or false news. The researchers surveyed two groups, a standard sample of U.S. adults and a set of X (formerly Twitter) users who had shared links to fake news sites. 

There was a clear takeaway across both groups, said Rand. “People — even people who actually share low-quality information on Twitter — prefer to share accurate information when given the chance.” This preference isn’t limited to people’s perceptions of accuracy. The pattern also held when the researchers considered the objective veracity of information. 

However, even a widespread desire for the truth doesn’t mean it will rise to the top online. When they studied the relative success of 1,000 news articles from 40 outlets and integrated 20,000 accuracy ratings, Rand and his coauthors found that on known misinformation sites, more implausible stories generate higher engagement. Mainstream outlets show the opposite trend, with more plausible articles garnering more engagement.

While many researchers have assumed that preferential engagement with inaccurate headlines means consumers prefer news that is novel and attention-grabbing, even if it’s false, Rand and his colleagues found that this pattern can arise even when consumers strictly prefer accurate news. 

This mismatch between consumer preferences and content performance backs up the model’s prediction that unethical news producers could use responsive strategies and microtargeting to push falsehoods to a broader audience. 

Avoiding the misinformation trap

For consumers to find the truth in a sea of falsehoods, said Rand, we all need to slow down and focus when we’re taking in the news. 

“People should be paying attention to the content and to their previous encounters with the producer of the content,” he said. “They should reflect on whether the content itself is accurate and remember the quality of information they've seen in the past from the producer.”

One red flag that an outlet might be trying to manipulate consumers? Very high engagement on obviously implausible articles. This pattern could suggest that the outlet uses responsive transmission strategies like microtargeting to spread misinformation.

Right now, the deck may be stacked in misinformation producers’ favor — but as Rand and his team found, close attention puts consumers back in control. Moving forward, they plan to research interventions to help readers and viewers spot the signs of misinformation and reduce its spread.

About the MIT Sloan School of Management

The MIT Sloan School of Management is where smart, independent leaders come together to solve problems, create new organizations, and improve the world. Learn more at mitsloan.mit.edu.