Boys stand next to a poster about the Ebola virus during a training session by Sierra Leone's national soccer team.Luc Gnago / Reuters / Corbis

The unprecedented scale of the Ebola outbreak that began in 2014 led not only to the inevitable frantic media coverage, both good and bad, but also to a meta-conversation about health media, whether it’s good or bad, responsible or irresponsible, scientific or sensationalist.

Did the media overreact to the outbreak when cases appeared in the United States? Did journalists help fuel panic? Did the media overreact about the level of panic the media was actually causing? And so on and so on, in a dizzying ouroboros of analysis.

Through all of this, it’s hard to know what people are actually taking away from the articles they read and the news broadcasts they see. In a new study, researchers at Princeton University investigated how at-risk people feel after learning about a disease, and how that is linked to what information they remember about it.

A group of 460 participants read an article about meningococcal disease with information about its symptoms (“severe headaches that have a rapid onset,” “uncomfortable and painful stiff neck,” “sensitivity to natural and artificial light,” and “a skin rash that could cover the whole body”) as well as risk factors, preventive steps to take, and how it’s diagnosed. Some people also read an article emphasizing that only one in 100,000 people in the U.S. gets the disease each year, while some read an article that emphasized that the mortality rate could be as high as 40 percent. The latter article, while providing mostly the same information on symptoms and such as the first, was designed to make people feel more at-risk.

Then, participants listened to an audio clip of a “radio host” interviewing a fake employee for the Centers for Disease Control and Prevention, who repeated some of the facts from the article the participants read, but not all of them.

While people were always better at remembering the information that was repeated in the audio clip, since they’d been exposed to it twice, the researchers found that people in the high-risk condition tended to forget more of the other facts—the ones that hadn’t been included in the audio clip—than people in the other group did.

The odd thing is, the high-risk people were more likely to forget information that was related to what the expert was talking about on the radio. So if the CDC expert mentioned the headaches and the rash, for example, the listener would be more likely to forget the other symptoms. But they might still remember the risk factors, or other things unrelated to the symptoms.

This sort of simplification happens when people are remembering previously learned information when listening to someone speak about the same topic. (It’s called “socially shared retrieval-induced forgetting,” if you want to get technical.) Another study, from 2014, found that health messages tend to get condensed the more they’re shared from person to person, and to become less accurate.

And the more a subject is being discussed in the media, the more opportunities there are for these things to happen. People read stories that may emphasize some facts over others, they talk amongst themselves, and they share things on social media, likely leaving a lot of nuance—and sometimes accuracy—by the wayside. Especially if they feel like they’re at risk.

When there’s an outbreak, and people are worried about themselves and their loved ones, this is probably unavoidable to a certain degree. But if reporters, government officials, and everyone else spreading health messages try not to provoke more than the necessary anxiety about a disease, this research suggests people will be more likely to remember the whole picture.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.