The term “fake news” is being used so broadly that Dominique Brossard, professor of life sciences communication at the University of Wisconsin-Madison, felt compelled to define it at the annual meeting of the American Association for the Advancement of Science on Saturday.

Fake news generally is false information packaged as a real news story to influence people, she said. However, in the context of science, things get “murkier and unclear.”

That's because science has always been plagued by sensationalism. The media, in an effort to simplify a finding, tends to exaggerate it, and many reporters lack the experience to accurately assess scientific claims. “Journalists are not all well-trained to assess the validity of a study,” Broussad said. “They are trying to find the human interest and the hope — a headline like: ‘New study brings hope to families with Alzheimer's.’”

This dynamic has also given rise to what we might call “content science” — lightweight studies that sell so well they even get mistakenly recycled.

“The problem in the science realm is deciding where is the line between bad science reporting and fake news,” Broussard said.

“Journalists are not all well-trained to assess the validity of a study.”

Okay, so what to do about it?

First, scientists need to get media training and take responsibility for communicating about their own work, Broussard said.

Second, institutions need to get better about monitoring news about the studies they produce, she said, and be vigilant about correcting any misconceptions.

Finally, she said, search engines should remove retracted studies from their search results. The prime example is the infamous discredited 1998 study about vaccines and autism by Andrew Wakefield.

The blog Retraction Watch reports between 500 and 600 retractions of inaccurate or false studies a year, Broussard said. While it's an important watch dog, more is needed.