Why People Believe Misinformation, Even After It's Corrected

Feb. 8, 2012 — -- Have you seen the photo of the dog that's as big as a horse? How about the deer on top of a telephone pole? And do you know about the Hollywood actor who needed emergency medical help because of a gerbil that went where no gerbil had gone before?

That's all a bunch of bunk, or course. But we've heard those stories, or seen those photos, so many times that they have become a part of our world, even if they are totally false.

These days we are bombarded with information, much of it incorrect, and long after the political campaigns are over a lot of it will still be buried in the part of our brain where we store our memories. And new research shows that the more intensely we believe something to be true, the more likely it will resurface in the future, even if we have learned it was false.

Cognitive psychologist Andrew Butler of Duke University, a memory and learning specialist, hopes to figure out a way to help us purge our brains of false data, and he's a little encouraged. But it's probably not going to be easy.

Butler's latest research project, conducted with psychologist Lisa Fazio of Carnegie Mellon University and Elizabeth Marsh of Duke, found that it's possible to correct misinformation, but the correction may not last much more than a week.

Give it a little time, and that dog will be as big as a horse again.

Fifty students participated in the study, in which they were asked 120 basic science questions (What is stored in a camel's hump? What organ in the human body cleans the bloodstream and produces urine? What class of animals is the closest living relative of the dinosaurs?) The students also ranked their level of confidence in their answers, and they were really sure they had it right, at least some of the time.

But in most cases they were dead wrong. And here's the finding that Butler described in a telephone interview as "totally surprising."

The more strongly they believed they were right, the more efficient they were at accepting and remembering a correction.

"That flies in the face of a lot of memory theory," Butler said. According to memory theory, the brain throws up a wall of interference to protect a "deeply entrenched" idea or factoid, even if it is wrong. So a person who is highly confident that his understanding is correct should fight any effort to prove it wrong. But that didn't happen here.

Half the participants took the same test again immediately after learning the correct answers. And most of those who were so sure they had been right answered the question correctly when asked, for instance, about the closest living relative of dinosaurs. It's birds, not reptiles. And they knew, the second time around, that the kidney cleans the bloodstream and produces urine, not the bladder.

In fact, they corrected their mistakes 86 percent of the time on the retest. And they were more likely to get it right the second time if they had really believed their previous answers were correct than if they had less confidence in their facts.

The other half of the participants waited one week to take the test a second time. The same pattern persisted, but by then they only corrected their errors 56 percent of the time.

In just one week, nearly half the time they regressed to their cherished, but untrue, answers.

"It seems like a relatively transient thing," Butler said. "Our results indicate that over time, you are going to shift back to that misconception that you had before."