How do people form political beliefs? When will they change their minds? When will actual facts matter? A recent study, conducted by political scientist Brendan Nyhan of Dartmouth College and two co-authors, offers some clues.

One group of participants was provided with a 2009 news article in which Sarah Palin claimed that the Barack Obama administration’s Affordable Care Act created death panels and that these panels included bureaucrats authorized to decide whether seniors were “worthy of health care.” A separate group was given the same news story, but with an appended correction saying that “nonpartisan health care experts have concluded that Palin is wrong.”

The study’s big question: Would the correction have any effect? Would people who saw the correction be less likely to believe that the Affordable Care Act calls for death panels?

Not surprisingly, the correction was more likely to convince people who viewed Palin unfavorably than those who had a high opinion of her. Notably, the correction also tended to sway the participants who liked Palin but who didn’t have a lot of political knowledge (as measured by their answers to general questions, such as how many terms a president may serve).

Here’s the most interesting finding in the study. Those who viewed Palin favorably, and who also had a lot of political knowledge, were not persuaded by the correction. On the contrary, it made them more likely to believe Palin was right.

This finding presents an intriguing puzzle. While the correction tended to convince Palin supporters who lacked political knowledge that she was wrong about death panels, it generally failed to persuade Palin supporters who had such knowledge. For those supporters, the correction actually backfired. How come?

There are two explanations, and they tell us a lot about current political controversies — and about why the United States and other nations remain so badly polarized. The first is that if you know a lot about politics, you are more likely to be emotionally invested in what you believe. Efforts to undermine or dislodge those beliefs might well upset you and therefore backfire. The second explanation is that if you have a lot of political knowledge, you are more likely to think you know what is really true, and it will be pretty hard for people to convince you otherwise.

The general lesson is both straightforward and disturbing. People who know a lot, and who trust a particular messenger, might well be impervious to factual corrections, even if what they believe turns out to be false.

It is important to distinguish between two kinds of political “validators”: the expected and the surprising. For informed listeners, certain messengers will be hopelessly ineffective and maybe even counterproductive, not because of what they say, but because of who they are. When Democrat congresswoman Nancy Pelosi, the House minority leader, argues that climate change is a serious problem, or when Republican congressman John Boehner, the House speaker, contends that increases in the minimum wage will reduce employment, their messages are expected and to that extent uninformative. Skeptics are more likely to yawn than to shift their own views.

But some political validators are surprising. If Boehner suddenly announced that climate change is an urgent problem that needed to be addressed, or if former Secretary of State Hillary Clinton proclaimed that increases in the minimum wage are a terrible idea because they would reduce employment, skeptics might well stop and notice, even if they are well-informed.

Consider Republican Sen. Rob Portman’s recent expression of support for same-sex marriage. That statement was newsworthy, and may have moved people, precisely because he was a Republican. (It is interesting to wonder whether the message was strengthened or weakened by the knowledge that Portman’s son is gay.)

There is a clear implication here for those who provide information. Balanced presentations may divide people even more sharply than before, and factual corrections may backfire. When people begin with clear convictions, efforts to correct their errors may have the perverse effect of entrenching them — at least if the messengers aren’t seen as credible.

The most important lesson may be for those who receive political information. Sure, it is important to consider the source, but the content matters, as well. Our favorite messengers are sometimes wrong and our least favorite messengers are sometimes right. It’s sometimes worthwhile to pay a lot more attention to what is being said, and a lot less to the identity of the person who is saying it.

Cass R. Sunstein, the Robert Walmsley University professor at Harvard Law School, is a Bloomberg View columnist. The opinions expressed are his own.

In a time of both misinformation and too much information, quality journalism is more crucial than ever.
By subscribing, you can help us get the story right.