Let’s start with a little quiz: How attractive do you think you are, on a scale of 1 to 10? (We’ll get to politics soon, I promise.)

After you’ve given your answer, suppose you get good news. Some observers have been studying your appearance. In their view, you are a lot better looking than you think. Maybe you said 5; if so, the observers said 7.

Next question: After being informed about the observers’ rating, will you significantly revise your own?

If you’re like most people, you certainly will. You’ll give it a healthy boost.

Now suppose you get some bad news instead: The observers actually concluded that you are a lot less attractive than you think. If you said 5, the observers said 3. After being informed about that, will you change your rating a lot?

If you’re like most people, you won’t. You’ll stick pretty close to 5.

This is a demonstration of the “good news-bad news effect,” named and discovered by economists David Eil and Justin M. Rao. A technical note: This is similar to “confirmation bias,” but it isn’t the same thing. People respond more to good news than to bad news, regardless of whether it confirms their previous convictions.

In general, people are a lot more willing to revise their opinions after getting good news than after getting bad news, at least with respect to a wide variety of personal issues. For many of us, bad news seems like fake news.

How long do you expect to live? What’s the likelihood that you’ll get cancer or heart disease? How probable is it that you’ll be trapped in an elevator, or that you’ll find some mice in your basement?

Tali Sharot of the University College London has shown that for such questions, and many others, good news is more likely to alter people’s views than bad news.

What does that have to do with contemporary political issues? Along with Sharot and other collaborators, I have been exploring exactly that question. One of our initial findings, to be published soon in the Cornell Law Review, involves climate change.

In our survey, Americans who believe that climate change is real, but who are not particularly worried about it, estimated, on average, that unless further regulatory action is taken, the average U.S. temperature will increase by 2 degrees Celsius by 2100.

When they received good news, to the effect that scientists have recently concluded that the planet will get a lot less hot than previously thought, they cut their previous estimate by nearly one-third — to about 1.44 C.

But when they received bad news, to the effect that scientists have recently concluded that the planet will get a lot hotter than previously thought, their estimates did not change at all!

That’s the good news-bad news effect in action. It’s a lot like people’s pattern of responses to the how-attractive-are-you question.

The same effect is almost certainly at work for other political issues, and it helps to explain why some beliefs seem impervious to new information.

For example, people on the left may not be much influenced by evidence that the Affordable Care Act is causing some serious problems, or that big increases in the minimum wage are increasing unemployment. On those issues, good news will be a lot more credible.

Many Trump supporters are unlikely to be greatly affected by information about the repeated constitutional violations committed by Joe Arpaio, the former Arizona sheriff who was just pardoned, or by reports that the president’s regulatory policies are endangering public safety and health.

In this light, the current administration’s eagerness to claim that negative reports are fake news starts to look simultaneously smart and sinister. Whenever the White House labels bad news as fake news, it is playing on, and working to intensify, a widespread psychological inclination.

Can anything be done? Our work on climate change offers one upbeat message. People in the middle — moderates on the climate-change question — were equally affected by good news and bad news. We discovered that they were willing to listen to new scientific evidence, regardless of whether the evidence suggested that the problem was less or more worrisome than scientists had previously thought.

A growing body of work also suggests that people are far more willing to believe bad news if it comes from people they trust, or from people who seem like them. The source of the news may matter a lot more than the content.

Suppose, for example, that some of Trump’s closest advisers disassociate themselves from his statements and actions, or suggest that he has gone badly off course.

If and when that happens, it will remain possible for his supporters to dismiss the bad news — but doing so will become a lot harder.

Cass R. Sunstein is a Bloomberg View columnist and the Robert Walmsley University Professor at Harvard. He is the author of “#Republic: Divided Democracy in the Age of Social Media” and a co-author of “Nudge: Improving Decisions About Health, Wealth and Happiness.”

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.