When the Brain Stalls at Disjunction Junction

(By George Winkens -- Istockphoto)
By Shankar Vedantam
Monday, January 1, 2007

Should the United States pull out of Iraq, or increase the number of troops on the ground? Should you break up with that person you've been dating for eight years -- or propose marriage? Should you sell that old house, or hold on to it?

The start of a new year is typically a time we revisit big questions -- and big questions often do not have obviously right or wrong answers. So this is a good time to remind ourselves of the disjunction effect.

Don't know what that is? Sure you do. Have you ever delayed making a decision as you waited for more information to come along, only to find when the information arrives that it is irrelevant to your decision?

The disjunction effect describes the curious phenomenon in which people desperately seek information they do not actually need.

Think about a difficult medical decision you have to make for yourself or a loved one. There are many situations in which better information can indeed lead to better decisions, but there are also lots of cases in which more information does not make things any easier.

"What you can do is learn about the situation, but what you really need to do is learn about yourself," said Baruch Fischhoff, a psychologist at Carnegie Mellon University who is interested in how the pursuit of information in some situations can be self-defeating and even risky. "If you say we need absolute certainty before we do anything, you are guaranteeing paralysis, because you are hiding behind the uncertainty."

Cleverly designed experiments show that people will accept enormous financial and emotional costs in the pursuit of information, even when that information has no bearing on their final decisions.

Students at Stanford University, for example, were once divided into three groups. Students in one group were told they had just finished a difficult exam and had passed. In another group, all were told they had failed. Psychologists asked the students whether they wanted to go on a vacation. A little more than half the students in both groups said yes -- for different reasons. The group that passed wanted to celebrate. The group that failed wanted to get away from it all.

Psychologists then painted an identical scenario to a third group, except that these students were told the results of their exam were not known. Fewer than one in three students said they were willing to buy the vacation tickets right away, even though a delay meant an increase in price and ultimately made no difference -- equal numbers of students would go on the vacation whether they passed or failed.

"People are uncomfortable focusing on just the act and outcome," said Eldar Shafir, the psychologist who ran the experiment. "They want a script and narrative of why they are doing something. It's different packing a suit whether you are going to a funeral or a wedding, even though it is the same suit and the same suitcase."

The research underlines the importance of thinking about what information you need to make a decision, and not just seeking information in an open-ended manner. This is true whether the question has international implications -- how to fight global warming or what the U.S. military should do in Iraq -- or is intensely personal. The alternative is not just paralysis but the risk of being misled and manipulated by information that does not matter.

Let's say you are buying a new car and you have decided that the things you care about are price and efficiency. Once you get to the showroom to negotiate over the price of the model you like, the salesman tells you he is not sure whether the price includes the high-tech stereo system that is on the floor model. Would you like him to check? Of course, you say.

As he goes off to talk with his manager, you now suddenly care about the sound system. When the salesman comes back with the "good news" that the sound system is included, a piece of information you would have previously considered irrelevant now makes it more likely you will buy the car at the salesman's price. Information that didn't matter has overruled information that does.

"People wait for information when they shouldn't, and once they wait, they infer from waiting that the information matters," said Shafir, who is now at Princeton University.

Another experiment asked highly trained nurses in Canada whether they would be willing to donate a kidney to a relative who needed a transplant. Fewer than half said yes. Another group was given a different scenario. Those people were told that a relative needed a kidney but that doctors were not sure whether they were eligible donors. Would they be willing to take a test to find out? Nearly three-quarters of the nurses agreed to take the test -- far more than the number willing to donate a kidney in the first place.

Some of these people doubtless hoped the test would come back negative, which would spare them the difficult decision. But as the next stage of the experiment showed, such passivity can be harmful: All volunteers were told the test was positive, meaning they were eligible donors. More than two-thirds now said they were willing to donate the kidney. The very act of seeking information that should have been irrelevant, in other words, biased people into changing their decision.

"We don't know how much weight to give to different features of a decision," Shafir said. "How much should I care about the grade or the stereo? The salesman saying 'Let me check,' or the doctor saying 'Let me test' -- it may alter the weight of the information."

© 2007 The Washington Post Company