Amos Tversky died in 1996. In 2002, Daniel Kahneman experienced the academic equivalent of a conquering general’s triumphal parade: He was awarded the Prize in Economic Sciences in Memory of Alfred Nobel. He is probably the only winner in the history of the prize who never took so much as a single class in economics.
The amazing thing is that the Science article, which sent shock waves out in every direction, is such a modest thing on its face. Kahneman and Tversky didn’t say anything about rationality. They didn’t call Homo economicus a myth. All they did was lay out solid research that revealed some of the heuristics—the rules of thumb—that Gut uses to make judgments, such as guessing how old Gandhi was when he died or whether it’s safe to drive to work. Today, Kahneman thinks that’s one reason that the article was as influential as it was. There was no grand theorizing, only research so solid it would withstand countless challenges in the years ahead.
Like the paper itself, the three rules of thumb it revealed were admirably simple and clear. The first—the Anchoring Rule—we’ve already discussed. The second is what psychologists call the “representativeness heuristic,” which I’ll call the Rule of Typical Things. And finally, there is the “availability heuristic,” or the Example Rule, which is by far the most important of the three in shaping our perceptions and reactions to risk.
THE RULE OF TYPICAL THINGS
Linda is thirty-one years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.
How likely is it that Linda
• is a teacher in elementary school?
• works in a bookstore and takes yoga classes?
• is active in the feminist movement?
• is a psychiatric social worker?
• is a member of the League of Women Voters?
• is a bank teller?
• is an insurance salesperson?
• is a bank teller and is active in the feminist movement?
Now, please rank these descriptions from most to least likely.
This is one of the most famous quizzes in psychology. When Kahneman and Tversky wrote the profile of “Linda” almost forty years ago, they intended to make it strongly match people’s image of an active feminist (an image that likely stood out a little more strongly at the time). Some of the descriptions on the list seem to be right on target. A member of the League of Women Voters? Yes, that fits. So it’s very likely true and it will certainly be at or near the top of the list. Active in the feminist movement? Absolutely. It will also rank highly. But an insurance salesperson? A bank teller? There’s nothing in the profile of Linda that specifically suggests either of these is correct, so people taking this quiz rank them at or near the bottom of the list.
That’s simple enough, but what about the final description of Linda as a bank teller who is also active in the feminist movement? Almost everyone who takes this quiz feels that, yes, this seems at least somewhat likely— certainly more likely than Linda being an insurance salesperson or a bank teller. When Kahneman and Tversky gave this quiz to undergraduate students, 89 percent decided it was more likely that Linda is a bank teller who is active in the feminist movement than that she is a bank teller alone.
But if you stop and think about it, that makes no sense. How can it be more likely that Linda is a bank teller and a feminist than that she is solely a bank teller? If it turns out to be true that she is a bank teller and a feminist, then she is a bank teller—so the two descriptions have to be, at a minimum, equally likely. What’s more, there is always the possibility that Linda is a bank teller but not a feminist. So it has to be true that it is more likely that she is a bank teller alone than that she is a bank teller and a feminist. It’s simple logic—but very few people see it.
So Kahneman and Tversky stripped the quiz down and tried again. They had students read the same profile of Linda. But then they simply asked whether it is more likely that Linda is (a) a bank teller or (b) a bank teller who is active in the feminist movement?
Here, the logic is laid bare. Kahneman and Tversky were sure people would spot it and correct their intuition. But they were wrong. Almost exactly the same percentage of students—85 percent—said it is more likely that Linda is a bank teller and a feminist than a bank teller only.
Kahneman and Tversky also put both versions of the “Linda problem,” as they called it, under the noses of experts trained in logic and statistics. When the experts answered the original question, with its long list of distracting details, they got it just as wrong as the undergraduates. But when they were given the two-line version, it was as if someone had elbowed them in the ribs. Head stepped in to correct Gut and the error rate plunged. When the scientist and essayist Stephen Jay Gould took the test, he realized what logic—his Head—told him was the right answer. But that didn’t change what intuition—his Gut—insisted was true. “I know [the right answer],” he recounted, “yet a little homunculus in my head continues to jump up and down, shouting at me—‘but she can’t just be a bank teller; read the description. ’”
What’s happening here is simple and powerful. One tool Gut uses to make judgments is the Rule of Typical Things. The typical summer day is hot and sunny, so how likely is a particular summer day to be hot and sunny? Very. That’s a simple example based on a simple notion of what’s “typical,” but we are capable of forming very complex images of typicality— such as that of a “typical” feminist or a “typical” bank teller. We make these sorts of judgments all the time and we’re scarcely aware of them for the good reason that they usually work, and that makes the Rule of Typical Things an effective way to simplify complex situations and come up with reliable snap judgments.
Or at least, it usually is. The Linda problem demonstrates one way the Rule of Typical Things can go wrong. When there’s something “typical” involved, our intuition is triggered. It just feels right. And as always with intuitive feelings, we tend to go with them even when doing so flies in the face of logic and evidence. It’s not just ordinary people who fall into this trap, incidentally. When Kahneman and Tversky asked a group of doctors to judge probabilities in a medical situation, the Rule of Typical Things kicked in and most of the doctors chose intuition over logic.
Another problem is that the Rule of Typical Things is only as good as our knowledge of what is “typical.” One belief about typicality that is unfortunately common in Western countries, particularly the United States, involves black men: The typical black man is a criminal and the typical criminal is a black man. Some people believe this consciously. Others who consciously reject this stereotype nonetheless believe it unconsciously—as even many black men do. Imagine someone who believes this—consciously or not—walking down a city sidewalk. A black man approaches. Instantaneously, this person’s Gut will use the Rule of Typical Things to conclude that there is a good chance this black man is a criminal. If Head does not intervene, the person will experience anxiety and consider crossing the street. But even if Head does put a stop to this nonsense, that nagging worry will remain—which may produce the uneasy body language black men so often encounter on sidewalks.