So another type of mental tool that we rely on to make sense of the world is what have been called heuristics. These are informal cognate procedures for solving a variety of everyday life problems involving inference and judgement. They are mental shortcuts or rules of thumb that provide generally useful answers to judgment problems. One of the ones that we'll talk about is the availability heuristic. That assesses how readily certain instances come to mind. It's used for judging the frequency or probability of some type of event. And the representativeness heuristic, which categorizes something by how similar it is to our conception of the typical member of the category. It's used for judging probability and for inferring causality by assessing how similar an effect is to some possible cause. Which of these American states do you think has more tornadoes each year, Nebraska or Kansas? Bet you want to say Kansas, don't you? Nevermind that the tornado that you're thinking about never happened. Most people, though, do think there are more tornadoes in Kansas. The ready availability of a Kansas tornado, which happens to be fictional, makes us intuitively feel that there are more tornadoes in Kansas. And by the way there aren't, about the same in both states. The logic behind the availability heuristic, such as it is, is this. If I can quickly recall some there must be a lot of them. If I can't come up with things very easily, there probably aren't so many. That gives the right answer for whether it's Russia or Norway that has produced the more famous novelists. It also works for which country has started more wars of aggression, Germany or Switzerland? But are there more words with R in the first position or in the third position? Most people say the first position. It's easier to come up with words with R in the first position because one of ways we store words in memory is by their first letter. We don't store our words by their third letters. In fact, however, there are more words with R in the third position. Are there more suicides or homicides in the U.S.? It's easier to come up with homicides, isn't it? And most people think there are more homicides, but suicides outnumber homicides three to two. Do more people die in fires or drownings? It's easier to come up with fires, but, in fact, there are more drownings. Does it seem to you that the rate of crime has increased or decreased in recent years? No matter what's happening to the crime rate, people usually say it's increasing. There are always lots of crime examples available to us. Easy to think of crimes, must be getting worse. In fact, crime decreased more or less steadily for the 25 year period from 1990 to 2015. Most people thought it was increasing during that period. Notice that the availability heuristic makes us overestimate our contributions to joint projects. The stuff I did is more available to memory than the stuff you did. This can be a source of conflict in many kinds of relationships, including marriage. In one study married couples were asked how much they contributed to various projects, how much, for example, they contributed to breakfast. A woman said, I contribute 100%. I buy the food, I cook the food, I set the table, I clear the table, I clean the dishes. And the husband said, I do 25% because I feed the cat. Please read the following description. 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 antinuclear demonstrations. Which of the following seems more likely? Linda is a bank teller or Linda is a bank teller who is active in the feminist movement? We'll talk about Linda a little later. Recall that the representativeness heuristic is used for categorizing something by how similar it is to our conception of the typical member of the category. And we use it for assessing causality by judging how similar an effect is to some possible cause. One of the number sequences below is random, the other is not. Which do you think is random? It's the top sequence that's random. It's one of the first two I selected from a random number table. The second one looks more random because it is more representative of random sequence. It just so happens that our conception of randomness is faulty, which by the way gets us in all kinds of trouble. Actual random sequences have too many long strings, 00000 and too many repetitive patterns, 0101010101. The representativeness heuristic is perfectly reasonable tool to use to decide whether a pattern is random or not. The problem isn't the mental process, it's the conception of the category that's wrong. If I told you the student I know is shy and a poetry lover, do you think it's more likely that the person is a student of Chinese literature or a student of business administration? Even on a campus like mine where you actually can study Chinese literature, there are vastly more business administration students than Chinese literature students, so betting on Chinese literature would be very risky. I'd choose business administration. Lots of them are poetry lovers, I'm sure. Back to Linda, I asked you to read about. If you said she's more likely to be a bank teller who is active in the feminist movement, you're in good company. 85% of college students say that, but it's logically impossible. A and B can't be more likely than just A by itself. The problem shows how powerful the representativeness heuristic can be. A bank teller who's active in the feminist movement is much more representative of someone like Linda than just being a bank teller. But there can't be more feminist bank tellers than bank tellers or more female bank tellers than bank tellers or more red-headed bank tellers than bank tellers. So returning to the last lesson on prediction, why do we tend to ignore base rates in favor of individuating information about an object or person? Partly because we apply the representativeness heuristic so easily to the individuating information. Interviews are compelling because we convert the information we're getting into a stereotype of some kind, a category or set of categories. Go-getter, good sense of humor, solid judgement probably. The base rate information, the information about the person's history of achievement is pale in comparison to our vivid image of a person that we can classify so easily by using the representativeness heuristic. A number of years ago I was talking to a friend from graduate school about how successful our fellow graduate students had been in their research careers. As we thought about these people we begun to realize that we were awfully surprised about a lot of them. Some of them did much better than we thought, some did much worse. So we began to say, well, what were we basing our predictions on here? And it became clear that the basis was how representative our friends were of our stereotype of a good psychologist, verbally clever, knowledgeable, insightful about people. And then we began to think, what could we have based our judgment on? Could we have done better? And then we realized, yes, we could. We should have looked at the base rate for the person. The more good research they did in graduate school, the more good research later in life. And this reminds me of one the most useful things a psychologist can tell you. The best predictor of future behavior is past behavior. You're not going to do better than that. No other information you can get is likely to be a better predictor. No amount of interviewing, no psychological test you can give, no promises on the part of the person that they're turning over a new leaf. It's past behavior that you want to pay attention to. The representativeness heuristic is often used for making causal judgements. Causes should be represented of effects with respect to their size. The world changed when Lee Harvey Oswald killed JFK. I don't know whether Oswald acted alone or not, but I do know that the conviction that many people that it had to be a conspiracy was fueled by the sense that he was too small a cause for such a gigantic effect. There must have been a large conspiracy. There used to be a medical principle that diseases could be treated by something that resembled the disease or something that resembled the opposite of the disease. Jaundice turns people yellow, so we should treat jaundice with yellow things. The lungs of a fox, which has great respiratory powers, can be used for respiratory illness. The representativeness heuristic is the basis for many home remedies that are still in use. In the next segment, we'll be talking about two particular kinds of errors which are very broad and general, and which can get us in a lot of trouble.
