At the end of the vacation, all pictures and videos will be destroyed. Furthermore, you will swallow a potion that will wipe out all your memories of the vacation. How would this prospect affect your vacation plans? How much would you be willing to pay for it, relative to a normally memorable vacation? While I have not formally studied the reactions to this scenario, my impression from discussing it with people is that the elimination of memories greatly reduces the value of the experience. In some cases, people treat themselves as they would treat another amnesic, choosing to maximize overall pleasure by returning to a place where they have been happy in the past. However, some people say that they would not bother to go at all, revealing that they care only about their remembering self, and care less about their amnesic experiencing self than about an amnesic stranger. Many point out that they would not send either themselves or another amnesic to climb mountains or trek through the jungle—because these experiences are mostly painful in real time and gain value from the expectation that both the pain and the joy of reaching the goal will be memorable. For another thought experiment, imagine you face a painful operation during which you will remain conscious. You are told you will scream in pain and beg the surgeon to stop. However, you are promised an amnesia- inducing drug that will completely wipe out any memory of the episode. How do you feel about such a prospect? Here again, my informal observation is that most people are remarkably indifferent to the pains of their experiencing self. Some say they don’t care at all. Others share my feeling, which is that I feel pity for my suffering self but not more than I would feel for a stranger in pain. Odd as it may seem, I am my remembering self, and the experiencing self, who does my living, is like a stranger to me. SPEAKING OF LIFE AS A STORY “He is desperately trying to protect the narrative of a life of integrity, which is endangered by the latest episode.” “The length to which he was willing to go for a one-night encounter is a sign of total duration neglect.” “You seem to be devoting your entire vacation to the construction of memories. Perhaps you should put away the camera and enjoy the moment, even if it is not very memorable?” “She is an Alzheimer’s patient. She no longer maintains a narrative of her life, but her experiencing self is still sensitive to beauty and gentleness.”
37 Experienced Well-Being When I became interested in the study of well-being about fifteen years ago, I quickly found out that almost everything that was known about the subject drew on the answers of millions of people to minor variations of a survey question, which was generally accepted as a measure of happiness. The question is clearly addressed to your remembering self, which is invited to think about your life: All things considered, how satisfied are you with your life as a whole these days? Having come to the topic of well-being from the study of the mistaken memories of colonoscopies and painfully cold hands, I was naturally suspicious of global satisfaction with life as a valid measure of well-being. As the remembering self had not proved to be a good witness in my experiments, I focused on the well-being of the experiencing self. I proposed that it made sense to say that “Helen was happy in the month of March” if she spent most of her time engaged in activities that she would rather continue than stop, little time in situations she wished to escape, and—very important because life is short— not too much time in a neutral state in which she would not care either way. There are many different experiences we would rather continue than stop, including both mental and physical pleasures. One of the examples I had in mind for a situation that Helen would wish to continue is total absorption in a task, which Mihaly Csikszentmihalyi calls flow—a state that some artists
experience in their creative moments and that many other people achieve when enthralled by a film, a book, or a crossword puzzle: interruptions are not welcome in any of these situations. I also had memories of a happy early childhood in which I always cried when my mother came to tear me away from my toys to take me to the park, and cried again when she took me away from the swings and the slide. The resistance to interruption was a sign I had been having a good time, both with my toys and with the swings. I proposed to measure Helen’s objective happiness precisely as we assessed the experience of the two colonoscopy patients, by evaluating a profile of the well-being she experienced over successive moments of her life. In this I was following Edgeworth’s hedonimeter method of a century earlier. In my initial enthusiasm for this approach, I was inclined to dismiss Helen’s remembering self as an error-prone witness to the actual well-being of her experiencing self. I suspected this position was too extreme, which it turned out to be, but it was a good start. EXPERIENCED WELL-BEING I assembled “a dream team” that included three other psychologists of different specialties and one economist, and we set out together to develop a measure of the well-being of the experiencing self. A continuous record of experience was unfortunately impossible—a person cannot live normally while constantly reporting her experiences. The closest alternative was experience sampling, a method that Csikszentmihalyi had invented. Technology has advanced since its first uses. Experience sampling is now implemented by programming an individual’s cell phone to beep or vibrate at random intervals during the day. The phone then presents a brief menu of questions about what the respondent was doing and who was with her when she was interrupted. The participant is also shown rating scales to report the intensity of various feelings: happiness, tension, anger, worry, engagement, physical pain, and others. Experience sampling is expensive and burdensome (although less disturbing than most people initially expect; answering the questions takes very little time). A more practical alternative was needed, so we developed a method that we called the Day Reconstruction Method (DRM). We hoped it would approximate the results of experience sampling and provide additional information about the way people spend their time. Participants (all women, in the early studies) were invited to a two-hour session. We
first asked them to relive the previous day in detail, breaking it up into episodes like scenes in a film. Later, they answered menus of questions about each episode, based on the experience-sampling method. They selected activities in which they were engaged from a list and indicated the one to which they paid most attention. They also listed the individuals they had been with, and rated the intensity of several feelings on separate 0–6 scales (0 = the absence of the feeling; 6 = most intense feeling). Our method drew on evidence that people who are able to retrieve a past situation in detail are also able to relive the feelings that accompanied it, even experiencing their earlier physiological indications of emotion. We assumed that our participants would fairly accurately recover the feeling of a prototypical moment of the episode. Several comparisons with experience sampling confirmed the validity of the DRM. Because the participants also reported the times at which episodes began and ended, we were able to compute a duration-weighted measure of their feeling during the entire waking day. Longer episodes counted more than short episodes in our summary measure of daily affect. Our questionnaire also included measures of life satisfaction, which we interpreted as the satisfaction of the remembering self. We used the DRM to study the determinants of both emotional well-being and life satisfaction in several thousand women in the United States, France, and Denmark. The experience of a moment or an episode is not easily represented by a single happiness value. There are many variants of positive feelings, including love, joy, engagement, hope, amusement, and many others. Negative emotions also come in many varieties, including anger, shame, depression, and loneliness. Although positive and negative emotions exist at the same time, it is possible to classify most moments of life as ultimately positive or negative. We could identify unpleasant episodes by comparing the ratings of positive and negative adjectives. We called an episode unpleasant if a negative feeling was assigned a higher rating than all the positive feelings. We found that American women spent about 19% of the time in an unpleasant state, somewhat higher than French women (16%) or Danish women (14%). We called the percentage of time that an individual spends in an unpleasant state the U-index. For example, an individual who spent 4 hours of a 16-hour waking day in an unpleasant state would have a U-index of 25%. The appeal of the U-index is that it is based not on a rating scale but
on an objective measurement of time. If the U-index for a population drops from 20% to 18%, you can infer that the total time that the population spent in emotional discomfort or pain has diminished by a tenth. A striking observation was the extent of inequality in the distribution of emotional pain. About half our participants reported going through an entire day without experiencing an unpleasant episode. On the other hand, a significant minority of the population experienced considerable emotional distress for much of the day. It appears that a small fraction of the population does most of the suffering—whether because of physical or mental illness, an unhappy temperament, or the misfortunes and personal tragedies in their life. A U-index can also be computed for activities. For example, we can measure the proportion of time that people spend in a negative emotional state while commuting, working, or interacting with their parents, spouses, or children. For 1,000 American women in a Midwestern city, the U-index was 29% for the morning commute, 27% for work, 24% for child care, 18% for housework, 12% for socializing, 12% for TV watching, and 5% for sex. The U-index was higher by about 6% on weekdays than it was on weekends, mostly because on weekends people spend less time in activities they dislike and do not suffer the tension and stress associated with work. The biggest surprise was the emotional experience of the time spent with one’s children, which for American women was slightly less enjoyable than doing housework. Here we found one of the few contrasts between French and American women: Frenchwomen spend less time with their children but enjoy it more, perhaps because they have more access to child care and spend less of the afternoon driving children to various activities. An individual’s mood at any moment depends on her temperament and overall happiness, but emotional well-being also fluctuates considerably over the day and the week. The mood of the moment depends primarily on the current situation. Mood at work, for example, is largely unaffected by the factors that influence general job satisfaction, including benefits and status. More important are situational factors such as an opportunity to socialize with coworkers, exposure to loud noise, time pressure (a significant source of negative affect), and the immediate presence of a boss (in our first study, the only thing that was worse than being alone). Attention is key. Our emotional state is largely determined by what we attend to, and we are normally focused on our current activity and
immediate environment. There are exceptions, where the quality of subjective experience is dominated by recurrent thoughts rather than by the events of the moment. When happily in love, we may feel joy even when caught in traffic, and if grieving, we may remain depressed when watching a funny movie. In normal circumstances, however, we draw pleasure and pain from what is happening at the moment, if we attend to it. To get pleasure from eating, for example, you must notice that you are doing it. We found that French and American women spent about the same amount of time eating, but for Frenchwomen, eating was twice as likely to be focal as it was for American women. The Americans were far more prone to combine eating with other activities, and their pleasure from eating was correspondingly diluted. These observations have implications for both individuals and society. The use of time is one of the areas of life over which people have some control. Few individuals can will themselves to have a sunnier disposition, but some may be able to arrange their lives to spend less of their day commuting, and more time doing things they enjoy with people they like. The feelings associated with different activities suggest that another way to improve experience is to switch time from passive leisure, such as TV watching, to more active forms of leisure, including socializing and exercise. From the social perspective, improved transportation for the labor force, availability of child care for working women, and improved socializing opportunities for the elderly may be relatively efficient ways to reduce the U-index of society—even a reduction by 1% would be a significant achievement, amounting to millions of hours of avoided suffering. Combined national surveys of time use and of experienced well- being can inform social policy in multiple ways. The economist on our team, Alan Krueger, took the lead in an effort to introduce elements of this method into national statistics. Measures of experienced well-being are now routinely used in large-scale national surveys in the United States, Canada, and Europe, and the Gallup World Poll has extended these measurements to millions of respondents in the United States and in more than 150 countries. The polls elicit reports of the emotions experienced during the previous day, though in less detail than the DRM. The gigantic samples allow extremely fine analyses, which have confirmed the importance of situational factors, physical health, and social
contact in experienced well-being. Not surprisingly, a headache will make a person miserable, and the second best predictor of the feelings of a day is whether a person did or did not have contacts with friends or relatives. It is only a slight exaggeration to say that happiness is the experience of spending time with people you love and who love you. The Gallup data permit a comparison of two aspects of well-being: the well-being that people experience as they live their lives the judgment they make when they evaluate their life Gallup’s life evaluation is measured by a question known as the Cantril Self-Anchoring Striving Scale: Please imagine a ladder with steps numbered from zero at the bottom to 10 at the top. The top of the ladder represents the best possible life for you and the bottom of the ladder represents the worst possible life for you. On which step of the ladder would you say you personally feel you stand at this time? Some aspects of life have more effect on the evaluation of one’s life than on the experience of living. Educational attainment is an example. More education is associated with higher evaluation of one’s life, but not with greater experienced well-being. Indeed, at least in the United States, the more educated tend to report higher stress. On the other hand, ill health has a much stronger adverse effect on experienced well-being than on life evaluation. Living with children also imposes a significant cost in the currency of daily feelings—reports of stress and anger are common among parents, but the adverse effects on life evaluation are smaller. Religious participation also has relatively greater favorable impact on both positive affect and stress reduction than on life evaluation. Surprisingly, however, religion provides no reduction of feelings of depression or worry. An analysis of more than 450,000 responses to the Gallup-Healthways Well-Being Index, a daily survey of 1,000 Americans, provides a surprisingly definite answer to the most frequently asked question in well- being research: Can money buy happiness? The conclusion is that being poor makes one miserable, and that being rich may enhance one’s life satisfaction, but does not (on average) improve experienced well-being. Severe poverty amplifies the experienced effects of other misfortunes of life. In particular, illness is much worse for the very poor than for those who are more comfortable. A headache increases the proportion reporting
sadness and worry from 19% to 38% for individuals in the top two-thirds of the income distribution. The corresponding numbers for the poorest tenth are 38% and 70%—a higher baseline level and a much larger increase. Significant differences between the very poor and others are also found for the effects of divorce and loneliness. Furthermore, the beneficial effects of the weekend on experienced well-being are significantly smaller for the very poor than for most everyone else. The satiation level beyond which experienced well-being no longer increases was a household income of about $75,000 in high-cost areas (it could be less in areas where the cost of living is lower). The average increase of experienced well-being associated with incomes beyond that level was precisely zero. This is surprising because higher income undoubtedly permits the purchase of many pleasures, including vacations in interesting places and opera tickets, as well as an improved living environment. Why do these added pleasures not show up in reports of emotional experience? A plausible interpretation is that higher income is associated with a reduced ability to enjoy the small pleasures of life. There is suggestive evidence in favor of this idea: priming students with the idea of wealth reduces the pleasure their face expresses as they eat a bar of chocolate! There is a clear contrast between the effects of income on experienced well-being and on life satisfaction. Higher income brings with it higher satisfaction, well beyond the point at which it ceases to have any positive effect on experience. The general conclusion is as clear for well-being as it was for colonoscopies: people’s evaluations of their lives and their actual experience may be related, but they are also different. Life satisfaction is not a flawed measure of their experienced well-being, as I thought some years ago. It is something else entirely. SPEAKING OF EXPERIENCED WELL-BEING “The objective of policy should be to reduce human suffering. We aim for a lower U-index in society. Dealing with depression and extreme poverty should be a priority.” “The easiest way to increase happiness is to control your use of time. Can you find more time to do the things you enjoy doing?” “Beyond the satiation level of income, you can buy more pleasurable experiences, but you will lose some of your ability to enjoy the less expensive ones.”
38 Thinking About Life Figure 16 is taken from an analysis by Andrew Clark, Ed Diener, and Yannis Georgellis of the German Socio-Economic Panel, in which the same respondents were asked every year about their satisfaction with their life. Respondents also reported major changes that had occurred in their circumstances during the preceding year. The graph shows the level of satisfaction reported by people around the time they got married.
Figure 16 The graph reliably evokes nervous laughter from audiences, and the nervousness is easy to understand: after all, people who decide to get married do so either because they expect it will make them happier or because they hope that making a tie permanent will maintain the present state of bliss. In the useful term introduced by Daniel Gilbert and Timothy Wilson, the decision to get married reflects, for many people, a massive error of affective forecasting. On their wedding day, the bride and the groom know that the rate of divorce is high and that the incidence of marital disappointment is even higher, but they do not believe that these statistics apply to them. The startling news of figure 16 is the steep decline of life satisfaction. The graph is commonly interpreted as tracing a process of adaptation, in which the early joys of marriage quickly disappear as the experiences become routine. However, another approach is possible, which focuses on heuristics of judgment. Here we ask what happens in people’s minds when they are asked to evaluate their life. The questions “How satisfied are you with your life as a whole?” and “How happy are you these days?” are not as simple as “What is your telephone number?” How do survey participants
manage to answer such questions in a few seconds, as all do? It will help to think of this as another judgment. As is also the case for other questions, some people may have a ready-made answer, which they had produced on another occasion in which they evaluated their life. Others, probably the majority, do not quickly find a response to the exact question they were asked, and automatically make their task easier by substituting the answer to another question. System 1 is at work. When we look at figure 16 in this light, it takes on a different meaning. The answers to many simple questions can be substituted for a global evaluation of life. You remember the study in which students who had just been asked how many dates they had in the previous month reported their “happiness these days” as if dating was the only significant fact in their life. In another well-known experiment in the same vein, Norbert Schwarz and his colleagues invited subjects to the lab to complete a questionnaire on life satisfaction. Before they began that task, however, he asked them to photocopy a sheet of paper for him. Half the respondents found a dime on the copying machine, planted there by the experimenter. The minor lucky incident caused a marked improvement in subjects’ reported satisfaction with their life as a whole! A mood heuristic is one way to answer life- satisfaction questions. The dating survey and the coin-on-the-machine experiment demonstrated, as intended, that the responses to global well-being questions should be taken with a grain of salt. But of course your current mood is not the only thing that comes to mind when you are asked to evaluate your life. You are likely to be reminded of significant events in your recent past or near future; of recurrent concerns, such as the health of a spouse or the bad company that your teenager keeps; of important achievements and painful failures. A few ideas that are relevant to the question will occur to you; many others will not. Even when it is not influenced by completely irrelevant accidents such as the coin on the machine, the score that you quickly assign to your life is determined by a small sample of highly available ideas, not by a careful weighting of the domains of your life. People who recently married, or are expecting to marry in the near future, are likely to retrieve that fact when asked a general question about their life. Because marriage is almost always voluntary in the United States, almost everyone who is reminded of his or her recent or forthcoming marriage will be happy with the idea. Attention is the key to the puzzle. Figure 16 can be
read as a graph of the likelihood that people will think of their recent or forthcoming marriage when asked about their life. The salience of this thought is bound to diminish with the passage of time, as its novelty wanes. The figure shows an unusually high level of life satisfaction that lasts two or three years around the event of marriage. However, if this apparent surge reflects the time course of a heuristic for answering the question, there is little we can learn from it about either happiness or about the process of adaptation to marriage. We cannot infer from it that a tide of raised happiness lasts for several years and gradually recedes. Even people who are happy to be reminded of their marriage when asked a question about their life are not necessarily happier the rest of the time. Unless they think happy thoughts about their marriage during much of their day, it will not directly influence their happiness. Even newlyweds who are lucky enough to enjoy a state of happy preoccupation with their love will eventually return to earth, and their experienced well-being will again depend, as it does for the rest of us, on the environment and activities of the present moment. In the DRM studies, there was no overall difference in experienced well- being between women who lived with a mate and women who did not. The details of how the two groups used their time explained the finding. Women who have a mate spend less time alone, but also much less time with friends. They spend more time making love, which is wonderful, but also more time doing housework, preparing food, and caring for children, all relatively unpopular activities. And of course, the large amount of time married women spend with their husband is much more pleasant for some than for others. Experienced well-being is on average unaffected by marriage, not because marriage makes no difference to happiness but because it changes some aspects of life for the better and others for the worse. One reason for the low correlations between individuals’ circumstances and their satisfaction with life is that both experienced happiness and life satisfaction are largely determined by the genetics of temperament. A disposition for well-being is as heritable as height or intelligence, as demonstrated by studies of twins separated at birth. People who appear equally fortunate vary greatly in how happy they are. In some instances, as in the case of marriage, the correlations with well-being are low because of
balancing effects. The same situation may be good for some people and bad for others, and new circumstances have both benefits and costs. In other cases, such as high income, the effects on life satisfaction are generally positive, but the picture is complicated by the fact that some people care much more about money than others do. A large-scale study of the impact of higher education, which was conducted for another purpose, revealed striking evidence of the lifelong effects of the goals that young people set for themselves. The relevant data were drawn from questionnaires collected in 1995–1997 from approximately 12,000 people who had started their higher education in elite schools in 1976. When they were 17 or 18, the participants had filled out a questionnaire in which they rated the goal of “being very well-off financially” on a 4-point scale ranging from “not important” to “essential.” The questionnaire they completed twenty years later included measures of their income in 1995, as well as a global measure of life satisfaction. Goals make a large difference. Nineteen years after they stated their financial aspirations, many of the people who wanted a high income had achieved it. Among the 597 physicians and other medical professionals in the sample, for example, each additional point on the money-importance scale was associated with an increment of over $14,000 of job income in 1995 dollars! Nonworking married women were also likely to have satisfied their financial ambitions. Each point on the scale translated into more than $12,000 of added household income for these women, evidently through the earnings of their spouse. The importance that people attached to income at age 18 also anticipated their satisfaction with their income as adults. We compared life satisfaction in a high-income group (more than $200,000 household income) to a low-to moderate-income group (less than $50,000). The effect of income on life satisfaction was larger for those who had listed being well-off financially as an essential goal: .57 point on a 5-point scale. The corresponding difference for those who had indicated that money was not important was only .12. The people who wanted money and got it were significantly more satisfied than average; those who wanted money and didn’t get it were significantly more dissatisfied. The same principle applies to other goals—one recipe for a dissatisfied adulthood is setting goals that are especially difficult to attain. Measured by life satisfaction 20 years later, the least promising goal that a young person could have was “becoming accomplished in a performing
art.” Teenagers’ goals influence what happens to them, where they end up, and how satisfied they are. In part because of these findings I have changed my mind about the definition of well-being. The goals that people set for themselves are so important to what they do and how they feel about it that an exclusive focus on experienced well-being is not tenable. We cannot hold a concept of well- being that ignores what people want. On the other hand, it is also true that a concept of well-being that ignores how people feel as they live and focuses only on how they feel when they think about their life is also untenable. We must accept the complexities of a hybrid view, in which the well-being of both selves is considered. THE FOCUSING ILLUSION We can infer from the speed with which people respond to questions about their life, and from the effects of current mood on their responses, that they do not engage in a careful examination when they evaluate their life. They must be using heuristics, which are examples of both substitution and WYSIATI. Although their view of their life was influenced by a question about dating or by a coin on the copying machine, the participants in these studies did not forget that there is more to life than dating or feeling lucky. The concept of happiness is not suddenly changed by finding a dime, but System 1 readily substitutes a small part of it for the whole of it. Any aspect of life to which attention is directed will loom large in a global evaluation. This is the essence of the focusing illusion, which can be described in a single sentence: Nothing in life is as important as you think it is when you are thinking about it. The origin of this idea was a family debate about moving from California to Princeton, in which my wife claimed that people are happier in California than on the East Coast. I argued that climate is demonstrably not an important determinant of well-being—the Scandinavian countries are probably the happiest in the world. I observed that permanent life circumstances have little effect on well-being and tried in vain to convince my wife that her intuitions about the happiness of Californians were an error of affective forecasting. A short time later, with this debate still on my mind, I participated in a workshop about the social science of global warming. A colleague made an
argument that was based on his view of the well-being of the population of planet Earth in the next century. I argued that it was preposterous to forecast what it would be like to live on a warmer planet when we did not even know what it is like to live in California. Soon after that exchange, my colleague David Schkade and I were granted research funds to study two questions: Are people who live in California happier than others? and What are the popular beliefs about the relative happiness of Californians? We recruited large samples of students at major state universities in California, Ohio, and Michigan. From some of them we obtained a detailed report of their satisfaction with various aspects of their lives. From others we obtained a prediction of how someone “with your interests and values” who lived elsewhere would complete the same questionnaire. As we analyzed the data, it became obvious that I had won the family argument. As expected, the students in the two regions differed greatly in their attitude to their climate: the Californians enjoyed their climate and the Midwesterners despised theirs. But climate was not an important determinant of well-being. Indeed, there was no difference whatsoever between the life satisfaction of students in California and in the Midwest. We also found that my wife was not alone in her belief that Californians enjoy greater well-being than others. The students in both regions shared the same mistaken view, and we were able to trace their error to an exaggerated belief in the importance of climate. We described the error as a focusing illusion. The essence of the focusing illusion is WYSIATI, giving too much weight to the climate, too little to all the other determinants of well-being. To appreciate how strong this illusion is, take a few seconds to consider the question: How much pleasure do you get from your car? An answer came to your mind immediately; you know how much you like and enjoy your car. Now examine a different question: “When do you get pleasure from your car?” The answer to this question may surprise you, but it is straightforward: you get pleasure (or displeasure) from your car when you think about your car, which is probably not very often. Under normal circumstances, you do not spend much time thinking about your car when you are driving it. You think of other things as you drive, and your mood is determined by whatever you think about. Here again, when you tried to rate
how much you enjoyed your car, you actually answered a much narrower question: “How much pleasure do you get from your car when you think about it?” The substitution caused you to ignore the fact that you rarely think about your car, a form of duration neglect. The upshot is a focusing illusion. If you like your car, you are likely to exaggerate the pleasure you derive from it, which will mislead you when you think of the virtues of your current vehicle as well as when you contemplate buying a new one. A similar bias distorts judgments of the happiness of Californians. When asked about the happiness of Californians, you probably conjure an image of someone attending to a distinctive aspect of the California experience, such as hiking in the summer or admiring the mild winter weather. The focusing illusion arises because Californians actually spend little time attending to these aspects of their life. Moreover, long-term Californians are unlikely to be reminded of the climate when asked for a global evaluation of their life. If you have been there all your life and do not travel much, living in California is like having ten toes: nice, but not something one thinks much about. Thoughts of any aspect of life are more likely to be salient if a contrasting alternative is highly available. People who recently moved to California will respond differently. Consider an enterprising soul who moved from Ohio to seek happiness in a better climate. For a few years following the move, a question about his satisfaction with life will probably remind him of the move and also evoke thoughts of the contrasting climates in the two states. The comparison will surely favor California, and the attention to that aspect of life may distort its true weight in experience. However, the focusing illusion can also bring comfort. Whether or not the individual is actually happier after the move, he will report himself happier, because thoughts of the climate will make him believe that he is. The focusing illusion can cause people to be wrong about their present state of well-being as well as about the happiness of others, and about their own happiness in the future. What proportion of the day do paraplegics spend in a bad mood? This question almost certainly made you think of a paraplegic who is currently thinking about some aspect of his condition. Your guess about a paraplegic’s mood is therefore likely to be accurate in the early days after a crippling accident; for some time after the event, accident victims think of little else. But over time, with few exceptions, attention is withdrawn from a
new situation as it becomes more familiar. The main exceptions are chronic pain, constant exposure to loud noise, and severe depression. Pain and noise are biologically set to be signals that attract attention, and depression involves a self-reinforcing cycle of miserable thoughts. There is therefore no adaptation to these conditions. Paraplegia, however, is not one of the exceptions: detailed observations show that paraplegics are in a fairly good mood more than half of the time as early as one month following their accident—though their mood is certainly somber when they think about their situation. Most of the time, however, paraplegics work, read, enjoy jokes and friends, and get angry when they read about politics in the newspaper. When they are involved in any of these activities, they are not much different from anyone else, and we can expect the experienced well- being of paraplegics to be near normal much of the time. Adaptation to a new situation, whether good or bad, consists in large part of thinking less and less about it. In that sense, most long-term circumstances of life, including paraplegia and marriage, are part-time states that one inhabits only when one attends to them. One of the privileges of teaching at Princeton is the opportunity to guide bright undergraduates through a research thesis. And one of my favorite experiences in this vein was a project in which Beruria Cohn collected and analyzed data from a survey firm that asked respondents to estimate the proportion of time that paraplegics spend in a bad mood. She split her respondents into two groups: some were told that the crippling accident had occurred a month earlier, some a year earlier. In addition, each respondent indicated whether he or she knew a paraplegic personally. The two groups agreed closely in their judgment about the recent paraplegics: those who knew a paraplegic estimated 75% bad mood; those who had to imagine a paraplegic said 70%. In contrast, the two groups differed sharply in their estimates of the mood of paraplegics a year after the accidents: those who knew a paraplegic offered 41% as their estimate of the time in that bad mood. The estimates of those who were not personally acquainted with a paraplegic averaged 68%. Evidently, those who knew a paraplegic had observed the gradual withdrawal of attention from the condition, but others did not forecast that this adaptation would occur. Judgments about the mood of lottery winners one month and one year after the event showed exactly the same pattern.
We can expect the life satisfaction of paraplegics and those afflicted by other chronic and burdensome conditions to be low relative to their experienced well-being, because the request to evaluate their lives will inevitably remind them of the life of others and of the life they used to lead. Consistent with this idea, recent studies of colostomy patients have produced dramatic inconsistencies between the patients’ experienced well- being and their evaluations of their lives. Experience sampling shows no difference in experienced happiness between these patients and a healthy population. Yet colostomy patients would be willing to trade away years of their life for a shorter life without the colostomy. Furthermore, patients whose colostomy has been reversed remember their time in this condition as awful, and they would give up even more of their remaining life not to have to return to it. Here it appears that the remembering self is subject to a massive focusing illusion about the life that the experiencing self endures quite comfortably. Daniel Gilbert and Timothy Wilson introduced the word miswanting to describe bad choices that arise from errors of affective forecasting. This word deserves to be in everyday language. The focusing illusion (which Gilbert and Wilson call focalism) is a rich source of miswanting. In particular, it makes us prone to exaggerate the effect of significant purchases or changed circumstances on our future well-being. Compare two commitments that will change some aspects of your life: buying a comfortable new car and joining a group that meets weekly, perhaps a poker or book club. Both experiences will be novel and exciting at the start. The crucial difference is that you will eventually pay little attention to the car as you drive it, but you will always attend to the social interaction to which you committed yourself. By WYSIATI, you are likely to exaggerate the long-term benefits of the car, but you are not likely to make the same mistake for a social gathering or for inherently attention- demanding activities such as playing tennis or learning to play the cello. The focusing illusion creates a bias in favor of goods and experiences that are initially exciting, even if they will eventually lose their appeal. Time is neglected, causing experiences that will retain their attention value in the long term to be appreciated less than they deserve to be. TIME AND TIME AGAIN
The role of time has been a refrain in this part of the book. It is logical to describe the life of the experiencing self as a series of moments, each with a value. The value of an episode—I have called it a hedonimeter total—is simply the sum of the values of its moments. But this is not how the mind represents episodes. The remembering self, as I have described it, also tells stories and makes choices, and neither the stories nor the choices properly represent time. In storytelling mode, an episode is represented by a few critical moments, especially the beginning, the peak, and the end. Duration is neglected. We saw this focus on singular moments both in the cold-hand situation and in Violetta’s story. We saw a different form of duration neglect in prospect theory, in which a state is represented by the transition to it. Winning a lottery yields a new state of wealth that will endure for some time, but decision utility corresponds to the anticipated intensity of the reaction to the news that one has won. The withdrawal of attention and other adaptations to the new state are neglected, as only that thin slice of time is considered. The same focus on the transition to the new state and the same neglect of time and adaptation are found in forecasts of the reaction to chronic diseases, and of course in the focusing illusion. The mistake that people make in the focusing illusion involves attention to selected moments and neglect of what happens at other times. The mind is good with stories, but it does not appear to be well designed for the processing of time. During the last ten years we have learned many new facts about happiness. But we have also learned that the word happiness does not have a simple meaning and should not be used as if it does. Sometimes scientific progress leaves us more puzzled than we were before. SPEAKING OF THINKING ABOUT LIFE “She thought that buying a fancy car would make her happier, but it turned out to be an error of affective forecasting.” “His car broke down on the way to work this morning and he’s in a foul mood. This is not a good day to ask him about his job satisfaction!” “She looks quite cheerful most of the time, but when she is asked she says she is very unhappy. The question must make her think of her recent divorce.” “Buying a larger house may not make us happier in the long term. We could be suffering from a focusing illusion.”
“He has chosen to split his time between two cities. Probably a serious case of miswanting.”
Conclusions I began this book by introducing two fictitious characters, spent some time discussing two species, and ended with two selves. The two characters were the intuitive System 1, which does the fast thinking, and the effortful and slower System 2, which does the slow thinking, monitors System 1, and maintains control as best it can within its limited resources. The two species were the fictitious Econs, who live in the land of theory, and the Humans, who act in the real world. The two selves are the experiencing self, which does the living, and the remembering self, which keeps score and makes the choices. In this final chapter I consider some applications of the three distinctions, taking them in reverse order. TWO SELVES The possibility of conflicts between the remembering self and the interests of the experiencing self turned out to be a harder problem than I initially thought. In an early experiment, the cold-hand study, the combination of duration neglect and the peak-end rule led to choices that were manifestly absurd. Why would people willingly expose themselves to unnecessary pain? Our subjects left the choice to their remembering self, preferring to repeat the trial that left the better memory, although it involved more pain. Choosing by the quality of the memory may be justified in extreme cases, for example when post-traumatic stress is a possibility, but the cold-hand experience was not traumatic. An objective observer making the choice for someone else would undoubtedly choose the short exposure, favoring the sufferer’s experiencing self. The choices that people made on their own behalf are fairly described as mistakes. Duration neglect and the peak-end rule in the evaluation of stories, both at the opera and in judgments of Jen’s life, are equally indefensible. It does not make sense to evaluate an entire life by its last moments, or to give no weight to duration in deciding which life is more desirable.
The remembering self is a construction of System 2. However, the distinctive features of the way it evaluates episodes and lives are characteristics of our memory. Duration neglect and the peak-end rule originate in System 1 and do not necessarily correspond to the values of System 2. We believe that duration is important, but our memory tells us it is not. The rules that govern the evaluation of the past are poor guides for decision making, because time does matter. The central fact of our existence is that time is the ultimate finite resource, but the remembering self ignores that reality. The neglect of duration combined with the peak-end rule causes a bias that favors a short period of intense joy over a long period of moderate happiness. The mirror image of the same bias makes us fear a short period of intense but tolerable suffering more than we fear a much longer period of moderate pain. Duration neglect also makes us prone to accept a long period of mild unpleasantness because the end will be better, and it favors giving up an opportunity for a long happy period if it is likely to have a poor ending. To drive the same idea to the point of discomfort, consider the common admonition, “Don’t do it, you will regret it.” The advice sounds wise because anticipated regret is the verdict of the remembering self and we are inclined to accept such judgments as final and conclusive. We should not forget, however, that the perspective of the remembering self is not always correct. An objective observer of the hedonimeter profile, with the interests of the experiencing self in mind, might well offer different advice. The remembering self’s neglect of duration, its exaggerated emphasis on peaks and ends, and its susceptibility to hindsight combine to yield distorted reflections of our actual experience. In contrast, the duration-weighted conception of well-being treats all moments of life alike, memorable or not. Some moments end up weighted more than others, either because they are memorable or because they are important. The time that people spend dwelling on a memorable moment should be included in its duration, adding to its weight. A moment can also gain importance by altering the experience of subsequent moments. For example, an hour spent practicing the violin may enhance the experience of many hours of playing or listening to music years later. Similarly, a brief awful event that causes PTSD should be weighted by the total duration of the long-term misery it causes. In the duration-weighted perspective, we can determine only after the fact that a moment is memorable or meaningful. The statements “I will always remember …” or “this is a
meaningful moment” should be taken as promises or predictions, which can be false—and often are—even when uttered with complete sincerity. It is a good bet that many of the things we say we will always remember will be long forgotten ten years later. The logic of duration weighting is compelling, but it cannot be considered a complete theory of well-being because individuals identify with their remembering self and care about their story. A theory of well- being that ignores what people want cannot be sustained. On the other hand, a theory that ignores what actually happens in people’s lives and focuses exclusively on what they think about their life is not tenable either. The remembering self and the experiencing self must both be considered, because their interests do not always coincide. Philosophers could struggle with these questions for a long time. The issue of which of the two selves matters more is not a question only for philosophers; it has implications for policies in several domains, notably medicine and welfare. Consider the investment that should be made in the treatment of various medical conditions, including blindness, deafness, or kidney failure. Should the investments be determined by how much people fear these conditions? Should investments be guided by the suffering that patients actually experience? Or should they follow the intensity of the patients’ desire to be relieved from their condition and by the sacrifices that they would be willing to make to achieve that relief? The ranking of blindness and deafness, or of colostomy and dialysis, might well be different depending on which measure of the severity of suffering is used. No easy solution is in sight, but the issue is too important to be ignored. The possibility of using measures of well-being as indicators to guide government policies has attracted considerable recent interest, both among academics and in several governments in Europe. It is now conceivable, as it was not even a few years ago, that an index of the amount of suffering in society will someday be included in national statistics, along with measures of unemployment, physical disability, and income. This project has come a long way. ECONS AND HUMANS In everyday speech, we call people reasonable if it is possible to reason with them, if their beliefs are generally in tune with reality, and if their preferences are in line with their interests and their values. The word
rational conveys an image of greater deliberation, more calculation, and less warmth, but in common language a rational person is certainly reasonable. For economists and decision theorists, the adjective has an altogether different meaning. The only test of rationality is not whether a person’s beliefs and preferences are reasonable, but whether they are internally consistent. A rational person can believe in ghosts so long as all her other beliefs are consistent with the existence of ghosts. A rational person can prefer being hated over being loved, so long as his preferences are consistent. Rationality is logical coherence—reasonable or not. Econs are rational by this definition, but there is overwhelming evidence that Humans cannot be. An Econ would not be susceptible to priming, WYSIATI, narrow framing, the inside view, or preference reversals, which Humans cannot consistently avoid. The definition of rationality as coherence is impossibly restrictive; it demands adherence to rules of logic that a finite mind is not able to implement. Reasonable people cannot be rational by that definition, but they should not be branded as irrational for that reason. Irrational is a strong word, which connotes impulsivity, emotionality, and a stubborn resistance to reasonable argument. I often cringe when my work with Amos is credited with demonstrating that human choices are irrational, when in fact our research only showed that Humans are not well described by the rational-agent model. Although Humans are not irrational, they often need help to make more accurate judgments and better decisions, and in some cases policies and institutions can provide that help. These claims may seem innocuous, but they are in fact quite controversial. As interpreted by the important Chicago school of economics, faith in human rationality is closely linked to an ideology in which it is unnecessary and even immoral to protect people against their choices. Rational people should be free, and they should be responsible for taking care of themselves. Milton Friedman, the leading figure in that school, expressed this view in the title of one of his popular books: Free to Choose. The assumption that agents are rational provides the intellectual foundation for the libertarian approach to public policy: do not interfere with the individual’s right to choose, unless the choices harm others. Libertarian policies are further bolstered by admiration for the efficiency of markets in allocating goods to the people who are willing to pay the most
for them. A famous example of the Chicago approach is titled A Theory of Rational Addiction; it explains how a rational agent with a strong preference for intense and immediate gratification may make the rational decision to accept future addiction as a consequence. I once heard Gary Becker, one of the authors of that article, who is also a Nobel laureate of the Chicago school, argue in a lighter vein, but not entirely as a joke, that we should consider the possibility of explaining the so-called obesity epidemic by people’s belief that a cure for diabetes will soon become available. He was making a valuable point: when we observe people acting in ways that seem odd, we should first examine the possibility that they have a good reason to do what they do. Psychological interpretations should only be invoked when the reasons become implausible—which Becker’s explanation of obesity probably is. In a nation of Econs, government should keep out of the way, allowing the Econs to act as they choose, so long as they do not harm others. If a motorcycle rider chooses to ride without a helmet, a libertarian will support his right to do so. Citizens know what they are doing, even when they choose not to save for their old age, or when they expose themselves to addictive substances. There is sometimes a hard edge to this position: elderly people who did not save enough for retirement get little more sympathy than someone who complains about the bill after consuming a large meal at a restaurant. Much is therefore at stake in the debate between the Chicago school and the behavioral economists, who reject the extreme form of the rational-agent model. Freedom is not a contested value; all the participants in the debate are in favor of it. But life is more complex for behavioral economists than for true believers in human rationality. No behavioral economist favors a state that will force its citizens to eat a balanced diet and to watch only television programs that are good for the soul. For behavioral economists, however, freedom has a cost, which is borne by individuals who make bad choices, and by a society that feels obligated to help them. The decision of whether or not to protect individuals against their mistakes therefore presents a dilemma for behavioral economists. The economists of the Chicago school do not face that problem, because rational agents do not make mistakes. For adherents of this school, freedom is free of charge. In 2008 the economist Richard Thaler and the jurist Cass Sunstein teamed up to write a book, Nudge, which quickly became an international
bestseller and the bible of behavioral economics. Their book introduced several new words into the language, including Econs and Humans. It also presented a set of solutions to the dilemma of how to help people make good decisions without curtailing their freedom. Thaler and Sunstein advocate a position of libertarian paternalism, in which the state and other institutions are allowed to nudge people to make decisions that serve their own long-term interests. The designation of joining a pension plan as the default option is an example of a nudge. It is difficult to argue that anyone’s freedom is diminished by being automatically enrolled in the plan, when they merely have to check a box to opt out. As we saw earlier, the framing of the individual’s decision—Thaler and Sunstein call it choice architecture —has a huge effect on the outcome. The nudge is based on sound psychology, which I described earlier. The default option is naturally perceived as the normal choice. Deviating from the normal choice is an act of commission, which requires more effortful deliberation, takes on more responsibility, and is more likely to evoke regret than doing nothing. These are powerful forces that may guide the decision of someone who is otherwise unsure of what to do. Humans, more than Econs, also need protection from others who deliberately exploit their weaknesses—and especially the quirks of System 1 and the laziness of System 2. Rational agents are assumed to make important decisions carefully, and to use all the information that is provided to them. An Econ will read and understand the fine print of a contract before signing it, but Humans usually do not. An unscrupulous firm that designs contracts that customers will routinely sign without reading has considerable legal leeway in hiding important information in plain sight. A pernicious implication of the rational-agent model in its extreme form is that customers are assumed to need no protection beyond ensuring that the relevant information is disclosed. The size of the print and the complexity of the language in the disclosure are not considered relevant—an Econ knows how to deal with small print when it matters. In contrast, the recommendations of Nudge require firms to offer contracts that are sufficiently simple to be read and understood by Human customers. It is a good sign that some of these recommendations have encountered significant opposition from firms whose profits might suffer if their customers were better informed. A world in which firms compete by offering better
products is preferable to one in which the winner is the firm that is best at obfuscation. A remarkable feature of libertarian paternalism is its appeal across a broad political spectrum. The flagship example of behavioral policy, called Save More Tomorrow, was sponsored in Congress by an unusual coalition that included extreme conservatives as well as liberals. Save More Tomorrow is a financial plan that firms can offer their employees. Those who sign on allow the employer to increase their contribution to their saving plan by a fixed proportion whenever they receive a raise. The increased saving rate is implemented automatically until the employee gives notice that she wants to opt out of it. This brilliant innovation, proposed by Richard Thaler and Shlomo Benartzi in 2003, has now improved the savings rate and brightened the future prospects of millions of workers. It is soundly based in the psychological principles that readers of this book will recognize. It avoids the resistance to an immediate loss by requiring no immediate change; by tying increased saving to pay raises, it turns losses into foregone gains, which are much easier to bear; and the feature of automaticity aligns the laziness of System 2 with the long-term interests of the workers. All this, of course, without compelling anyone to do anything he does not wish to do and without any misdirection or artifice. The appeal of libertarian paternalism has been recognized in many countries, including the UK and South Korea, and by politicians of many stripes, including Tories and the Democratic administration of President Obama. Indeed, Britain’s government has created a new small unit whose mission is to apply the principles of behavioral science to help the government better accomplish its goals. The official name for this group is the Behavioural Insight Team, but it is known both in and out of government simply as the Nudge Unit. Thaler is an adviser to this team. In a storybook sequel to the writing of Nudge, Sunstein was invited by President Obama to serve as administrator of the Office of Information and Regulatory Affairs, a position that gave him considerable opportunity to encourage the application of the lessons of psychology and behavioral economics in government agencies. The mission is described in the 2010 Report of the Office of Management and Budget. Readers of this book will appreciate the logic behind specific recommendations, including encouraging “clear, simple, salient, and meaningful disclosures.” They will also recognize background statements such as “presentation greatly matters;
if, for example, a potential outcome is framed as a loss, it may have more impact than if it is presented as a gain.” The example of a regulation about the framing of disclosures concerning fuel consumption was mentioned earlier. Additional applications that have been implemented include automatic enrollment in health insurance, a new version of the dietary guidelines that replaces the incomprehensible Food Pyramid with the powerful image of a Food Plate loaded with a balanced diet, and a rule formulated by the USDA that permits the inclusion of messages such as “90% fat-free” on the label of meat products, provided that the statement “10% fat” is also displayed “contiguous to, in lettering of the same color, size, and type as, and on the same color background as, the statement of lean percentage.” Humans, unlike Econs, need help to make good decisions, and there are informed and unintrusive ways to provide that help. TWO SYSTEMS This book has described the workings of the mind as an uneasy interaction between two fictitious characters: the automatic System 1 and the effortful System 2. You are now quite familiar with the personalities of the two systems and able to anticipate how they might respond in different situations. And of course you also remember that the two systems do not really exist in the brain or anywhere else. “System 1 does X” is a shortcut for “X occurs automatically.” And “System 2 is mobilized to do Y” is a shortcut for “arousal increases, pupils dilate, attention is focused, and activity Y is performed.” I hope you find the language of systems as helpful as I do, and that you have acquired an intuitive sense of how they work without getting confused by the question of whether they exist. Having delivered this necessary warning, I will continue to use the language to the end. The attentive System 2 is who we think we are. System 2 articulates judgments and makes choices, but it often endorses or rationalizes ideas and feelings that were generated by System 1. You may not know that you are optimistic about a project because something about its leader reminds you of your beloved sister, or that you dislike a person who looks vaguely like your dentist. If asked for an explanation, however, you will search your memory for presentable reasons and will certainly find some. Moreover, you will believe the story you make up. But System 2 is not merely an
apologist for System 1; it also prevents many foolish thoughts and inappropriate impulses from overt expression. The investment of attention improves performance in numerous activities—think of the risks of driving through a narrow space while your mind is wandering—and is essential to some tasks, including comparison, choice, and ordered reasoning. However, System 2 is not a paragon of rationality. Its abilities are limited and so is the knowledge to which it has access. We do not always think straight when we reason, and the errors are not always due to intrusive and incorrect intuitions. Often we make mistakes because we (our System 2) do not know any better. I have spent more time describing System 1, and have devoted many pages to errors of intuitive judgment and choice that I attribute to it. However, the relative number of pages is a poor indicator of the balance between the marvels and the flaws of intuitive thinking. System 1 is indeed the origin of much that we do wrong, but it is also the origin of most of what we do right—which is most of what we do. Our thoughts and actions are routinely guided by System 1 and generally are on the mark. One of the marvels is the rich and detailed model of our world that is maintained in associative memory: it distinguishes surprising from normal events in a fraction of a second, immediately generates an idea of what was expected instead of a surprise, and automatically searches for some causal interpretation of surprises and of events as they take place. Memory also holds the vast repertory of skills we have acquired in a lifetime of practice, which automatically produce adequate solutions to challenges as they arise, from walking around a large stone on the path to averting the incipient outburst of a customer. The acquisition of skills requires a regular environment, an adequate opportunity to practice, and rapid and unequivocal feedback about the correctness of thoughts and actions. When these conditions are fulfilled, skill eventually develops, and the intuitive judgments and choices that quickly come to mind will mostly be accurate. All this is the work of System 1, which means it occurs automatically and fast. A marker of skilled performance is the ability to deal with vast amounts of information swiftly and efficiently. When a challenge is encountered to which a skilled response is available, that response is evoked. What happens in the absence of skill? Sometimes, as in the problem 17 × 24 = ?, which calls for a specific answer, it is immediately apparent that System 2 must be called in. But it is rare for
System 1 to be dumbfounded. System 1 is not constrained by capacity limits and is profligate in its computations. When engaged in searching for an answer to one question, it simultaneously generates the answers to related questions, and it may substitute a response that more easily comes to mind for the one that was requested. In this conception of heuristics, the heuristic answer is not necessarily simpler or more frugal than the original question—it is only more accessible, computed more quickly and easily. The heuristic answers are not random, and they are often approximately correct. And sometimes they are quite wrong. System 1 registers the cognitive ease with which it processes information, but it does not generate a warning signal when it becomes unreliable. Intuitive answers come to mind quickly and confidently, whether they originate from skills or from heuristics. There is no simple way for System 2 to distinguish between a skilled and a heuristic response. Its only recourse is to slow down and attempt to construct an answer on its own, which it is reluctant to do because it is indolent. Many suggestions of System 1 are casually endorsed with minimal checking, as in the bat-and- ball problem. This is how System 1 acquires its bad reputation as the source of errors and biases. Its operative features, which include WYSIATI, intensity matching, and associative coherence, among others, give rise to predictable biases and to cognitive illusions such as anchoring, nonregressive predictions, overconfidence, and numerous others. What can be done about biases? How can we improve judgments and decisions, both our own and those of the institutions that we serve and that serve us? The short answer is that little can be achieved without a considerable investment of effort. As I know from experience, System 1 is not readily educable. Except for some effects that I attribute mostly to age, my intuitive thinking is just as prone to overconfidence, extreme predictions, and the planning fallacy as it was before I made a study of these issues. I have improved only in my ability to recognize situations in which errors are likely: “This number will be an anchor …,” “The decision could change if the problem is reframed …” And I have made much more progress in recognizing the errors of others than my own. The way to block errors that originate in System 1 is simple in principle: recognize the signs that you are in a cognitive minefield, slow down, and ask for reinforcement from System 2. This is how you will proceed when you next encounter the Müller-Lyer illusion. When you see lines with fins
pointing in different directions, you will recognize the situation as one in which you should not trust your impressions of length. Unfortunately, this sensible procedure is least likely to be applied when it is needed most. We would all like to have a warning bell that rings loudly whenever we are about to make a serious error, but no such bell is available, and cognitive illusions are generally more difficult to recognize than perceptual illusions. The voice of reason may be much fainter than the loud and clear voice of an erroneous intuition, and questioning your intuitions is unpleasant when you face the stress of a big decision. More doubt is the last thing you want when you are in trouble. The upshot is that it is much easier to identify a minefield when you observe others wandering into it than when you are about to do so. Observers are less cognitively busy and more open to information than actors. That was my reason for writing a book that is oriented to critics and gossipers rather than to decision makers. Organizations are better than individuals when it comes to avoiding errors, because they naturally think more slowly and have the power to impose orderly procedures. Organizations can institute and enforce the application of useful checklists, as well as more elaborate exercises, such as reference-class forecasting and the premortem. At least in part by providing a distinctive vocabulary, organizations can also encourage a culture in which people watch out for one another as they approach minefields. Whatever else it produces, an organization is a factory that manufactures judgments and decisions. Every factory must have ways to ensure the quality of its products in the initial design, in fabrication, and in final inspections. The corresponding stages in the production of decisions are the framing of the problem that is to be solved, the collection of relevant information leading to a decision, and reflection and review. An organization that seeks to improve its decision product should routinely look for efficiency improvements at each of these stages. The operative concept is routine. Constant quality control is an alternative to the wholesale reviews of processes that organizations commonly undertake in the wake of disasters. There is much to be done to improve decision making. One example out of many is the remarkable absence of systematic training for the essential skill of conducting efficient meetings. Ultimately, a richer language is essential to the skill of constructive criticism. Much like medicine, the identification of judgment errors is a diagnostic task, which requires a precise vocabulary. The name of a disease
is a hook to which all that is known about the disease is attached, including vulnerabilities, environmental factors, symptoms, prognosis, and care. Similarly, labels such as “anchoring effects,” “narrow framing,” or “excessive coherence” bring together in memory everything we know about a bias, its causes, its effects, and what can be done about it. There is a direct link from more precise gossip at the watercooler to better decisions. Decision makers are sometimes better able to imagine the voices of present gossipers and future critics than to hear the hesitant voice of their own doubts. They will make better choices when they trust their critics to be sophisticated and fair, and when they expect their decision to be judged by how it was made, not only by how it turned out.
Appendix A: Judgment Under Uncertainty: Heuristics and Biases fn1 Amos Tversky and Daniel Kahneman Many decisions are based on beliefs concerning the likelihood of uncertain events such as the outcome of an election, the guilt of a defendant, or the future value of the dollar. These beliefs are usually expressed in statements such as “I think that …,” “chances are …,” “it is unlikely that …,” and so forth. Occasionally, beliefs concerning uncertain events are expressed in numerical form as odds or subjective probabilities. What determines such beliefs? How do people assess the probability of an uncertain event or the value of an uncertain quantity? This article shows that people rely on a limited number of heuristic principles which reduce the complex tasks of assessing probabilities and predicting values to simpler judgmental operations. In general, these heuristics are quite useful, but sometimes they lead to severe and systematic errors. The subjective assessment of probability resembles the subjective assessment of physical quantities such as distance or size. These judgments are all based on data of limited validity, which are processed according to heuristic rules. For example, the apparent distance of an object is determined in part by its clarity. The more sharply the object is seen, the closer it appears to be. This rule has some validity, because in any given scene the more distant objects are seen less sharply than nearer objects. However, the reliance on this rule leads to systematic errors in the estimation of distance. Specifically, distances are often overestimated when visibility is poor because the contours of objects are blurred. On the other hand, distances are often underestimated when visibility is good because the objects are seen sharply. Thus, the reliance on clarity as an indication of distance leads to common biases. Such biases are also found in the intuitive judgment of probability. This article describes three heuristics that are employed to assess probabilities and to predict values. Biases to which
these heuristics lead are enumerated, and the applied and theoretical implications of these observations are discussed. REPRESENTATIVENESS Many of the probabilistic questions with which people are concerned belong to one of the following types: What is the probability that object A belongs to class B? What is the probability that event A originates from process B? What is the probability that process B will generate event A? In answering such questions, people typically rely on the representativeness heuristic, in which probabilities are evaluated by the degree to which A is representative of B, that is, by the degree to which A resembles B. For example, when A is highly representative of B, the probability that A originates from B is judged to be high. On the other hand, if A is not similar to B, the probability that A originates from B is judged to be low. For an illustration of judgment by representativeness, consider an individual who has been described by a former neighbor as follows: “Steve is very shy and withdrawn, invariably helpful, but with little interest in people, or in the world of reality. A meek and tidy soul, he has a need for order and structure, and a passion for detail.” How do people assess the probability that Steve is engaged in a particular occupation from a list of possibilities (for example, farmer, salesman, airline pilot, librarian, or physician)? How do people order these occupations from most to least likely? In the representativeness heuristic, the probability that Steve is a librarian, for example, is assessed by the degree to which he is representative of, or similar to, the stereotype of a librarian. Indeed, research with problems of this type has shown that people order the occupations by probability and by similarity in exactly the same way.1 This approach to the judgment of probability leads to serious errors, because similarity, or representativeness, is not influenced by several factors that should affect judgments of probability. Insensitivity to prior probability of outcomes. One of the factors that have no effect on representativeness but should have a major effect on probability is the prior probability, or base-rate frequency, of the outcomes. In the case of Steve, for example, the fact that there are many more farmers than librarians in the population should enter into any reasonable estimate of the probability that Steve is a librarian rather than a farmer. Considerations of base-rate frequency, however, do not affect the similarity
of Steve to the stereotypes of librarians and farmers. If people evaluate probability by representativeness, therefore, prior probabilities will be neglected. This hypothesis was tested in an experiment where prior probabilities were manipulated.2 Subjects were shown brief personality descriptions of several individuals, allegedly sampled at random from a group of 100 professionals—engineers and lawyers. The subjects were asked to assess, for each description, the probability that it belonged to an engineer rather than to a lawyer. In one experimental condition, subjects were told that the group from which the descriptions had been drawn consisted of 70 engineers and 30 lawyers. In another condition, subjects were told that the group consisted of 30 engineers and 70 lawyers. The odds that any particular description belongs to an engineer rather than to a lawyer should be higher in the first condition, where there is a majority of engineers, than in the second condition, where there is a majority of lawyers. Specifically, it can be shown by applying Bayes’ rule that the ratio of these odds should be (.7/.3)2, or 5.44, for each description. In a sharp violation of Bayes’ rule, the subjects in the two conditions produced essentially the same probability judgments. Apparently, subjects evaluated the likelihood that a particular description belonged to an engineer rather than to a lawyer by the degree to which this description was representative of the two stereotypes, with little or no regard for the prior probabilities of the categories. The subjects used prior probabilities correctly when they had no other information. In the absence of a personality sketch, they judged the probability that an unknown individual is an engineer to be .7 and .3, respectively, in the two base-rate conditions. However, prior probabilities were effectively ignored when a description was introduced, even when this description was totally uninformative. The responses to the following description illustrate this phenomenon: Dick is a 30-year-old man. He is married with no children. A man of high ability and high motivation, he promises to be quite successful in his field. He is well liked by his colleagues. This description was intended to convey no information relevant to the question of whether Dick is an engineer or a lawyer. Consequently, the probability that Dick is an engineer should equal the proportion of engineers in the group, as if no description had been given. The subjects, however, judged the probability of Dick being an engineer to be .5
regardless of whether the stated proportion of engineers in the group was .7 or .3. Evidently, people respond differently when given no evidence and when given worthless evidence. When no specific evidence is given, prior probabilities are properly utilized; when worthless evidence is given, prior probabilities are ignored.3 Insensitivity to sample size. To evaluate the probability of obtaining a particular result in a sample drawn from a specified population, people typically apply the representativeness heuristic. That is, they assess the likelihood of a sample result, for example, that the average height in a random sample of ten men will be 6 feet, by the similarity of this result to the corresponding parameter (that is, to the average height in the population of men). The similarity of a sample statistic to a population parameter does not depend on the size of the sample. Consequently, if probabilities are assessed by representativeness, then the judged probability of a sample statistic will be essentially independent of sample size. Indeed, when subjects assessed the distributions of average height for samples of various sizes, they produced identical distributions. For example, the probability of obtaining an average height greater than 6 feet was assigned the same value for samples of 1,000, 100, and 10 men.4 Moreover, subjects failed to appreciate the role of sample size even when it was emphasized in the formulation of the problem. Consider the following question: A certain town is served by two hospitals. In the larger hospital about 45 babies are born each day, and in the smaller hospital about 15 babies are born each day. As you know, about 50% of all babies are boys. However, the exact percentage varies from day to day. Sometimes it may be higher than 50%, sometimes lower. For a period of 1 year, each hospital recorded the days on which more than 60% of the babies born were boys. Which hospital do you think recorded more such days? The larger hospital (21) The smaller hospital (21) About the same (that is, within 5% of each other) (53) The values in parentheses are the number of undergraduate students who chose each answer. Most subjects judged the probability of obtaining more than 60% boys to be the same in the small and in the large hospital, presumably because these events are described by the same statistic and are therefore equally representative of the general population. In contrast, sampling theory entails that the expected number of days on which more than 60% of the babies are boys is much greater in the small hospital than in the large one, because a
large sample is less likely to stray from 50%. This fundamental notion of statistics is evidently not part of people’s repertoire of intuitions. A similar insensitivity to sample size has been reported in judgments of posterior probability, that is, of the probability that a sample has been drawn from one population rather than from another. Consider the following example: Imagine an urn filled with balls, of which ⅔ are of one color and ⅓ of another. One individual has drawn 5 balls from the urn, and found that 4 were red and 1 was white. Another individual has drawn 20 balls and found that 12 were red and 8 were white. Which of the two individuals should feel more confident that the urn contains ⅔ red balls and ⅓ white balls, rather than the opposite? What odds should each individual give? In this problem, the correct posterior odds are 8 to 1 for the 4:1 sample and 16 to 1 for the 12:8 sample, assuming equal prior probabilities. However, most people feel that the first sample provides much stronger evidence for the hypothesis that the urn is predominantly red, because the proportion of red balls is larger in the first than in the second sample. Here again, intuitive judgments are dominated by the sample proportion and are essentially unaffected by the size of the sample, which plays a crucial role in the determination of the actual posterior odds.5 In addition, intuitive estimates of posterior odds are far less extreme than the correct values. The underestimation of the impact of evidence has been observed repeatedly in problems of this type.6 It has been labeled “conservatism.” Misconceptions of chance. People expect that a sequence of events generated by a random process will represent the essential characteristics of that process even when the sequence is short. In considering tosses of a coin for heads or tails, for example, people regard the sequence H-T-H-T-T-H to be more likely than the sequence H-H-H-T-T-T, which does not appear random, and also more likely than the sequence H-H-H-H-T-H, which does not represent the fairness of the coin.7 Thus, people expect that the essential characteristics of the process will be represented, not only globally in the entire sequence, but also locally in each of its parts. A locally representative sequence, however, deviates systematically from chance expectation: it contains too many alternations and too few runs. Another consequence of the belief in local representativeness is the well-known gambler’s fallacy. After observing a long run of red on the roulette wheel, for example, most people erroneously believe that black is now due, presumably because the occurrence of black will result in a more representative sequence than the occurrence of an additional red. Chance is commonly viewed as a self-
correcting process in which a deviation in one direction induces a deviation in the opposite direction to restore the equilibrium. In fact, deviations are not “corrected” as a chance process unfolds, they are merely diluted. Misconceptions of chance are not limited to naive subjects. A study of the statistical intuitions of experienced research psychologists8 revealed a lingering belief in what may be called the “law of small numbers,” according to which even small samples are highly representative of the populations from which they are drawn. The responses of these investigators reflected the expectation that a valid hypothesis about a population will be represented by a statistically significant result in a sample with little regard for its size. As a consequence, the researchers put too much faith in the results of small samples and grossly overestimated the replicability of such results. In the actual conduct of research, this bias leads to the selection of samples of inadequate size and to overinterpretation of findings. Insensitivity to predictability. People are sometimes called upon to make such numerical predictions as the future value of a stock, the demand for a commodity, or the outcome of a football game. Such predictions are often made by representativeness. For example, suppose one is given a description of a company and is asked to predict its future profit. If the description of the company is very favorable, a very high profit will appear most representative of that description; if the description is mediocre, a mediocre performance will appear most representative. The degree to which the description is favorable is unaffected by the reliability of that description or by the degree to which it permits accurate prediction. Hence, if people predict solely in terms of the favorableness of the description, their predictions will be insensitive to the reliability of the evidence and to the expected accuracy of the prediction. This mode of judgment violates the normative statistical theory in which the extremeness and the range of predictions are controlled by considerations of predictability. When predictability is nil, the same prediction should be made in all cases. For example, if the descriptions of companies provide no information relevant to profit, then the same value (such as average profit) should be predicted for all companies. If predictability is perfect, of course, the values predicted will match the actual values and the range of predictions will equal the range of outcomes.
In general, the higher the predictability, the wider the range of predicted values. Several studies of numerical prediction have demonstrated that intuitive predictions violate this rule, and that subjects show little or no regard for considerations of predictability.9 In one of these studies, subjects were presented with several paragraphs, each describing the performance of a student teacher during a particular practice lesson. Some subjects were asked to evaluate the quality of the lesson described in the paragraph in percentile scores, relative to a specified population. Other subjects were asked to predict, also in percentile scores, the standing of each student teacher 5 years after the practice lesson. The judgments made under the two conditions were identical. That is, the prediction of a remote criterion (success of a teacher after 5 years) was identical to the evaluation of the information on which the prediction was based (the quality of the practice lesson). The students who made these predictions were undoubtedly aware of the limited predictability of teaching competence on the basis of a single trial lesson 5 years earlier; nevertheless, their predictions were as extreme as their evaluations. The illusion of validity. As we have seen, people often predict by selecting the outcome (for example, an occupation) that is most representative of the input (for example, the description of a person). The confidence they have in their prediction depends primarily on the degree of representativeness (that is, on the quality of the match between the selected outcome and the input) with little or no regard for the factors that limit predictive accuracy. Thus, people express great confidence in the prediction that a person is a librarian when given a description of his personality which matches the stereotype of librarians, even if the description is scanty, unreliable, or outdated. The unwarranted confidence which is produced by a good fit between the predicted outcome and the input information may be called the illusion of validity. This illusion persists even when the judge is aware of the factors that limit the accuracy of his predictions. It is a common observation that psychologists who conduct selection interviews often experience considerable confidence in their predictions, even when they know of the vast literature that shows selection interviews to be highly fallible. The continued reliance on the clinical interview for selection, despite repeated demonstrations of its inadequacy, amply attests to the strength of this effect.
The internal consistency of a pattern of inputs is a major determinant of one’s confidence in predictions based on these inputs. For example, people express more confidence in predicting the final grade point average of a student whose first-year record consists entirely of B’s than in predicting the grade point average of a student whose first-year record includes many A’s and C’s. Highly consistent patterns are most often observed when the input variables are highly redundant or correlated. Hence, people tend to have great confidence in predictions based on redundant input variables. However, an elementary result in the statistics of correlation asserts that, given input variables of stated validity, a prediction based on several such inputs can achieve higher accuracy when they are independent of each other than when they are redundant or correlated. Thus, redundancy among inputs decreases accuracy even as it increases confidence, and people are often confident in predictions that are quite likely to be off the mark.10 Misconceptions of regression. Suppose a large group of children has been examined on two equivalent versions of an aptitude test. If one selects ten children from among those who did best on one of the two versions, he will usually find their performance on the second version to be somewhat disappointing. Conversely, if one selects ten children from among those who did worst on one version, they will be found, on the average, to do somewhat better on the other version. More generally, consider two variables X and Y which have the same distribution. If one selects individuals whose average X score deviates from the mean of X by k units, then the average of their Y scores will usually deviate from the mean of Y by less than k units. These observations illustrate a general phenomenon known as regression toward the mean, which was first documented by Galton more than 100 years ago. In the normal course of life, one encounters many instances of regression toward the mean, in the comparison of the height of fathers and sons, of the intelligence of husbands and wives, or of the performance of individuals on consecutive examinations. Nevertheless, people do not develop correct intuitions about this phenomenon. First, they do not expect regression in many contexts where it is bound to occur. Second, when they recognize the occurrence of regression, they often invent spurious causal explanations for it.11 We suggest that the phenomenon of regression remains elusive because it is incompatible with the belief that the predicted outcome should be
maximally representative of the input, and, hence, that the value of the outcome variable should be as extreme as the value of the input variable. The failure to recognize the import of regression can have pernicious consequences, as illustrated by the following observation.12 In a discussion of flight training, experienced instructors noted that praise for an exceptionally smooth landing is typically followed by a poorer landing on the next try, while harsh criticism after a rough landing is usually followed by an improvement on the next try. The instructors concluded that verbal rewards are detrimental to learning, while verbal punishments are beneficial, contrary to accepted psychological doctrine. This conclusion is unwarranted because of the presence of regression toward the mean. As in other cases of repeated examination, an improvement will usually follow a poor performance and a deterioration will usually follow an outstanding performance, even if the instructor does not respond to the trainee’s achievement on the first attempt. Because the instructors had praised their trainees after good landings and admonished them after poor ones, they reached the erroneous and potentially harmful conclusion that punishment is more effective than reward. Thus, the failure to understand the effect of regression leads one to overestimate the effectiveness of punishment and to underestimate the effectiveness of reward. In social interaction, as well as in training, rewards are typically administered when performance is good, and punishments are typically administered when performance is poor. By regression alone, therefore, behavior is most likely to improve after punishment and most likely to deteriorate after reward. Consequently, the human condition is such that, by chance alone, one is most often rewarded for punishing others and most often punished for rewarding them. People are generally not aware of this contingency. In fact, the elusive role of regression in determining the apparent consequences of reward and punishment seems to have escaped the notice of students of this area. AVAILABILITY There are situations in which people assess the frequency of a class or the probability of an event by the ease with which instances or occurrences can be brought to mind. For example, one may assess the risk of heart attack among middle-aged people by recalling such occurrences among one’s acquaintances. Similarly, one may evaluate the probability that a given
business venture will fail by imagining various difficulties it could encounter. This judgmental heuristic is called availability. Availability is a useful clue for assessing frequency or probability, because instances of large classes are usually recalled better and faster than instances of less frequent classes. However, availability is affected by factors other than frequency and probability. Consequently, the reliance on availability leads to predictable biases, some of which are illustrated below. Biases due to the retrievability of instances. When the size of a class is judged by the availability of its instances, a class whose instances are easily retrieved will appear more numerous than a class of equal frequency whose instances are less retrievable. In an elementary demonstration of this effect, subjects heard a list of well-known personalities of both sexes and were subsequently asked to judge whether the list contained more names of men than of women. Different lists were presented to different groups of subjects. In some of the lists the men were relatively more famous than the women, and in others the women were relatively more famous than the men. In each of the lists, the subjects erroneously judged that the class (sex) that had the more famous personalities was the more numerous.13 In addition to familiarity, there are other factors, such as salience, which affect the retrievability of instances. For example, the impact of seeing a house burning on the subjective probability of such accidents is probably greater than the impact of reading about a fire in the local paper. Furthermore, recent occurrences are likely to be relatively more available than earlier occurrences. It is a common experience that the subjective probability of traffic accidents rises temporarily when one sees a car overturned by the side of the road. Biases due to the effectiveness of a search set. Suppose one samples a word (of three letters or more) at random from an English text. Is it more likely that the word starts with r or that r is the third letter? People approach this problem by recalling words that begin with r (road) and words that have r in the third position (car) and assess the relative frequency by the ease with which words of the two types come to mind. Because it is much easier to search for words by their first letter than by their third letter, most people judge words that begin with a given consonant to be more numerous than words in which the same consonant appears in the third position. They do so even for consonants, such as r or k, that are more frequent in the third position than in the first.14
Different tasks elicit different search sets. For example, suppose you are asked to rate the frequency with which abstract words (thought, love) and concrete words (door, water) appear in written English. A natural way to answer this question is to search for contexts in which the word could appear. It seems easier to think of contexts in which an abstract concept is mentioned (love in love stories) than to think of contexts in which a concrete word (such as door) is mentioned. If the frequency of words is judged by the availability of the contexts in which they appear, abstract words will be judged as relatively more numerous than concrete words. This bias has been observed in a recent study15 which showed that the judged frequency of occurrence of abstract words was much higher than that of concrete words, equated in objective frequency. Abstract words were also judged to appear in a much greater variety of contexts than concrete words. Biases of imaginability. Sometimes one has to assess the frequency of a class whose instances are not stored in memory but can be generated according to a given rule. In such situations, one typically generates several instances and evaluates frequency or probability by the ease with which the relevant instances can be constructed. However, the ease of constructing instances does not always reflect their actual frequency, and this mode of evaluation is prone to biases. To illustrate, consider a group of 10 people who form committees of k members, 2 ≤ k ≤ 8. How many different committees of k members can be formed? The correct answer to this problem is given by the binomial coefficient (10/k) which reaches a maximum of 252 for k = 5. Clearly, the number of committees of k members equals the number of committees of (10 − k) members, because any committee of k members defines a unique group of (10 − k) nonmembers. One way to answer this question without computation is to mentally construct committees of k members and to evaluate their number by the ease with which they come to mind. Committees of few members, say 2, are more available than committees of many members, say 8. The simplest scheme for the construction of committees is a partition of the group into disjoint sets. One readily sees that it is easy to construct five disjoint committees of 2 members, while it is impossible to generate even two disjoint committees of 8 members. Consequently, if frequency is assessed by imaginability, or by availability for construction, the small committees
will appear more numerous than larger committees, in contrast to the correct bell-shaped function. Indeed, when naive subjects were asked to estimate the number of distinct committees of various sizes, their estimates were a decreasing monotonic function of committee size.16 For example, the median estimate of the number of committees of 2 members was 70, while the estimate for committees of 8 members was 20 (the correct answer is 45 in both cases). Imaginability plays an important role in the evaluation of probabilities in real-life situations. The risk involved in an adventurous expedition, for example, is evaluated by imagining contingencies with which the expedition is not equipped to cope. If many such difficulties are vividly portrayed, the expedition can be made to appear exceedingly dangerous, although the ease with which disasters are imagined need not reflect their actual likelihood. Conversely, the risk involved in an undertaking may be grossly underestimated if some possible dangers are either difficult to conceive of, or simply do not come to mind. Illusory correlation. Chapman and Chapman17 have described an interesting bias in the judgment of the frequency with which two events co- occur. They presented naive judges with information concerning several hypothetical mental patients. The data for each patient consisted of a clinical diagnosis and a drawing of a person made by the patient. Later the judges estimated the frequency with which each diagnosis (such as paranoia or suspiciousness) had been accompanied by various features of the drawing (such as peculiar eyes). The subjects markedly overestimated the frequency of co-occurrence of natural associates, such as suspiciousness and peculiar eyes. This effect was labeled illusory correlation. In their erroneous judgments of the data to which they had been exposed, naive subjects “rediscovered” much of the common, but unfounded, clinical lore concerning the interpretation of the draw-a-person test. The illusory correlation effect was extremely resistant to contradictory data. It persisted even when the correlation between symptom and diagnosis was actually negative, and it prevented the judges from detecting relationships that were in fact present. Availability provides a natural account for the illusory-correlation effect. The judgment of how frequently two events co-occur could be based on the strength of the associative bond between them. When the association is strong, one is likely to conclude that the events have been frequently paired.
Consequently, strong associates will be judged to have occurred together frequently. According to this view, the illusory correlation between suspiciousness and peculiar drawing of the eyes, for example, is due to the fact that suspiciousness is more readily associated with the eyes than with any other part of the body. Lifelong experience has taught us that, in general, instances of large classes are recalled better and faster than instances of less frequent classes; that likely occurrences are easier to imagine than unlikely ones; and that the associative connections between events are strengthened when the events frequently co-occur. As a result, man has at his disposal a procedure (the availability heuristic) for estimating the numerosity of a class, the likelihood of an event, or the frequency of co-occurrences, by the ease with which the relevant mental operations of retrieval, construction, or association can be performed. However, as the preceding examples have demonstrated, this valuable estimation procedure results in systematic errors. ADJUSTMENT AND ANCHORING In many situations, people make estimates by starting from an initial value that is adjusted to yield the final answer. The initial value, or starting point, may be suggested by the formulation of the problem, or it may be the result of a partial computation. In either case, adjustments are typically insufficient.18 That is, different starting points yield different estimates, which are biased toward the initial values. We call this phenomenon anchoring. Insufficient adjustment. In a demonstration of the anchoring effect, subjects were asked to estimate various quantities, stated in percentages (for example, the percentage of African countries in the United Nations). For each quantity, a number between 0 and 100 was determined by spinning a wheel of fortune in the subjects’ presence. The subjects were instructed to indicate first whether that number was higher or lower than the value of the quantity, and then to estimate the value of the quantity by moving upward or downward from the given number. Different groups were given different numbers for each quantity, and these arbitrary numbers had a marked effect on estimates. For example, the median estimates of the percentage of African countries in the United Nations were 25 and 45 for groups that
received 10 and 65, respectively, as starting points. Payoffs for accuracy did not reduce the anchoring effect. Anchoring occurs not only when the starting point is given to the subject, but also when the subject bases his estimate on the result of some incomplete computation. A study of intuitive numerical estimation illustrates this effect. Two groups of high school students estimated, within 5 seconds, a numerical expression that was written on the blackboard. One group estimated the product 8×7×6×5×4×3×2×1 while another group estimated the product 1×2×3×4×5×6×7×8 To rapidly answer such questions, people may perform a few steps of computation and estimate the product by extrapolation or adjustment. Because adjustments are typically insufficient, this procedure should lead to underestimation. Furthermore, because the result of the first few steps of multiplication (performed from left to right) is higher in the descending sequence than in the ascending sequence, the former expression should be judged larger than the latter. Both predictions were confirmed. The median estimate for the ascending sequence was 512, while the median estimate for the descending sequence was 2,250. The correct answer is 40,320. Biases in the evaluation of conjunctive and disjunctive events. In a recent study by Bar-Hillel19 subjects were given the opportunity to bet on one of two events. Three types of events were used: (i) simple events, such as drawing a red marble from a bag containing 50% red marbles and 50% white marbles; (ii) conjunctive events, such as drawing a red marble seven times in succession, with replacement, from a bag containing 90% red marbles and 10% white marbles; and (iii) disjunctive events, such as drawing a red marble at least once in seven successive tries, with replacement, from a bag containing 10% red marbles and 90% white marbles. In this problem, a significant majority of subjects preferred to bet on the conjunctive event (the probability of which is .48) rather than on the simple event (the probability of which is .50). Subjects also preferred to bet on the simple event rather than on the disjunctive event, which has a probability of .52. Thus, most subjects bet on the less likely event in both
comparisons. This pattern of choices illustrates a general finding. Studies of choice among gambles and of judgments of probability indicate that people tend to overestimate the probability of conjunctive events20 and to underestimate the probability of disjunctive events. These biases are readily explained as effects of anchoring. The stated probability of the elementary event (success at any one stage) provides a natural starting point for the estimation of the probabilities of both conjunctive and disjunctive events. Since adjustment from the starting point is typically insufficient, the final estimates remain too close to the probabilities of the elementary events in both cases. Note that the overall probability of a conjunctive event is lower than the probability of each elementary event, whereas the overall probability of a disjunctive event is higher than the probability of each elementary event. As a consequence of anchoring, the overall probability will be overestimated in conjunctive problems and underestimated in disjunctive problems. Biases in the evaluation of compound events are particularly significant in the context of planning. The successful completion of an undertaking, such as the development of a new product, typically has a conjunctive character: for the undertaking to succeed, each of a series of events must occur. Even when each of these events is very likely, the overall probability of success can be quite low if the number of events is large. The general tendency to overestimate the probability of conjunctive events leads to unwarranted optimism in the evaluation of the likelihood that a plan will succeed or that a project will be completed on time. Conversely, disjunctive structures are typically encountered in the evaluation of risks. A complex system, such as a nuclear reactor or a human body, will malfunction if any of its essential components fails. Even when the likelihood of failure in each component is slight, the probability of an overall failure can be high if many components are involved. Because of anchoring, people will tend to underestimate the probabilities of failure in complex systems. Thus, the direction of the anchoring bias can sometimes be inferred from the structure of the event. The chain-like structure of conjunctions leads to overestimation, the funnel-like structure of disjunctions leads to underestimation. Anchoring in the assessment of subjective probability distributions. In decision analysis, experts are often required to express their beliefs about a quantity, such as the value of the Dow Jones average on a particular day, in
the form of a probability distribution. Such a distribution is usually constructed by asking the person to select values of the quantity that correspond to specified percentiles of his subjective probability distribution. For example, the judge may be asked to select a number, X90, such that his subjective probability that this number will be higher than the value of the Dow Jones average is .90. That is, he should select the value X90 so that he is just willing to accept 9 to 1 odds that the Dow Jones average will not exceed it. A subjective probability distribution for the value of the Dow Jones average can be constructed from several such judgments corresponding to different percentiles. By collecting subjective probability distributions for many different quantities, it is possible to test the judge for proper calibration. A judge is properly (or externally) calibrated in a set of problems if exactly Π% of the true values of the assessed quantities falls below his stated values of XΠ. For example, the true values should fall below X01 for 1% of the quantities and above X99 for 1% of the quantities. Thus, the true values should fall in the confidence interval between X01 and X99 on 98% of the problems. Several investigators21 have obtained probability distributions for many quantities from a large number of judges. These distributions indicated large and systematic departures from proper calibration. In most studies, the actual values of the assessed quantities are either smaller than X0l or greater than X99 for about 30% of the problems. That is, the subjects state overly narrow confidence intervals which reflect more certainty than is justified by their knowledge about the assessed quantities. This bias is common to naive and to sophisticated subjects, and it is not eliminated by introducing proper scoring rules, which provide incentives for external calibration. This effect is attributable, in part at least, to anchoring. To select X90 for the value of the Dow Jones average, for example, it is natural to begin by thinking about one’s best estimate of the Dow Jones and to adjust this value upward. If this adjustment—like most others—is insufficient, then X90 will not be sufficiently extreme. A similar anchoring effect will occur in the selection of X10, which is presumably obtained by adjusting one’s best estimate downward. Consequently, the confidence interval between X10 and X90 will be too narrow, and the assessed probability distribution will be too tight. In support of this interpretation it can be
shown that subjective probabilities are systematically altered by a procedure in which one’s best estimate does not serve as an anchor. Subjective probability distributions for a given quantity (the Dow Jones average) can be obtained in two different ways: (i) by asking the subject to select values of the Dow Jones that correspond to specified percentiles of his probability distribution and (ii) by asking the subject to assess the probabilities that the true value of the Dow Jones will exceed some specified values. The two procedures are formally equivalent and should yield identical distributions. However, they suggest different modes of adjustment from different anchors. In procedure (i), the natural starting point is one’s best estimate of the quantity. In procedure (ii), on the other hand, the subject may be anchored on the value stated in the question. Alternatively, he may be anchored on even odds, or a 50–50 chance, which is a natural starting point in the estimation of likelihood. In either case, procedure (ii) should yield less extreme odds than procedure (i). To contrast the two procedures, a set of 24 quantities (such as the air distance from New Delhi to Peking) was presented to a group of subjects who assessed either X10 or X90 for each problem. Another group of subjects received the median judgment of the first group for each of the 24 quantities. They were asked to assess the odds that each of the given values exceeded the true value of the relevant quantity. In the absence of any bias, the second group should retrieve the odds specified to the first group, that is, 9:1. However, if even odds or the stated value serve as anchors, the odds of the second group should be less extreme, that is, closer to 1:1. Indeed, the median odds stated by this group, across all problems, were 3:1. When the judgments of the two groups were tested for external calibration, it was found that subjects in the first group were too extreme, in accord with earlier studies. The events that they defined as having a probability of .10 actually obtained in 24% of the cases. In contrast, subjects in the second group were too conservative. Events to which they assigned an average probability of .34 actually obtained in 26% of the cases. These results illustrate the manner in which the degree of calibration depends on the procedure of elicitation. DISCUSSION
This article has been concerned with cognitive biases that stem from the reliance on judgmental heuristics. These biases are not attributable to motivational effects such as wishful thinking or the distortion of judgments by payoffs and penalties. Indeed, several of the severe errors of judgment reported earlier occurred despite the fact that subjects were encouraged to be accurate and were rewarded for the correct answers.22 The reliance on heuristics and the prevalence of biases are not restricted to laymen. Experienced researchers are also prone to the same biases— when they think intuitively. For example, the tendency to predict the outcome that best represents the data, with insufficient regard for prior probability, has been observed in the intuitive judgments of individuals who have had extensive training in statistics.23 Although the statistically sophisticated avoid elementary errors, such as the gambler’s fallacy, their intuitive judgments are liable to similar fallacies in more intricate and less transparent problems. It is not surprising that useful heuristics such as representativeness and availability are retained, even though they occasionally lead to errors in prediction or estimation. What is perhaps surprising is the failure of people to infer from lifelong experience such fundamental statistical rules as regression toward the mean, or the effect of sample size on sampling variability. Although everyone is exposed, in the normal course of life, to numerous examples from which these rules could have been induced, very few people discover the principles of sampling and regression on their own. Statistical principles are not learned from everyday experience because the relevant instances are not coded appropriately. For example, people do not discover that successive lines in a text differ more in average word length than do successive pages, because they simply do not attend to the average word length of individual lines or pages. Thus, people do not learn the relation between sample size and sampling variability, although the data for such learning are abundant. The lack of an appropriate code also explains why people usually do not detect the biases in their judgments of probability. A person could conceivably learn whether his judgments are externally calibrated by keeping a tally of the proportion of events that actually occur among those to which he assigns the same probability. However, it is not natural to group events by their judged probability. In the absence of such grouping it is impossible for an individual to discover, for example, that only 50% of the
Search
Read the Text Version
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
- 272
- 273
- 274
- 275
- 276
- 277
- 278
- 279
- 280
- 281
- 282
- 283
- 284
- 285
- 286
- 287
- 288
- 289
- 290
- 291
- 292
- 293
- 294
- 295
- 296
- 297
- 298
- 299
- 300
- 301
- 302
- 303
- 304
- 305
- 306
- 307
- 308
- 309
- 310
- 311
- 312
- 313
- 314
- 315
- 316
- 317
- 318
- 319
- 320
- 321
- 322
- 323
- 324
- 325
- 326
- 327
- 328
- 329
- 330
- 331
- 332
- 333
- 334
- 335
- 336
- 337
- 338
- 339
- 340
- 341
- 342
- 343
- 344
- 345
- 346
- 347
- 348
- 349
- 350
- 351
- 352
- 353
- 354
- 355
- 356
- 357
- 358
- 359
- 360
- 361
- 362
- 363
- 364
- 365
- 366
- 367
- 368
- 369
- 370
- 371
- 372
- 373
- 374
- 375
- 376
- 377
- 378
- 379
- 380
- 381
- 382
- 383
- 384
- 385
- 386
- 387
- 388
- 389
- 390
- 391
- 392
- 393
- 394
- 395
- 396
- 397
- 398
- 399
- 400
- 401
- 402
- 403
- 404
- 405
- 406
- 407
- 408
- 409
- 410
- 411
- 412
- 413
- 414
- 415
- 416
- 417
- 418
- 419
- 420
- 421
- 422
- 423
- 424
- 425
- 426
- 427
- 428
- 429
- 430
- 431
- 432
- 433
- 434
- 435
- 436
- 437
- 438
- 439
- 440
- 441
- 442
- 443
- 444
- 445
- 446
- 447
- 448
- 449
- 450
- 451
- 452
- 453
- 454
- 455
- 456
- 457
- 458
- 459
- 460
- 461
- 462
- 463
- 464
- 465
- 466
- 467
- 468
- 469
- 470
- 471
- 472
- 473
- 474
- 475
- 476
- 477
- 478
- 479
- 480
- 481
- 482
- 483
- 484
- 485
- 486
- 487
- 488
- 489
- 490
- 491
- 492
- 493
- 494
- 495
- 496
- 497
- 498
- 499
- 500
- 501
- 502
- 503
- 504
- 505
- 506
- 507
- 508
- 509
- 510
- 511
- 512
- 513
- 514
- 515
- 516
- 517
- 518
- 519
- 520
- 521
- 522
- 523
- 524
- 525
- 526
- 527
- 528
- 529
- 530
- 531
- 532
- 533
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 533
Pages:
