Even more troubling is what happens when people are confronted with their inconsistency: "You chose to save 200 lives for sure in one formulation and you chose to gamble rather than accept 400 deaths in the other. Now that you know these choices were inconsistent, how do you decide?" The answer is usually embarrassed silence. The intuitions that determined the original choice came from System 1 and had no more moral basis than did the preference for keeping 20 or the aversion to losing 30. Saving lives with certainty is good, deaths are bad. Most people find that their System 2 has no moral intuitions of its own to answer the question.

I am grateful to the great economist Thomas Schelling for my favorite example of a framing effect, which he described in his book Choice and Consequence. Schelling's book was written before our work on framing was published, and framing was not his main concern. He reported on his experience teaching a class at the Kennedy School at Harvard, in which Bon he linthe topic was child exemptions in the tax code. Schelling told his students that a standard exemption is allowed for each child, and that the amount of the exemption is independent of the taxpayer's income. He asked their opinion of the following proposition: Should the child exemption be larger for the rich than for the poor?

Your own intuitions are very likely the same as those of Schelling's students: they found the idea of favoring the rich by a larger exemption completely unacceptable.

Schelling then pointed out that the tax law is arbitrary. It assumes a childless family as the default case and reduces the tax by the amount of the exemption for each child. The tax law could of course be rewritten with another default case: a family with two children. In this formulation, families with fewer than the default number of children would pay a surcharge. Schelling now asked his students to report their view of another proposition: Should the childless poor pay as large a surcharge as the childless rich?

Here again you probably agree with the students' reaction to this idea, which they rejected with as much vehemence as the first. But Schelling showed his class that they could not logically reject both proposals. Set the two formulations next to each other. The difference between the tax due by a childless family and by a family with two children is described as a reduction of tax in the first version and as an increase in the second. If in the first version you want the poor to receive the same (or greater) benefit as the rich for having children, then you must want the poor to pay at least the same penalty as the rich for being childless.

We can recognize System 1 at work. It delivers an immediate response to any question about rich and poor: when in doubt, favor the poor. The surprising aspect of Schelling's problem is that this apparently simple moral rule does not work reliably. It generates contradictory answers to the same problem, depending on how that problem is framed. And of course you already know the question that comes next. Now that you have seen that your reactions to the problem are influenced by the frame, what is your answer to the question: How should the tax code treat the children of the rich and the poor?

Here again, you will probably find yourself dumbfounded. You have moral intuitions about differences between the rich and the poor, but these intuitions depend on an arbitrary reference point, and they are not about the real problem. This problem-the question about actual states of the world-is how much tax individual families should pay, how to fill the cells in the matrix of the tax code. You have no compelling moral intuitions to guide you in solving that problem. Your moral feelings are attached to frames, to descriptions of reality rather than to reality itself. The message about the nature of framing is stark: framing should not be viewed as an intervention that masks or distorts an underlying preference. At least in this instance-and also in the problems of the Asian disease and of surgery versus radiation for lung cancer-there is no underlying preference that is masked or distorted by the frame. Our preferences are about framed problems, and our moral intuitions are about descriptions, not about substance.Good FramesNot all frames are equal, and s Bon nd t="4%" wome frames are clearly better than alternative ways to describe (or to think about) the same thing. Consider the following pair of problems: A woman has bought two $80 tickets to the theater. When she arrives at the theater, she opens her wallet and discovers that the tickets are missing. Will she buy two more tickets to see the play?

A woman goes to the theater, intending to buy two tickets that cost $80 each. She arrives at the theater, opens her wallet, and discovers to her dismay that the $160 with which she was going to make the purchase is missing. She could use her credit card. Will she buy the tickets?

Respondents who see only one version of this problem reach different conclusions, depending on the frame. Most believe that the woman in the first story will go home without seeing the show if she has lost tickets, and most believe that she will charge tickets for the show if she has lost money.

The explanation should already be familiar-this problem involves mental accounting and the sunk-cost fallacy. The different frames evoke different mental accounts, and the significance of the loss depends on the account to which it is posted. When tickets to a particular show are lost, it is natural to post them to the account associated with that play. The cost appears to have doubled and may now be more than the experience is worth. In contrast, a loss of cash is charged to a "general revenue" account-the theater patron is slightly poorer than she had thought she was, and the question she is likely to ask herself is whether the small reduction in her disposable wealth will change her decision about paying for tickets. Most respondents thought it would not.

The version in which cash was lost leads to more reasonable decisions. It is a better frame because the loss, even if tickets were lost, is "sunk," and sunk costs should be ignored. History is irrelevant and the only issue that matters is the set of options the theater patron has now, and their likely consequences. Whatever she lost, the relevant fact is that she is less wealthy than she was before she opened her wallet. If the person who lost tickets were to ask for my advice, this is what I would say: "Would you have bought tickets if you had lost the equivalent amount of cash? If yes, go ahead and buy new ones." Broader frames and inclusive accounts generally lead to more rational decisions.

In the next example, two alternative frames evoke different mathematical intuitions, and one is much superior to the other. In an article titled "The MPG Illusion," which appeared in Science magazine in 2008, the psychologists Richard Larrick and Jack Soll identified a case in which passive acceptance of a misleading frame has substantial costs and serious policy consequences. Most car buyers list gas mileage as one of the factors that determine their choice; they know that high-mileage cars have lower operating costs. But the frame that has traditionally been used in the United States-miles per gallon-provides very poor guidance to the decisions of both individuals and policy makers. Consider two car owners who seek to reduce their costs: Adam switches from a gas-guzzler of 12 mpg to a slightly less voracious guzzler that runs at 14 mpg.

The environmentally virtuous Beth switches from a Bon ss es from 30 mpg car to one that runs at 40 mpg.

Suppose both drivers travel equal distances over a year. Who will save more gas by switching? You almost certainly share the widespread intuition that Beth's action is more significant than Adam's: she reduced mpg by 10 miles rather than 2, and by a third (from 30 to 40) rather than a sixth (from 12 to 14). Now engage your System 2 and work it out. If the two car owners both drive 10,000 miles, Adam will reduce his consumption from a scandalous 833 gallons to a still shocking 714 gallons, for a saving of 119 gallons. Beth's use of fuel will drop from 333 gallons to 250, saving only 83 gallons. The mpg frame is wrong, and it should be replaced by the gallons-per-mile frame (or liters-per100 kilometers, which is used in most other countries). As Larrick and Soll point out, the misleading intuitions fostered by the mpg frame are likely to mislead policy makers as well as car buyers.

Under President Obama, Cass Sunstein served as administrator of the Office of Information and Regulatory Affairs. With Richard Thaler, Sunstein coauthored Nudge, which is the basic manual for applying behavioral economics to policy. It was no accident that the "fuel economy and environment" sticker that will be displayed on every new car starting in 2013 will for the first time in the United States include the gallons-per-mile information. Unfortunately, the correct formulation will be in small print, along with the more familiar mpg information in large print, but the move is in the right direction. The five-year interval between the publication of "The MPG Illusion" and the implementation of a partial correction is probably a speed record for a significant application of psychological science to public policy.

A directive about organ donation in case of accidental death is noted on an individual's driver license in many countries. The formulation of that directive is another case in which one frame is clearly superior to the other. Few people would argue that the decision of whether or not to donate one's organs is unimportant, but there is strong evidence that most people make their choice thoughtlessly. The evidence comes from a comparison of the rate of organ donation in European countries, which reveals startling differences between neighboring and culturally similar countries. An article published in 2003 noted that the rate of organ donation was close to 100% in Austria but only 12% in Germany, 86% in Sweden but only 4% in Denmark.

These enormous differences are a framing effect, which is caused by the format of the critical question. The high-donation countries have an opt out form, where individuals who wish not to donate must check an appropriate box. Unless they take this simple action, they are considered willing donors. The low-contribution countries have an opt-in form: you must check a box to become a donor. That is all. The best single predictor of whether or not people will donate their organs is the designation of the default option that will be adopted without having to check a box.

Unlike other framing effects that have been traced to features of System 1, the organ donation effect is best explained by the laziness of System 2. People will check the box if they have already decided what they wish to do. If they are unprepared for the question, they have to make the effort of thinking whether they want to check the box. I imagine an organ donation form in which people are required to solve a mathematical problem in the box that corresponds to their decision. One of the boxes contains the problem 2 + 2 = ? The problem in the other box is 13 37 = ? The rate of donations would surely be swayed.When the role of formulation is acknowledged, a policy question arises: Which formulation should be adopted? In this case, the answer is straightforward. If you believe that a large supply of donated organs is good for society, you will not be neutral between a formulation that yields almost 100% donations and another formulation that elicits donations from 4% of drivers.

As we have seen again and again, an important choice is controlled by an utterly inconsequential feature of the situation. This is embarrassing-it is not how we would wish to make important decisions. Furthermore, it is not how we experience the workings of our mind, but the evidence for these cognitive illusions is undeniable.

Count that as a point against the rational-agent theory. A theory that is worthy of the name asserts that certain events are impossible-they will not happen if the theory is true. When an "impossible" event is observed, the theory is falsified. Theories can survive for a long time after conclusive evidence falsifies them, and the rational-agent model certainly survived the evidence we have seen, and much other evidence as well.

The case of organ donation shows that the debate about human rationality can have a large effect in the real world. A significant difference between believers in the rational-agent model and the skeptics who question it is that the believers simply take it for granted that the formulation of a choice cannot determine preferences on significant problems. They will not even be interested in investigating the problem-and so we are often left with inferior outcomes.

Skeptics about rationality are not surprised. They are trained to be sensitive to the power of inconsequential factors as determinants of preference-my hope is that readers of this book have acquired this sensitivity.Speaking of Frames and Reality

"They will feel better about what happened if they manage to frame the outcome in terms of how much money they kept rather than how much they lost."

"Let's reframe the problem by changing the reference point. Imagine we did not own it; how much would we think it is worth?"

"Charge the loss to your mental account of 'general revenue'-you will feel better!"

"They ask you to check the box to opt out of their mailing list. Their list would shrink if they asked you to check a box to opt in!"

Part 5

Two Selves

Two Selves The term utility has had two distinct meanings in its long history. Jeremy Bentham opened his Introduction to the Principles of Morals and Legislation with the famous sentence "Nature has placed mankind under the governance of two sovereign masters, pain and pleasure. It is for them alone to point out what we ought to do, as well as to determine what we shall do." In an awkward footnote, Bentham apologized for applying the word utility to these experiences, saying that he had been unable to find a better word. To distinguish Bentham's interpretation of the term, I will call it experienced utility.

For the last 100 years, economists have used the same word to mean something else. As economists and decision theorists apply the term, it means "wantability"-and I have called it decision utility. Expected utility theory, for example, is entirely about the rules of rationality that should govern decision utilities; it has nothing at all to say about hedonic experiences. Of course, the two concepts of utility will coincide if people want what they will enjoy, and enjoy what they chose for themselves-and this assumption of coincidence is implicit in the general idea that economic agents are rational. Rational agents are expected to know their tastes, both present and future, and they are supposed to make good decisions that will maximize these interests.Experienced UtilityMy fascination with the possible discrepancies between experienced utility and decision utility goes back a long way. While Amos and I were still working on prospect theory, I formulated a puzzle, which went like this: imagine an individual who receives one painful injection every day. There is no adaptation; the pain is the same day to day. Will people attach the same value to reducing the number of planned injections from 20 to 18 as from 6 to 4? Is there any justification for a distinction?

I did not collect data, because the outcome was evident. You can verify for yourself that you would pay more to reduce the number of injections by a third (from 6 to 4) than by one tenth (from 20 to 18). The decision utility of avoiding two injections is higher in the first case than in the second, and everyone will pay more for the first reduction than for the second. But this difference is absurd. If the pain does not change from day to day, what could justify assigning different utilities to a reduction of the total amount of pain by two injections, depending on the number of previous injections? In the terms we would use today, the puzzle introduced the idea that experienced utility could be measured by the number of injections. It also suggested that, at least in some cases, experienced utility is the criterion by which a decision should be assessed. A decision maker who pays different amounts to achieve the same gain of experienced utility (or be spared the same loss) is making a mistake. You may find this observation obvious, but in decision theory the only basis for judging that a decision is wrong is inconsistency with other preferences. Amos and I discussed the problem but we did not pursue it. Many years later, I returned to it.Experience and MemoryHow can experienced utility be measured? How should we answer questions such as "How much pain did Helen suffer during the medical procedure?" or "How much enjoyment did she get from her 20 minutes on the beach?" T Jon e t8221; T Jhe British economist Francis Edgeworth speculated about this topic in the nineteenth century and proposed the idea of a "hedonimeter," an imaginary instrument analogous to the devices used in weather-recording stations, which would measure the level of pleasure or pain that an individual experiences at any moment.

Experienced utility would vary, much as daily temperature or barometric pressure do, and the results would be plotted as a function of time. The answer to the question of how much pain or pleasure Helen experienced during her medical procedure or vacation would be the "area under the curve." Time plays a critical role in Edgeworth's conception. If Helen stays on the beach for 40 minutes instead of 20, and her enjoyment remains as intense, then the total experienced utility of that episode doubles, just as doubling the number of injections makes a course of injections twice as bad. This was Edgeworth's theory, and we now have a precise understanding of the conditions under which his theory holds.

The graphs in figure 15 show profiles of the experiences of two patients undergoing a painful colonoscopy, drawn from a study that Don Redelmeier and I designed together. Redelmeier, a physician and researcher at the University of Toronto, carried it out in the early 1990s. This procedure is now routinely administered with an anesthetic as well as an amnesic drug, but these drugs were not as widespread when our data were collected. The patients were prompted every 60 seconds to indicate the level of pain they experienced at the moment. The data shown are on a scale where zero is "no pain at all" and 10 is "intolerable pain." As you can see, the experience of each patient varied considerably during the procedure, which lasted 8 minutes for patient A and 24 minutes for patient B (the last reading of zero pain was recorded after the end of the procedure). A total of 154 patients participated in the experiment; the shortest procedure lasted 4 minutes, the longest 69 minutes.

Next, consider an easy question: Assuming that the two patients used the scale of pain similarly, which patient suffered more? No contest. There is general agreement that patient B had the worse time. Patient B spent at least as much time as patient A at any level of pain, and the "area under the curve" is clearly larger for B than for A. The key factor, of course, is that B's procedure lasted much longer. I will call the measures based on reports of momentary pain hedonimeter totals.

Figure 15

When the procedure was over, all participants were asked to rate "the total amount of pain" they had experienced during the procedure. The wording was intended to encourage them to think of the integral of the pain they had reported, reproducing the hedonimeter totals. Surprisingly, the patients did nothing of the kind. The statistical analysis revealed two findings, which illustrate a pattern we have observed in other experiments:Peak-end rule: The global retrospective rating was well predicted by the average of the level of pain reported at the worst moment of the experience and at its end.

Duration neglect: The duration of the procedure had no effect whatsoever on the ratings of total pain.

You can now apply these rules to the profiles of patients A and B. The worst rati Jon er soever on ng (8 on the 10-point scale) was the same for both patients, but the last rating before the end of the procedure was 7 for patient A and only 1 for patient B. The peak-end average was therefore 7.5 for patient A and only 4.5 for patient B. As expected, patient A retained a much worse memory of the episode than patient B. It was the bad luck of patient A that the procedure ended at a bad moment, leaving him with an unpleasant memory.

We now have an embarrassment of riches: two measures of experienced utility-the hedonimeter total and the retrospective assessment-that are systematically different. The hedonimeter totals are computed by an observer from an individual's report of the experience of moments. We call these judgments duration-weighted, because the computation of the "area under the curve" assigns equal weights to all moments: two minutes of pain at level 9 is twice as bad as one minute at the same level of pain. However, the findings of this experiment and others show that the retrospective assessments are insensitive to duration and weight two singular moments, the peak and the end, much more than others. So which should matter? What should the physician do? The choice has implications for medical practice. We noted that:If the objective is to reduce patients' memory of pain, lowering the peak intensity of pain could be more important than minimizing the duration of the procedure. By the same reasoning, gradual relief may be preferable to abrupt relief if patients retain a better memory when the pain at the end of the procedure is relatively mild.

If the objective is to reduce the amount of pain actually experienced, conducting the procedure swiftly may be appropriate even if doing so increases the peak pain intensity and leaves patients with an awful memory.

Which of the two objectives did you find most compelling? I have not conducted a proper survey, but my impression is that a strong majority will come down in favor of reducing the memory of pain. I find it helpful to think of this dilemma as a conflict of interests between two selves (which do not correspond to the two familiar systems). The experiencing self is the one that answers the question: "Does it hurt now?" The remembering self is the one that answers the question: "How was it, on the whole?" Memories are all we get to keep from our experience of living, and the only perspective that we can adopt as we think about our lives is therefore that of the remembering self.

A comment I heard from a member of the audience after a lecture illustrates the difficulty of distinguishing memories from experiences. He told of listening raptly to a long symphony on a disc that was scratched near the end, producing a shocking sound, and he reported that the bad ending "ruined the whole experience." But the experience was not actually ruined, only the memory of it. The experiencing self had had an experience that was almost entirely good, and the bad end could not undo it, because it had already happened. My questioner had assigned the entire episode a failing grade because it had ended very badly, but that grade effectively ignored 40 minutes of musical bliss. Does the actual experience count for nothing?

Confusing experience with the memory of it is a compelling cognitive illusion-and it is the substitution that makes us believe a past experience can be ruined. The experiencing self does not have a voice. The remembering self is sometimes wrong, but it is the one that keeps score and governs what we learn from living, and it is the one that makes decisions Jon thaperienci. What we learn from the past is to maximize the qualities of our future memories, not necessarily of our future experience. This is the tyranny of the remembering self.Which Self Should Count?To demonstrate the decision-making power of the remembering self, my colleagues and I designed an experiment, using a mild form of torture that I will call the cold-hand situation (its ugly technical name is cold-pressor). Participants are asked to hold their hand up to the wrist in painfully cold water until they are invited to remove it and are offered a warm towel. The subjects in our experiment used their free hand to control arrows on a keyboard to provide a continuous record of the pain they were enduring, a direct communication from their experiencing self. We chose a temperature that caused moderate but tolerable pain: the volunteer participants were of course free to remove their hand at any time, but none chose to do so.

Each participant endured two cold-hand episodes: The short episode consisted of 60 seconds of immersion in water at 14 Celsius, which is experienced as painfully cold, but not intolerable. At the end of the 60 seconds, the experimenter instructed the participant to remove his hand from the water and offered a warm towel.

The long episode lasted 90 seconds. Its first 60 seconds were identical to the short episode. The experimenter said nothing at all at the end of the 60 seconds. Instead he opened a valve that allowed slightly warmer water to flow into the tub. During the additional 30 seconds, the temperature of the water rose by roughly 1, just enough for most subjects to detect a slight decrease in the intensity of pain.

Our participants were told that they would have three cold-hand trials, but in fact they experienced only the short and the long episodes, each with a different hand. The trials were separated by seven minutes. Seven minutes after the second trial, the participants were given a choice about the third trial. They were told that one of their experiences would be repeated exactly, and were free to choose whether to repeat the experience they had had with their left hand or with their right hand. Of course, half the participants had the short trial with the left hand, half with the right; half had the short trial first, half began with the long, etc. This was a carefully controlled experiment.

The experiment was designed to create a conflict between the interests of the experiencing and the remembering selves, and also between experienced utility and decision utility. From the perspective of the experiencing self, the long trial was obviously worse. We expected the remembering self to have another opinion. The peak-end rule predicts a worse memory for the short than for the long trial, and duration neglect predicts that the difference between 90 seconds and 60 seconds of pain will be ignored. We therefore predicted that the participants would have a more favorable (or less unfavorable) memory of the long trial and choose to repeat it. They did. Fully 80% of the participants who reported that their pain diminished during the final phase of the longer episode opted to repeat it, thereby declaring themselves willing to suffer 30 seconds of needless pain in the anticipated third trial.

The subjects who preferred the long episode were not masochists and did not deliberately choose to expose themselves to the worse experience; they simply Jon the heigmade a mistake. If we had asked them, "Would you prefer a 90-second immersion or only the first part of it?" they would certainly have selected the short option. We did not use these words, however, and the subjects did what came naturally: they chose to repeat the episode of which they had the less aversive memory. The subjects knew quite well which of the two exposures was longer-we asked them-but they did not use that knowledge. Their decision was governed by a simple rule of intuitive choice: pick the option you like the most, or dislike the least. Rules of memory determined how much they disliked the two options, which in turn determined their choice. The cold-hand experiment, like my old injections puzzle, revealed a discrepancy between decision utility and experienced utility.

The preferences we observed in this experiment are another example of the less-is-more effect that we have encountered on previous occasions. One was Christopher Hsee's study in which adding dishes to a set of 24 dishes lowered the total value because some of the added dishes were broken. Another was Linda, the activist woman who is judged more likely to be a feminist bank teller than a bank teller. The similarity is not accidental. The same operating feature of System 1 accounts for all three situations: System 1 represents sets by averages, norms, and prototypes, not by sums. Each cold-hand episode is a set of moments, which the remembering self stores as a prototypical moment. This leads to a conflict. For an objective observer evaluating the episode from the reports of the experiencing self, what counts is the "area under the curve" that integrates pain over time; it has the nature of a sum. The memory that the remembering self keeps, in contrast, is a representative moment, strongly influenced by the peak and the end.

Of course, evolution could have designed animals' memory to store integrals, as it surely does in some cases. It is important for a squirrel to "know" the total amount of food it has stored, and a representation of the average size of the nuts would not be a good substitute. However, the integral of pain or pleasure over time may be less biologically significant. We know, for example, that rats show duration neglect for both pleasure and pain. In one experiment, rats were consistently exposed to a sequence in which the onset of a light signals that an electric shock will soon be delivered. The rats quickly learned to fear the light, and the intensity of their fear could be measured by several physiological responses. The main finding was that the duration of the shock has little or no effect on fear-all that matters is the painful intensity of the stimulus.