I have mentioned the term “miswanting” before. I picked it up when I took the free online class from Yale on how to be happy called, “The Science of Well-Being.” (I highly recommend it). In that class, the instructor referenced the paper “Miswanting: Some Problems in the Forecasting of Future Affective States” by Daniel T. Gilbert and Timothy D. Wilson. The link for the PDF was to Harvard’s website here.
I referenced that PDF on Substack Notes the other day and thought, “I wonder if anyone will read that,” remembering that the PDF is very difficult to read. It was oddly printed in landscape mode while being presented in portrait. You basically have to print it out in order to read it, and it’s got a bunch of grey spots from being copied poorly. I edited it to make it much more readable.
I’m not claiming it’s my work, and I’m sure the writers won’t agree with all my comments elsewhere on Substack. But, to make this more accessible, I thought it might be nice if I read it for you all. I hope you enjoy it.
The original PDF is available here. My cleaned-up PDF version is here.
8. Miswanting
Some Problems in the Forecasting of Future Affective States
DANIEL T. GILBERT AND TIMOTHY. D. WILSON1
"It would not be better if things happened to men just as they want."
Heraclitus, Fragments (500 b.C.)
Introduction
Like and want are among the first things children learn to say, and once they learn to say them, they never stop. Liking has to do with how a thing makes us feel, and wanting is, simply enough, a prediction of liking. When we say, “I like this doughnut,” we are letting others know that the doughnut currently under consumption is making us feel a bit better than before. When we say, “I want a doughnut,” we are making an abbreviated statement whose extended translation is something like, “Right now I'm not feeling quite as good as I might be, and I think fried dough will fix that.” Statements about wanting tend to be statements about those things that we believe will influence our sense of well-being, satisfaction, happiness, and contentment. Hence, when we say we want something, we are more or less promising that we will like it when we get it.
But promises are easier to make than to keep, and sometimes we get what we say we want and feel entirely unhappy about it. We order a cheeseburger only to find that it looks and smells precisely as cheeseburgers always look and smell, and despite that fact, we have absolutely no interest in eating it. We are perplexed and embarrassed by such mistakes and can only offer cunning explanations such as, "I guess I didn't really want a cheeseburger after all." Dining companions often consider such accounts inadequate. "If you didn't want the damned thing, then why did you get it?" they may ask, at which point we are usually forced to admit the truth, which is that we just do not know. We only know that it looks exactly like what we said we wanted, we are not going to eat it, and the waiter is not amused.
Although we tend to think of unhappiness as something that happens to us when we do not get what we want, much unhappiness is actually of the cheeseburger variety and has less to do with not getting what we want, and more to do with not wanting what we like. When wanting and liking are uncoordinated in this way you may say that a person has miswanted. The word sounds odd at first, but if wanting is indeed a prediction of liking, then it, like any prediction, can be wrong. When the things we want to happen do not improve our happiness, and when the things we want not to happen; do, it seems fair to say that we have wanted badly. Why should this happen to people as clever and handsome as us?
When wanting and liking are uncoordinated in this way you may say that a person has miswanted.
The Fundamentals of Miswanting
In a perfect world, wanting would cause trying, trying would cause getting, getting would cause liking, and this chapter would be missing all the words. Ours is apparently not such a place. How is it possible to get what we want and yet not like what we get? At least three problems bedevil our attempts to want well.
Imagining the Wrong Event
The fundamental problem, of course, is that the events we imagine when we are in the midst of a really good want are not precisely the events we experience when we are at the tail end of a really disappointing get. For instance, most of us are skeptical when we hear movie stars describe how relentless adoration can be a source of suffering, or when terminally ill patients insist that a dreaded disease has given their lives deeper meaning. We feel certain that we would be delighted in the first instance and devastated in the second because most of us have no idea what stardom or terminal illness actually entail. When we think of “adoring fans,” we tend to envision a cheering throng of admirers calling us back for an encore performance rather than a slightly demented autograph hound peeping through our bedroom window at midnight. When we think of “terminal illness,” we tend to envision ourselves wasting away in a hospital bed, connected to machines by plugs and tubes, rather than planting flowers in the hospice garden, surrounded by those we love. Terminal illness is not an event, but a class of events, and each member of the class unfolds in a different way. How much we like an event depends mightily on the details of its unfolding. When the imagined cheeseburger, (a half-pound of prime-aged beef) is not the experienced cheeseburger (three ounces of rubbery soy), it seems inevitable that our wanting and our liking will be poorly matched.
Given how varied a class of events can be, we might expect people prudently to refrain from directing their wants toward classes ("I don't know if I want a cheeseburger") and direct them instead toward particular, well-understood members of the class ("However, I know I don't want that cheeseburger"). Research suggests that people are not always so prudent and that when asked to make predictions about future events, they tend to imagine a particular event while making little provision for the possibility that the particular event they are imagining may not necessarily be the particular event they will be experiencing (Dunning, Griffin, Milojkovic, & Ross, 1990; Griffin, Dunning, & Ross, 1990; Griffin & Ross, 1991; Lord, Lepper, &.Mackie, 1984; Robinson, Keltner, Ward, & Ross, 1995).
When our spouse asks us to attend “a party” on Friday night, we instantly imagine a particular kind of party (e.g., a cocktail party in the penthouse of a downtown hotel with waiters in black ties carrying silver trays of hors d'oeuvres past a slightly bored harpist) and then estimate our reaction to that imagined event (e.g., yawn). We generally fail to consider how many different members constitute the class (e.g., birthday parties, orgies, wakes) and how different our reactions would be to each. So we tell our spouse that we would rather skip the party, our spouse naturally drags us along anyhow, and we have a truly marvelous time. Why? Because the party involves cheap beer and hula hoops rather than classical music and seaweed crackers. It is precisely our style, and we like what we previously did not want because the event we experienced (and liked) was not the event we imagined (and wanted to avoid).
Using the Wrong Theory
If imagining the wrong event were the sole cause of miswanting, then we would only miswant objects and experiences when the details of their unfolding were unknown to us. The fact is, people often want - and then fail to like - objects and experiences whose details they know quite well. Even when we know precisely the kind of party our spouse is hauling us to, or precisely the kind of cheeseburger this particular restaurant serves, we may still be surprised to find that we enjoy it a great deal more or less than we had anticipated. For example, Read and Loewenstein (1995) asked subjects to plan a menu by deciding which of several snacks they would eat when they returned to the laboratory on each of three consecutive Mondays (cf. Simonson 1990). Subjects tended to order a mixed plate that included instances of their favorite snack (“I'll have a Snickers bar on the first two Mondays”), as well as instances of their next favorite (“And tortilla chips on the third Monday”). Alas, when it actually came time to eat the snacks, subjects were not so pleased on the day when they arrived at the laboratory only to find themselves faced with a snack that was, well, not their favorite. Their disappointment was perfectly understandable. We should be disappointed when we do not get what we like most, and the only thing that seems hard to understand is why subjects wanted something that they knew perfectly well they did not like perfectly well?
Apparently, subjects in this study believed that variety is the spice of life - and in this case, they were wrong (cf. Kahneman & Snell, 1992). A Snickers with every meal is indeed a dull prospect for anyone, but a Snickers once a week is just about right. As such; Snickers-lovers are made less happy - and not more happy - when their weekly Snickers is replaced by a less desirable treat. Because subjects in this study had erroneous theories about their own need for variety over time, they miswanted tortilla chips when they planned their menu.
The moral of this ripping yarn about snack foods is that even when people have a perfect idea of what an event will entail (i.e., tortilla chips are deep-fried corn pancakes covered with salt - period), they may still have imperfect ideas about themselves and thus may make imperfect predictions about how they will react to the event: People who can imagine sun, surf, sand, and daiquiris in exquisite detail may still be surprised when their desert island vacation turns out to be a bust - not because they imagined this island and ended up on that one, but simply because they did not realize how much they require daily structure, intellectual stimulation, or regular infusions of Pop Tarts. Highbrows fall asleep at the ballet, pacifists find themselves strangely excited by a glimpse of world-class wrestling, and tough guys in leather jackets are occasionally caught making clucking sounds by the duck pond. To the extent that we have incorrect theories about who we are, we may also have incorrect beliefs about what we will like.
Misinterpreting Feelings
If we could imagine events exactly as they were actually to unfold, and if we had complete and accurate knowledge of our relevant tastes and attitudes, could we necessarily avoid miswanting? Unfortunately not. When we imagine a future event, we normally have an affective reaction to its mental representation (imagining one's spouse happily entwined with the mail carrier usually illustrates this fact convincingly), and we naturally take this affective reaction to the mental representation of the event as a proxy for the affective reaction we might have to the event itself. If the mere thought of a mate's infidelity makes us feel slightly nauseous, then we have every reason to suppose that the real thing would end in an upchuck. Our affective reactions to imaginary events are, in a sense, experiential previews of our affective reactions to the events themselves, and they figure prominently in our predictions of future liking. Few of us need to consult a cookbook to know that we should avoid any event involving liver and maple syrup. That funny feeling right here is information enough (see Forgas, 1995; Schwarz & Clore, 1983; Schwarz, 1990).
Wantings, then, are based on three ingredients: the particular details that we imagine when we consider a future event, our beliefs about the ways in which people like us are likely to react to such events, and the "gut reactions" we experience when we imagine the event. Just as the first two of these ingredients can lead us to miswant, so too can the third. How so?
Wantings, then, are based on three ingredients: the particular details that we imagine when we consider a future event, our beliefs about the ways in which people like us are likely to react to such events, and the "gut reactions" we experience when we imagine the event.
The crux of the problem is that the feelings we experience when we imagine a future event are not necessarily or solely caused by that act of imagination. We may feel enormously excited when we contemplate spending next Sunday at the circus, and thus we may drop buckets of money on ringside tickets without realizing that the good news we received about our aging uncle's miraculous recovery from psoriasis just moments before we purchased our ticket has contaminated our affective reaction to the thought of dancing elephants (Wilson & Brekke, 1994).
Come Sunday, we may find ourselves bored to tears beneath the big top, wondering why we paid good money to see a herd of clowns in a little car. Our miswanting in this case would not have been a result of having imagined the wrong event ("Oh, I was thinking of a flea circus") nor of having had a false conception of ourselves ("Why did I think I liked men in floppy shoes?"). Rather, we would have miswanted because when we initially thought about the circus we felt excited, and we took that fact as information about the circus rather than as information about Uncle Frank's remission. Feelings do not say where they came from, and thus it is all too easy for us to attribute them to the wrong source.
Feelings do not say where they came from, and thus it is all too easy for us to attribute them to the wrong source.
Experimental demonstrations of this home truth abound. People may mistakenly believe that their lives are empty when, in fact, their gloomy mood is a consequence of rain (Schwarz & Clore, 1983); they may mistakenly believe that a person is attractive when, in fact, their pounding pulse is being caused by the swaying of a suspension bridge (Duttoq & Aron, 1974); and so on. Because we cannot always tell if the feelings we are having as we imagine an event are being caused solely by that imagining, we may use these feelings as proxies for future liking, and hence, miswant.
Thinking and Feeling
We ordinarily experience both thoughts and feelings when we imagine a future event, and these influence our wantings to different extents, under different circumstances, and with different results. Sometimes, our affective reactions to an imagined event provide an excellent basis for wanting, but our cognitive reactions muck things up. This seems to be what happened when Wilson et al. (1993) offered college students a reproduction of an impressionist painting or a poster of a cat with a humorous caption.
Before making their choices, some students were asked to think about why they liked or disliked each poster ("deep thinkers") and others were not ("shallow thinkers"). When the experimenters phoned the students later and asked how much they liked their new objet d'art, the deep thinkers were the least satisfied. Presumably, the shallow thinkers used their current affective reaction as the basis for their decision and ended up liking the posters they had chosen.
Deep thinkers, on the other hand, had some badly mistaken theories about their own aesthetic preferences (“Now that I think about it, the olive green in the Monet is rather drab, whereas the cat poster is bright and cheery”), and when they allowed these cognitive reactions to overrule their affective reactions, they inevitably miswanted.
At other times, however, our cognitive reactions can provide an excellent basis for our wantings, and our affective reactions may lead us astray. For example, Gilbert, Gill, and Wilson (1998) asked shoppers at a grocery store to write down all the items they had come to buy, and allowed some shoppers to retain that list. Next, they asked some shoppers to eat a quarter pound of blueberry muffins before entering the store. As shoppers exited the store, the experimenters examined their cash register receipts. When shoppers were deprived of their shopping lists, those who had eaten blueberry muffins bought fewer unwanted items than did those who had not eaten any muffins. Presumably, when these listless shoppers encountered items in the store, they had more positive affective reactions to the items when they were unfed (“The marshmallow cookies look so delicious!”) than they did when they were well-fed (“I never want to eat again”) and thus were more inclined to buy items that they had not intended to buy.
Shoppers who had their lists in hand, however, were unaffected by the blueberry muffins, and bought no more unwanted items when they were unfed than when they were well-fed. These listful shoppers surely had the same affective reactions as did their listless counterparts, but because they had in hand a copy of A Theory About What I Will Want in the Future (aka a grocery list), they were able to avoid basing their choices on their affective reactions and thus they were able to avoid miswanting.
It seems, then, that feelings sometimes serve us better than theories, and theories sometimes serve us better than feelings. Alas, sometimes neither serves us well at all. Gilbert et al. (1998) asked college students to predict how much they would enjoy eating a bite of spaghetti the next morning or the next evening. Some of the students were hungry when they made these predictions, others were not; Some students were allowed to think deeply about their predictions, and others were distracted while they made their predictions.
When the students were distracted, they relied on their gut feelings to make their predictions, and thus the hungriest students naturally predicted that they would like spaghetti more the next day than did the less hungry students.
Notably, the time of day at which the spaghetti was to be eaten made no difference to them at all. When students were allowed to think deeply, however, they relied on their theories to make their predictions, and thus they predicted that they would enjoy spaghetti (which is generally considered a more appropriate dinner than breakfast) more the next evening than they would the next morning. Notably, the students' current hunger made no difference to them at all. Finally, when students were actually brought to the laboratory in the morning or evening and given a bite of spaghetti, neither the extent of their hunger the day before nor the time of day at which the spaghetti was eaten had a measurable influence on their enjoyment of the food. In other words, students relied on their cognitive reactions when they could, their affective reactions otherwise, and in this instance, neither of these reactions to the imagined event enabled them to want correctly.
Miswanting Over Time
What do spaghetti, cheeseburgers, marshmallow cookies, tortilla chips, and Snickers bars have in common? They are objects that can be wanted today and liked tomorrow, but once that liking occurs, they quickly become a trivial bit of personal history that only our thighs remember. Each of these objects can be experienced, but none of these experiences has enduring emotional consequences, and thus none provides an opportunity for us to think about how people might want or miswant in the long run. When we want a bite of pecan pie or a warm shower or a sexy kiss, it is not because we think these things will change us in some significant way, but because we think they will be perfectly lovely for as long as they last.
On the other hand, when we want a promotion or a wedding or a college degree, it is not so much because we believe these things will improve our lives at the moment we attain them, but because we think they will provide emotional rewards that will, persist long enough to repay the effort we spent in their pursuit. Significant events are supposed to have significant emotional consequences, and the duration of these consequences matters a lot.
If it is difficult to know whether we will be happy 15 minutes after eating a bite of spaghetti, it is all the more difficult to know whether we will be happy 15 months after a divorce or 15 years after a marriage. Gilbert, Pinel, Wilson, Blumberg, and Wheatley (1998) have suggested that, people tend to overestimate the duration of their emotional reactions to future events - especially negative events - and that this can lead them to miswant in the long term. For example, Gilbert et al. (1998) asked assistant professors to predict how happy they would be in general a few years after achieving or failing to achieve tenure at their current university, and they also measured the general happiness of those former assistant professors who had or had not achieved tenure at the same institution. Although assistant professors believed that the tenure decision would dramatically influence their general happiness for many years to come (and hence desperately wanted tenure), the former assistant professors who had not achieved tenure were no less happy than the former assistant professors who had.
Similarly, Gilbert et al. (1998) asked voters in a gubernatorial election to predict how happy they would generally be a month after an election. Voters believed that they would be significantly happier a month after the election if their candidate won than if their candidate lost. As it turned out, a month after the election, the losers and winners were just as happy as they had been before the election (see Brickman, Coates, & Janoff-Bulman, 1978; Taylor, 1983; 1996; Wortman & Silver, 1989).
Do not misunderstand: Those assistant professors who were promoted and those voters whose candidate triumphed were surely happier about the event, and were surely happier for some time after the event, than were those who lost their jobs or who backed the incumbent governor, who lost hers. But after just a little while - a much littler while than the assistant professors and voters had themselves predicted - the emotional traces of these events had evaporated (see Suh, Diener, & Fujita, 1996). What might cause people to overestimate the enduring emotional impact of such events?
Focalism: The Invisible Future
When asked how we might feel a year after losing our left hand, we tend to imagine the immediate emotional impact of this calamity (“No more clapping, no more shoe tying - I’d be sad”). What we do not do is go on to calculate the impact of the dental appointments, foreign films; job promotions, freak snowstorms, and Snickers bars that will inevitably fill the year that follows our unhanding. Rather, we naturally focus on the event whose emotional impact we are trying to gauge and then make some provision for the passage of time (“I guess a year later I'd be a little less sad”). But how we will feel in general a year after losing a hand, and how we will feel about losing a hand a year after the loss, are not the same thing.
Predicting the latter may be relatively simple, but predicting the former requires that we estimate the combined impact of the focal event and all the nonfocal events that follow it. Put another way, our general happiness some time after an event is influenced by just two things: (a) the event, and (b) everything else. If we estimate that happiness by considering only the event then we are ignoring some of the most powerful determinants of our future well-being (see Loewenstein & Schkade, in press; Schkade & Kahneman, 1997).
Wilson, Wheatley, Meyers, Gilbert, and Axsom (1998) demonstrated how focalism (the failure to consider the consequences of nonfocal events when making predictions about the ultimate affective impact of focal events) can give rise to the durability bias and hence promote miswanting. College students were asked to predict their happiness the day after their football team won or lost an important game. Some students were also asked to complete a “future diary” in which they listed the events that they thought would occur in the 3 days after the game. Those students who completed the diary, and who were thus most likely to consider the impact of future nonfocal events when making their predictions, made less extreme predictions about their general happiness - predictions that turned out to be more accurate when their overall happiness was measured the day after the game.
It seems that merely considering the emotional impact of an event can lead us to overestimate that impact, simply because we do not also consider other impactful events as well. Focalism is an especially vexing problem because avoiding it seems to require that we do the impossible, namely, consider the impact of every event before estimating the impact of any event. If we think of happiness as a general state that is determined by innumerable events, it does indeed seem likely that no single event will have the power to influence our general happiness for very long.
Indeed, those events that seem to make a big difference (e.g., moving to a new country) tend to be those that give rise to many other events, which suggests that the ramifications of an event - that is, the sheer number of experiences it alters - may be the best predictor of its ultimate emotional impact. Although few parents would believe it, the death of a spouse may have more impact than the death of a child, simply because the former produces more changes in one's life than does the latter (see Lehman et al., 1993). In any case, it seems quite clear that focusing on an event can cause us to overestimate the duration of its influence on our happiness, and, hence, to miswant.
Immune Neglect: The Invisible Shield
Many shrewd observers of the human condition have remarked on people's extraordinary ability to change the way they feel simply by changing the way they think. When circumstances threaten our psychological well-being, we execute an assortment of cognitive strategies, tactics, and maneuvers that are designed to prevent, limit, or repair the damage (e.g., Festinger, 1957; Freud, 1937; Steele, 1988; Taylor & Brown, 1988; Vaillant, 1993; Westen, 1994).
These maneuvers usually have two properties. First, they work like a charm, enabling all of us to be well above average in all the ways that count. Second and more important, we tend not to know we are executing them, and what looks like rationalization to the giggling onlooker feels very much like rational reasoning to us. Taken together, the mechanisms that protect the sources of our psychological well-being (e.g., our sense of competence, integrity, and worth) in the face of assault constitute a psychological immune system that seems to be both powerful and invisible to the person it serves.
If our happiness is, in fact, defended by an invisible shield, then it is easy to see why we overestimate our vulnerability to the slings and arrows of outrageous fortune. Recall that voters in the Gilbert et al. (1988) study overestimated the duration of their emotional reactions to their candidate's electoral triumph or defeat.
Interestingly, voters in that study were also asked to predict how their opinions of the candidates would change once one was elected, and their answers may tell us something about why they overestimated the durability of their emotions. Although voters flatly denied that the outcome of the election would change their opinions of the candidates by even a hair, a month after the election, those voters whose candidate had lost had experienced an unforeseen transformation: Although the new governor had yet to take office, had yet to perform an official act, and had yet to make a substantive speech, those who had voted against him had a significantly higher opinion of him than they had had a month earlier. It seems that those voters overestimated the duration of their disappointment because they did not realize that once they were stuck with a governor whom they had not wanted, their psychological immune systems would help them locate 16 new reasons to like him anyway.
Gilbert et al. (1998) provided direct experimental evidence of immune neglect: the tendency for people to fail to consider how readily their psychological immune systems will vitiate their despair. Students were given the opportunity to apply for an exciting and lucrative position as an ice cream taster in a model business. The application process included answering several questions before a video camera while judges watched from another room. The situation was arranged such that if students were rejected, their psychological immune systems would have much more work to do in one condition than the other.
Specifically, students in the “difficult rationalization” condition were shown a number of highly relevant questions and were told that while answering these questions they would be observed by a panel of judges, who would then vote on the student's appropriateness for the job. Unless the judges unanimously disapproved of the student, he or she would be offered the job. In the “easy rationalization” condition, students were shown a number of largely irrelevant questions and were told that while answering these questions they would be observed by a single judge who would solely determine whether or not they were offered the job. Students in each condition predicted how they would feel if they were rejected, and how they would feel 10 minutes later. All participants then answered the relevant or irrelevant questions before the video camera and were promptly rejected. Their happiness was measured immediately following the rejection and then again 10 minutes later.
As the top (left) part of Figure 8.1 shows, the students believed they would be much less happy immediately following rejection than they actually turned out to be. But as the bottom (right) part of Figure 8.1 shows, the more interesting effect occurred 10 minutes later; Not only were all the students happier than they expected to be 10 minutes after being rejected, but they were happier when they had been rejected by a solo judge who had heard them answer irrelevant questions than when they had been rejected by a panel of judges who had heard them answer irrelevant questions.
This difference reveals the work of the psychological immune system, which should have found it easier to heal the wounds of rejection in the easy rationalization condition (“One guy doesn't think I'm competent. So what? Maybe I look like his ex-roommate, or maybe he's biased against Southerners, or maybe he just didn't have enough information to go on”) than in the difficult rationalization condition (“An entire group of judges agreed on the basis of adequate information that I'm not smart enough to taste ice cream? Yikes!”). The important point is that the students did not anticipate this difference, which suggests that when they looked into their emotional futures, they saw only the pain of rejection. What they did not consider was the ease or difficulty with which their psychological immune systems would dispatch their malaise.
Immune neglect can have important interpersonal consequences too. For example, few of us would expect to come undone if an irritated motorist shouted a few choice words about our parentage as we crossed against the light, but we might well expect to be shocked and dismayed if a good friend did the same. We expect an insulting remark from a stranger to be less painful than an insulting remark from a friend, and thus we might naturally expect the former to have less enduring emotional consequences than the latter. Gilbert and Lieberman (1998) asked pairs of college students to evaluate each other's personalities on the basis of brief autobiographies; in which they had explicitly been asked to describe some embarrassing incidents. Some students were told that they would work together as a team later in the experiment ("partners") and others were told that they would never meet ("strangers").
The students were asked to predict how they would feel a few minutes after finding out that the other student had read their autobiography and given them a very negative evaluation, and indeed, they predicted that they would feel worse if the negative evaluation came from their partner than from a stranger. In fact, the students were considerably happier after receiving a negative evaluation from their partner than a from a stranger, and they even forgave their partners more readily than they forgave strangers. Why should this have happened?
Once again, the invisibility of the psychological immune system seems to explain these paradoxical results. Most of us find it rather uncomfortable to interact with people we do not like, and so we are highly motivated to like those with whom we must interact (Darley & Berscheid, 1967). Our psychological immune systems work much harder to help us find ways to forgive our partner's transgressions (“My partner probably didn't realize that the embarrassing incident I wrote about in my autobiography was a unique occurrence, and now that I think of it, I'd probably have made the same negative evaluation myself if I were in the same position”) than to forgive the transgressions of strangers. The insulted students' psychological immune systems did what they were designed to do by enabling them to feel happy about working with someone who had evaluated them negatively. What is interesting, of course, is that the students were unable to predict this outcome just 10 minutes before it happened. Like most of us, they blithely predicted that a big pain would last longer than a little one, unaware that big pains often evoke remedies that little pains do not. Broken legs hurt so much that they cry out to be fixed, whereas trick knees are often allowed to go on hurting for a lifetime.
Immune neglect can cause us to miswant by causing us to fear and avoid outcomes that will not, in the long run, hinder our happiness. But one ironic consequence of the failure to anticipate the operation of the psychological immune system is that we may inadvertently do things that impair its operation, thereby undermining our own hidden talent for happiness. For example, if given the opportunity to shop at a store that allows customers to return merchandise for any reason and another store at which all sales are final, most of us would patronize the first rather than the second - and we might even be willing to pay a bit more just so we could have the luxury of changing our minds later on. We firmly believe that bridges ought to be there for crossing and recrossing, and our aversion to burning them is probably wise in many respects. But if keeping one's options open is wise in many respects, it is not wise in all respects, because open options have the unfortunate consequence of paralyzing the psychological immune system. As dissonance theorists have long noted, it is the firm commitment to a single course of action that most effectively triggers attempts to justify it.
Gilbert and Jenkins (1998) gave college students a short course in black-and-white photography. The students took photographs of their favorite people and places on campus and were then taught how to develop their photographs. After students had printed their two favorite photographs, they were asked to donate one of them to the experimenter's “photography project.” Some students were told that the donated photograph would be mailed to England that evening, whereas others were told that the photograph would not be mailed for 5 days. Students in this latter condition were told that if they changed their minds about which photograph to keep after they made the donation, they could swap the chosen for the donated photograph anytime before it was mailed. When the students' happiness with their photographs was measured 2 days later, those whose decisions were reversible did not like the chosen photograph as much as did those students whose decisions were irreversible. This makes sense inasmuch as these students were probably still in the process, of deciding which photograph they would keep, and thus they did not yet have a final outcome with which their psychological immune systems could help them feel happy. But interestingly, 9 days later, the irreversible deciders were still happier with their photographs than were the reversible deciders - despite the fact that the reversible deciders' “swapping opportunity” had expired days ago and their unchosen photograph was irrevocably winging its way across the Atlantic. It seems that merely having had a brief opportunity to change their minds prevented reversible deciders from ever exercising their hidden talent for happiness.
All of this work on immune neglect leads to one conclusion: Our tendency to neglect the operation of the immune system when anticipating the future can have unhappy consequences. We often want one thing so much more than another that we willingly incur enormous costs in our attempts to avoid the unwanted event. We may spend little time with our children and neglect our hobbies while putting in long hours at the office because we are convinced that keeping our current job will be better than being forced to find a new one. What we fail to realize is that while the thing we wanted to experience is in some ways better than the thing we wanted to avoid, it is probably worse in others, and should we fail to achieve what we wanted, our psychological immune systems will quickly help us locate the ways in which the thing we got was better than the thing we were aiming for. As the man who narrowly missed the opportunity to franchise the first McDonalds restaurant (and hence narrowly missed the opportunity to become a billionaire) noted many decades later, “I believe it turned out for the best” (Van Gelder, 1997). If we do indeed have a greater talent for happiness than we recognize, then our ignorance of this talent may cause us to pay a steeper price for future experiences than we should.
Conclusions
The naïve psychology of happiness is simple: We want, we try, we get, we like. And then, with the help of television commercials, we want some more. Wants are underwritten by our beliefs about the relation between getting and liking, and in this sense they are prescriptions for action. They tell us what to do with our time by telling us what to aim for and what to avoid, and we allow ourselves to be steered by them because we trust that they are, by and large, correct. Most of us feel certain that if we could experience all the events and only the events we want to experience, happiness would inevitably follow.
The research discussed in this chapter suggests that there are at least two flaws in the naïve analysis of happiness. First, our wants are, like any other prediction, susceptible to error. We may misconstrue events, misunderstand ourselves, misinterpret our feelings - and any of these mistakes can be a cause of miswanting. In short, things do not always feel the way we expect them to feel.
Second, even if we could predict how much we would like an event when it happened, we might still be unable to predict how that event would affect us in the long run. One reason is that our general happiness is influenced by a multitude of events. It is impossible to consider all of these influences every time we consider one of them, of course, but unless we do just that, we have little hope of correctly predicting the future states that are their conjoint products.
A second reason why we have trouble predicting the enduring emotional consequences of an event is that liking does not follow from getting so much as it accommodates it. Although our initial emotional reaction to an event is usually based on those properties of the event that caused us to aim for it or avoid it in the first place, once a particular outcome is achieved, we have an uncanny ability to reconstrue it in terms of its most sanguine properties. Because we do not recognize how easily we can reconstrue events in this way, we anticipate more enduring reactions than we often have.
“In the world there are only two tragedies,” wrote Oscar Wilde (1893). “One is not getting what one wants, and the other is getting it.” We all chuckle and nod knowingly when we hear this clever quip, but not one of us believes it for a moment. Rather, our chuckling and nodding are licensed by a serene certainty that the things we run after will, in fact, bring us far greater happiness than the things we run from.
The research discussed in this chapter does not suggest that all ends are emotionally equivalent or that all desires are misdirected. Rather, it merely suggests that if we could know the future, we still might not know how much we would like it when we got there. The psychological mechanisms that keep us from this knowledge are many, and a better understanding of them seems well worth wanting.
For references, please see the PDFs linked to at the beginning of this article.
The writing of this chapter was supported by research grant RO1-MH56075 from the National Institute of Mental Health to Daniel T. Gilbert and Timothy D. Wilson. We thank Joe Forgas for his insightful comments on an earlier version of this chapter, and the other authors in this volume for helpful discussions of these issues. Correspondence concerning this work should be addressed to Daniel T. Gilbert, Department of Psychology, Harvard University, William James Hall, Cambridge, MA 02138 E-mail: dtg@wjh.harvard.edu.
Share this post