CURRENT DI RE CTIONS IN PSYCHO LOGICAL S CIENCE

Affective Forecasting
Knowing What to Want
Timothy D. Wilson1 and Daniel T. Gilbert2
1

University of Virginia, 2Harvard University

ABSTRACTPeople

base many decisions on affective forecasts, predictions about their emotional reactions to future
events. They often display an impact bias, overestimating
the intensity and duration of their emotional reactions to
such events. One cause of the impact bias is focalism, the
tendency to underestimate the extent to which other events
will influence our thoughts and feelings. Another is peoples
failure to anticipate how quickly they will make sense of
things that happen to them in a way that speeds emotional
recovery. This is especially true when predicting reactions
to negative events: People fail to anticipate how quickly
they will cope psychologically with such events in ways that
speed their recovery from them. Several implications are
discussed, such as the tendency for people to attribute their
unexpected resilience to external agents.

KEYWORDSaffective forecasting; prediction; emotion; sense

making
Many cultures have myths in which people can make their wishes
come true. The story of Aladdin and his lamp is best known to
readers of the Arabian Nights (and to Disney fans); in Irish legends, it is leprechauns who make wishes come true; whereas in a
Chinese fable it is an obliging dragon that has the head of a camel,
the eyes of a hare, the neck of a snake, the claws of an eagle, and
the ears of a buffalo (McNeil, 2003).
Common to these myths is the notion that if people (perhaps
with the help of a genie) could make their wishes come true, they
would achieve everlasting happiness. Sometimes, however,
people are disappointed by the very things they think they want.
Research on affective forecasting has shown that people routinely
mispredict how much pleasure or displeasure future events will
bring and, as a result, sometimes work to bring about events that
do not maximize their happiness.

Address correspondence to Timothy D. Wilson, P.O. Box 400400, 102

Gilmer Hall, University of Virginia, Charlottesville, VA 22904-4400,
e-mail: twilson@virginia.edu, or to Daniel Gilbert, Department of
Psychology, William James Hall, 33 Kirkland Street, Harvard University, Cambridge, MA, 02138, e-mail: gilbert@wjh.harvard.edu.

Volume 14Number 3

These mispredictions can take a number of forms. People can

be wrong about how positive or negative their reactions to future
events will be, particularly if what unfolds is different from what
they had imagined. Prospective dog owners might predict that
Rover will bring nothing but joy because they picture a faithful
companion who obediently fetches the newspaper each morning
instead of an obstinate beast who chews shoes and demands 6:00a.m. walks in the freezing rain. Generally, however, humans are
adept at predicting whether events are likely to be pleasant or
unpleasant. Even a rat can readily learn that pressing one bar will
produce a food pellet and another an electric shock and will vote
with its paws for the more pleasant option. People know that a root
beer will be more pleasant than a root canal.
People are less adept at predicting the intensity and duration of
their future emotional reactions. Occasionally they underestimate intensity and duration; this may happen, for example, when
a person is in a cold emotional state at the time of prediction and
is trying to imagine being in a hot emotional state in the future.
Satiated shoppers underestimate how much they will want ice
cream later in the week, and addicts who have just injected heroin
underestimate how much they will crave the drug when they are
deprived of it later (Gilbert, Gill, & Wilson, 2002; Loewenstein,
ODonoghue, & Rabin, 2003).
THE IMPACT BIAS

More common than underestimating future emotional reactions,

however, is the impact bias, whereby people overestimate the
intensity and duration of their emotional reactions to future
eventseven when they know what the future event is likely to
entail and they are not in a particularly hot or cold emotional
state at the time of making their forecast. This error has been
found repeatedly in a variety of populations and contexts. College
students overestimated how happy or unhappy they would be
after being assigned to a desirable or undesirable dormitory (see
Fig. 1), people overestimated how unhappy they would be 2
months after the dissolution of a romantic relationship, untenured college professors overestimated how unhappy they would
be 5 years after being denied tenure, women overestimated how
unhappy they would be upon receiving unwanted results from a

Copyright r 2005 American Psychological Society

131

Affective Forecasting

SENSE MAKING AND PEOPLES IGNORANCE OF IT

7
Predicted
Actual

Happiness Rating

6
5
4
3
2
1

Desirable

Undesirable

Desirability of Dormitory
Fig. 1. College students predicted and actual levels of happiness after
dormitory assignments. Participants predicted what their overall level
of happiness would be a year later if they were randomly assigned to a desirable or undesirable dormitory (on a 7-point scale, with 1 5 unhappy and
7 5 happy). Students predicted that their dormitory assignment would
have a large positive or negative impact on their overall happiness (solid
bars); but a year later, those living in undesirable and desirable dormitories
were at nearly identical levels of happiness (open bars). Adapted from
Dunn, Wilson, & Gilbert (2003).

pregnancy test, and so on (see Loewenstein et al., 2003; Mellers

& McGraw, 2001; Wilson & Gilbert, 2003). The impact bias is
important because, when deciding what to work for, people need
to predict not only the valence (positivity or negativity) of their
emotional reactions (Will I feel good or bad?), but also the
intensity and duration of these reactions (e.g., Will I feel good
for a few seconds or a few months?). If consumers overestimate
the intensity and duration of the pleasure they will get from
purchasing a new car, for example, they may be better off
spending their money in some other way.
One cause of the impact bias is focalism, the tendency to
overestimate how much we will think about the event in the future
and to underestimate the extent to which other events will influence our thoughts and feelings (Schkade & Kahneman, 1998;
Wilson, Wheatley, Meyers, Gilbert, & Axsom, 2000). When
football fans think about how they will feel after their favorite
team wins an important game, for example, they are likely to focus
exclusively on the game and neglect to think about the many other
thingssuch as upcoming deadlines at work, the need to get the
car fixed, or a visit from old family friendsthat will influence
their thoughts and feelings. Focalism is a straightforward and, we
suspect, quite common source of the impact bias. It can be corrected, to some degree, by asking people to think carefully about
the many other events that will demand their attention in the
future; studies have found that this exercise tempers peoples
predictions about the impact of a victory or loss by their favorite
football team on their happiness (Wilson et al., 2000).

132

Another cause of the impact bias is that forecasters fail to recognize how readily they will make sense of novel or unexpected
events once they happen. Research across a variety of fields
suggests that such events trigger four processes in sequence:
attention, reaction, explanation, and adaptation.
 First, people are especially likely to attend to events that are
self-relevant but poorly understood. For example, a student
who unexpectedly receives an A on an important exam will
initially think about little else.
 Second, people react emotionally to self-relevant, poorly
understood events. The student who receives an unexpected
A will initially feel overjoyed.
 Third, people attempt to explain or make sense of self-relevant, poorly understood events. For example, the overjoyed
student will begin to search for reasons why she received a
better-than-expected grade.
 Fourth, by making sense of events, people adapt emotionally
to them. Once the student has explained the reasons for her
grade, she will think about her achievement less and experience less happiness when she does think about it. The event
will come to be seen as more normal and inevitable then it
actually was, and hence it will lose some of the emotional
power that it had when it still seemed extraordinary.
These four processes may seem relatively uncontroversial to
psychologists, but research suggests that people neglect to take
them into account when forecasting their future emotions. In
particular, because the processes by which people explain or
make sense of unexpected events are often quick and nonconscious, people do not recognize beforehand that such processes
will occur; thus they do not consider how quickly their tendency
to explain events will reduce the impact of those events. When a
student tries to predict how she will feel if she receives an unexpected A, she has little trouble imagining herself feeling
overjoyed but a lot of trouble imagining herself explaining the
event in a way that makes it seem ordinary and predictable.

The Pleasure of Uncertainty About Positive Events

If making sense of positive events reduces the duration of the
pleasure they cause, then inhibiting the sense-making process
should prolong peoples pleasure. In one study, for example,
students who were studying in a library were unexpectedly given
an index card with a dollar coin attached, and results showed that
they were in a better mood 5 minutes later if the text on the card
made it difficult rather than easy for them to explain why they had
received the money. Yet people did not anticipate this effect; in
fact, forecaster participants predicted that they would be
happier if the card made explanation easy rather than difficult
(Wilson, Centerbar, Kermer, & Gilbert, 2005). People do not
realize how quickly they will make sense of unexpected positive

Volume 14Number 3

Timothy D. Wilson and Daniel T. Gilbert

events and how doing so will make their positive emotions dissipate.

A Pleasure Paradox
Most organisms avoid that which has previously caused them
pain and approach that which has previously given them pleasure. Humans are better at this than most other animals because
they do more than merely associate stimuli with their affective
consequences. People are naive scientists who explain events to
themselves, and the sophisticated causal theories people generate allow them to pursue pleasures and avoid pains with an
unusual degree of success. But an ironic consequence of this
inveterate sense making is that events tend to lose some of their
hedonic impact as they become more sensible. People work to
understand events so that they can repeat the good ones and avoid
repeating the bad ones, but in understanding these events people
may reduce their ability to be moved by them. True, some explanations of events make people feel better than other explanations do; taking credit for a major success is more pleasurable,
for example, than attributing it to luck. Independent of the favorability of the explanation, however, sense making hastens
emotional recovery from events. Things are rarely as good or
bad as people expect them to be because people do not realize
that by explaining the things that happen to them, they drain
these things of the hedonic qualities that caused them to focus on
the events in the first place.
NEGATIVE EVENTS: MOTIVATED SENSE MAKING

People are motivated to recover from negative emotional events,

and the kind of sense making they engage in often involves
coping, psychological defenses, and rationalization. Like the
physiological immune system that fights threats to physical
health, people have a psychological immune system that fights
threats to emotional well-being. These defenses have been well
documented by social and personality psychologists and include
dissonance reduction, motivated reasoning, self-serving attributions, self-affirmation, and positive illusions.
A feature that all these defenses have in common is that they
are largely unconscious, and in fact are more effective by operating behind the mental scenes. When trying to cope with a romantic breakup, for example, people usually will not be able
deliberately and consciously to adopt a more negative view of
their partner in order to make themselves feel better. Instead, the
ex-partner will come to seem less suitable, with no awareness that
ones own psychological immune system was responsible for this
shift in view. Because people are generally unaware of the operation of these defenses, they tend not to take them into account
when predicting their future emotional reactionsan oversight
we have termed immune neglect.
In one study, for example, participants who failed to get a desirable job were less upset 10 minutes later when the failure was

Volume 14Number 3

attributable to a single capricious interviewer (easy to rationalize: The guys a jerk) rather than to a team of interviewers
(difficult to rationalize: How could they all dislike me?). In
another study, participants were less upset when they received
negative personality feedback from a computer (easy to rationalize: Computers make mistakes) than from a clinician (difficult to rationalize: How could I have scored so badly on the
personality test?). In both cases, people had stronger reactions
when unexpected negative events were difficult to rationalize and
explain, but in both cases they failed to anticipate that this would
happen (Gilbert, Pinel, Wilson, Blumberg, & Wheatley, 1998).
Implications of Immune Neglect
Peoples failure to anticipate their natural tendency to make the
best of bad outcomes has a number of consequences:
 Because people do not recognize that they have reduced the
impact of negative events by explaining and rationalizing
them, they sometimes attribute their unexpected resilience to
the work of powerful, insightful, and benevolent external
agents (Gilbert, Brown, Pinel, & Wilson, 2000). For example,
employees who are transferred to undesirable locations might
be surprised by how happy they are; by failing to recognize
that they produced their own happiness with nonconscious
coping and defensive processes, they might attribute their
good fortune to the guiding hand of an external agent, such as
God.
 When people make a decision that is difficult to reverse, such
as buying a sweater from a store with a no returns policy,
they are strongly motivated to rationalize the decision and
make the best of it. When people can more easily undo a
decision, such as buying a sweater they can return, they are
less motivated to rationalize their choice, because they can
always change their minds. Consequently people are often
happier with irrevocable choices because they do the psychological work necessary to rationalize what they cant
undo. Because people do not realize in advance that they will
work harder to rationalize irreversible decisions, however,
they often avoid the binding commitments that would actually increase their satisfaction (Gilbert & Ebert, 2002). For
example, many people pay more to purchase clothing from
stores with a liberal return policy, when they would more
satisfied with clothes they bought that they could not return.
 Not surprisingly, people believe that major traumas will have
a more enduring emotional impact than minor ones will.
Because people are more strongly motivated to make sense of
major traumas than minor ones, however, the pain of minor
traumas can sometimes last longer than more serious ones. It
seems like it would be worse, for example, to be insulted by
a close friend than a stranger. Because people are more
motivated to cope with (and perhaps rationalize) the insult
from the friend, however, they may recover from it more
quickly (Gilbert, Lieberman, Morewedge, & Wilson, 2004).

133

Affective Forecasting

 It is well-known that people weigh potential losses more

heavily than corresponding gains, which often leads to economically illogical decisions. Kermer, Driver-Linn, Wilson,
and Gilbert (2005), for example, found that most people refused a gamble in which they had a 50% of winning $5 and
a 50% chance of losing only $3, demonstrating classic loss
aversion. Loss aversion seems to involve a faulty affective
forecast: Although participants predicted that losing a gamble would have a larger emotional impact than winning, they
were wrong; the magnitude of unhappiness caused by losing
was no greater than the magnitude of happiness caused by
winning (Kermer et al., 2005).
SUMMARY AND FUTURE DIRECTIONS

Affective forecasts are important because people base many

decisions on them. Decisions about who to marry, what career to
pursue, and whether to donate money to the local homeless
shelter are based, at least in part, on predictions about how these
decisions will make one feel. To the extent that peoples predictions about what will make them happy are flawed, people fail at
maximizing their happiness.
One unanswered question is whether the impact bias is advantageous in some way. It could be argued that exaggerating the
impact of emotional events serves as a motivator, making people
work hard to obtain things that they predict will have large positive
consequences and avoid things that they predict will have large
negative consequences. It may be, however, that overestimating
the impact of negative events creates unnecessary dread and
anxiety about the future. And there are other costs to affectiveforecasting errors. People suffering from debilitating digestive
disorders who underestimate how quickly they will adapt to an
ostomy bag might make less-than-optimal treatment decisions.
People who overestimate the positive emotional impact of undergoing cosmetic surgery might be too willing to get an extreme
makeover. Finding ways to increase the accuracy of affective
forecasts is a worthy enterprisethough not, we suspect, a particularly easy one (Ubel et al., 2001). It is difficult to place oneself
in the future and imagine what it will be like to have made sense of
an event that, in the present, seems extraordinary. Such mental
time traveling, however, might ultimately lead to better decisions.

Recommended Reading
Gilbert, D.T., Driver-Linn, E., & Wilson, T.D. (2002). The trouble with
Vronsky: Impact bias in the forecasting of future affective states. In
L.F. Barrett & P. Salovey (Eds.), The wisdom in feeling: Psychological processes in emotional intelligence (pp. 114143). New
York: Guilford.
Loewenstein, G., ODonoghue, T. and Rabin, M. (2003). (See References)
Mellers, B.A., & McGraw, A.P. (2001). (See References)
Wilson, T.D. (2002). Strangers to ourselves: Discovering the adaptive
unconscious. Cambridge, MA: Harvard University Press.

134

Wilson, T.D., & Gilbert, D.T. (2003). (See References)

AcknowledgmentsMuch of the research discussed in this

article was supported by research grant #RO1-MH56075 from
the National Institute of Mental Health to the authors.

REFERENCES
Dunn, E.W., Wilson, T.D., & Gilbert, D.T. (2003). Location, location,
location: The misprediction of satisfaction in housing lotteries.
Personality and Social Psychology Bulletin, 29, 14211432.
Gilbert, D.T., Brown, R.A., Pinel, E.C., & Wilson, T.D. (2000). The illusion of external agency. Journal of Personality and Social Psychology, 79, 690700.
Gilbert, D.T., & Ebert, J.E. (2002). Decisions and revisions: The affective forecasting of changeable outcomes. Journal of Personality
and Social Psychology, 82, 503514.
Gilbert, D.T., Gill, M., & Wilson, T.D. (2002). The future is now: Temporal correction in affective forecasting. Organizational Behavior
and Human Decision Processes, 88, 430444.
Gilbert, D.T., Lieberman, M.D., Morewedge, C., & Wilson, T.D. (2004).
The peculiar longevity of things not so bad. Psychological Science,
15, 1419.
Gilbert, D.T., Pinel, E.C., Wilson, T.D., Blumberg, S.J., & Wheatley, T.P.
(1998). Immune neglect: A source of durability bias in affective forecasting. Journal of Personality and Social Psychology, 75,
617638.
Kermer, D.A., Driver-Linn, E., Wilson, T.D., & Gilbert, D.T. (2005). Loss
aversion applies to predictions more than experience. Unpublished
raw data, University of Virginia.
Loewenstein, G., ODonoghue, T., & Rabin, M. (2003). Projection bias in
predicting future utility. Quarterly Journal of Economics, 118,
12091248.
Mellers, B.A., & McGraw, A.P. (2001). Anticipated emotions as guides
to choice. Current Directions in Psychological Science, 10,
210214.
McNeil, D.G. Jr. (2003, April 29). Dragons, a brief history in long miles.
New York Times (p. F2).
Schkade, D.A., & Kahneman, D. (1998). Does living in California make
people happy? A focusing illusion in judgments of life satisfaction.
Psychological Science, 9, 340346.
Ubel, P.A., Loewenstein, G., Hershey, J., Baron, J., Mohr, T., Asch, D., &
Jepson, C. (2001). Do nonpatients underestimate the quality of life
associated with chronic health conditions because of a focusing
illusion? Medical Decision Making, 21, 190199.
Wilson, T.D., Centerbar, D.B., Kermer, D.A., & Gilbert, D.T. (2005). The
pleasures of uncertainty: Prolonging positive moods in ways people
do not anticipate. Journal of Personality and Social Psychology, 88,
521.
Wilson, T.D., & Gilbert, D.T. (2003). Affective forecasting. In M.P. Zanna
(Ed.), Advances in experimental social psychology (Vol. 35, pp.
345411). San Diego, CA: Academic Press.
Wilson, T.D., Wheatley, T.P., Meyers, J.M., Gilbert, D.T., & Axsom,
D. (2000). Focalism: A Source of durability bias in affective
forecasting. Journal of Personality and Social Psychology, 78,
821836.

Volume 14Number 3