The motivating function of thinking about the future: Expectations versus fantasies

Abstract

Two forms of thinking about the future are distinguished: expectations versus fantasies. Positive expectations (judging a desired future as likely) predicted high effort and successful performance, but the reverse was true for positive fantasies (experiencing one's thoughts and mental images about a desired future positively). Participants were graduates looking for a job (Study 1), students with a crush on a peer of the opposite sex (Study 2), undergraduates anticipating an exam (Study 3), and patients undergoing hip-replacement surgery (Study 4). Effort and performance were measured weeks or months (up to 2 years) after expectations and fantasies had been assessed. Implications for the self-regulation of effort and performance are discussed.

Full text

The Motivating Function of Thinking About the Future:
Expectations Versus Fantasies
Gabriele Oettingen
New York University and University of Hamburg Doris Mayer
University of Hamburg
Two forms of thinking about the future are distinguished: expectations versus fantasies. Positive
expectations (judging a desired future as likely) predicted high effort and successful performance, but the
reverse was true for positive fantasies (experiencing one’s thoughts and mental images about a desired
future positively). Participants were graduates looking for a job (Study 1), students with a crush on a peer
of the opposite sex (Study 2), undergraduates anticipating an exam (Study 3), and patients undergoing
hip-replacement surgery (Study 4). Effort and performance were measured weeks or months (up to 2
years) after expectations and fantasies had been assessed. Implications for the self-regulation of effort and
performance are discussed.
Research consistently finds that optimistic thinking about the
future fosters motivation and successful performance, whereas
pessimistic thinking dampens motivation and successful perfor-
mance (Bandura, 1997; Heckhausen, 1991; Seligman, 1991; Tay-
lor & Brown, 1988). Optimistic thinking is associated with suc-
cessful cognitive and self-regulatory problem solving, with
prosocial and helping behavior, with setting high standards and
aspirations, and with indicators of mental health—all of these
factors are essential for well-being and personality development.
Optimistic thinking even hinders the emergence of acute and
chronic disease and slows its progress (C. Peterson & Bossio,
1991; Scheier & Carver, 1992; Taylor, Kemeny, Reed, Bower, &
Gruenewald, 2000).
Such beliefs about the future, or expectancy judgments, are
conceptualized as self-efficacy expectations (i.e., whether one can
perform a certain behavior in its relevant context; Bandura, 1997),
as outcome expectations (i.e., whether performing a certain behav-
ior will lead to the desired outcome; Bandura, 1997), as general
expectations (i.e., whether a certain event will occur, thereby
encompassing both efficacy and outcome expectations; Heck-
hausen, 1991; Oettingen & Wadden, 1991; Reed, Kemeny, Taylor,
& Visscher, 1999), or as generalized expectations (i.e., whether the
future in general will be positive or negative; Scheier & Carver,
1992). Generalized expectations are also indirectly measured
through a person’s habitual use of causal attributions for past
events (Abramson, Seligman, & Teasdale, 1978). When attribu-
tions to positive events are more stable and global than attributions
to negative events, expectations are said to be optimistic: Positive
events can be expected to last longer and generalize more across
situations than negative events (C. Peterson & Seligman, 1984).
Expectations and performance are reciprocally determined
(Bandura, 1978). High expectations of success lead to strong
performance, and strong performance leads to high expectations of
success. Expectations are thus informed by past experiences and
thereby represent a person’s performance history (Bandura, 1977,
1997; Mischel, 1973; Mischel, Cantor, & Feldman, 1996; Olson,
Roese, & Zanna, 1996). Observed performances of others, persua-
sive messages received by respected others, and experienced levels
of arousal during performance are also known to influence expec-
tations (Bandura, 1997).
Alternative forms of thinking positively about the future, which
are based less on past experiences, and thus are less informed by
a person’s performance history, seem to be less beneficial for
effortful action, performance, and well being. For example, wish-
ful thinking and other avoidant coping styles are linked to lower
effort, performance, and well-being compared with planning and
confrontative coping styles (e.g., Holahan & Moos, 1986; Lengua
& Sandler, 1996; Ried, Dubow, & Carey, 1995). For example,
avoidance coping in cancer patients at diagnosis (e.g., “I try not to
think about it”) predicted disease progression 1 year later (Epping-
Jordan, Compas, & Howell, 1994), and in a recent prospective
study (over 2 years), HIV-negative and HIV-positive caregivers
high in cognitive avoidance showed higher levels of physical
symptoms (Billings, Folkman, Acree, & Moskowitz, 2000). More-
over, avoiding or lacking information about upcoming medical
procedures is less beneficial than mentally facing the painful future
events, for both children and adults (L. Peterson, Oliver, & Sal-
dana, 1997; Suls & Wan, 1989; Taylor & Clark, 1986).
Gabriele Oettingen, Department of Psychology, New York University
and Department of Psychology, University of Hamburg, Hamburg, Ger-
many; Doris Mayer, Department of Psychology, University of Hamburg.
Preparation of this article was supported by German Science Foundation
Grant Oe-237/1-1 awarded to Gabriele Oettingen. We are grateful to
Kathrin Doll, Conny Ho¨hle, Christiane Schreiber, and Katherine M. Shar-
key for their assistance in collecting and analyzing the data and to Peter M.
Gollwitzer, Heidi Grant-Pillow, Christie L. K. Kawada, Hyeon-ju Pak,
Diana Rus, Paul Silvia, and two anonymous reviewers for making valuable
comments on previous versions of the article.
Correspondence concerning this article should be addressed to Gabriele
Oettingen, Department of Psychology, New York University, 6 Washing-
ton Place, 7th Floor, New York, New York 10003. E-mail: gabriele
.oettingen@nyu.edu
Journal of Personality and Social Psychology Copyright 2002 by the American Psychological Association, Inc.
2002, Vol. 83, No. 5, 1198–1212 0022-3514/02/$5.00 DOI: 10.1037//0022-3514.83.5.1198
1198

Wishful thinking about future stressors is linked to neuroticism
(Bolger, 1990), sadness, and anger (Spirito, Stark, & Tyc, 1994),
and impedes the mastery of impending problems, especially when
mastery cannot be achieved with ignorance, but demands vigilance
and effortful action (Carver, Scheier, & Weintraub, 1989; Lazarus,
1983). Students who reported to habitually deny stressful events
felt more threatened by an upcoming exam than students who used
less denial (Carver & Scheier, 1994).
Experimental research on how self-regulatory thought affects
task completion further suggests that positive thoughts are not
always beneficial for effort and performance. Goodhart (1986)
reported unfavorable effects of positive task-related thoughts or
images on success in solving anagrams. The findings, however,
were only observed in people who had not been asked to judge
their past performance prior to solving the anagram (i.e., for those
participants whose spontaneous images had not been overridden
by their expectancy judgments based on the facts of the past). An
earlier study yielded similar findings (Sherman, Skov, Hervitz, &
Stock, 1981). Participants imagining hypothetical failure before an
anagram task did better than people imagining hypothetical suc-
cess. Yet this was true only if participants did not have to make
expectancy judgments of failure or success prior to the anagram
test. If they did formulate their performance expectations for the
anagram tasks, participants in the success condition fared better
than those in the failure condition. Finally, Spencer and Norem
(1996) demonstrated that images of perfect mastery hurt perfor-
mance in high-achieving students.
Expectations Versus Fantasies as Different Forms of
Thinking About the Future
These studies suggest that there might be two forms of
thinking about the future with different effects on motivation
and performance. More specifically, beliefs about the future
(expectations) should be differentiated from images (fantasies)
depicting future events. The difference between beliefs and
images was first suggested by William James (1890/1950):
“Everyone knows the difference between imagining a thing and
believing in its existence, between supposing a proposition and
acquiescing in its truth”(p. 283). James’s differentiation be-
tween believing and imagining pertained to events of the past or
the present. Following his reasoning, we differentiate two kinds
of thinking about the future: beliefs (expectancy judgments)
that assess probability of occurrence, and images (fantasies)
that contain future events per se as they appear in the stream of
thought. Positive expectancy judgments, then, are beliefs that a
desired event is likely to occur; positive fantasies about the
future, in contrast, are defined as positively experienced images
of future desired events that emerge in the stream of thought.
Positive fantasies, as used here, differs from what Lewin
(1926) and Mahler (1933) called “Zauberdenken”(i.e., thoughts
depicting actions and events that violate known natural laws),
and it resembles what Klinger (1971, 1978) named “day-
dreams,”that is, thoughts pertaining to immediate or longer
range desires including instrumental activities to attain the
desired future.
Expectations Versus Fantasies and Their Relation to
Motivation and Performance
Because positive expectations judge future events’likelihood by
applying past facts to future events (Bandura, 1977, 1997; Mischel,
1973), they are a valid basis for strong behavioral investment.
Positive fantasies, to the contrary, embellish future events regard-
less of past performance and probability of future occurrences
(Klinger, 1990; Singer, 1966). Therefore, they fail to be a solid
basis for acting. Moreover, positive fantasies about desired future
outcomes and about effortlessly moving toward them should se-
duce a person to mentally enjoy the desired future in the here and
now, thereby yielding little motivation to implement the desired
future in actuality. Finally, because positive fantasies conceal the
necessity to act toward attaining the desired future, they will
prevent a person from preparing for upcoming obstacles and
temptations, and from planning how to overcome them (Oettingen,
1996, 2000; Oettingen, Pak, & Schnetter, 2001). Lacking prepa-
ratory action and planning should further compromise success
(Gollwitzer, 1999; Taylor, Pham, Rivkin, & Armor, 1998).
Positive fantasy about a desired future can be respondent versus
operant (i.e., impulsive vs. volitional; Klinger 1978) or coherent
versus incoherent (a well-integrated scene vs. shreds of thoughts;
Klinger, 1978), and the perspective can be Ior me (James, 1890/
1950; Klinger, 1978; Watkins, 1976). Of most importance, posi-
tive fantasy can pertain to mentally enjoying future outcomes, and
to mentally enjoying a future smooth and effortless progress to-
ward that outcome. In other words, the positive versus negative
tone of one’s fantasies about the future can be based on mentally
experiencing having attained the outcome, moving smoothly to-
ward it, or both. Regardless of whether positive fantasy is respon-
dent or operant, coherent or incoherent, focuses on the me or on the
experiencing I, and whether it is outcome based or process based,
it should lead to comparatively little effort and performance.
If, however, individuals begin questioning the unrestricted en-
joyment of the desired outcome and its smooth attainment in more
negatively toned fantasies, the desired future is no longer experi-
enced as merely enjoyable but as something to be achieved (Oet-
tingen, 1996, 2000; Oettingen et al., 2001). People start to lay out
the road to success, prepare for setbacks and hindrances, and at the
end will exert effort and show persistence. In sum, whereas pos-
itive expectations of success predict effortful action and successful
performance, positive fantasies about the future should predict the
reverse.
Related Approaches
Positive fantasies, whether outcome or process, should be a
motivational burden, because they hamper motivation to imple-
ment the desired future and conceal the steps needed for its
attainment. Thus, the present work on positive versus negative
fantasies differs from research on outcome versus process simu-
lations (Pham & Taylor, 1999; Taylor et al., 1998), which finds
superior effort and performance after process simulations (rehears-
ing the cumbersome steps needed to reach a set goal, e.g., getting
an A) than after outcome simulations (rehearsing the moment of
getting an A). Rather than focusing on the differential effects of
process simulations versus outcome simulations, the present ap-
proach focuses on the positivity of thoughts and images about the
1199
EXPECTATIONS VERSUS FANTASIES

future and postulates that positive fantasies (outcome and process)
are a motivational hindrance.
Further, positive fantasies need to be distinguished from illusory
optimism (Taylor & Brown, 1988; Taylor et al., 2000; Schneider,
2001; Wright, 2000). Because positive fantasies are mute to future
events’reality, they cannot be taken as an indicator of illusory
optimism. Only expectations can be illusory, because they assess
the future events’reality. This assessment of reality, then, can be
more or less realistic (accurate) or illusory (inaccurate).
Finally, positive fantasies differ from avoidant coping, among
them denial (Carver et al., 1989) and wishful thinking (Folkman &
Lazarus, 1985). Denial refers to people refusing to believe that
stressful events have happened, acting as if they did not happen,
and trying to convince themselves that they are not real (Carver et
al., 1989), and wishful thinking pertains to wishing a stressful
event or its emotional consequences away, or wishing to be a
stronger person (Folkman & Lazarus, 1985). Positive fantasies, to
the contrary, assess the experienced positivity of mental images
about a not yet reached future, rather than reports about wishing
past or ongoing stressful events to vanish.
Focus of the Present Studies
Four studies examine the relation between positive future
thought in terms of expectations versus fantasies, to effort and
successful performance. All four studies refer to changing the
status quo toward mastering a developmental task or life task.
Developmental tasks or life tasks arise at certain periods of the life
span (Havighurst, 1948/1972). They are contextual demands con-
strued as desires, such as finishing a dissertation or finding a
marriage partner (Cantor & Harlow, 1994; Cantor & Kihlstrom,
1987). People should thus find it easy to fantasize about a positive
future implied in mastering the life task.
Moreover, life tasks are an ideal topic for the present studies,
because they are hard to master. Fantasies about successfully
solving such hard tasks should seduce people to mentally indulge
in the successful future instead of taking on the hardships of
implementing it. Thus, life tasks are prone to show a negative
relation between positive fantasies and success. To the contrary, in
easy-to-master tasks that can be solved effortlessly and immedi-
ately (e.g., I stretch my hand to obtain a piece of cake), positive
fantasies should provide enough of an implicit pull effect to trigger
necessary action. The same is true for choices between promising
and available alternatives (e.g., I decide to take one offer rather
than the other). In sum, only when desired outcomes are hard to
come by, should positive fantasies seduce a person to mentally
enjoy the desired future in the here and now instead of effortfully
achieving the outcome in reality. These considerations are in line
with the work on delay of gratification (Mischel, 1974), where
consummatory mental images fostered consummatory action for
the immediate small reward (e.g., one marshmallow), but spoiled
attaining the delayed large reward (two marshmallows).
Participants in Studies 1–3 are young adults. Study 1 pertains to
the life task of entering professional life, Study 2 to the life task of
finding a partner, and Study 3 to the life task of academic achieve-
ment. Participants in Study 4 are elderly adults, and the life task
refers to recovering from a major health threat (i.e., hip-
replacement surgery). In each of the four studies, life task-related
expectations and fantasies were measured long before we assessed
effort and actual success in attaining the desired future. We op-
erationalized expectations of success by the subjective likelihood
of successfully mastering relevant life tasks, and we measured the
positivity of fantasies by self-report and semiprojective techniques.
Effort and success were assessed by self-report and objective
measures. Finally, content analyses of participants’thoughts and
images tested to what extent subjectively experienced positive
fantasies contained idealization of a perfect future at the expense
of considering future hardships.
We hypothesized that positive expectations about reaching the
desired future predict strong effort and much success, but positive
fantasies about successfully reaching the desired future predict
weak effort and little success. We also tested to what extent effort
mediated a possible relation between positive fantasies and little
success.
Study 1: Expectations Versus Fantasies and Entering
Professional Life
Students in their last year of graduate studies at a German
university were asked to rate the likelihood of obtaining an ade-
quate job. Then they produced positive and negative thoughts and
images pertaining to entering professional life, and indicated how
often they had recently entertained such positive and negative
thoughts and images. Two years later, we measured actual success
by the number of job offers participants had received, and by the
salary they were currently earning.
During the last 15 years, unemployment has been a chronic
problem in Germany. Even though academics are suffering less
than those with lower education, 16% of unemployed academics
never had a job in their field (Labor Exchange, 1997). Expecta-
tions have been found to play an important role in regaining
employment. Individuals of different ages and educational back-
grounds who had recently lost their job, but were confident about
finding an adequate job after a short period of time, found such a
position earlier than their less confident peers (Holmes & Werbel,
1992). And individuals who had been out of work for more than a
year, but who still had high expectations for finding a new posi-
tion, did get a new job faster than those with low expectations of
success (Schaufeli & Van Yperen, 1993). An intervention program
for strengthening self-efficacy expectations helped individuals
who had been out of work for more than 1.5 years to receive more
job offers than members of a no-treatment control group (Eden &
Aviram, 1993). Finally, graduating students with high expectations
of success received more job offers than students with low expec-
tations of success (Schaufeli & Van Yperen, 1993; Steffy, Shaw,
& Noe, 1989).
Though the role of positive expectations of success in finding
work is amply demonstrated, the role of fantasy has been ne-
glected. We hypothesized that positive fantasies about successfully
and effortlessly attaining the first step on one’s career ladder
should curb the striving for job offers and limit success more than
less positive thoughts picturing hindrances and hardships in find-
ing a first job.
Method
Participants and procedure. Eighty-three male students of Ludwig-
Maximilians-University, Munich, Germany participated in the initial as-
1200 OETTINGEN AND MAYER

sessment. On average, students were 26 years old (SD ⫽1.86). All students
wanted to complete their academic education within the next year, but had
no job offer yet. Participants were invited to take part in a study on the
transition into professional life. They were informed in great detail about
the study’s guaranteed anonymity, procedure, and their voluntary partici-
pation. We asked them for their parents’addresses and informed them
about a second questionnaire to be answered in the future. Students
completed the forms at home. Two years later, we could contact 40 of these
students (48% of the original sample) to ask them about their successes in
finding a job.
Measuring career-related expectations versus fantasies. In the present
study, success was defined as getting adequate job offers. Therefore, as a
measure of job-related expectations, we asked the following question:
“How probable do you think it is that you’ll find an adequate position in
your field?”This measure of general expectations encompasses self-
efficacy expectations, outcome expectations, and expectations concerning
person-external factors (e.g., the economic situation). We also assessed the
incentive value to get adequate job offers by the following item: “How
important is it for you to find an adequate position in your field?”The
10-point answering scales ranged from not at all probable to very probable
and from not at all important to very important, respectively.
To measure the experienced tone of fantasies, we asked participants to
state whether “during their everyday life they already had experienced
positive thoughts, images, or fantasies on the subject of transition into work
life, graduating from university, looking for and finding a job.”If so, they
were told: “Please now generate these positive thoughts, images, or fanta-
sies and write them down.”A half page with lines was prepared for
participants to write on. Thereafter, participants had to indicate: “How
frequently did you experience such thoughts and images?”The 10-point
scale ranged from very rarely to very often. Participants were then given
the exact same instructions with respect to negative fantasies (i.e., only the
word positive was replaced by the word negative). To arrive at an overall
scale of experienced tone of career-related future fantasies, we subtracted
reported frequencies of negatively toned fantasies from those of positively
toned fantasies, arriving at a scale ranging from ⫺9to⫹9.
Dependent variables: Number of job offers, amount of salary, and
number of applications. Two years after our first assessment, we ap-
proached participants to check on their successes in finding a job. We
mailed a short letter to their parents’addresses, asking them to forward the
mail. The letter asked participants to indicate how many job offers in their
field they had received over the last 2 years. We also inquired about their
current income (for those who did not have a job, the score was set at zero)
and about the total number of applications they had sent out.
Results
Descriptive analyses. The two predictor variables—expecta-
tions and fantasies—correlated positively (r⫽.31, p⬍.05;
expectation: M⫽5.18, SD ⫽2.44; fantasy: M⫽2.55,
SD ⫽6.82). Incentive value (M⫽7.33, SD ⫽1.95) showed a
moderately positive correlation with expectations (r⫽.24, p⬍
.15) and no correlation with fantasies (r⫽⫺.03).
The dependent variables—number of job offers received and
amount of salary earned—correlated positively with each other
(r⫽.67, p⬍.001) and positively with the number of applications
sent out (r⫽.33 and r⫽.31, respectively, both ps⬍.05). Of
the 40 participants, 13 (33%) were left without a job offer, 12
(30%) received one job offer, and 15 (37%) had gotten more than
one job offer. Participants had on average sent out eight job
applications (SD ⫽9.52).
Expectations versus fantasies and performance. Partial corre-
lations controlling for the other predictor variable showed that
students with high expectations of success received comparatively
more job offers and earned more money; students experiencing
positive fantasies, to the contrary, received comparatively fewer
job offers and earned less money (see Table 1). Partial correlations
tended to be higher than raw correlations (shown in parentheses in
Table 1), suggesting that the two predictor variables acted as
mutual suppressors.
Participants experiencing positive fantasies had not tried as hard
as students who also permitted negative thoughts about entering
professional life. They reported sending out fewer job applications
(see Table 1). One could argue that reporting fewer applications is
Table 1
Success and Effort as Predicted by Expectation and Fantasy: Study 1, Study 2, Study 3, and Study 4
Success and effort
Index Positivity Scale Negativity Scale
Expectation Fantasy Expectation Fantasy Expectation Fantasy
Study 1
No. of job offers .41*** (.30*) ⫺.39** (⫺.26*)
Amount of salary .33* (.25) ⫺.29* (⫺.19)
No. of applications ⫺.04 (⫺.17) ⫺.40** (⫺.43**)
Study 2
Intimate relationship .55*** (.53***) ⫺.23* (⫺.10)
Confession of love .11 (.07) ⫺.21* (⫺.20)
Study 3
Course grades .21** (.17) ⫺.16* (⫺.10) .22** (.17) ⫺.19* (⫺.13) .19* (.17) .11 (.05)
Study effort .20* (.12) ⫺.25** (⫺.19*) .20* (.12) ⫺.27** (⫺.22*) .17* (.12) .19* (.15)
Study 4
Hip joint motion .27* (.10) ⫺.43*** (⫺.37**) .24* (.10) ⫺.39** (⫺.33**) .27* (.10) .44*** (.37**)
Walking on stairs .37** (.25*) ⫺.36** (⫺.23*) .35** (.25*) ⫺.33** (⫺.22) .36** (.25*) .35** (.23*)
General recovery .30* (.20) ⫺.31* (⫺.21) .28* (.20) ⫺.27* (⫺.18) .31* (.20) .32** (.22*)
Note. Partial correlation coefficients controlled for the other predictor variable; raw correlation coefficients are shown in parentheses.
*p⬍.05. ** p⬍.01. *** p⬍.001.
1201
EXPECTATIONS VERSUS FANTASIES

simply a self-protective attribution (i.e., attribution to low effort
after failure). However, the reported number of applications cor-
related positively with actually achieved success (rs⬎.31).
To check for potential interaction effects, expectations and
fantasies were entered in the first step of a regression analysis, and
the interaction between the two variables in the second step.
Dependent variables were the two measures of success in finding
a job (number of job offers received, amount of salary) and
instrumental activities to find a job (number of applications sent
out). There were no significant interaction effects (Fs⬍1.46, ps⬎
.24), indicating that the negative relations between fantasy and the
three dependent variables apply to students with low and high
expectations alike.
Incentive value. Finally, we repeated our analyses, using in-
centive value as a third predictor variable in the first step of our
regression equation, and the interaction between incentive value
and fantasies as well as between incentive value and expectations,
in the second step. Incentive value did not affect the data pattern
nor did it qualify as a significant predictor (all ps⬎.13). Further,
we did not observe significant interaction effects (ps⬎.28), save
with respect to the dependent variable number of job offers re-
ceived ( p⬍.01), where the negative relation to fantasies was more
pronounced in participants for whom it was very important to find
an adequate job.
Number of applications as a mediator. Baron and Kenny’s
(1986) modified version of the Sobel (1982) test (see Kenny,
Kashy, & Bolger, 1998, p. 260) revealed that the relation between
fantasy and number of job offers, and the relation between fantasy
and amount of salary, were only partly and not significantly
mediated by number of applications (Z⫽1.21, p⬍.24; Z⫽1.27,
p⬍.21).
Discussion
Confirming the facilitating relation of positive expectations and
success in finding work (e.g., Eden & Aviram, 1993; Schaufeli &
Van Yperen, 1993), the present results show that by comparison,
participants with positive expectations over a period of 2 years had
received more job offers and enjoyed higher salaries. However, we
add a new observation to this literature. Participants who reported
frequently experiencing positive fantasies about their transition
into professional life were less successful in their job search over
a period of 2 years. They sent out fewer applications, were offered
fewer jobs, and ultimately earned less money than students who
reported frequently experiencing fantasies that picture entering
professional life in a more negative tone.
Although participants with positive expectations were more
successful in mastering the life task of entering professional life,
they had not sent out more job applications than those with
negative expectations. As high expectations of success reflect an
already successful record, participants with high expectations of
success trusting in a promising start of their professional future
may have felt little need to send out many applications: A few
select applications should suffice for reaching success. Indeed,
number of job applications rather moderately correlated with suc-
cess (number of job offers: r⫽.33; amount of salary: r⫽.31).
This relatively low correlation also explains why the negative
relation between positively experienced fantasies and actual suc-
cess was only partly mediated by fewer instrumental activities (i.e.,
number of applications).
Whereas Study 1 focused on the life task of entering profes-
sional life, Study 2 pertains to the life task of entering a romantic
relationship. Furthermore, in Study 1 we asked participants to
reproduce relevant fantasies and then report how often they had
such fantasies in the recent past. Because such frequency reports
may be distorted by problems of retrieval (Ericsson & Simon,
1993), in Study 2 we measured fantasies by a semiprojective
technique asking participants to produce relevant fantasies and rate
these fantasies for their experienced positivity versus negativity
(Oettingen & Wadden, 1991). This method combines the advan-
tages of projective (or operant; McClelland, 1980) tests with the
strengths of questionnaire (or respondent) methods.
Study 2: Expectations Versus Fantasies and Entering a
Romantic Relationship
American college students who had a crush on a person of the
opposite sex, but whom they were not dating yet, indicated their
expectations of starting a relationship with their “crushee.”Partic-
ipants then had to identify with the main characters of eight
scenarios that depicted potential opportunities to interact with their
crushee. These scenarios, provided by us, ended prematurely and
participants fantasized them to their completion, writing down
their thoughts and images. After each scenario, participants rated
the tone of their thoughts and images on a bipolar scale ranging
from very negative to very positive. Four months later, we assessed
students’success in starting an intimate relationship with their
crushee.
Romantic love (for a summary, see Bierhoff, 1991; Hatfield,
1988; Sternberg & Barnes, 1988) is “a state of intense absorption
in another”(Hatfield & Walster, 1978, p. 9), accompanied by a
permanent, sometimes even insistent indulgence in images and
fantasies (Brehm, 1988; Tennov, 1979). A person in this state of
love is dominated by indescribable happiness when imagining that
love is returned. Sharon S. Brehm (1988) summarized this central
aspect of being in love as follows: “the core of passionate love lies
in the capacity to construct in one’s imagination an elaborated
vision of a future state of perfect happiness”(p. 253). Success in
passionate love is often defined as “union with the other”(Hat-
field, 1988, p. 193) or as “oneness”(Tennov, 1979, p. 120).
Therefore, in the present study, romantic success is measured as
starting an intimate relationship with the beloved person.
College students who positively fantasize about their romantic
success should be less likely to actively approach the adored
person and have their love less successfully reciprocated, than
those who allow their fantasies to cover the problems of obtaining
the crushee’s affection. However, positive expectations of roman-
tic success, reflecting romantic achievement in the past (Carnelley
& Janoff-Bulman, 1992), should be a good predictor for future
romantic achievement.
Method
Participants and procedure. Two months after the beginning of fall
semester, we asked American college students in a large introductory class:
“Do you have a crush on someone of the opposite sex with whom you do
not have a relationship yet or do not go out with yet?”Students had to
1202 OETTINGEN AND MAYER

answer yes or no. A total of 103 students reporting such a crush participated
in the study. Upon coming to the laboratory, they were informed about
their guaranteed anonymity, the procedure, and their voluntary participa-
tion. We also told them that we would contact them again with some brief
questions at the end of spring term. Students, separated by partitions, were
tested 4 to 6 at a time. They first completed a questionnaire assessing
expectations and fantasies related to their crushee. If students had more
than one crush, they were supposed to focus on the one person that they
were in love with the most. Five months later, shortly before the summer
break, we asked 83 participants (80% of the original sample) whether they
had started an intimate relationship with their crushee. The 83 students (40
males, and 43 females) had an average age of 19.6 years (SD ⫽1.5), the
youngest participant was 17 and the oldest was 28 years. Most (70%) of the
students were freshmen or sophomores and 23% were juniors; only 7%
were seniors.
Three quarters of the students either had a crush on an acquaintance or
a friend; 17% had a crush on a close friend, and only 8% were in love with
a stranger. Three quarters of these romantic involvements had lasted less
than 3 months, a quarter more than 3 months, and 5 students had a crush
that had lasted for more than 2 years.
Measuring romantic expectations and fantasies. Success was defined
as starting an intimate relationship with the crushee. Therefore, the assess-
ment of expectations had to encompass relevant self-efficacy and outcome
expectations (Bandura, 1997) as well as expectations about external factors
(e.g., participant and crushee living in the same dorm). The question for
measuring expectations read: “How likely do you think it is that you will
get involved with HIM or HER?”The response scale ranged from 0% to
100%. We also assessed incentive value to get involved with the crushee:
“How important is it to you that you get involved with HIM or HER?”The
scale ranged from 1 (not at all important)to7(very important).
As with expectations, fantasies pertained to a future with or without the
crushee. We asked participants to imagine themselves as the main character
in eight scenarios provided by us. These scenarios were incomplete and the
students had to fantasize them to their completion, writing down their
individual endings. Each of these scenarios presented a situation in which
the student’s passionate love either could be reciprocated or not. There
were separate versions for male and female students, differing only in that
we called their crushees SHE or HE, respectively. A typical scenario read:
“You arrive early to class and sit down. SHE also arrives early and looks
for a seat. As SHE approaches your aisle, you imagine...”Another
example is: “You are at a party. While you are talking to HIM you see a
girl whom you believe HE might like, coming into the room. As she
approaches the two of you, you imagine...”
After imagining one scenario and writing down their thoughts and
images, and before addressing the next, participants responded to a bipolar
scale ranging from 1 (very negative)to7(very positive). As the internal
consistency of the eight scenarios was high (Cronbach’s
␣
⫽.77), we
created an index of experienced positivity.
Dependent variables: Intimate relationship and confession of love. In
April, shortly before the summer break and about 5 months (M⫽4.48
months, SD ⫽0.28) after assessing expectations and fantasies, we mailed
a short notice to participants asking them a number of questions with
regard to their then-crushee. First, they were to state whether they had
started an intimate relationship. The respective two items read: “Have you
been sexually involved with HIM or HER?”and “Have you gone some-
where with HIM or HER as ‘more than just friends’?”Participants re-
sponded to each question with yes or no. Second, we wanted to learn to
what extent students themselves had actively contributed to initiating a
relationship by confessing their love to the crushee. The item was twofold.
We first asked: “Did the two of you ever discuss your having a crush on
HIM or HER?”This question was to be answered with a yes orano.
Participants answering yes were asked: “Who was the initiator of this
discussion?”Here the 7-point response scale ranged from completely me to
completely HIM or HER. For participants who had answered no,we
assumed a score of least possible initiative (i.e., completely HIM or HER).
Results
Descriptive analyses. Expectations and fantasies correlated
moderately positively (r⫽.21, p⫽.06; expectations: M⫽44.52,
SD ⫽23.36; fantasies: M⫽4.26, SD ⫽1.04). Fantasies and
incentive value did not differ by gender, Fs(1, 81) ⬍1.87, ps⬎
.17, ds⬍.30, but expectations tended to be higher for men than for
women, F(1, 81) ⫽3.13, p⫽.08, d⬍.40 (as measures of effect
size for mean differences, throughout this article, we use Cohen’s,
1988, d). Incentive value (M⫽4.36, SD ⫽1.20) correlated with
expectations moderately positively (r⫽.25, p⫽.03), but there
was no substantial correlation with fantasies (r⫽⫺.03). Fantasies
and expectations did not differ by age, Fs(1, 81) ⬍1.86, ps⬎.17,
ds⬍.30, but younger participants tended to estimate their getting
involved with their crushees as comparatively more important,
F(1, 81) ⫽3.10, p⫽.08, d⬍.39. Finally, fantasies and expec-
tations did not differ between the 83 participants who could be
contacted again in the spring and the 20 participants with whom
we were not able to follow up, Fs(1, 101) ⬍.26, ps⬎.61, ds⬍
.13, but participants who could be contacted again estimated their
getting involved with their crushees as comparatively less impor-
tant, F(1, 101) ⫽4.84, p⫽.03, d⬍.59. The two groups of
participants did not differ in gender, or years in school, Fs(1,
101) ⬍1.82, ps⬎.18, ds⬍.34.
Success in establishing an intimate relationship was determined
by combining the items “Have you been sexually involved with
HIM or HER?”and “Have you gone somewhere with HIM or HER
as ‘more than just friends’?”Six participants had missing data on
one of these items and thus were omitted from the respective
analyses. For the remaining 77 participants, the two items “sexual
relationship”(M⫽0.18, SD ⫽0.39) and “going out as more than
just friends”(M⫽0.29, SD ⫽0.45) showed a positive correlation
(r⫽.60, p⬍.001). Finally, the dependent variable of establishing
an intimate relationship correlated positively (r⫽.48, p⬍.001)
with confession of love (M⫽1.60, SD ⫽1.49).
The predictive power of expectations versus fantasies. Stu-
dents expecting success were more likely to start an intimate
relationship with their crushee. Those who experienced positive
fantasies about future romantic success, to the contrary, were less
likely. They also were less likely to confess their love (see Table
1). Comparing partial correlations controlling for the other predic-
tor variable and respective raw correlations (in parentheses) shows
that expectations tended to suppress the fantasy romantic success
relation.
Interaction effects between expectations and fantasies were not
significant for starting an intimate relationship. However, for ad-
mitting love there was a nearly significant interaction effect (p⫽
.06), indicating that participants with positive fantasies were es-
pecially reluctant to admit their love when they had low expecta-
tions of success.
Gender and incentive value. Controlling for gender and incen-
tive value did not change the findings nor did the two variables
emerge as significant predictors. We observed no further signifi-
cant interaction effects (ps⬍.31) indicating that the reverse
relation of expectation and fantasy to romantic success and con-
1203
EXPECTATIONS VERSUS FANTASIES

fession of love is valid for both male and female students as well
as for crushees with low and high incentive.
Confession of love as a mediator. Baron and Kenny’s (1986)
modified version of the Sobel (1982) test showed that the relation
between fantasy and starting an intimate relationship was partly,
though not quite significantly, mediated by confession of love
(Z⫽1.65, p⬍.11).
Discussion
Participants who entertained positive future thoughts in the form
of expectations were more likely to start an intimate relationship
with the adored person; participants who entertained positive fu-
ture thoughts in the form of fantasies were less likely to be
successful. We had also asked whether students had confessed
their love to their crushee between November and April. Such an
effort to discuss their crush openly is a sign of decisiveness that
shows participants’readiness to effortfully achieve their desired
future. Students imagining a positive romantic future refrained
from such decisiveness.
Expectations were unrelated to confession of love. Thus, present
observations parallel findings in Study 1 (professional success),
where expectations only negligibly predicted active efforts to
realize the desired future (number of job applications). As success
experienced in past love relations is the basis of high expectations
of romantic success (Carnelley & Janoff-Bulman, 1992), such high
expectations might signal that the crushee will reciprocate one’s
love anyway. Therefore, participants with high expectations of
romantic success might have waited until their crushee will come
forward, rather than initiate contact themselves. Also, next to
confessing one’s love, participants might have used other viable
strategies to have their love reciprocated, and thus confession of
love only partially mediated the relation between fantasies and
actual success.
Study 2 replicated the findings of Study 1 in the interpersonal
domain, with a larger sample showing less attrition, and for a
semiprojective measure of fantasy. However, both studies mea-
sured success by self-reports. In Study 3, therefore, we assessed
success by more objective measures: We obtained actual course
grades of students who previously had noted their academic ex-
pectations and fantasies. Further, Study 3 measured the experi-
enced positivity of fantasies not using a bipolar scale, but two
scales, one positivity scale and one negativity scale. We wanted to
investigate whether the presence of positive fantasies goes along
with the absence of negative thoughts, and whether it is the
presence of positive thoughts, the absence of negative thoughts, or
both that determines the negative relation to motivation and
performance.
Study 3: Expectations Versus Fantasies
and Academic Success
Shortly before their midterm exam, American college students,
enrolled in an introductory psychology class, indicated the grade
they would like to obtain in the course, and the likelihood that they
would reach this grade. We then assessed their grade-related
fantasies. As the dependent variable, we measured study effort and
the change in course grades from midterm to the end of the course.
Like professional and romantic success, academic success has
been found to benefit from positive expectations. High expecta-
tions of success predict successful achievement in students of
different ages and different academic backgrounds, and with re-
spect to a variety of indicators (e.g., standardized tests, course
grades, solving specific tasks, application of learning strategies;
Schunk, 1982; Zimmerman & Martinez-Pons, 1992; summary by
Bandura, 1997). Positively experienced fantasies, to the contrary,
should restrain effort and the preparing for hardships and obsta-
cles. Thus, students entertaining positive fantasies might improve
little in their schoolwork from midterm to final.
Method
Participants and procedure. Two days before their midterm exam, 117
college students in an introductory psychology class filled out a question-
naire assessing their academic expectations and fantasies for how they
might be doing in the course. Students (63% male and 37% female; mean
age ⫽19.5 years, SD ⫽0.99) were informed about the procedure, guar-
anteed anonymity, and voluntary participation. We also obtained their
permission to record their course grades. Students completed the forms in
class. Their midterm and cumulative final grades were collected to deter-
mine change in academic achievement from midterm to final (a period of 6
weeks).
Measuring expectations versus fantasies. To measure academic expec-
tations, we first asked participants: “What grade would you personally
consider a success in this course? Please circle the grade which first comes
to your mind.”The answering scale ranged from Fto A⫹. Participants
were then asked: “How probable do you think it is that you will obtain the
grade that you circled in question number 1? Please circle the percentage
which first comes to your mind.”The percentage scale ranged from 10%
to 100%. We also assessed incentive value: “How important is it for you
to obtain the grade circled in question number 1?”The scale ranged from 1
(not at all important)to7(very important).
To measure academic fantasies, we used a similar procedure as in
Study 2. Participants had to complete the following scenario: “You have
already taken your exams and today is the day the course grades are posted.
As you are walking toward the building that the board is in...”After
completing the story, participants indicated how positive these thoughts
were on a 7-point scale ranging from not at all positive to very positive, and
how negative these thoughts were on a scale ranging from not at all
negative to very negative. We observed a strong negative correlation
between the two items (r⫽⫺.82, p⬍.001), and thus combined them into
a fantasy index by subtracting the negativity scale from the positivity scale
(arriving at an index ranging from ⫺6to⫹6). However, we also analyzed
the findings for the positivity and the negativity scales separately.
Dependent variables: Study effort and course grades. Study effort was
assessed at the last class meeting before the final exam. We first asked
“How many hours per week did you spend preparing for your class
since the past midterm exam? Please indicate the exact number of
hours: ”. Students then indicated: “How hard did you try to prepare
for your class since the past midterm exam?”The 7-point scale ranged
from I did not try hard at all to I tried very hard. Finally, students reported
whether they had done optional work: “Did you write one or more extra-
credit papers or essays for your class since the past midterm exam?,”
answering with a yes orano. These three variables correlated moderately
high (Cronbach’s
␣
⫽.57), and were combined to an index labeled study
effort. Using each item separately yielded the same pattern of results as the
items combined for all analyses reported below.
Participants’midterm and final grades (in letter grades ranging from 1
[A⫹]to13[F]) were recorded. As a dependent variable, we used the
improvement in course grades from midterm to the final course grades.
Over this 6-week period, 54 (46.2%) of the students improved in their
1204 OETTINGEN AND MAYER

grades, 24 students (20.4%) kept their grade, and 39 (33.3%) decreased in
their grades.
Results
Descriptive analyses. The two predictor variables—expecta-
tions and fantasies—correlated positively (index: r⫽.33, p⬍
.001; positivity: r⫽.32, p⬍.001; negativity: r⫽⫺.30, p⬍
.001). Means of expectations and fantasies were above the mid-
points of the scales (expectations: M⫽6.68, SD ⫽1.69; fantasies:
index: M⫽0.21, SD ⫽3.08; positivity: M⫽3.94, SD ⫽1.65;
negativity: M⫽3.74, SD ⫽1.59). Incentive value (M⫽5.50,
SD ⫽1.06) showed a moderately positive correlation with expec-
tations (r⫽.28, p⬍.01), and no substantial correlation with
fantasies (index, r⫽⫺.01; positivity, r⫽⫺.02; negativity, r⫽
⫺.01). The mean desired grade (M⫽5.72, SD ⫽1.33) correlated
with incentive value positively (r⫽.24, p⬍.01), but there was no
significant correlation with expectations (r⫽⫺.06) or fantasies
(index: r⫽.03; positivity: r⫽⫺.01; negativity: r⫽⫺.06). Male
and female students did not differ in their fantasies, incentive
value, or desired grades, all Fs(1, 115) ⬍1.75, ps⬎.18, ds⬍.24.
Expectations of success were higher for women than for men
(women: M⫽7.09, SD ⫽1.66; men: M⫽6.45, SD ⫽1.67), F(1,
115) ⫽4.09, p⬍.05, d⬍.39. The dependent variables—course
grades (M⫽0.23, SD ⫽2.35) and study effort—correlated
positively (r⫽.35, p⬍.001; different indicators of study effort
are hours studied per week: M⫽4.16, SD ⫽2.60; degree of
preparation: M⫽4.14, SD ⫽1.58; extra credit: M⫽0.32,
SD ⫽0.47).
The predictive power of expectations versus fantasies. Expec-
tations of success predicted comparatively high course grades and
strong effort, whereas positive fantasies tended to predict compar-
atively low course grades and weak effort (see Table 1). The index,
the positivity scale, and the negativity scale all yielded the same
pattern of results, indicating that the presence of positive fantasies
predicts low course grades and weak effort as does the absence of
negative fantasies, though the latter does not quite reach signifi-
cance for course grades. Subsequent regression analyses yielded
no significant interaction between expectation and fantasy (ps⬎
.14), indicating that positive fantasy negatively predicts academic
performance and study effort in students with low and high ex-
pectations alike. As in the previous studies, mutual suppressor
effects of expectation and fantasy emerged, indicated by slightly
higher coefficients for partial than for raw correlations (see
Table 1, raw correlations in parentheses).
Gender and incentive value. For both course grades and effort,
the inclusion of gender and incentive value left the pattern of
results unchanged. Incentive value, though, positively predicted
effort (
␤
ⱖ.30, p⫽.001), indicating that students who deemed
academic success as important reported investing more effort in
their studies. Interaction effects between expectations and gender,
and expectations and incentive, as well as between fantasy and
gender, and fantasy and incentive predicting effort and course
grades were all nonsignificant (ps⬎.26 and ⬎.14, respectively).
Effort as mediator. Baron and Kenny’s (1986) modified ver-
sion of the Sobel (1982) test revealed that the relation between
fantasy and subsequent performance was significantly mediated by
effort (index: Z⫽2.16, p⬍.04; positivity: Z⫽2.30, p⬍.02;
negativity: Z⫽1.85, p⬍.06).
Discussion
The reverse relation of expectations and fantasies to success was
observed for objective performance criteria: course grades. Fur-
ther, analyses using all three fantasy scales, the positivity scale, the
negativity scale, and the difference between the positivity scale
and the negativity scale (index) showed the predicted pattern of
results. These results rest on a strong negative correlation between
the positivity and the negativity scales, indicating that participants
experiencing their fantasies as positive did not experience them as
negative, and vice-versa.
The predictive relation between positive fantasy and low per-
formance was mediated by intense studying. Apparently, intense
studying was a valid indicator of effort and effort was an effective
means to excelling one’s grades. Assuming that in Studies 1 and 2,
we were less successful in finding a valid indicator of effort
(number of applications and confession of love), the findings of the
three studies point at effort as a mediator of the relation between
positive fantasy and low performance.
So far we have analyzed the role of thinking about the future in
terms of expectation versus fantasy for mastering life tasks in the
professional, interpersonal, and academic domain. Participants of
the three studies were young adults. Participants of the next study,
then, are older adults, and the life task pertains to the health
domain. More specifically, inpatients just admitted for hip-
replacement surgery were assessed for their recovery-related ex-
pectations and fantasies. Two weeks after surgery, shortly before
patients left the hospital, we asked physical therapists to report on
the degree of patients’recovery.
Moreover, in Study 4, we content analyzed whether subjective
experiences of positive fantasies are reflected in participants’writ-
ings containing idealization of a perfect future at the expense of
considering future hardships. Two independent raters to whom the
hypotheses were unknown scored to what extent participants in
their written responses already enjoyed their perfect recovery or
questioned such a relieving outcome, and to what extent partici-
pants enjoyed a smooth road to recovery versus questioned such an
unencumbered road to success. We hypothesized that positively
experienced fantasies will focus on a perfect outcome and a
smooth process rather than on an imperfect outcome and a cum-
bersome process.
Study 4: Expectations Versus Fantasies and Recovery
From Surgery
Total hip-replacement surgery is a commonly performed and
accepted procedure in patients with osteoarthritis of the hip. Os-
teoarthritis is the most frequent joint disorder, and osteoarthritis of
the hip joint is a particular problem in the elderly (Gogia, Chris-
tensen, & Schmidt, 1994). In surgery, affected bone and cartilage
is removed and replaced with an artificial joint made of metal and
plastic. Functional disability and pain in the absence of primary
and secondary preventive measures are the two major indications
for total hip-replacement surgery (Verbrugge, 1990). As surgery is
elective and is supposed to reduce disability and dependency,
patients should be prone to generate fantasies about a future with
improved functional activity and reduced pain. At the same time,
patients should vary in their expectations of effective recovery, as
their expectations are based on individual illness histories.
1205
EXPECTATIONS VERSUS FANTASIES

Supporting our contention that high expectations of success
promote recovery from hip-replacement surgery, patients’expec-
tations to gain high functional ability have been found to precede
satisfaction with the outcome of hip-replacement surgery (Man-
cuso, Salvati, Johanson, Peterson, & Charlson, 1997), and high
efficacy beliefs predicted reduced pain and joint inflammation in
patients with rheumatoid arthritis (O’Leary, Shoor, Lorig, & Hol-
man, 1988). High expectations of success were followed by com-
pliance with the medical regimen and improved health after car-
diac transplant surgery even after controlling for preoperative
health status (Leedham, Meyerowitz, Muirhead, & Frist, 1995),
and expectations about a positive future in general predicted re-
covery from coronary artery bypass surgery (Scheier & Carver,
1992; Scheier et al., 1989). Finally, high recovery-related efficacy
expectations fostered mastery of various acute and chronic ill-
nesses (Bandura, 1997), and optimistic attributional style preceded
good health in the short run (C. Peterson, 1988) and over the life
span (C. Peterson, Seligman, & Vaillant, 1988).
Two correlational studies support the notion of positive fanta-
sies negatively predicting health outcomes. First, in a prospective
study with obese women who had just enrolled in a weight-loss
program, those patients who at pretreatment had generated highly
positive fantasies about successfully losing weight, lost less weight
during the next year than those who also had permitted negative
weight-related thoughts and images (Oettingen & Wadden, 1991).
To the contrary, patients with positive expectations of successfully
losing weight did much better than those with negative expecta-
tions. In another prospective study, testing 50 patients diagnosed
with breast cancer, positive fantasies (assessed by the Short Form
of the Imaginal Processes Inventory; Huba, Singer, Aneshensel, &
Antrobus, 1982) predicted high disease activity over the period
of 1 year (Jensen, 1987). In line with these findings, we hypoth-
esized that positive fantasies about successful recovery from hip-
replacement surgery predict low success, because they conceal the
necessity of still having to reach convalescence and of engag-
ing in the painful and cumbersome procedures demanded by
rehabilitation.
Method
Participants and procedure. Sixty-seven patients consecutively admit-
ted to a hospital in Germany for their first hip-replacement surgery were
invited to the study. Fifty-eight patients agreed to participate. Seven
patients were excluded because they reported not to have left bed several
weeks before admission, and staying in bed leads to atrophy in muscles,
which severely impedes postsurgery recovery. The remaining 51 patients
(41% male and 59% female) were on average 65.7 (SD ⫽8.66) years old,
ranging from 49 to 84 years. Body Mass Index (BMI ⫽Weight in
kg/[height in m]
2
) was on average 27.5 (SD ⫽4.10). Thirty-six patients
were overweight (BMI ⱖ25; 70.6% of the total sample), and of these 36
patients, 14 patients were obese (BMI ⱖ30; 27.5% of the total sample).
Doctors examined each patient’s hip in comparison with the average of
patients being admitted for hip-replacement surgery using a scale ranging
from 1 (very bad)to5(very good; M ⫽2.78, SD ⫽1.01).
Measuring health-related expectations and fantasies. Recovery-
related expectations and fantasies were assessed at the day of admission to
the hospital (i.e., 1 day before surgery). Expectations were measured by the
following four questions: (a) “How likely do you think it is, that 2 weeks
after surgery you will be able to go for a brief walk using an assistive
cane?”; (b) “How likely do you think it is, that 2 weeks after surgery you
will be walking on stairs up and down with the help of an assistive cane?”;
(c) “How functionally able do you think you will be 3 months after
surgery?”; and (d) “To what extent do you think you will be without pain 3
months after surgery?”Scales ranged from 1 (not at all likely)to5(very
likely), and from 1 (not at all)to5(fully), respectively. Cronbach’s alpha
for the four items was .75; they were combined into a single index of
expectations of successful recovery.
We assessed fantasies by having participants imagine themselves in five
scenarios. As in Studies 2 and 3, the scenarios were interrupted in the
middle, and patients had to imagine them to their completion. The first
scenario read as follows: “You wake up after surgery in the wake-up room.
You feel your body increasingly clearly...”The other four scenarios
pertained to going to the hospital’s shopping area to buy a newspaper, to
take a walk with friends, to tidying up the living room and kitchen, and to
preparing a dinner invitation. The last three of the scenarios were to be
imagined as happening 3 months after surgery.
After imagining each story to their completion and writing down their
thoughts and images, patients indicated both how positive and how nega-
tive their thoughts and images were. The scales ranged from 1 (not at all
positive)to5(very positive) and from 1 (not at all negative)to5(very
negative), respectively. Cronbach’s alpha for the positivity scale was .77,
for the negativity scale .68. The two scales correlated highly negatively
(r⫽⫺.83), and thus were combined into a fantasy index (by subtracting
the negativity scale from the positivity scale) ranging from ⫺4to⫹4.
Cronbach’s alpha for the five scenarios was .75.
Dependent variables: Range of hip joint motion, walking on stairs, and
general recovery. Two weeks after surgery, when all patients were still in
the hospital, physical therapists reported about (a) the functional status of
the hip in terms of degree of hip joint motion, (b) the functional status of
the hip in terms of patients’ability to walk on stairs, and (c) patients’
general recovery. There were 14 physical therapists that worked with the
patients. We asked each patient’s primary physical therapist, because
primary physical therapists could most closely follow their patients’course
of recovery.
First, physical therapists rated the range of hip joint motion, a classic
indicator of functional recovery after hip-replacement surgery (Gogia et al.,
1994). They were asked to report the range of the hip joint’s motion: “How
well can the hip joint of the patient be moved? When it comes to: (a)
‘abduction,’(b) ‘extension,’and (c) ‘flexion’.”Answer scales ranged
from 1 (badly)to5(well). The three items showed high internal consis-
tency (Cronbach’s
␣
⫽.77), and were therefore combined to an overall
scale of functional status of the hip (i.e., range of motion). Another classic
indicator of functional ability of the hip joint after hip-replacement surgery
is the extent to which patients can walk on stairs (Dekker, Boot, van der
Woude, & Bijlsma, 1992). Physical therapists answered the question “How
many stairs can the patient walk upstairs and downstairs?”on a 5-point
scale ranging from none at all to three stairs. Finally, as a measure of
general recovery, physical therapists indicated: “How well does the patient
do in comparison to other patients in his or her recovery from hip-
replacement surgery?”with respect to (a) pain, (b) muscular strength, (c)
functional status of the hip, and (d) general well-being. Scales ranged
from 1 (very badly)to5(very well). Cronbach’s alpha was .89, and
therefore we combined the items in an overall index of general recovery.
Content analyses: Idealizing versus questioning outcome and process.
Two independent raters evaluated the written responses to each of the five
scenarios with respect to the extent participants in their written thoughts
and images (a) idealized the positive outcome or questioned such a positive
outcome, and (b) idealized an effortless process to reach the desired
outcome or questioned an unencumbered way to recovery. Rating scales
ranged from ⫺5(total questioning)to0(neither questioning nor idealiz-
ing)to⫹5(total idealizing).
With respect to outcome, the highest score for idealizing (score of ⫹5)
was coded when the response solely focused on positive aspects of the
outcome without entailing any lessening or impairment (e.g., in response to
the scenario of the patient going for a walk with friends, 1 participant
1206 OETTINGEN AND MAYER

wrote: “I am so happy that I am well, it is easy for me to keep up with my
friends on the walk, they are completely amazed how I am doing”). The
highest score for questioning a positive outcome (score of ⫺5) was given
when a response entailed curtailment regarding the future outcome (e.g., in
response to the above mentioned scenario, 1 participant wrote: “Will there
be serious consequences of the surgery? What if I have to slow down in the
middle of the walk? Will I have to wait in the restaurant?”).
With respect to process, the highest score for idealizing (score of ⫹5)
was coded when the way to the desired outcome was described as effortless
and unencumbered and lacking any hindrances, pains, temptations, or
setbacks (e.g., in response to the scenario of the patient buying a newspaper
in the hospital shop, 1 participant wrote: “I am walking on the stairways
downwards without help, and I walk easily and quickly to the newspaper
stand”). The highest score for questioning the process (score of ⫺5) was
coded when the way to recovery entailed troublesome aspects such as
obstacles, temptations, or set-backs, or when pain and effort were men-
tioned (e.g., in response to the scenario above, 1 participant wrote: “Iam
trying to walk to the door first, using my cane. But how shall I open the
door? Uh, and then walking until the elevator? How would I ever do it?”).
Interrater reliability was high (r⫽.90), and ratings regarding outcome
and process were highly correlated (r⫽.73; Cronbach’s
␣
of the 10 scales
was .75). We created an overall index of idealizing versus questioning
containing the ratings of outcome as well as process. High scores on this
overall index reflect descriptions of perfect recovery that is effortless and
painlessly achieved, low scores reflect descriptions of a fair recovery that
is attained with effort and pain.
Results
Descriptive analyses. Expectations (M⫽4.00, SD ⫽0.78)
and fantasies (index: M⫽3.05, SD ⫽0.89, positivity: M⫽4.44,
SD ⫽0.51, negativity: M⫽1.39, SD ⫽0.42) correlated positively
(index: r⫽.37, p⬍.01, positivity: r⫽.34, p⬍.01, negativity:
r⫽⫺.37, p⬍.01). BMI (M⫽27.49, SD ⫽4.11) correlated
negatively with expectations (r⫽⫺.31, p⬍.01), as well as with
fantasies (index: r⫽⫺.27, p⬍.03; positivity: r⫽⫺.25, p⬍.04;
negativity: r⫽.26, p⬍.03). Functional status of the hip before
surgery as estimated by the doctors (M⫽2.78, SD ⫽1.01) did not
correlate significantly with expectations (r⫽.01, p⫽.44), fan-
tasies (index: r⫽.04, p⫽.38; positivity: r⫽.04, p⫽.37;
negativity: r⫽⫺.03, p⫽.40), or BMI (r⫽.07, p⫽.29).
Expectations, fantasies, and functional status of the hip before
surgery did not differ between men and women, all Fs(1,
49) ⬍1.64, ps⬎.20, ds⬍.37, and between participants of
different age, all Fs(1, 49) ⬍1.16, ps⬎.28, ds⬍.31. However,
men tended to have higher BMI than women, F(1,49) ⫽2.60, p⬎
.11, d⬍.47.
The dependent variables of motion of the hip joint (M⫽3.67,
SD ⫽0.74), walking on stairs (M⫽4.24, SD ⫽0.91), and general
recovery (M⫽3.79, SD ⫽0.81) correlated positively (hip joint
motion with walking on stairs: r⫽.43, p⬍.01; general recovery:
r⫽.70, p⬍.01; walking on stairs with general recovery: r⫽.48,
p⬍.01).
The predictive power of expectations versus fantasies. Positive
expectation predicted successful recovery from hip-replacement sur-
gery, and the reverse was true for positive fantasy. This finding
held whether recovery was measured by highly specific measures,
that is, by patients’hip joint motion and their capability of walking
on stairs, or by a general measure including pain, muscular
strength, functional status of the hip, and patients’well-being. It
held for all three fantasy scales, the index, as well as for the
positivity and negativity scales separately (see Table 1). Like in
Study 3, the experience of more positive than negative fantasies,
the presence of positive fantasies as well as the absence of negative
fantasies were all linked to comparatively less success. Interaction
effects between expectation and fantasy were not significant ( ps⬎
.26). Mutual suppressor effects with respect to all dependent
variables and for the index, the positivity, and the negativity scales
are indicated by partial correlations tending to be higher than raw
correlations (shown in parentheses in Table 1).
Gender, weight, and presurgery hip condition. Regression
analyses revealed that gender, weight (BMI), and presurgery hip
condition were not significant predictors of hip joint motion (see
Table 2, left column), nor did they together explain a significant
portion of the variance (the three variables together explained 8%
of the variance, p⫽.26). However, gender and presurgery hip
condition significantly predicted walking on stairs (see Table 2,
middle column) with men and better-off patients being more
successful (the three variables together explained 30% of the
variance, p⫽.001). Finally, gender, weight, and presurgery hip
condition all significantly predicted general recovery (see Table 2,
right column) with men, patients with lower weight, and initially
better-off patients recovering more successfully (the three vari-
ables together explaining 22% of the variance, p⫽.007). Of most
importance, positive expectations remained a near-significant pos-
itive predictor, and positive fantasies remained a significant neg-
Table 2
Recovery From Hip-Replacement Surgery as Predicted by Gender, Weight, and Presurgery Hip
Condition As Well As Expectation and Fantasy: Study 4
Predictor variable
Hip joint motion Walking on stairs General recovery
␤
Fp
␤
Fp
␤
Fp
Step 1
Gender .23 2.51 .12 .40 9.68 .003 .31 5.40 .03
Weight (BMI) ⫺.16 1.13 .29 ⫺.05 0.18 .69 ⫺.33 5.97 .02
Presurgery hip condition .16 1.29 .26 .40 10.23 .003 .30 5.23 .03
Step 2
Expectation .23 2.59 .11 .35 7.73 .008 .23 2.81 .10
Fantasy ⫺.47 11.00 .002 ⫺.34 7.35 .01 ⫺.34 6.32 .02
Note. BMI ⫽Body Mass Index.
1207
EXPECTATIONS VERSUS FANTASIES

ative predictor even after gender, weight, and presurgery hip
condition had been controlled (see Table 2).
No interaction effects were significant (hip joint motion: ps⬎
.15; walking on stairs: ps⬎.34; general recovery: ps⬎.16),
indicating that the positive relation between expectations and the
three indicators of success, as well as the negative relation between
fantasies and success, holds for men and women, patients of
different weight, and patients of different presurgery hip condi-
tions alike.
Content analysis: Idealization versus questioning. Mean ide-
alization ratings were above the middle of the scale (outcome plus
process: M⫽1.23, SD ⫽1.56; outcome: M⫽1.69, SD ⫽1.80;
process: M⫽.78, SD ⫽1.54), and participants idealized outcome
more than process, t(50) ⫽5.15,p⫽.001. Of most importance,
the correlations between the experienced positivity of fantasies and
the ratings of idealization were significantly positive, and this was
true for idealization of outcome plus process as well as for out-
come and process analyzed separately (see Table 3). Apparently,
positively experienced fantasies entail descriptions of perfect fu-
tures that are effortlessly achieved, whereas comparatively more
negatively experienced fantasies question a perfect future and its
unencumbered realization. Partial correlations tended to be lower
than raw correlations (in parentheses, Table 3), indicating that part
of the positive relation between experienced fantasies and ideali-
zation was due to expectations. But even after controlling for
expectations, the relation between experienced fantasy and ratings
of idealization stayed significant (ps⬍.01).
Expectations also correlated positively with idealization. How-
ever, after controlling for experienced positivity of fantasies, the
correlations significantly decreased (ts⬎2.40, ps⬍.01; see Table
3), indicating that the positive correlation between ratings of
idealization and subjective expectations was mostly due to the
experienced positivity of fantasies. In other words, ratings of
idealization are linked to the positivity of subjective fantasies
rather than to the positivity of expectations.
Finally, when determining the predictive value of expectation,
fantasy, and rated idealization for recovery, the patterns of results
for expectations and fantasy stayed untouched, whereas the rela-
tion between idealization and recovery was nonsignificant (see
Table 4). Apparently, the experienced positivity of fantasies rather
than the observable contents of these fantasies are responsible for
the negative relation between fantasy and successfully realizing
one’s desired future. Finally, all patterns of results reported above
apply regardless of whether fantasies are measured by the index,
the positivity scale, or the negativity scale.
Discussion
The reverse relation of expectation and fantasy to actual success
was replicated in the health domain, with older participants, with
independent raters (physical therapists) recording specific and
general criteria of success, and with gender and other possibly
confounding variables (i.e., weight, presurgery functional hip con-
dition) statistically controlled. As in Study 3, the presence of
positive thoughts and the absence of negative thoughts were pre-
cursors of less success, as was the index, demonstrating that the
more lopsided (i.e., the more positive than negative) participants
experienced their fantasies, the worse the recovery.
Content analyses revealed that although people idealized out-
come more than process, idealization of process was common.
Further, both idealized outcome and idealized process positively
related to experienced positivity of fantasies. Thus positive fanta-
sies contain both outcome and process in its idealized form, that is,
successful achievement as well as effortless and unencumbered
progress to attaining the desired future.
But even though the correlations between the tone of subjec-
tively experienced fantasies and ratings of idealization were sig-
nificant (rs ranged from .31 to .36, ps⬍.01), they still leave
variance unexplained. Patients, although entertaining negative fan-
tasies, might have felt hesitant to explicate these fully in writing,
because such negative fantasies might cover content that is socially
undesirable (e.g., expressing anger, suffering pain, or feeling
shame), scary (e.g., having nightmares, feeling lonely, being de-
pendent), or can hurt the feelings of members of the hospital staff
(e.g., the doctor making a mistake; the health care provider being
unfriendly). Conversely, patients might have entertained positive
fantasies, but did not want to express them fully in their writings,
fearing that they then might not come true or lead to ridicule from
the interviewer. Else, participants varying in acceptance of their
daydreams might have differed in how they conveyed the affective
tone in their writings, and individuals with poor attention control
( Huba et al., 1982; Singer & Antrobus, 1963) might have written
more ambiguous texts producing less reliable ratings. Though such
factors may have blurred the correlation between rated idealization
and experienced tone of fantasies, the correlation suggests that
positively experienced fantasies are based on imagining successful
outcomes and unencumbered process to reach these outcomes.
And of importance, it was the subjectively experienced positive
fantasies rather than the expressed idealization as picked up by the
raters that predicted low recovery.
General Discussion
On the basis of William James’s (1890/1950) distinction be-
tween beliefs and images, we explored the predictive power of
thinking about the future in terms of expectations versus fantasies.
Across four studies, we observed that positive expectations pre-
dicted higher motivation and more success than negative expecta-
tions, but positive fantasies predicted lower motivation and less
success than negative fantasies.
Experienced positivity of fantasies predicted weak effort and
low success over diverse periods of time (from 2 weeks to 2 years),
for different measures of fantasies (frequency ratings, semiprojec-
tive techniques), for positivity as well as negativity ratings, for
different life domains (professional, interpersonal, academic,
Table 3
Relations Between Expectation, Fantasy, and Ratings of
Idealized Future: Study 4
Thinking about
the future Outcome and
process Outcome Process
Expectation .10 (.24*) .07 (.21) .13 (.25*)
Fantasy .36** (.41***) .34** (.39**) .31** (.38**)
Note. Partial correlation coefficients controlled for the other predictor
variable; raw correlation coefficients are shown in parentheses.
*p⬍.05. ** p⬍.01. *** p⬍.001.
1208 OETTINGEN AND MAYER

health), in samples of different ages (young adults, elderly), in
samples tested in the laboratory and in institutions (classroom,
hospital), and in samples embedded in different countries (Ger-
many, United States). The findings also support the notion that
weak effort mediates the relation between positive fantasy and
little success.
All four studies pertained to the mastery of life tasks (Cantor &
Kihlstrom, 1987). As failures to solve impending life tasks affect
the mastery of subsequent life tasks (Havighurst, 1948/1972), our
findings imply that positive fantasies can have long-term costs
with respect to a person’s personality development (Cantor &
Harlow, 1994). For example, the first job is the basis for successive
job positions a person holds during his or her life (Super & Hall,
1978), with both status and wages increasing over time. Thus our
findings of positive fantasies predicting low success in entering
professional life imply that positive fantasies may be problematic
not only for the life task of beginning a career. Moreover, an
enduring transitional period between academic and professional
life should adversely affect relations with parents and friends,
financial and living conditions, and the advent of starting a family.
Similarly, it should strain well-being, high self-esteem, and pro-
fessional self-efficacy.
Fantasy-Performance Link: Person Variables
The present research does not speak to person variables affect-
ing the relation between fantasy and performance. For example,
acceptance of one’s own daydreams correlates with reports of
frequent positive daydreams (both measured by the Imaginal Pro-
cesses Inventory; Huba et al., 1981; Singer & Antrobus, 1963).
Thus, by facilitating the emergence of positive fantasies, accep-
tance of daydreams might hinder successful performance. Further,
person variables might mediate the fantasy-performance link. For
example, hardiness (readiness to confront problems and to change
them through effortful action; Kobasa, Maddi, & Kahn, 1982)
might be hampered by positive fantasy thus reducing successful per-
formance. Finally, person variables might moderate the fantasy-
performance link. For example, repressors (i.e., individuals who
notice negative stimuli but do not experience them as negative;
Bonanno & Singer, 1993; Weinberger, Schwartz, & Davidson,
1979) might not show the link between fantasy and performance,
because they do not experience imagined hardships in a negative
way. Indeed, Jensen (1987) found in his study with cancer patients
that positive fantasies relate to low recovery more in nonrepressors
than in repressors.
Implications for Research on Thinking About the Future
Expectation-fantasy link. In all four studies, high expectations
of success related positively to the positivity of experienced fan-
tasies. High expectations of success might have facilitated positive
fantasies about the future (Klinger, 1977), and positive fantasies
(thoughts and images) might have raised respective expectations of
success (Anderson & Godfrey, 1987; summary by Tversky &
Koehler, 1994). Regardless of how the positive relation between
expectations and fantasies emerged, the observed mutual suppres-
sor effects suggest that future research should benefit from mea-
suring both expectations and fantasies, because they will predict
behavior most clearly when the other type of thinking about the
future is assessed and (in case of a positive correlation) statistically
controlled.
The mutual suppressor effects further suggest that a combina-
tion of positive expectations and negative fantasies might be
particularly motivating. Future studies need to investigate when
and how people entertaining positive expectations spontaneously
generate negative fantasies geared at preparing themselves for
upcoming difficulties and hindrances. Aspinwall (1998) reports
that people with illusory optimism readily turn their focus to
negative information when the negative information is serving
their goal pursuit, suggesting that people who feel confident about
attaining their goals might be in a good position to imagine
relevant hindrances and hardships.
Potential benefits of positive fantasy. Positive fantasies do not
always need to limit a person’s effort and successful performance.
When it comes to mentally exploring one’s possibilities to grow
and one’s opportunities to act—that is, when one wants to discover
one’s possible selves (Markus & Nurius, 1986)—positive fantasies
might become useful because they are not closely tied to the
hardships of reality. Fredrickson (1998, 2001), in her broaden-and-
build theory of positive emotions, finds that emotions such as joy,
interest, contentment, pride, and love broaden people’s thought-
action repertoires, widen their array of thoughts and possible
actions, and thus build their personal resources. Fredrickson ex-
plicitly contrasts this broaden-and-build function of positive emo-
tions with models on specific action tendencies that see negative
emotions promoting quick and decisive actions. Positive fantasies
should help people mentally experience the various possibilities
and opportunities the future might bring, thereby broadening and
building a person’s thought-action repertoire; negative fantasies
orienting the person to the necessities of the here and now, then,
help translate such a repertoire into effortful action and achieving
success.
Table 4
Recovery From Hip-Replacement Surgery as Predicted by Expectation, Fantasy, and Ratings of
Idealized Future: Study 4
Predictor variable
Hip joint motion Walking on stairs General recovery
␤
Fp
␤
Fp
␤
Fp
Expectation 0.26 3.51 .07 0.37 6.79 .01 0.29 4.19 .05
Fantasy ⫺0.50 11.14 .002 ⫺0.44 8.46 .006 ⫺0.42 7.61 .008
Rated idealization 0.09 0.42 .52 0.17 1.36 .25 0.25 2.99 .09
1209
EXPECTATIONS VERSUS FANTASIES

Potential harm of negative fantasy. Just as positive fantasies
might at times be beneficial, so can negative fantasies be detri-
mental. Ruminative coping with depressed mood, for example, is
a risk factor for severe and prolonged periods of depressive mood
(Nolen-Hoeksema, 1991; Nolen-Hoeksema, Parker, & Larson,
1994). Ruminative thoughts to depression have been conceptual-
ized as thoughts that focus one’s attention on one’s depressive
symptoms and their personal consequences, thus prolonging and
extending the miserable current status quo into the far future. Thus
ruminative thoughts make people behave passively and keep them
in extensive worries.
In contrast, negative fantasies about a desired future depict
potential problems and setbacks related to an improved future.
Though negative in tone, they should fail to be linked to depression
and passivity, because they pertain to a constructive road to a
better future. These negative fantasies about a desired future, then,
seem to have the reverse effects than ruminative thoughts about
one’s present depressive mood. Future studies have to disentangle
the links between negative fantasies about a desired future, rumi-
native thoughts, and various indicators of depression.
Culture and future thought. Finally, the motivational conse-
quences of positive fantasies might depend on the cultural context.
People in cultures in which their cultural products (e.g., myths,
legends, language) foster a long-term time perspective as well as
tight cultural values (i.e., favor many norms that apply across
situations and over time; Triandis, 1996) might benefit from pos-
itive fantasies about a distant future (Oettingen, 1997), because
they should help to mentally escape the dire and restricted reality.
Conversely, in such cultures, people might benefit less from pos-
itive expectations as the basis of behavior, because cultural values
and norms largely determine action (who interacts with whom,
when, in what way, and with what kind of outcome; Oettingen,
1997; Triandis, 1996). Thus research that assumes positive expec-
tations as the basis for motivation and action (summary by Oet-
tingen & Gollwitzer, 2001) might not apply to highly norm-
oriented cultural contexts. Here, positive fantasies might play a
comparatively more important role for motivation and action.
Conclusion
Four studies investigated the predictive power of thinking about
the future in terms of positive expectations versus positive fanta-
sies. As positive expectations reflect past successes, they signal
that investment in the future will pay off. Positive fantasies, to the
contrary, lead people to mentally enjoy the desired future in the
here and now, and thus curb investment and future success. In
studies on four different life tasks, positive expectations predicted
high effort and performance, whereas positive fantasies predicted
low effort and performance. Content analyses showed that positive
fantasies are linked to idealizing a desired future outcome as well
as the process to get there, and that it is the experience of high
positivity generated by thoughts and images that predicts low
effort and little success.
References
Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned
helplessness in humans: Critique and reformulation. Journal of Abnor-
mal Psychology, 87, 49–74.
Anderson, C. A., & Godfrey, S. A. (1987). Thoughts about actions: The
effects of specificity and availability of imagined behavioral scripts on
expectations about oneself and others. Social Cognition, 5, 238–258.
Aspinwall, L. G. (1998). Rethinking the role of positive affect in self-
regulation. Motivation and Emotion, 22, 1–32.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral
change. Psychological Review, 84, 191–215.
Bandura, A. (1978). The self system in reciprocal determinism. American
Psychologist, 33, 344–358.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York:
Freeman.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable
distinction in social psychological research: Conceptual, strategic, and
statistical considerations. Journal of Personality and Social Psychol-
ogy, 51, 1173–1182.
Bierhoff, H. W. (1991). Twenty years of research on love: Theory, results,
and prospects for the future. The German Journal of Psychology, 15,
95–117.
Billings, D. W., Folkman, S., Acree, M., & Moskowitz, J. T. (2000).
Coping and physical health during caregiving: The roles of positive and
negative affect. Journal of Personality and Social Psychology, 79,
131–142.
Bolger, N. (1990). Coping as personality process: A prospective study.
Journal of Personality and Social Psychology, 59, 525–537.
Bonanno, G. A., & Singer, J. L. (1993). Controlling one’s stream of
thought through perceptual and reflective processing. In D. M. Wegner
& J. W. Pennebaker (Eds.), Handbook of mental control (pp. 149–170).
Englewood Cliffs, NJ: Prentice Hall.
Brehm, S. S. (1988). Passionate love. In R. J. Sternberg & M. L. Barnes
(Eds.), The psychology of love (pp. 232–263). New Haven, CT and
London: Yale University Press.
Cantor, N., & Harlow, R. E. (1994). Social intelligence and personality:
Flexible life-task pursuit. In R. Sternberg & P. Ruzgis (Eds.), Person-
ality and intelligence (pp. 137–168). Cambridge, England: Cambridge
University Press.
Cantor, N., & Kihlstrom, J. F. (1987). Personality and social intelligence.
Englewood Cliffs, NJ: Prentice Hall.
Carnelley, K. B., & Janoff-Bulman, R. (1992). Optimism about love
relationships: General vs. specific lessons from one’s personal experi-
ences. Journal of Social and Personal Relationships, 9, 5–20.
Carver, C. S., & Scheier, M. F. (1994). Situational coping and coping
dispositions in a stressful transaction. Journal of Personality and Social
Psychology, 66, 184–195.
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping
strategies: A theoretically based approach. Journal of Personality and
Social Psychology, 56, 267–283.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Erlbaum.
Dekker, J., Boot, B., van der Woude, L. H. V., & Bijlsma, J. W. J. (1992).
Pain and disability in ostereoarthritis: A review of biobehavioral mech-
anisms. Journal of Behavioral Medicine, 15, 189–214.
Eden, D., & Aviram, A. (1993). Self-efficacy training to speed reemploy-
ment: Helping people to help themselves. Journal of Applied Psychol-
ogy, 78, 352–360.
Epping-Jordan, J. E., Compas, B. E., & Howell, D. C. (1994). Predictors of
cancer progression in young adult men and women: Avoidance, intrusive
thoughts, and psychological symptoms. Health Psychology, 13, 539–
547.
Ericsson, K. A., & Simon, H. A. (1993). Protocol analysis: Verbal reports
as data. Cambridge, MA: MIT Press.
Folkman, S., & Lazarus, R. (1985). If it changes it must be a process: Study
of emotion and coping during three stages of a college examination.
Journal of Personality and Social Psychology, 48, 150–170.
1210 OETTINGEN AND MAYER

Fredrickson, B. L. (1998). What good are positive emotions? Review of
General Psychology, 25, 364–372.
Fredrickson, B. L. (2001). The role of positive emotions in positive
psychology: The broaden-and-build theory of positive emotions. Amer-
ican Psychologist, 56, 218–226.
Gogia, P. P., Christensen, C. M., & Schmidt, C. (1994). Total hip replace-
ment in patients with osteoarthritis of the hip: Improvement in pain and
functional status. Orthopedics, 17, 145–150.
Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of
simple plans. American Psychologist, 54, 493–503.
Goodhart, D. E. (1986). The effects of positive and negative thinking on
performance in an achievement situation. Journal of Personality and
Social Psychology, 51, 117–124.
Hatfield, E. (1988). Passionate and companionate love. In R. J. Sternberg
& M. L. Barnes (Eds.), The psychology of love (pp. 191–217). New
Haven, CT and London: Yale University Press.
Hatfield, E., & Walster, G. W. (1978). A new look at love. Reading, MA:
Addison-Wesley.
Havighurst, R. J. (1972). Developmental tasks and education. New York:
David McKay. (Original work published 1948)
Heckhausen, H. (1991). Motivation and action. Heidelberg, Germany:
Springer.
Holahan, C. J., & Moos, R. H. (1986). Personality, coping, and family
resources in stress resistance: A longitudinal analysis. Journal of Per-
sonality and Social Psychology, 51, 389–395.
Holmes, B. H., & Werbel, J. D. (1992). Finding work after job loss: The
role of coping resources. Journal of Employment Counseling, 29, 22–29.
Huba, G. J., Singer, J. L., Aneshensel, C. S., & Antrobus, J. S. (1982). The
short imaginal processes inventory. Port Huron, MI: Research Psychol-
ogist Press.
James, W. (1950). The principles of psychology. New York: Dover. (Orig-
inal work published 1890)
Jensen, M. R. (1987). Psychobiological factors predicting the course of
breast cancer. Journal of Personality, 55, 317–342.
Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social
psychology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The
handbook of social psychology (Vol. 1, pp. 233–265). New York:
McGraw-Hill.
Klinger, E. (1971). Structure and functions of fantasy. New York: Wiley.
Klinger, E. (1977). Meaning and void: Inner experience and the incentives
in people’s lives. Minneapolis, MN: University of Minnesota Press.
Klinger, E. (1978). Modes of normal conscious flow. In K. S. Pope & J. L.
Singer (Eds.), The stream of consciousness: Scientific investigations into
the flow of human experience (pp. 225–258). New York: Plenum Press.
Klinger, E. (1990). Daydreaming: Using waking fantasy and imagery for
self-knowledge and creativity. Los Angeles: Tarcher.
Kobasa, S. C., Maddi, S., & Kahn, S. (1982). Hardiness and health: A
prospective study. Journal of Personality and Social Psychology, 42,
168–177.
Labor Exchange (1997). Statistics information. Hamburg, Germany: Glo-
bus Kartendienst.
Lazarus, R. S. (1983). The costs and benefits of denial. In S. Bresnitz (Ed.),
Denial of stress (pp. 1–30). New York: International Universities Press.
Leedham, B., Meyerowitz, B. E., Muirhead, J., & Frist, W. H. (1995).
Positive expectations predict health after heart transplantation. Health
Psychology, 14, 74–79.
Lengua, L., & Sandler, I. (1996). Self-regulation as a moderator of the
relation between coping and symptomatology in children of divorce.
Journal of Abnormal Child Psychology, 24, 681–701.
Lewin, K. (1926). Vorsatz, Wille und Bedu¨rfnis [Intention, will, and need].
Psychologische Forschung, 7, 330–385.
Mahler, W. (1933). Ersatzhandlungen verschiedenen Realita¨tsgrades
[Compensatory action based on different degrees of reality]. Psycholo-
gische Forschung, 18, 27–89.
Mancuso, C. A., Salvati, E. A., Johanson, N. A., Peterson, M. G. E., &
Charlson, M. E. (1997). Patients’expectations and satisfaction with total
hip arthroplasty. The Journal of Arthroplasty, 12, 387–396.
Markus, H., & Nurius, P. (1986). Possible selves. American Psycholo-
gist, 41, 954–969.
McClelland, D. C. (1980). Motive dispositions: The merits of operant and
respondent measures. In L. Wheeler (Ed.), Review of personality and
social psychology (Vol. 1, pp. 10–41). Beverly Hills, CA: Sage.
Mischel, W. (1973). Toward a cognitive social learning reconceptualiza-
tion of personality. Psychological Review, 80, 252–283.
Mischel, W. (1974). Processes in delay of gratification. In L. Berkowitz
(Ed.), Advances in experimental social psychology (Vol. 7, pp. 249–
292). New York: Academic Press.
Mischel, W., Cantor, N., & Feldman, S. (1996). Principles of self-regula-
tion: The nature of willpower and self-control. In E. T. Higgins & A. W.
Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp.
329–360). New York: Guilford Press.
Nolen-Hoeksema, S. (1991). Responses to depression and their effects on
the duration of depressive episodes. Journal of Abnormal Psychology,
100, 569–582.
Nolen-Hoeksema, S., Parker, L. E., & Larson, J. (1994). Ruminative
coping with depressed mood following loss. Journal of Personality and
Social Psychology, 67, 92–104.
Oettingen, G. (1996). Positive fantasy and motivation. In P. M. Gollwitzer
& J. A. Bargh (Eds.), The psychology of action: Linking cognition and
motivation to behavior (pp. 236–259). New York: Guilford Press.
Oettingen, G. (1997). Culture and future thought. Culture & Psychology, 3,
353–381.
Oettingen, G. (2000). Expectancy effects on behavior depend on self-
regulatory thought. Social Cognition, 18, 101–129.
Oettingen, G., & Gollwitzer, P. M. (2001). Goal setting and goal striving.
In A. Tesser & N. Schwarz (Eds.), Intraindividual processes. Volume 1
of the Blackwell handbook in social psychology (pp. 329–347). Oxford,
England: Blackwell.
Oettingen, G., Pak, H., & Schnetter, K. (2001). Self-regulation of goal
setting: Turning free fantasies about the future into binding goals.
Journal of Personality and Social Psychology, 80, 736–753.
Oettingen, G., & Wadden, T. A. (1991). Expectation, fantasy, and weight
loss: Is the impact of positive thinking always positive? Cognitive
Therapy and Research, 15, 167–175.
O’Leary, A., Shoor, S., Lorig, K., & Holman, H. R. (1988). A cognitive-
behavioral treatment for rheumatoid arthritis. Health Psychology, 7,
527–544.
Olson, J. M., Roese, N. J., & Zanna, M. P. (1996). Expectancies. In E. T.
Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of
basic principles (pp. 211–238). New York: Guilford Press.
Peterson, C. (1988). Explanatory style as a risk factor for illness. Cognitive
Therapy and Research, 12, 119–132.
Peterson, C., & Bossio, L. M. (1991) Health and optimism: New research
on the relationship between positive thinking and physical well-being.
New York: Free Press.
Peterson, C., & Seligman, M. E. P. (1984). Causal explanations as a risk
factor for depression: Theory and evidence. Psychological Review, 91,
347–374.
Peterson, C., Seligman, M. E. P., & Vaillant, G. E. (1988). Pessimistic
explanatory style is a risk factor for physical illness: A thirty-five-year
longitudinal study. Journal of Personality and Social Psychology, 55,
23–27.
Peterson, L., Oliver, K. K., & Saldana, L. (1997). Children’s coping with
stressful medical procedures. In S. A. Wolchik & I. N. Sandler (Eds.),
Handbook of children’s coping: Linking theory and intervention (pp.
333–360). New York: Plenum Press.
Pham, L. B., & Taylor, S. E. (1999). From thought to action: Effects of
1211
EXPECTATIONS VERSUS FANTASIES

process- versus outcome-based mental simulations on performance. Per-
sonality and Social Psychology Bulletin, 25, 250–260.
Reed, G. M., Kemeny, M. E., Taylor, S. E., & Visscher, B. R. (1999).
Negative HIV-specific expectancies and AIDS-related bereavement as
predictors of symptom onset in asymptomatic HIV-positive gay men.
Health Psychology, 18, 354–363.
Ried, G. J., Dubow, E. F., & Carey, T. C. (1995). Developmental and
situational differences in coping among children and adolescents with
diabetes. Journal of Applied Developmental Psychology, 16, 529–554.
Schaufeli, W. B., & Van Yperen, N. W. (1993). Success and failure in the
labor market. Journal of Organizational Behavior, 14, 559–572.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psycho-
logical and physical well-being: Theoretical overview and empirical
update. Cognitive Therapy and Research, 16, 201–228.
Scheier, M. F., Matthews, K. A., Owens, J. F., Magovern, G. J., Sr.,
Lefebvre, R. C., Abbott, R. A., & Carver, C. S. (1989). Dispositional
optimism and recovery from coronary artery bypass surgery: The ben-
eficial effects on physical and psychological well-being. Journal of
Personality and Social Psychology, 57, 1024–1040.
Schneider, S. L. (2001). In search of realistic optimism: Meaning, knowl-
edge, and warm fuzziness. American Psychologist, 56, 250–263.
Schunk, D. H. (1982). Effects of effort attributional feedback on children’s
perceived self-efficacy and achievement. Journal of Educational Psy-
chology, 74, 548–556.
Seligman, M. E. P. (1991). Learned optimism. New York: Knopf.
Sherman, S. J., Skov, R. B., Hervitz, E. F., & Stock, C. B. (1981). The
effects of explaining hypothetical future events: From possibility to
probability to actuality and beyond. Journal of Experimental Social
Psychology, 17, 142–158.
Singer, J. L. (1966). Daydreaming. New York: Random House.
Singer, J. L., & Antrobus, J. S. (1963). A factor-analytic study of day-
dreaming and conceptually-related cognitive and personality variables.
Perceptual and Motor Skills, 17, 187–209.
Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural
equations models. In S. Leinhart (Ed.), Sociological methodology (pp.
290–312). San Francisco: Jossey-Bass.
Spencer, S. M., & Norem, J. K. (1996). Reflection and distraction: Defen-
sive pessimism, strategic optimism, and performance. Personality and
Social Psychology Bulletin, 22, 354–365.
Spirito, A., Stark, L., & Tyc, V. (1994). Stressors and coping strategies
described during hospitalization by chronically ill children. Journal of
Clinical Child Psychology, 23, 314–322.
Steffy, B. D., Shaw, K. N., & Noe, A. W. (1989). Antecedents and
consequences of job search behaviors. Journal of Vocational Behav-
ior, 35, 254–269.
Sternberg, R. J., & Barnes, M. L. (Eds.). (1988). The psychology of love.
New Haven, CT: Yale University Press.
Suls, J., & Wan, C. K. (1989). Effects of sensory and procedural informa-
tion on coping with stressful medical procedures and pain: A meta-
analysis. Journal of Consulting and Clinical Psychology, 57, 372–379.
Super, D. E., & Hall, D. T. (1978). Career development: Exploration and
planning. Annual Review of Psychology, 29, 333–372.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social
psychological perspective on mental health. Psychological Bulletin, 103,
193–210.
Taylor, S. E., & Clark, L. F. (1986). Does information improve adjustment
to noxious events? In M. J. Saks & L. Saxe (Eds.), Advances in applied
social psychology (Vol. 3, pp. 1–28). Hillsdale, NJ: Erlbaum.
Taylor, S. E., Kemeny, M. E., Reed, G. M., Bower, J. E., & Gruenewald,
T. L. (2000). Psychological resources, positive illusions, and health.
American Psychologist, 55, 99–109.
Taylor, S. E., Pham, L. B., Rivkin, I. D., & Armor, D. A. (1998).
Harnessing the imagination: Mental simulation, self-regulation, and cop-
ing. American Psychologist, 53, 429–439.
Tennov, D. (1979). Love and limerence. New York: Stein & Day.
Triandis, H. C. (1996). The psychological measurement of cultural syn-
dromes. American Psychologist, 51, 407–415.
Tversky, A., & Koehler, D. J. (1994). Support theory: A nonextensional
representation of subjective probability. Psychological Review, 101,
547–567.
Verbrugge, L. M. (1990). Disability. Rheumatic Disease Clinic, 16, 741–
761.
Watkins, M. M. (1976). Waking dreams. New York: Gordon and Breach.
Weinberger, D. A., Schwartz, G., & Davidson, R. (1979). Low-anxious,
high-anxious and repressive coping styles: Psychometric patterns and
behavioral and physiological responses to stress. Journal of Abnormal
Psychology, 88, 369–380.
Wright, S. S. (2000). Looking at the self in a rose-colored mirror: Unre-
alistically positive self-views and academic performance. Journal of
Social and Clinical Psychology, 19, 451–462.
Zimmerman, B. J., & Martinez-Pons, M. (1992). Perceptions of efficacy
and strategy use in the self-regulation of learning. In D. H. Schunk & J.
Meece (Eds.), Student perceptions in the classroom: Causes and con-
sequences (pp. 185–207). Hillsdale, NJ: Erlbaum.
Received December 13, 2001
Revision received May 21, 2002
Accepted May 24, 2002 䡲
1212 OETTINGEN AND MAYER