Download ACR 2010 Special Session Proposal

ACR 2010 Special Session Proposal
Submission Type: Special Session
Submission Title: Standards and Construction of Hedonic Value
Co-Chairs: Meng Zhu & Carey Morewedge (Carnegie Mellon University)
Address: 5000 Forbes Ave, Pittsburgh, PA 15217, Tel: 412-780-5380
E-mail: [email protected]; [email protected]
Papers: Each speaker (underlined below) has agreed to present in the proposed special session
1. The AB Identification Survey: Identifying Absolute versus Relative Determinants of
Happiness
Christopher K. Hsee & Adelle X. Yang (University of Chicago)
2. Social Standards and the Construction of Hedonic Value
Meng Zhu1, Carey Morewedge1, Daniel Gilbert2, Kristian Myrseth3, Karim Kassam2, & Timothy
Wilson4 (1Carnegie Mellon University, 2Harvard University, 3EMST, 4University of Virginia )
3. Absolute and Relative Choices of Maximizers and Satisficers
Kimberlee Weaver1, Norbert Schwarz2, Kim Daniloski1, & Keenan Cottone2 (1Virginia Tech,
2
University of Michigan)
4. Integrate or separate future pain? The impact of current pain
Eduardo B. Andrade1 , Marco Aurelio Bianchini2 , Newton Lucchiari 2 (1University of California,
Berkeley, 2 Universidade Federal de Santa Catarina, Brazil)
Special Session Proposal
Choices are based on the anticipated utility of experiences—their hedonic value. Correctly
anticipating hedonic value is much like hitting a moving target while riding a galloping horse, as (1)
anticipated value is constructed by comparison between experiences and imagined, present, and previous
experiences, and (2) forecasters inaccurately predict to which standards experiences will be compared
(Gilbert & Wilson, 2007; Kahneman & Miller, 1986; Parducci, 1995). Although standards do influence
the hedonic value of experiences (Medvec, Madey, & Gilovich, 1995; Zellner, Rohm, Bassetti, & Parker,
2003), the factors that determine the strength of their influence are ambiguous and their influence is often
overestimated (Hsee & Zhang, 2004; Novemsky & Ratner, 2003).
This symposium takes a close look at the relationship between standards and the construction of
hedonic value. In an attempt to gain better understanding of the interplay between standards and hedonic
experiences, the papers examine how judgment of and choices between different hedonic experiences are
influenced by comparison to absolute standards (i.e., experiences that are inherently evaluable, such as
pain) and relative standards (i.e., experiences that have to be evaluated based on external reference
information, such as social comparison).
Hsee and Yang introduce a simple method to distinguish between absolute vs. relative standards
that affect happiness. They propose the AB Identification Survey (ABIS), a simple and practical method
to identify whether a given variable exerts an absolute effect on happiness or affects happiness only
through social comparison. The authors validated the method and used it on a variety of naturallyoccurring variables. Zhu, Morewedge, Gilbert, Myrseth, Kassam and Wilson show that relative
standards (i.e., other people) serve as a potent standard of comparison. They show that social standards
evoke hedonic contrasts in situations where non-social standards don’t. They demonstrate that social
comparison is sufficient to lead to suboptimal decisions, e.g., choosing low-value, difficult-to-compare
options rather than higher-value, easy-to-compare options. Weaver, Schwarz, Daniloski and Cottone
further examine the differential influence of absolute vs. relative standards on two types of consumers:
maximizers who are concerned with obtaining the best outcome, and satisficers who are more likely to
choose the first available option that meets a threshold. They find that maximizers are more likely to
choose objectively inferior products and outcomes when they are paired with higher relative standing than
satisficers. Finally, Andrade, Bianchini and Lucchiari examine the impact of current pain, an abosulte
standard that is inherently evaluable, on people’s preference for integration or separateion of future
experiences, in this case, dental treatments. They demonstrate that patients are more likely to integrate
future pain. However, the preference for integration of pain is reduced when a “sample” of pain is
provided directly before patients make their choice.
Taken together, the four papers (all in advanced stages) in this session elucidate the ambiguous
relationship between standards and hedonic value, noting when and why predictions and experiences
often diverge and lead to suboptimal decisions and fruitless behavior. As the session integrates diverse
research, it is expected to appeal to a broad audience, including those interested in happiness and wellbeing, consumer behavior, hedonic value, contrast and assimilation, social comparison and social
influences, and judgments and decision making. Talks will be kept brief to allow ample time for audience
interaction.
Paper Abstracts
The AB Identification Survey: Identifying Absolute versus Relative Determinants of Happiness
Christopher K. Hsee & Adelle X. Yang (University of Chicago)
Short Abstract. This article introduces a simple survey method to distinguish between two types of
variables that affect happiness – type A, which exerts an absolute effect on happiness, and type B, which
affects happiness only through social comparison. The authors validated the method and used it on a
variety of naturally-occurring variables.
Long Abstract. Some variables are inherently evaluable and others are inherently inevaluable. To say that
a variable is inherently evaluable (for simplicity, call it a type A variable), we mean that human beings
have an innate, shared and stable scale to assess which level of the variable is desirable and which is not
(ambient temperature, amount of sleep, etc); to say that a variable is inherently inevaluable (type B), we
mean that human beings have no innate scale to gauge its desirability and in order to evaluate its
desirability one must rely on external reference information, such as social comparison (the size of a
diamond, the horsepower of a car, etc).
The distinction between type A and type B variables carries important social implications. For
instance, improving the value of a type A variable for all the members in a society will absolutely raise
the happiness of all, whereas improving the value of a type B variable for all the members in a society is
merely a zero sum game and will not raise the happiness of all. Therefore from a congregate point of
view, policymakers should direct the resources to improving type A variables in order to increase
happiness across time.
In this paper we propose the AB Identification Survey (ABIS), a simple and practical method to
identify whether a given a variable is type A or type B. The procedure of the ABIS is as follows. Suppose
that we wish to identify whether a given variable X, say men’s height, is type A or type B, assuming that
greater X is more desirable and that the average X value in the given population is x m, for example, the
average height of men is 5’9”. ABIS consists of two versions, the JE version and the simulated-SE
version. Each version is simple and contains only two questions. The JE version reads as follows:
1.
How tall are you? ____
2.
As you may know, some men are tall and others are short, and men’s average height is
about 5’9”. Given that, how do you feel about your height? Give a number between 1 (very unhappy)
and 7 (very happy): _____
The simulated-SE version reads as follows:
1.
How tall are you? ____
2.
Suppose that you were living in a society where every man were of your height. Given
that, how would you feel about your height? Give a number between 1 (very unhappy) and 7 (very
happy): _____
Recruit a sample of men (whose height distribution should be representative of that of the target
population) and ask half the sample to answer the JE version of ABIS and the other half the simulated-SE
version.
After data are collected, run two regressions. One regresses the happiness ratings from the JE
version of ABIS on the target variable X (e.g., height) and control variables, if any. The other regresses
the happiness ratings from the Simulated-SE version of ABIS on X and control variables, if any. Let βJE
and βSimulated-SE denote the estimated regression coefficients of X from the two models, respectively. Then
we can simply use the following formula to identify whether X is type A or type B: AB coefficient =
βSimulated-SE/βJE.
βSimulated-SE reflects the non-comparative effect of X, and βJE reflects the total effect of X.
Generally speaking, The AB coefficient reflects the size of the non-comparative effect relative to the total
effect. The greater the AB coefficient, the more X is a type A variable; the smaller the AB coefficient, the
more X is a type B variable.
We report three studies. The first study is a validation study. It compares the results of the ABIS
with the results of a cleaner but less practical experimental method and shows that the two studies
produce very similar results. In addition, we report two studies to show how to apply this method to
identifying a series of naturally occurring variables as type A or type B. The first survey was conducted
among 453 undergraduate students, the second among 352 business students with work experiences. Our
surveys have revealed interesting findings . For undergraduate students, the two top type A variables are
vacation time and winter dorm temperature; the two top type B variables are home size and height
(meter). For business students, top type A variables were insomnia and vacation time, and top type B
variables are income and car price.
The ABIS can be used to inform policymakers where to invest resources. From the perspective of
an individual, improving either type A variables or type B variables will increase happiness: “I will be
happier than you if I can afford more comfortable temperature or I can afford more expensive jewelry.”
For that individual, pursuing both type A and type B variables is rational. But from the perspective of a
society, improving type B variables is a zero-sum game and only improving type A variables can improve
the happiness of all. Since this distinction applies only at the societal level but not at the individual level,
policymakers, rather than individuals, are better suited to distinguish type A and type B variables. ABIS is
an instrument they can use in this endeavor.
Social Standards and the Construction of Hedonic Value
Meng Zhu1, Carey Morewedge1, Daniel Gilbert2, Kristian Myrseth3, Karim Kassam2, & Timothy
Wilson4 (1Carnegie Mellon University, 2Harvard University, 3EMST, 4University of Virginia )
Short Abstract. We argue that hedonic contrast effects occur when experiences are unengaging or when
comparisons are easy to make. We demonstrate that as social comparison is automatic, social standards
can evoke hedonic contrasts in situations where non-social standard don’t, and we show that social
comparison can lead to suboptimal decisions.
Long Abstract. A contrast effect occurs when a standard of comparison influences judgment of a target
stimulus, such that the target is considered less like the standard when in its presence than when judged in
the absence of that standard. Contrast effects are prevalent in noetic (information based judgments) and
psychophysics, yet the evidence for hedonic contrast effects has been mixed (Novemsky & Ratner 2003).
We develop an attentional model to explain when and why comparative value influences hedonic value.
Specifically, we propose that hedonic contrast effects are particularly elusive in experience because
forecasters fail to account for (1) attentional requirements of comparison, as noticing the difference
between a standard and a target requires considerable cognitive resources; (2) different attentional
requirement of forecasting vs. experience—forecasters’ mental representations of experiences consume
fewer resources than having the forecasted experience. Thus, we predict that comparative value will
influence hedonic experiences when experiences are unengaging or when comparisons are easy to make.
We report four experiments that tested our theory by examining how social comparison, which is
automatic, well-practiced and often occurs without conscious awareness, influences hedonic value and
decision making. We predict that social standards will evoke hedonic contrast effects in situations where
nonsocial standard don’t.
Study 1 tested whether social standards produce a greater influence on hedonic value and whether
forecasters can correctly predict this difference. Participants were presented two food items, Lay’s Potato
Chips (target food) and another item (standard food), either Godiva Raspberry Chocolate (better standard)
or King Oscar canned sardines (worse standard). We manipulated social comparison by telling half of the
participants that they will be paired with another participant in the same experiment session and each of
them will be randomly assigned to sample different food item (partner condition). The other half of the
participants were simply assigned to evaluate one of the two food items (no-partner condition). All
participants were then assigned by computer to evaluate the target food (potato chips). Experiencers
reported how much they enjoyed eating chips and forecasted predicted how much they would enjoy eating
chips. As predicted, experiencers in the partner condition rated chips more enjoyable when their partner
was assigned to sardines as compared to chocolate, whereas in the no-partner condition, experiencers
enjoyed chips the same no matter what the alternative food was. However, forecasters predicted that
hedonic contrasts would occur in both the partner and no-partner conditions.
To test whether social standards are more influential because comparing to social standards
requires fewer attentional resources, Study 2 employed a 2(cognitive load: low vs. high) x 2(standard
food: chocolate vs. sardines) design. All participants were paired with another participant and assigned to
evaluate chips. To manipulate cognitive load, participants were asked to remember a series of number and
letter, either B4 (low load condition) or 8K3YW17L (high load condition) while performing the food
evaluation task. In the low load condition, participants enjoyed chips more when their partner was
assigned to sardines than chocolate. However, in the high load condition, enjoyment of chips was the
same regardless of the standard. These results are consistent with our attention based explanation showing
that hedonic contrasts are attenuated when experiencers don’t have enough cognitive resource to pay
attention to other people and compare their experiences. Study 3 examined whether hedonic contrast
occurs because people compare their experience to the alternative enjoyed by another person or because
they compare their imagined happiness to another person. In doing so, we manipulated the modality of
standards. We predict that social standards of a different modality will have a weaker effect on the target
experience, as it is more difficult to compare to standards of a different modality rather than standards of
the same modality. All participants were assigned to eat potato chips and their partners were assigned to
experience a superior or inferior alternative. Half of the participants were told that their partners were
experiencing either chocolate or sardines (same modality condition); whereas others were told that their
partners were listening to a sound clip of either jungle waterfall or vacuum cleaner (different modality
condition), which was rated in a pre-test as being as pleasant as chocolate or sardines. We found that
hedonic contrast occurred only when partners were experiencing easy-to compare same-modality
alternatives, thus supporting our theorization that social standards produce a greater influence on hedonic
value because social comparisons are easier not because they are fundamentally different.
Study 4 directly tested whether social comparison can lead to suboptimal decisions. We predict
that a superior social standard can lead people to choose a lower –value, difficult-to-compare option than
a higher-value, easy-to-compare option. Upon arrival, participants were paired with another participant
and completed an animal trivial. Participants were then asked to choose a prize from a set of four prizes—
two from a superior category (chocolates: Godiva & Hershey’s) and two from an inferior category (snack
bars: Granola Bar & Power Bar). Results from a separate pretest indicated that Godiva was ranked as the
best prize, Hershey’s was ranked as the 2nd best, Granola Bar was ranked as the 3rd best, and Power Bar
was ranked as the worst prize to get for participating in a psychology experiment. In the social
comparison condition, participants were told that their partner was randomly assigned to choose first and
Godiva was no longer available after being chosen by the partner. In the control, participants was simply
asked to choose a prize and told that Godiva was out of stock. Consistent with our prediction, we found
that participants in the social condition were more likely to choose the inferior Granola Bar instead of
Hershey’s than the control.
Absolute and Relative Choices of Maximizers and Satisficers
Kimberlee Weaver1, Norbert Schwarz2, Kim Daniloski1, & Keenan Cottone2 (1Virginia Tech,
2
University of Michigan)
Short Abstract. Three studies examining tradeoffs between absolute and relative options for consumer
goods show that maximizers, decision makers who are concerned with obtaining the best outcome, are
more likely to choose objectively inferior products and outcomes when they are paired with higher
relative standing than satisficers.
Long Abstract. According to the theory of bounded rationality, cognitive and situational limitations can
prevent individuals from optimizing or “maximizing” the outcome of many decisions (Simon 1955).
Recently, Schwartz et al. (2002) extended these theoretical insights into an individual differences scale.
“Maximizers” are consumers who exhaustively examine every possibility before deciding which choice is
best, while “satisficers” are more likely to choose the first available option that meets a threshold of
acceptability. In the current paper, we investigate a new question that past work has not considered:
whether maximizers and satisficers not only differ in the amount of effort they put forth in trying to
achieve their standards, but whether they also differ in the type of standard they choose to pursue in
consumer choice situations. Past work has remained agnostic about which type of standard maximizers
and satisficers emphasize when making decisions (Schwartz et al. 2002; Simon 1955). However, there
has been an implicit assumption that maximizers and satisficers are aspiring toward the same standard,
and that the major difference between the two types of decision makers is in the amount of effort they
expend in trying to achieve it, with maximizers expending more effort and satisficers truncating the
search process at an earlier point. In contrast to this implicit assumption, our major hypothesis in the
current paper is that maximizers and satisficers may actually aspire toward different standards; with
maximizers being more interested in pursuing the relative best, and satisficers more interested in pursuing
the absolute best.
Generally speaking, there are two types of standards that consumers can use when judging what is
“best”: absolute standards and relative standards. To illustrate this distinction, imagine that a consumer
can choose to live in one of two possible worlds. In World A, she will drive a car that is a medium
quality brand rated 7/10 in parts and performance by Automotive Weekly. Nearly all of her
acquaintances will drive cars that are luxury models rated 9/10. In World B, she will drive a car that is a
fair quality brand, rated 6/10 in parts and performance by Automotive Weekly. Nearly all of her
acquaintances will drive cars that are low quality models rated 4/10. Consumers who aspire toward an
absolute standard search for the objectively best outcome and would likely choose to live in World A,
since the car the consumer will have is rated more highly than the car they will have in World B (i.e., 7/10
in parts and performance versus 6/10). On the other hand, consumers who aspire toward a relative
standard search for an outcome that is best in the relative sense that it exceeds what others have. They
would thus prefer to live in World B since the car the consumer will have is better than the cars driven by
others (i.e., 6/10 on parts and performance versus 4/10) (Solnick and Hemenway 2005).
Results from three studies confirmed hypotheses by showing that maximizers pursue relative
standards more than satisficers. Specifically, study 1 showed that maximizers were more willing than
satisficers to accept objectively worse products or outcomes when they were paired with superior social
position (e.g., Option B above). Studies 2 and 3 sought to establish the psychological mechanism driving
the effect by manipulating the public visibility of the decision outcome. If maximizers’ emphasis on
social comparison reflects a concern for information about the absolute standards as proposed by past
work (Iyengar et al. 2006; Schwartz et al. 2002), then they should choose the same option regardless of
whether the outcome of their decision will be publicly visible or not. In contrast, if maximizers are
motivated by social rivalry, then the standards they use may shift from public to private contexts. Both
studies showed interactions between the maximizer/satisficer variable and the public visibility variable,
supporting the social rivalry explanation. Maximizers accepted objectively inferior products that were
paired with superior social position more than satisficers in public (e.g., house, car). However, the two
groups of decision makers were equally likely to maximize absolute quality when then outcome was
private and known only to the decision maker (e.g., pajamas, brand of mattress). The final study also
indicated that maximizers prefer counterfeit products –those that have low internal quality, but have the
outside appearance of high quality items—more than satisficers.
In conclusion, the current results examine a new theoretical question that past work has not
considered and show that maximizers and satisficers not only differ in the amount of effort they put forth
in trying to achieve their standards, but also differ in the type of standard they choose to pursue in
consumer choice situations, with maximizers emphasizing relative standards and satisficers emphasizing
absolute standards. In examining this question we bridge two literatures, the first on maximizing and
satisficing in decision making and the other on relative and absolute choices, which has been studied in
the decision making, psychology and consumer behavior literatures (e.g., Easterlin, 1974; Huguet, 2009;
Zhou & Soman 2003).
Integrate or separate future pain? The impact of current pain
Eduardo B. Andrade1 , Marco Aurelio Bianchini2 , Newton Lucchiari 2 (1University of California,
Berkeley, 2 Universidade Federal de Santa Catarina, Brazil)
Short Abstract. From plastic to dental surgeries, patients often have the option of undergoing future
painful treatments in either a single, or in two or more, periods. In a field experiment, we investigate the
extent to which patients’ current level of pain influences preference for integration or separation of future
dental treatments.
Long Abstract. It is often the case that health-related treatments can take place either in one single, or in
two or more, periods (e.g., extract two wisdom teeth). When no clinically dominant option is available,
the doctor allows the patient to decide whether to undergo the unpleasant experience all at once or to
separate it into multiple events. Although several considerations are at stake, the decision is in part a
matter of one’s ability to project and willingness to deal with the future painful experience. Interestingly,
patients are often not experiencing any level of pain at the time of the decision (e.g., a teenager who,
comfortably seated at the plastic surgeon’s office, decides to undergo breast and nose surgeries all at
once). Two questions emerge from this type of scenario: (a) do patients prefer to integrate or to separate
future painful treatments? And (b) does the current level of pain influence their preference?
In line with previous theoretical suggestions (Kahneman and Tversky 1979; Thaler 1985), we
hypothesize that patients are more likely to integrate future pain. However, this preference, we argue, is
partially due to the fact that they are often not experiencing any pain at the time of the decision. When
pain is experienced at the time of the decision, patients might question their capacity to endure the future
unpleasantness all at once (Linville and Fischer 1991). Thus, the preference for integration of pain could
be reduced when a “sample” of pain is provided directly before patients make their choice.
We tested this possibility in a field experiment in a public dental clinic. Sixty-two patients with
periodontal disease in both sides of the mouth (50% male) underwent a sequence of two standard and
required dental examinations: one painless (i.e., general clinical exam) and one painful (i.e., pocketing
and bleeding on probing exam). Our manipulation simply consisted of randomizing the order of
examinations. Half of the patients were randomized to a painless-painful order whereas the order was
reversed for the other half. After each dental examination, the patients were asked whether they preferred
to have the future periodontal therapy (i.e., scaling and root planing in both sides of the mouth) in either
one single intervention or in two interventions performed one week apart from one another. The results
show that, in general, patients were more likely to integrate than to separate the future painful experience.
However, the preference for integration diminished significantly when they experienced mild levels of
pain at the time of the decision. Whereas patients, on average, preferred to integrate future pain after a
painless examination, preference for integration did not differ from chance when the decision was made
after a painful examination. The effect emerged when comparisons were made within and between
subjects.
Given that patients’ currently experienced pain can so easily influence their choice, health
professionals should be aware of such effects when delegating the choice of future painful experiences to
patients. Finally, it is an open question whether visceral information about pain at the time of decision can
help patients make choices about integration versus separation with which they will, in retrospect, be
more satisfied.
References
Easterlin, R. A. (1974), “Does Economic Growth Improve the Human Lot? Some Empirical Evidence,”
in Nations and Households in Economic Growth, P.A. David and M.W. Reder, eds. New York
and London: Academic Press.
Huguet P, Dumas F, Marsh H, Wheeler L, Seaton M, Nezlek J, Suls J, & Régner I. (2009), “Clarifying the
Role of Social Comparison in the Big-Fish-Little-Pond Effect (BFLPE): An Integrative Study,”
Journal of Personality and Social Psychology, 97 (July), 156-170.
Iyengar, S. S. Wells, R. E., & Schwartz, B. (2006). Doing Better But Feeling Worse: Looking for the
“Best” Job Undermines Satisfaction. Psychological Science, 17 (February), 143-150.
Linville, P., & Fischer, G. (1991). Preferences for separating or combining events. Journal of Personality
and Social Psychology, 60, 5–23
Loewenstein, G., O'Donoghue, T. and Rabin, M. (2003). Projection bias in predicting future utility.
Quarterly Journal of Economics, 118, 1209-1248.
Gilbert, D. T., & Wilson, T. D. (2007). Prospection: Experiencing the future. Science, 317, 1351-1354.
Hsee, C. K. & Zhang, J. (2004). Distinction bias: Misprediction and mischoice due to joint evaluation.
Journal of Personality and Social Psychology, 86, 680-695.
Kahneman, D., & Miller, D. T. (1986). Norm theory: Comparing reality to its alternatives. Psychological
Review, 93, 136–153.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision. Econometrica, 47(2),
263-291.
Medvec, V. H., Madey, S. F., & Gilovich, T. (1995). When less is more: Counterfactual thinking and
satisfaction among Olympic medalists. Journal of Personality and Social Psychology, 69, 603610.
Novemsky, N., & Ratner, R. K. (2003). The time course and impact of consumers' erroneous beliefs about
hedonic contrast effects. Journal of Consumer Research, 29, 507-516.
Parducci, A. (1995). Happiness, pleasure, and judgment: The contextual theory and its applications.
Mahwah, NJ: Lawrence Erlbaum Associates.
Schwartz, B. Ward, A. Monterosso, J. Lyubomirsky, S. White, K., Lehman, D. R. (2002). Maximizing
versus satisficing: Happiness is a matter of choice. Journal of Personality and Social Psychology,
83 (November), 1178-1197.
Simon, H. A. (1955), “A behavioral model of rational choice,” Quarterly Journal of Economics, 59, 99118.
Solnick, S. J., & Hemenway, D. (2005), “Are positional concerns stronger in some domains than in
others?” American Economic Review, 95, 147-151.
Thaler, R. H. (1985), Mental accounting and consumer choice, Marketing Science, 4 199-214.
Zellner, D. A., Rohm, E. A., Bassetti, T. L., and Parker, S. (2003). Compared to what? Effects of
categorization on hedonic contrast. Psychonomic Bulletin & Review, 10, 468-473.
Zhou, R. & Soman D. (2003), “Looking Back: Exploring the Psychology of Queuing and the Effect of
the Number of People Behind,” Journal of Consumer Research, 29 (March), 517-530.