Choosing for others and its relation to information search

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Contents lists available at ScienceDirect

Organizational Behavior and Human Decision Processes journal homepage: www.elsevier.com/locate/obhdp

Choosing for others and its relation to information search Yi Liu a b c

a,1

, Evan Polman

b,⁎,1

a

, Yongfang Liu , Jiangli Jiao

T

c

School of Psychology and Cognitive Science, East China Normal University, China Wisconsin School of Business, University of Wisconsin-Madison, United States College of Educational Science, Xinjiang Normal University, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Self-other decision making Information search Regulatory focus Social distance

When people make choices, they both identify their options and research the unique details that comprise their options. Respectively, these two search behaviors are called alternative- and attribute-search. The literature treats these separate information search behaviors as a trade-off: Choosing to examine extant alternatives (alternative-search) means suffering the costs of not analyzing the details of alternatives (attribute-search), and vice versa. Here, we found that in choices people make for others, they search for more alternatives and more attributes than in choices people make for themselves. Moreover, we found that when people face a trade-off between searching for alternatives and attributes, people choosing for others will favor alternatives, whereas people choosing for themselves will favor attributes. Thus, we found that the pursuit of information is different when people choose for others (vs. themselves), suggesting a novel pivot to a range of areas in decision making where the alternative-attribute trade-off is ubiquitous.

A growing stream of research demonstrates that people choose differently when they choose for others than for themselves (Gorlin & Dhar, 2012; Hamilton & Thompson, 2007; Liu, Campbell, Fitzsimons, & Fitzsimons, 2013; Tu, Shaw, & Fishbach, 2016; Tunney & Ziegler, 2015). In some instances, the existing research finds that when deciding for others, people make less biased choices than when they decide for themselves. For example, people who choose for others demonstrate less intertemporal discounting effect, decoy effect, omission bias, betrayal bias, post-decisional distortion, choice overload, ego depletion, and loss aversion (Andersson, Holm, Tyran & Wengström, 2014; Gershoff & Koehler, 2011; Helgadóttir, 2015; Lu & Xie, 2014; Polman & Emich, 2011; Polman & Vohs, 2016; Polman, 2010, 2012a, 2012b; Pronin, Olivola, & Kennedy, 2008; Ziegler & Tunney, 2012; ZikmundFisher, Sarr, Fagerlin, & Ubel, 2006). That is, holding a decision constant – such as making a choice with the same options – past research has found that in some cases people make less biased choices for others. Consider one eye-opening study (Mata, Fiedler, Ferreira, & Almeida, 2013) that measured how well participants respond to the classic batand-ball problem, which goes: “A bat and a ball cost $1.10 in total. The bat costs $1 more than the ball. How much does the ball cost?” (despite the problem seems easy, a little over half of people tend to get it wrong, answering 10 cents when the correct answer is 5 cents; for a review, see Frederick, 2005). In their examination of the problem (and similar others), Mata et al. discovered that some people were more likely to

⁎

1

solve the problem when the problem belonged to someone else. Why? Mata et al. revealed that because people believe they are less biased than others (Pronin, 2008), their confidence in others’ ability for identifying the correct answer is relatively low, hence they subsequently scrutinize others’ choices more. Does this mean that people engage in more information seeking when making choices for others? We believe so. Specifically, we predict that when making choices for others (vs. for themselves), decision makers search for more information: They will search for more options, and search for more details that comprise their options. Several broad lines of research hint at the idea that people decide more thoroughly on behalf of others. For example, research shows that people sometimes help others more than they help themselves. At the extreme, research has found that people behave more assertively and less forgivingly toward transgressors who offend their friends than transgressors who offend them, personally (Green, Burnette, & Davis, 2008). Termed third-party (non)forgiveness, Kennedy and Ames (2013) found that people are more likely to protest on behalf of others’ misfortunes than on their own. In another example, research has found that people pay more to stop other people’s pain than what they pay to stop their own pain (Crockett, Kurth-Nelson, Siegel, Dayan, & Dolan, 2014). In this vein, it would appear as though people occasionally prioritize others’ well-being over their own, which might extend to prioritizing others’ choices too. In support, research has shown that for

Corresponding author. E-mail address: [email protected] (E. Polman). The author contributed equally to this research. This research was supported by the National Social Science Foundation of China (17CMZ017, 15ZDB121).

https://doi.org/10.1016/j.obhdp.2018.05.005 Received 16 May 2017; Received in revised form 9 May 2018; Accepted 14 May 2018 0749-5978/ © 2018 Elsevier Inc. All rights reserved.

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

information (the extent that people search for information about options; i.e., attribute-search). In this manner, we also focused on a theoretically-derived mediator, regulatory focus – which brings to bear a new prediction concerning alternative- and attribute-search. Research has shown that alternative- and attribute-search are thought of as a trade-off, where favoring one strategy means shying away from the other strategy (Payne, Bettman, & Johnson, 1993). And one of the fundamental questions since the beginning of research on decision making has been when and why people search for more (new) options or search for more (new) information about their options, before making a final choice (Simon, 1959). For example, before deciding what home to buy, people will search for many homes (seek out options) and they will research and pursue information about each home (seek out information about their options). Past research has identified some of the factors that affect people’s preferences for attributes and/or alternatives. For example, people focus more on attributes as task complexity or time pressure increases (Payne, 1976; Payne, Bettman, & Johnson, 1988) and people focus more on alternative-options as their expertise increases (Bettman & Park, 1980). In the midst of making a choice, people’s preferences for alternatives and attributes change as well, from one to the other: In the early stages, decision makers prefer to focus on attributes, whereas in the later stages, they prefer to focus on alternative-options (Bettman & Park, 1980). While a great deal of research has been dedicated to the determinants of information search strategies, little attention has been devoted to the processes underlying why decision makers will favor alternatives over attributes, and vice versa, why decision makers will favor attributes over alternatives. In this paper, we examine a self-other decision making account of behavior in alternative- and attribute-search decisions, and focus on how information search is higher overall when people make choices for others. Furthermore, we test why decision makers would favor one kind of information over another. In this vein, Polman (2012a, 2012b), Kuhn (2015), and Liu, Wang, Yao, Yang, and Wang (2017) have established that choosing for others puts people into a more promotion (vs. prevention) focused state, whereas choosing for the self puts people into a more prevention (vs. promotion) focused state. In specific terms, a promotion focus is related to seeking growth and development, and motivates decision makers to achieve positive outcomes; in contrast, a prevention focus is related to seeking safety and security, and motivates decision makers to avoid negative outcomes (Higgins, 1998). For instance, with respect to goal pursuit, the different motivations of promotion and prevention focus (along with their respective sensitivities to gains and losses) result in systematically different preferences for the types of strategies people use to complete tasks or make decisions. Preferring eagerness-related strategies, promotion focused individuals are concerned with achieving “hits” and avoiding “misses” (errors of omission); in contrast, prevention focused individuals prefer vigilancerelated strategies and are concerned with achieving “correct rejections” and avoiding “false hits” (errors of commission; Crowe & Higgins, 1997). It is in the service of these different goals that we believe distinct information search strategies would emerge. Specifically, we predict that a promotion focus is related to favoring alternative-options over attributes, and that a prevention focus is related to favoring attributes over alternative-options. In support of this prediction, it is well-established that a prevention focus directs attention toward more local, concrete, and detailed forms of information, whereas a promotion focus directs attention toward more global, abstract, and generalized forms of information (Forster & Higgins, 2005; Semin, Higgins, de Montes, Estourget, & Valencia, 2005; Zhu & Meyers-Levy, 2007). By dint of regulatory focus, this difference is thought to stem from the efficacy of local processing for avoiding an error of commission and, likewise, of global processing for avoiding an error of omission (Liberman, Idson, Camacho, & Higgins, 1999; Pham & Higgins, 2005). To illustrate, imagine you are hungry and looking in the fridge for something to eat:

intertemporal choices, people choose more valuable later-larger rewards when choosing for others than when choosing for themselves (Pronin et al., 2008). Likewise, people also value their close friends’ possessions more than their own possessions (Greenstein & Xu, 2015), and, people are happier spending money on others than they are on themselves (Dunn, Aknin, & Norton, 2008). Finally, as a general rule when it comes to making choices for others (vs. for the self), people are more creative, idealistic, pleasure-seeking, and variety-seeking (Choi, Kim, Choi, & Yi, 2006; Laran, 2010; Lu, Liu, & Fang, 2016; Lu, Xie, & Xu, 2013; Polman & Emich, 2011) – indeed people profess to enjoy making choices more for others than for themselves (Polman & Vohs, 2016), though there are exceptions to this rule, such as when people make life-and-death decisions for others (Botti & Iyengar, 2006). Admittedly, in these examples and in most real-life situations, the documented effects could be a result of other phenomena besides people searching for more information when choosing for others. Thus, it is unclear whether people who choose for others (vs. themselves) unequivocally seek out more information. Encouragingly, the link between self-other decision making and information search is less opaque in the relatively mature research on resource allocation in economic games (Engel, 2011). Germane to the current research, participants in one study were asked to choose between massage packages and allocate two massages between themselves and a close friend (Tu et al., 2016; Study 4a). Before making their choice, participants could elect to see one piece of information about the massage packages (e.g., their own massage duration or their friend’s massage duration). The authors found that 14% of the participants chose to see their own massage details, whereas almost twice as many participants chose to see their friend’s massage details. Unfortunately, it is not possible to conduct a statistical analysis on these participants’ preferences (these are descriptive statistics in a single sample/condition of participants), but it is encouraging that a higher number of participants sought out information about others’ choice-options than they did about their own. Indeed, these findings suggest that when people choose for others (vs. themselves), they may consider more information in the form of more choice-alternatives and choice-attributes. Entirely consistent with this view, making a choice for someone else has been theorized as an instance of accountability (Chang, Chuang, Cheng, & Huang, 2012; Lu, Liang, & Duan, 2017; Tetlock, 1992). And research by Lee, Herr, Kardes, and Kim (1999) showed that under the sway of accountability, people examine more information, employ multiple search criteria, and employ more compensatory choice strategies. Thus, while research is suggestive that decision makers have a larger appetite for information when choosing for others, this relation has not been directly tested. By examining this possible link, the present research resembles though differs from existing research on the difference between choices people make for themselves and the advice they give for others’ choices (e.g., Blunden & Gino, 2018; Dana & Cain, 2015). In particular, it has been found that people tend to give advice with less indifference (most fervor; Danziger, Montal, & Barkan, 2012) on account of strongly weighting some information-attributes over others. This type of lexicographic weighting leads to polarizing options in a choice-set, thereby establishing a clearly favored option that an advice-giving individual will correspondingly recommend – which stands in contrast to a more balanced set of preferences that is thought to underscore choices that people make for the self (Kray & Gonzalez, 1999; Kray, 2000). To be sure, our focus here is not on advice-giving nor on information-weighting, but on decisions people make for others and if the amount of information sought is higher by people making these decisions, compared to people making decisions for themselves. We therefore tested whether choosing for others increases the extent to which decision makers search for more information. In addition, in our research, we examine not only the amount of overall information that people search for, but also the type of information that people search for, in the form of alternative-information (the extent that people search for options; i.e., alternative-search) and attribute66

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

The closer you look (the more you “zoom in”) the more you can see if a particular product is harmful (such as the mold on an orange, or the passed-expiration date on a yogurt); in contrast, the farther out you are (the more you “zoom out”) the more you can see all of the potential available options at once (such as the box of chocolates at the back of the fridge, or the cheese on one of the shelves on the door, or perhaps even items located outside of the fridge). In this vein, shifting from prevention to promotion focus is akin to using the zoom lens of a camera, by focusing on the fine details one can identify (and avoid eating) a “false” snack, whereas by focusing on the overall summary impression one can identify (and avoid overlooking) a “missed” snack. If this link between regulatory focus and local/global processing is applied to information in a choice (for contrast, the link between regulatory focus and local/global processing is typically applied to perceptual/visual, linguistic, or novel/familiar information), one would predict that a prevention focused decision maker attends to more of the fine details of a choice, whereas a promotion focused decision maker attends to more of the extant alternatives to a choice. Indeed, the fine details are precisely the attributes that configure to form the alternatives in toto. Although this prediction is mainly derived from the regulatory focus literature, it is also consistent with the existing literature on psychological distance. For example, Liberman, Trope, and Stephan (2007) speculated that for temporally distant choices, decision makers would search more for alternative-options than attributes. And in light of work that has shown that dimensions of distance operate psychologically as correlated, interchangeable, and substitutable (Fiedler, Jung, Wänke, & Alexopoulos, 2012; Huang, Burtch, Hong, & Polman, 2016; Maglio, Trope, & Liberman, 2013), we might expect this pattern to extend to social distance, which scholars have operationalized as the difference between making choices for the self versus others (Cho, Khan, & Dhar, 2013; Hamilton & Thompson, 2007; Polman & Emich, 2011; Zhao & Xie, 2011).

making a choice, one group can consume more alternatives and attributes in general, yet consume less of one kind in particular (when making a choice between seeking attributes and alternatives). In terms of information search, we propose that in the absence of such a choice, the group with the larger appetite will indeed consume more information (more alternatives and/or more attributes), yet in the presence of such a choice, the group with the larger appetite may consume more of one type of information and less of another. Although this food consumption example helps to illustrate our hypotheses, it is encouraging that existing research in psychology demonstrates the same pattern with information consumption: In a study on cognitive closure, participants with low need for closure sought out more information overall (i.e., sought out a higher number of alternatives and attributes) compared to participants with high need for closure; however, despite seeking more information, participants with low need for closure also sought out fewer attributes compared to participants with high need for closure (Choi, Koo, Choi, & Auh, 2008). Thus, when it comes to information, it is theoretically and empirically possible to consume more overall, yet less of one type of information. In more support, research has found that people who prefer seeking alternative-options over attributes tend to seek out more information overall (Klayman, 1983; Payne et al., 1988; Payne, 1976). In keeping with our focus on information search, we conducted seven studies that test our two hypotheses. In Studies 1a/b and 2, we examined, separately, whether participants choosing for others will search for more alternatives (Studies 1a/b) and attributes (Study 2) compared to participants choosing for themselves. In particular, in Study 1b, we replicated our findings from Study 1a among incentivized participants who sought restaurant-alternatives in a choice that was incentive-compatible (receipt of a gift-card for participants’ chosen restaurant). Next, in Studies 3 and 4, we provide further evidence while ruling out an account that is based on a potentially greater feeling of preference-uncertainty that manifests among people who make choices for others (vs. themselves). Specifically, in Study 3, we measured how much participants would pay for seeing another attribute, and separately, another alternative. While in Study 4, we measured how much participants want to see information after making a choice. As we explain, the findings in both of these studies cast doubt on a preferenceuncertainty explanation. Finally, in Studies 5 and 6, we examined attributes and alternatives, in tandem, which allowed us to investigate individuals’ preference for seeking attributes versus alternatives. In particular, in keeping with the most recent updates in mediation analysis (Fiedler, Harris, & Schott, 2018), in Study 6, we found that when participants’ regulatory focus is included as a separate variable in a regression model, it absorbs a significant part of the variance between self-other decision making and information search (consistent with a mediation model, though not exclusively). For each study, we conducted a sensitivity power analysis (described in each study’s method section). This analysis indicates the minimum effect size that could be detected, given the respective study’s sample size, at 80% power with an alpha criterion of .05 (two-tailed). In all of our studies, the documented effect size exceeded the minimum effect size that can be detected. A summary of our studies’ results are presented in Table 1.

1. Hypotheses and studies overview In a nutshell, the literature on regulatory focus reveals parallels with alternative- and attribute-search. In our research, we develop a theoretical and empirical link between self-other decision making, regulatory focus, and alternative/attribute-search, which highlights why and when decision makers pursue more information, and what kind of information they pursue. Formally then, our research contains two hypotheses: (1) people who choose for others seek more information (more attributes and more alternatives) than people who choose for themselves; and (2) people who choose for others will prefer alternatives over attributes, whereas people who choose for themselves will prefer attributes over alternatives. To illustrate and clarify our hypotheses, consider an example based on consuming hotdogs and cheeseburgers at a BBQ. For the first hypothesis, we are proposing that one group of people will consume more food at the BBQ than another group of people. Which is to say that, one group has a greater appetite and will therefore consume more hotdogs and cheeseburgers than the other group. However, because there are different types of food at the BBQ, and because people like some foods more than others, we propose in the second hypothesis that when people can choose between foods (i.e., choose between hotdogs and cheeseburgers), an interaction pattern can emerge, which is to say that the group that consumes more food overall does not as a matter of fact always consume both more hotdogs and more cheeseburgers than the other group. To illustrate, if the group that consumes more food overall likes hotdogs more than cheeseburgers, and vice versa, the other group likes cheeseburgers more than hotdogs, then all things equal, the group that consumes more food overall (say, an average of 3 hotdogs and 1 cheeseburger) may consume more hotdogs yet fewer cheeseburgers compared to the other group and what it consumes (say, an average of 1 hotdog and 2 cheeseburgers). This example highlights that before

2. Studies 1a and 1b In Studies 1a and 1b, we examined whether participants choosing for others will search for more alternatives (hunt for more options) than participants choosing for themselves. The two studies followed similar procedures, except that in Study 1b participants had an incentive to gainfully make a choice. 2.1. Method In Study 1a, 119 undergraduate students (74 female, mean 67

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

Thus, for the same choice in which participants made a decision for either themselves or someone else, they elected to do more research – by investigating other alternatives – when making a choice for others (vs. themselves). This provides evidence that decision makers explore more options when making choices for others – they “zoom out” and consider new possibilities. Importantly, this behavior was not predicted by duration spent making choices, which did not differ by condition: in Study 1a, t(117) = 1.15, p = .252; in Study 1b, t(148) = 0.58, p = .558 (a point to which we return in the General Discussion, because it implies that participants making choices for others spent less time per alternative than participants making choices for themselves). In sum, Studies 1a and 1b found support for the hypothesis that when making decisions for others, people consider more choice-options. In doing so, the studies also demonstrate a novel take on choicesets and choice overload. Most research in decision making tends to provide participants with all of the possible choice-alternatives that they can choose, at the choice-outset. For example, research on choice overload provides participants with a pre-set small or large number of options, and then measures how satisfied participants are with their choice (Iyengar & Lepper, 2000). However, in our studies, participants could create their own sets (up to a limit of 10 options), which resembles how people ordinarily make choices – that is, people are rarely provided with a tabular or pre-set array of options; rather, they more often have to find and identify the options themselves. In this respect, our studies are different from past studies, though the results do square with previous work: Polman (2012a) found that people choosing for others were more satisfied with their choices after choosing among a large (vs. small) number of options. In complement, we found that people choosing for others will seek out more options in kind.

Table 1 Preference for alternatives and attributes in choices for the self and others. Choice context

Study 1a Study 1b Study 2 Study 3 Study 4 Study 5 Study 6

Restaurant Options Restaurant Options College Courses Dating Profiles College Courses Job Options College Courses

Alternatives

Attributes

Self choice

Other choice

Self choice

4.16 (1.47) 6.11 (1.67)

5.02 (1.69) 6.63 (1.63)

¥7.09 (4.44)

41.8% 3.00 (1.19)

¥10.99 (5.56)

65.9% 4.25 (1.38)

Effect size Other choice d = 0.54 d = 0.50

5.24 (1.51) ¥5.54 (2.95) 5.94 (1.56) 58.2% 3.98 (1.40)

6.46 (1.54) ¥7.63 (5.09) 6.75 (1.66) 34.1% 3.33 (1.48)

d = 0.81 f = 0.12 d = 0.51 w = 0.51 f = 0.50

Note: Standard deviations are in parentheses.

age = 20.9) participated in exchange for extra credit. With this sample size, our test is large enough to detect a medium effect, d = 0.52 at 80% power. In Study 1b, 150 undergraduate students (91 female, mean age = 19.2) participated in exchange for extra credit. With this sample size, our test is large enough to detect a similar effect, d = 0.46 at 80% power. In both studies, participants were randomly assigned to either make a choice for themselves or someone else. Specifically, participants chose between ten restaurants, where they saw information about one restaurant at a time. Each restaurant (and its corresponding information; viz. price, location, menu, environment) was presented in randomized order. Following the initial display of one restaurant (and its corresponding information), participants were asked whether they were ready to make a choice or whether they wanted to see another restaurant alternative. They were shown subsequent alternatives (one by one) until they decided to make a final choice (or until they had seen all ten alternatives). How many alternatives participants elected to see furnished our dependent measure – in addition we measured how much time participants spent making their choice. To create an incentive-compatible context in Study 1b, participants were told they would receive a ¥30 gift-card for the restaurant they choose for themselves. In a similar manner, participants choosing for others were told that their friend (for whom they are choosing) would receive a ¥30 gift-card for the chosen restaurant. Before beginning their choice, we asked participants in this condition to indicate a friend, for whom they would choose a restaurant. However, at the end of the study, we simply gave the gift-card to participants in this condition to give to their friend (for whom they chose) since asking for friends’ contact information would violate ethics board approval (that said, because we gave participants the gift-card after making their choice for others, this feature of the study should not affect the results). Thus, Study 1b comprised incentivized participants who sought alternatives in a choice that was incentive-compatible.

3. Study 2 In this study, we examined whether participants choosing for others will search for more attributes (seek out more information about each of their options) compared to participants choosing for themselves. 3.1. Method 124 undergraduate students (71 female, mean age = 20.3) participated in the experiment in exchange for extra credit. With this sample size, our test is large enough to detect a medium effect, d = 0.51 at 80% power. Participants were randomly assigned to make a decision (a choice of college course) for either the self or someone else (a classmate who cannot come back to school on time). Specifically, participants chose between two elective college courses with ten attributes describing each course (viz. pass rate, assignment difficulty, practical usefulness, depth of course materials, clarity of lectures, how interesting the course is, test difficulty, compatibility with interests, location convenience, time convenience). For each attribute, we created three levels (e.g., low, medium, high pass rate) and assigned one level of each attribute to each course randomly (thus the courses were effectively equal in desirability). Following the initial display of one attribute (for both courses), participants were asked whether they were ready to make a choice or whether they wanted to see another attribute. They were shown subsequent attributes (one by one in random order – i.e., participants did not know in advance which attribute would come next) until they decided to make a final choice (or until they had seen all ten attributes). How many attributes participants elected to see furnished our dependent measure – in addition we measured how much time participants spent making their choice.

2.2. Results and discussion In Study 1a, we found participants making choices for others considered more alternatives (M = 5.02; SD = 1.69) than participants making choices for themselves (M = 4.16; SD = 1.47), t(117) = 2.93, p = .004, d = 0.54. Likewise, we found the same result when participants made an incentive-compatible choice: In Study 1b, participants making choices for others also considered more alternatives (M = 6.93; SD = 1.63) than participants making choices for themselves (M = 6.11; SD = 1.67), t(148) = 3.07, p = .003, d = 0.50. In fact, despite the financial incentive, Study 1b’s effect size was nearly identical to Study 1a’s.

3.2. Results and discussion Participants making choices for others pursued more attributes (M = 6.46; SD = 1.54) than participants making choices for themselves (M = 5.24; SD = 1.51), t(122) = 4.47, p < .001, d = 0.81. Thus, not 68

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

to see another, fifth attribute that describes in more detail the available profiles. In addition, we asked participants how much they would pay to see another, fifth profile that contains information that comprises the attributes. Thus, in this study, we asked participants how much they would pay to see another attribute, and separately another alternative (to control for potential order effects, these questions were randomized among participants).

unlike Studies 1a and 1b, all participants in this study confronted the same choice of options – the only difference was whether they were making a choice for themselves or someone else. And we found that participants elected to do more research on their choice, by investigating more attributes, when making a choice for others (vs. themselves). This provides evidence that decision makers pursue more details about their options when they choose for others – they “dig in” deeper into their options. Viewed this way, our findings provide a replication of similar findings in the literature. For example, in an experiment that measured how much people distort information before making a choice, Polman (2010) found that, as a side-effect to distortion, participants examined more attributes when making choices for others (vs. for themselves). Likewise, as noted, Tu et al. (2016) found that more participants were interested in learning about the details of their friend’s massage than about their own. It bears noting that neither of these findings were formally predicted by the respective authors, which is to say that the information sought by participants in these studies is incidental to the authors’ main contribution. Still, it should be encouraging that these findings are consistent with our own findings. Finally, like Studies 1a and 1b, search behavior was not predicted by duration spent making choices, which did not differ by condition, t (122) = 1.35, p = .180, implying that participants making choices for others spent less time per attribute than participants making choices for themselves.

4.2. Results and discussion We conducted a one-way MANOVA with attribute-information price and alternative-information price as dependent variables, which yielded a significant difference between decisions for the self and decisions for a friend, F(2, 185) = 14.67, p < .001, ηp2 = 0.14. Followup tests indicated that participants choosing for others were willing to pay more for both attributes and alternatives compared to participants choosing for themselves. Specifically, the average willingness to pay for another attribute was ¥7.63 (SD = 5.09) for participants deciding for others, and ¥5.54 (SD = 2.95) for participants deciding for themselves, t(151.49) = 3.46, p = .001, d = 0.50 (the variances were unequal so we conducted an unequal variances t test). Likewise, the average willingness to pay for another alternative-option was ¥10.99 (SD = 5.56) for participants deciding for others, and ¥7.09 (SD = 4.44) for participants deciding for themselves, t(178.90) = 5.32, p < .001, d = 0.76 (as in the previous test, the variances were unequal so we conducted an unequal variances t test). Overall, participants deciding for others were willing to pay over 40% more money in exchange of seeing more information than participants deciding for themselves. A post-hoc consideration of our findings reveals that Study 3 tests and rules out an explanation that could plausibly account for our findings so far. Specifically, because people almost certainly know their own preferences better than others’ preferences, people may be able to make decisions for themselves faster (based on less information). Said differently, each piece of choice-information could be less diagnostic or less helpful when the choice is for someone else (vs. for the self). To illustrate, if Selby really likes oyster happy hours and wants to go to a seafood restaurant, then when making a restaurant-choice for herself, she may converge almost instantly on a seafood restaurant and easily dismiss (or avoid searching for) other alternatives. Whereas if Selby is choosing a restaurant for Otto, and is unclear about Otto’s preferences, then she may not curb the search of restaurants so soon, since there is no stand-out preference. In this vein, individual pieces of information could be considered less diagnostic when choosing for others, because the information does not shift decision makers that much closer to making a decision – which means that more pieces of information are sought to make a choice. If true, however, we would expect that participants pay less for information concerning choices for others – i.e., if the piecemeal information is less diagnostic and less helpful then participants should pay less for it – yet we find the opposite, which suggests that a different explanation is at play. That said, this is merely a post-hoc observation of Study 3; thus we conducted Study 4 to further test the role of preference-uncertainty.

4. Study 3 Study 3 sought to replicate the findings observed thus far with a different dependent measure. Instead of measuring how many alternatives and attributes participants elect to see before making their choice, we asked participants in this study how much they would pay to see another alternative or attribute. This is not unlike measuring how much participants value information – in that indicating a higher willingness to pay for information would suggest that participants value information and want more of it. This particular measure can be viewed as a laboratory analogue of paying for information, such as paying money for management consulting services, or for receiving a second medical opinion, or, as in the case of some dating websites, for receiving more access to potential dating partners. In this vein, we employed a different choice context from the previous studies – one where participants were asked to make choices about dating partners. This particular choice is well-suited to a context in which people make choices for others. For example, it was recently discovered that when it comes to creating dating profiles, users’ friends are better at creating profiles for users than users are themselves (White, Sutherland, & Burton, 2017). Moreover, people especially like to help others by playing the role of “matchmaker” and finding their friends a dating partner (Anik & Norton, 2014). 4.1. Method 188 undergraduates participated in the experiment in exchange for extra credit. With this sample size, our test is large enough to detect a small-to-medium effect, d = 0.41 at 80% power. Participants were randomly assigned to either make a choice for themselves or someone else (a friend). In the former, “self-choice” condition, participants viewed four dating profiles/partners described by four attributes each. Participants indicated their preferred partner-gender before seeing the profiles; next we showed them the profiles, all at once, with their preferred partner-gender (besides the photo, the male and female profiles were identical). In the latter, “other-choice” condition, participants were asked to imagine a friend who is choosing dating partners on a dating website (we subsequently asked what partner-gender the friend prefers), and then similar to the “self-choice” condition, participants saw four dating partners/profiles described by four attributes each. In both conditions, we asked participants how much they would pay

5. Study 4 As noted in Study 3, when making choices for others, a desire for more information could be attributed to people’s relative uncertainty of others’ preferences. To test this explanation, we carried out a study in which information diagnosticity should be the same across choices for the self and others. Specifically, we asked participants how much choice-information they want to see after making their choice. To be sure, acquiring information after a choice can still be useful or influential – it can bolster a decision maker’s confidence (or foster doubt) – but because the information follows the choice, it cannot be used to mitigate uncertainty that people might have of preferences (their own or others’ preferences) before making a choice. In other words, before a 69

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

6.1. Method

choice is made, information is acquired in the service of matching a choice with preferences (if preferences are initially uncertain then acquiring more information may be helpful to finding a match); however, after a choice is made, information is no longer serviceable in this way. Thus, while acquiring information after making a choice may be useful, its efficacy to match preferences with a choice-option is immediately rendered moot upon selection of a choice-option.

182 undergraduates participated in the experiment in exchange for extra credit. With this sample size, our test is large enough to detect a small-to-medium effect in a contingency table analysis, w = 0.21 at 80% power. Half of the participants were instructed to make a decision for the self concerning a job search website, and the other half were instructed to make the same decision but for a friend using the website. Specifically, participants were told that they could pay money to see more information about the available jobs. Similar to Study 3, participants saw four jobs described by four attributes each. And participants were instructed that they could pay to view a fifth attribute that would further describe the jobs, or view a fifth job that would be described by the attributes. Unlike Study 3, we asked participants which type of information they want to see: the fifth attribute or the fifth job alternative. Thus, our test is a chi-square test between target (choosing for self vs. friend) and preferred type of information (attribute vs. alternative).

5.1. Method 144 undergraduates (83 female, mean age = 18.9) participated in the experiment in exchange for extra credit. With this sample size, our test is large enough to detect a small-to-medium effect, d = 0.47 at 80% power. This study followed a similar procedure to Study 2, in which we examined whether participants choosing for others will search for more attributes (seek out more information about each of their options) compared to participants choosing for themselves. In extension of Study 2, we added a measure that assessed how much participants would like to see more information about their choice. We asked this question after participants made their choice; participants responded from 1 (not at all) to 7 (a lot).

6.2. Results and discussion We conducted a 2 (target: self/friend) × 2 (information type: alternative/attribute) chi-square test, which yielded a significant crossover interaction, χ2 (1, N = 182) = 10.70, p = .001, w = 0.24. Participants making choices for others were more likely to choose the alternative-option (65.9%) over the attribute (34.1%). In contrast, participants making choices for themselves were more likely to choose the attribute (58.2%) over the alternative-option (41.8%). Thus, in a relatively strong way – the odds ratio is 2.70 (95% confidence interval = [1.48, 4.93], p < .001) – we found that among participants choosing for others, the odds of choosing an alternative-option are 2.70 times larger than the odds among participants choosing for themselves; and vice versa for choosing an attribute. In other words, when faced with a choice between seeing an attribute versus an alternative-option, people choosing for others favored the alternative-option, whereas people choosing for themselves favored the attribute.

5.2. Results and discussion In line with Study 2, participants making choices for others pursued more attributes (M = 6.75; SD = 1.66) than participants making choices for themselves (M = 5.94; SD = 1.56), t(142) = 3.01, p = .003, d = 0.51. We next conducted the same test on the extent that participants want to see additional choice-information after making their choice. We found that participants making choices for others wanted to see more information (M = 2.36; SD = 1.27) than participants making choices for themselves (M = 1.81; SD = 1.04), t(142) = 2.87, p = .005, d = 0.48. Altogether, these results leave some degree of ambiguity about the uncertainty explanation – while it stands to reason that decision makers prefer more information the more uncertain they are before making a choice, it does not appear to exclusively drive the results in our studies inasmuch as participants continue to want information after making a choice for others, as well as report a higher willingness to pay for information for others’ choices (as found in Study 3).

7. Study 6 In the previous study, participants could pick only one piece of information. By contrast, in Study 6, we allowed participants to choose new information multiple times during a single choice, in which they could see both more attributes and more alternatives. Thus, in this study, we are able to test both of our hypotheses at once: We measured how many attributes and alternatives participants elected to see before making a final choice, and we measured participants’ relative favorability for attributes or alternatives. As before, we predict that participants choosing for others will search for more information than participants who choose for themselves – i.e., a main effect of target (self/ other). However, in line with Study 5, and our investigation of regulatory focus, we predict that the information-type preference will manifest in decisions where participants can search for both alternatives and attributes: Specifically, we tested whether a higher promotion focus (activated by choosing for others) will lead to favoring information about alternative-options over attributes, and vice versa, a higher prevention focus (activated by choosing for the self) will lead to favoring attributes over alternative-options. Thus, when both types of information (alternatives and attributes) can be considered, participants choosing for others will prefer to research more alternative-options than attributes, and vice versa, participants choosing for themselves will prefer to research more attributes than alternative-options – i.e., an interaction effect between target and information type.

6. Study 5 Although the preceding studies provide evidence consistent with our first hypothesis, they did not directly examine whether decision makers have different preferences for alternative-options and attributes when making self-other choices. In Study 5, we measured participants’ preferences for seeking alternatives and attributes by allowing them to research information about both alternatives and attributes. We predicted that people seek more information, overall, when choosing for others vs. themselves (our first hypothesis). Going a step further, we also predicted that people have a preference for a certain type of information when choosing for others (our second hypothesis). As the literature on self-other decision making suggests, when people choose for others, they focus more on promotion focused features, whereas when people choose for themselves, they focus more on prevention focused features (Kuhn, 2015; Liu et al., 2017; Polman, 2012a, 2012b). To the extent that alternative-options represent promotion focused features, and attributes represent prevention focused features, we predicted that people may prefer seeking alternative-options (over attributes) when choosing for others, and likewise, people may prefer seeking attributes (over alternative-options) when choosing for themselves.

7.1. Method 125 undergraduates participated in the experiment in exchange for 70

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

extra credit. With this sample size, our test is large enough to detect a small-to-medium effect in a 2 × 2 mixed-design ANOVA, f = 0.13 at 80% power. Half of the participants were instructed to make a decision (a choice of college course) for the self, and the other half were instructed to make a decision on behalf of someone else (a classmate who cannot come back to school on time). The experiment is a 2 (target: self/other) × 2 (information type: attribute/alternative) mixed-design, with target as a between-subjects factor and information type as a within-subjects factor. Participants initially received three attributes each about three college courses and were asked whether they were ready to make a final choice, or whether they wanted additional information and if so, what kind of information (attribute or alternative). Participants who chose to see more information were shown the requested information (attribute or alternative) until they were ready to make a final choice, or until they had seen all attributes and alternatives. In all, there were ten courses that were described by ten attributes each (we used the same attributes and levels from Study 2). After making a final choice, participants responded to the following two questions assessing promotion focus and prevention focus in terms of committing an error of omission versus an error of commission (Crowe & Higgins, 1997), respectively, “In the process of selection, I was concerned with missing a good course” and “In the process of selection, I was concerned with choosing a bad course.” Participants answered the counter-balanced questions from 1 (not at all) to 7 (extremely). How many alternatives and attributes participants elected to see furnished our dependent measures.

Table 2 Indirect effects of self-other decision making via regulatory focus on information search, Study 6. Alternatives

Attributes

Alternatives + Attributes

Promotion Focus

0.50* [0.26, 0.85]

0.46* [0.19, 0.89]

Prevention Focus

0.06† [−0.22, 0.04] 0.53* [0.28, 0.86]

0.04† [−0.13, 0.27] 0.34* [0.13, 0.74] 0.09† [−0.28, 0.07] 0.39* [0.14, 0.76]

Promotion Focus (with Prevention Focus Covariate) Prevention Focus (with Promotion Focus Covariate)

0.03† [−0.15, 0.06]

0.40* [0.15, 0.81] 0.44* [0.21, 0.83]

0.42* [0.15, 0.78]

Note: Asterisks indicate that 95% confidence interval does not include zero, and daggers indicate a “swapped” analysis.

via the prevention focus measure (95% CI = [0.13, 0.74]; point estimate = 0.34; the direct effect was not significant). Encouragingly, we conducted four different robustness checks that provide further support for the interrelations between self-other decision making, regulatory focus, and information search. First, in a separate set of bootstrapping analyses (see Table 2), we carried out the above tests with participants’ companion regulatory focus as a covariate. Which is to say, we tested (1) whether promotion focus absorbs a significant portion of the variance between target and alternativesearch while controlling for prevention focus, and (2) whether prevention focus absorbs a significant portion of the variance between target and attribute-search while controlling for promotion focus. For these respective indirect effects, results indicated the same pattern as above – the 95% confidence intervals (with 1000 samples) were [0.28, 0.86] and [0.14, 0.76], and the direct effects were not significant, ps > .25. Second, we conducted all of the above bootstrapping analyses with the promotion focus and prevention focus measures swapped for each other. As Table 2 indicates, in these “swapped” tests, the results of the indirect effects are no longer significant, which suggests that alternative-search is indeed closely related to promotion focus (and not prevention focus), and that attribute-search is likewise related to prevention focus (and not promotion focus). Third, we also examined information search as an omnibus variable (the sum of alternatives and attributes that participants sought). Again, as Table 2 indicates, we found that both promotion focus and prevention focus absorb a significant portion of the variance, which is to be expected because the omnibus variable is a composite score of the number of alternatives plus attributes sought, and as a result it is very highly correlated with each of the separate alterative-search and attribute-search variables (r = .68 and r = .70, respectively). Thus, when alternative- and attribute-search are considered together as an omnibus measure, the indirect effects of self-other decision making via promotion and prevention focus remain significant. Notably, promotion focus appears to have a stronger relation with information search than does prevention focus, which is consistent with (1) our studies’ findings showing that people search for more information when choosing for others, and (2) previous research showing that when people choose for others they tend to be more promotion focused (vs. prevention focused; Polman 2012a, 2012b). Furthermore, this pattern squares with other work that has found that people who tend to prefer searching for alternative-options over attributes also tend to examine more information overall (Choi et al., 2008; Klayman, 1983; Payne et al., 1988; Payne, 1976), which is akin to our finding that people who choose for others prefer searching for alternative-options over attributes, and likewise examine more information overall. As a final robustness check, we conducted a separate study among 137 participants that followed the same procedure as Study 6 except

7.2. Results and discussion We conducted a 2 (target: self/other) × 2 (information type: alternative/attribute) mixed-design ANOVA, which yielded a marginally significant main effect of target, F(1, 123) = 1.87, p = .093. Across both information types, participants making choices for others sought more information overall (M = 7.58, SD = 2.16) than participants making choices for themselves (M = 6.98, SD = 1.77). The main effect of information type was not significant, indicating that across all participants (i.e., across participants in both of the “self” and “other” conditions), one type of information was not preferred over the other. More importantly, we found the predicted interaction between target and information type which demonstrates the self-other difference in preference for seeing alternative-options versus attributes, F(1, 123) = 31.16, p < .001, f = 0.50. Participants making choices for others considered more alternative-options (M = 4.25, SD = 1.38) than attributes (M = 3.33, SD = 1.48), F(1, 123) = 12.93, p = .0005, d = 0.64. In contrast, participants making choices for themselves pursued more attributes (M = 3.98, SD = 1.40) than alternative-options (M = 3.00, SD = 1.19), F(1, 123) = 17.64, p < .001, d = 0.75. We also found that promotion focus was higher among participants making choices for others (M = 5.15, SD = 1.21) than for themselves (M = 4.22, SD = 1.44), F(1, 123) = 15.49, p < .001, d = 0.70. Moreover, prevention focus was higher among participants making choices for themselves (M = 4.93, SD = 1.21) than for others (M = 4.23, SD = 1.56), F(1, 123) = 7.77, p = .006, d = 0.50. In light of these results and our predictions, we next tested if promotion focus and prevention focus can separately account for a significant portion of the variance between self-other decision making and, respectively, alternative-search and attribute-search, which would be consistent with a mediation model, among other causal models. We adopted a bootstrap approach to testing a mediation model suggested by Preacher and Hayes (2004). This pair of analyses (see Table 2), using 1000 bootstrapped samples, rendered (1) a 95% confidence interval that did not include zero for the indirect effect of self-other decision making on alternative-search via the promotion focus measure (95% CI = [0.26, 0.85]; point estimate = 0.50; the direct effect was not significant); and (2) a 95% confidence interval that did not include zero for the indirect effect of self-other decision making on attribute-search 71

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

causal mechanisms that contribute to decision makers’ different preferences for different information types. For example, besides regulatory focus, one might also consider personality characteristics of decision makers who are choosing for others. Some research on dispositions and choices people make for others has been carried out already, among for example, decision makers who are depressed (GarciaRetamero, Okan, & Maldonado, 2015), or have higher vs. lower selfesteem (Wray & Stone, 2005), or an independent vs. interdependent self-construal (Polman & Vohs, 2016). In terms of information search, personalities such as maximizing and need for cognition seem especially relevant, and fruitful topics of future research. While we focused on regulatory focus, a worthwhile issue is the role of motivations and uncertainty when people are searching for information while making choices for others. For instance, in choices for others, decision makers could be (1) attempting to identify an option that the majority of others would like, perhaps because of the relative uncertainty of others’ preferences or fear of regret; (2) attempting to choose a defensible, easy-to-justify (e.g., less blameworthy) option; or (3) attempting to garner favor from others. The latter motivation is consistent with work on regulatory focus, where choosing for others may be viewed as an opportunity for affirming the ideal-self. Called the Michelangelo phenomenon, others are considered active participants in the self’s personal development and goal pursuit, where movement toward the ideal-self is described as receiving others’ affirmation or approval (Drigotas, Rusbult, Wieselquist, & Whitton, 1999). Quite possibly, among the ways of receiving others’ affirmation is making efficacious choices for those others. Like choosing a restaurant for a friend, and then finding out that it has become his or her favorite restaurant, people may develop positive meta-perceptions (perceptions of what others think of the self; Kenny, 1994), which are precisely what have been found to yield success toward attaining one’s ideal-self (Rusbult, Finkel, & Kumashiro, 2009). All three of the above motivations imply that choices that people make for others might be different for different kinds of others, such as strangers or close others. Our research focused on choices for friends because we think it is likely that such choices are more common than choices people make for relatively distant others. Thus, we focused on choices that we consider to be high on external validity – the most frequent choice that people often make for others: their friends. In further support of this intuition – that people make more choices for close (vs. distant) others – Laran (2010) examined choices people made for friends versus family, and did not find any differences. Nevertheless, it is possible that choosing for different others (e.g., choosing for employees or colleagues) might raise additional considerations, and thus change people’s choices. In this vein, through the lens of psychological (social) distance, one might view our focus on friends as a conservative test, being that decisions for the self will resemble decisions made for others, when others are close to the self (Polman & Emich, 2011). Thus, quite possibly, as people make decisions for increasingly distant others, the gap in terms of the extent of information search might widen at first and then gradually taper off – illustrating a logarithmic pattern that is thought to undergird a variety of effects in the psychological distance literature (Maglio et al., 2013; Polman, Effron, & Thomas, 2018). Future research might examine whether our findings differ depending on the particular person for whom people are making decisions. Not unlike how research in personality has shown that people with different dispositions make different decisions for themselves (e.g., maximizers versus satisficers; Parker, De Bruin, & Fischhoff, 2007), research could investigate whether people make different decisions for others who have different personalities. Here, our main contribution is to demonstrate the possible role of regulatory focus, but future research should deepen our understanding of how people choose for others by exploring its mechanisms and boundary conditions in greater detail. When it comes to making a choice – like choosing an ice cream flavor, or a research project to pursue, or a house to buy – decision makers typically explore their alternatives, and pursue information

that we used different measures of regulatory focus. Participants completed the Regulatory Focus Questionnaire (RFQ; Higgins et al., 2001) that measures promotion focus (α = .73) and prevention focus (α = .80). This study’s results mirrored those in Study 6, thus demonstrating convergent evidence with a different measure of regulatory focus (see Appendix for results to this study). In sum, the mediation model can be ascertained. That said, despite our robustness checks it remains an open question if regulatory focus unequivocally mediates the relation between self-other decision making and information search (for recent discussion on interpreting mediation models, see Fiedler et al., 2018). For instance, the regulatory focus measures that we used could just as well assess other concepts that are related to, though different from promotion and prevention focus. Moreover, other causal models besides a purely mediation model could be consistent with the data; and, other variables besides regulatory focus may also provide support for a mediation model on both theoretical and empirical grounds. As noted in our introduction, there is a wide range of documented self-other differences in decision making in the literature, which could mean that the self-other difference in information search is very likely multiply determined. Here, our main contribution is to demonstrate a novel phenomenon and to suggest a mechanism, but future research should deepen our understanding of the phenomenon by exploring underlying mechanisms. 8. General discussion People decide for others in ways that are inconsistent with the decisions they make for themselves, with occasionally unfortunate results for the self. For example, Amanatullah and Morris (2010) found women negotiate higher salaries for others than they do for themselves. The notion of people being unable to help themselves as much as they help others seems puzzling when their decisions are identical (when their options are the same). Shedding literal light, fMRI research has established that choices for others are made differently, as evidenced by different parts of the brain that activate when people choose for others versus themselves (Jung, Sul, & Kim, 2013). In line with this account, we found that people use different search strategies when making choices for others. And possibly, people may be able to imagine their own choices as though they are others’ choices, and reap the benefits such as a higher salary. In support, people who merely imagined themselves as stock traders experienced less loss aversion (SokolHessner et al., 2009); while similar work on enclothed cognition has found that when wearing a white lab coat (as if one is a scientist), people’s performance on the Stoop task increased (Adam & Galinsky, 2012). Considering this, people may be able to imagine their own problems as belonging to others – not unlike the convergence in how people imagine their future selves and others (Pronin et al., 2008). Taking this idea a step further, we ran a post-test that measured 514 undergraduates’ answers to the recently updated cognitive reflection task (CRT-2; a battery of challenging questions that measure “miserly information processing;” Thomson & Oppenheimer, 2016, p. 99), only we tested a condition in which participants were asked to “imagine you are advising someone and solving this problem on their behalf.” We compared this group of advisor-participants to a group of participants who simply answered the questions for themselves, and we found that advisor-participants answered more of the questions correctly (M = 67.2%, SD = 23.8%) compared to the other participants (M = 58.4%, SD = 23.1%), t(5 1 2) = 4.24, p < .001. In this regard, a path to self-help might include a dose of other-help. In fact, it is perhaps no coincidence that when it comes to offering advice, people sometimes say “here’s my two cents” yet when people ask for others’ advice, they might say “penny for your thoughts.” In line with our results, people can offer more to others. In our research we found regulatory focus to explain, statistically, what information people decide to pursue. Although this provides potential evidence of a causal mechanism, there are perhaps multiple 72

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

account of behavior in alternative- and attribute-search decisions, and focused on how information search is higher when people make choices for others, compared to when people make their own choices. Of import, this prediction assumes that the same decision maker is capable of alternative- and attribute-search – the sole difference is whether people are making a decision for themselves or for others. Thus, more generally speaking, our question asks whether people may help themselves (by pursuing more information) in the same fruitful way that they help others. From a research viewpoint, establishing a relationship between selfother decision making, regulatory focus, and alternative/attributesearch could set in motion future work that focuses on understanding these variables and illuminating their potential relationships with other psychological constructs. A broad range of behaviors has been identified to stem from regulatory focus. For example, regulatory focus is related to preference construction, attitudes, self-concept, social influence, emotions, visual attention, approach-avoidance, learning goal orientations, and group dynamics – to name a few (Higgins, 2005). In stark contrast, despite the clear relevance of psychology to self-other decision making, relatively little is known about how people make choices for others. However, by connecting self-other decision making and alternative/attribute-search to regulatory focus, we open the door to possible links between these areas and those investigated in the regulatory focus literature. In a similar vein, research on psychological distance is a mature literature that spans two decades, and one dimension of psychological distance is social distance. As noted, researchers have operationalized social distance as self-other decision making (Cho et al., 2013; Hamilton & Thompson, 2007; Polman & Emich, 2011; Zhao & Xie, 2011). Because the dimensions of psychological distance have been shown to affect a wide range of outcomes (Liberman, Trope, & Wakslak, 2007), and further, that the dimensions of distance operate psychologically as correlated, interchangeable, and substitutable (such that what is true for one distance holds true for each; Fiedler et al., 2012; Huang et al., 2016; Maglio et al., 2013), a link connecting self-other decision making (in terms of social distance) with alternative/attribute-search provides both areas with a well-developed network to various psychological constructs. In the spirit of future research, we encourage researchers to “dig in” to this network of findings while also “zoom out” and hunt for new ones.

about their alternatives before committing to one option over others. For instance, when deciding on a research project that has inconclusive results, a scientist may scrap the idea in favor of pursuing a different idea, or s/he may continue with the idea, and collect more data in order to arrive at a more conclusive result. In this example, the scientist is facing a trade-off between alternative- and attribute-search. Our research shows that choosing for others may help resolve some of the tension between alternative- and attribute-search. However, another way to view our findings is that people who choose for others are merely trifling with the information comprising their choices, seeing as how participants in our studies spent the same amount of time making their own decisions as participants making decisions for others. In other words, the participants choosing for others spent less time on each piece of information, suggesting their attention directed to each piece of information is smaller too. However, rather than see this as evidence that people are not carefully making choices for others, we suspect this pattern is a logical extension of the alternative/attribute-search tradeoff and of information search more generally – that the more information people consider, the less time they spend considering each individual piece of information. Consider that in our studies if the participants choosing for others felt less motivated, involved, or engaged then we would expect them to look at less information, or spend less time on their decisions, or spend less money for the information. However, we observed none of these behaviors. On the contrary, we found that decision makers preferred to search for more information when choosing for others, across a range of choices including an incentivized choice. The present investigation targeted information search through the lens of self-other decision making to examine two questions: how much and what information people pursue when making choices for others. At first glance, our findings could appear mixed, because we found in Studies 2 and 4 that participants sought more attributes when making choices for others (vs. themselves). Yet in Studies 5 and 6 we found that participants favored seeking alternative-options over attributes when making choices for others, to the extent that participants in Study 6 sought more attributes when making choices for themselves (vs. others), p = .015, which stands in contrast to the main finding in Studies 2 and 4. That said, there is a non-trivial difference between these studies and Study 6: Studies 2 and 4 only allowed participants to choose how many attributes to view, whereas Study 6 contained a choice between alternative- and attribute-search. In this vein, both results are consistent with our hypotheses – people choosing for others will tend to seek more information compared to people choosing for themselves (this addresses how much information people pursue; see Studies 1–4, and 6). However, when the information can take different forms and comprise a choice between different types of information, then it is possible for people to favor one kind of information over another (this addresses what information people pursue; see Studies 5 and 6). Study 5, in particular, clearly demonstrates the different information-type preference that decision makers have: When participants can choose only one type of information, the majority choose the alternative-option when choosing for others, whereas the majority choose the attribute when choosing for the self. In sum, when people choose for others they will indeed seek more information (generally speaking); moreover, when provided with a particular choice between types of information, they will seek information in a selective fashion, by favoring alternative-options over attributes.

Appendix 137 undergraduates (86 female, mean age = 19.7) participated in the experiment in exchange for extra credit. With this sample size, our test is large enough to detect a small-to-medium effect in a 2 × 2 mixed-design ANOVA, f = 0.24 at 80% power. Half of the participants were instructed to make a decision (a choice of college course) for the self, and the other half were instructed to make a decision on behalf of someone else (a classmate who cannot come back to school on time). The experiment is a 2 (target: self/other) × 2 (information type: attribute/alternative) mixed-design, with target as a between-subjects factor and information type as a within-subjects factor. Participants initially received three attributes each about three college courses and were asked whether they were ready to make a final choice, or whether they wanted additional information and if so, what kind of information (attribute or alternative). Participants who chose to see more information were shown the requested information (attribute or alternative) until they were ready to make a final choice, or until they had seen all attributes and alternatives. In all, there were ten courses that were described by ten attributes each (we used the same attributes and levels from Studies 2 and 6). After making a final choice, participants completed the Regulatory Focus Questionnaire (RFQ; Higgins et al., 2001) that measures promotion focus (α = .73) and prevention focus (α = .80). How many alternatives and attributes participants elected to see furnished our dependent measures – in addition we measured how

9. Conclusion Given the extensive interest in alternative- and attribute-search decisions, it is surprising that the basic processes underlying these decisions are still poorly understood. Why do some decision makers pursue more alternative-options and why do others pursue more attributes? Are those differences due to personality variables or situational variables? In this paper, we examined a self-other decision making 73

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

much time participants spent making their choice.

Cho, E. K., Khan, U., & Dhar, R. (2013). Comparing apples to apples or apples to oranges: The role of mental representation in choice difficulty. Journal of Marketing Research, 50(4), 505–516. Choi, J., Kim, B. K., Choi, I., & Yi, Y. (2006). Variety-seeking tendency in choice for others: Interpersonal and intrapersonal causes. Journal of Consumer Research, 32(4), 590–595. Choi, J. A., Koo, M., Choi, I., & Auh, S. (2008). Need for cognitive closure and information search strategy. Psychology & Marketing, 25(11), 1027–1042. Crockett, M. J., Kurth-Nelson, Z., Siegel, J. Z., Dayan, P., & Dolan, R. J. (2014). Harm to others outweighs harm to self in moral decision making. Proceedings of the National Academy of Sciences, 111(48), 17320–17325. Crowe, E., & Higgins, E. T. (1997). Regulatory focus and strategic inclinations: Promotion and prevention in decision-making. Organizational Behavior and Human Decision Processes, 69(2), 117–132. Dana, J., & Cain, D. M. (2015). Advice versus choice. Current Opinion in Psychology, 6, 173–176. Danziger, S., Montal, R., & Barkan, R. (2012). Idealistic advice and pragmatic choice: A psychological distance account. Journal of Personality and Social Psychology, 102(6), 1105–1117. Drigotas, S. M., Rusbult, C. E., Wieselquist, J., & Whitton, S. W. (1999). Close partner as sculptor of the ideal self: Behavioral affirmation and the Michelangelo phenomenon. Journal of Personality and Social Psychology, 77(2), 293–323. Dunn, E. W., Aknin, L. B., & Norton, M. I. (2008). Spending money on others promotes happiness. Science, 319(5870), 1687–1688. Engel, C. (2011). Dictator games: A meta study. Experimental Economics, 14(4), 583–610. Fiedler, K., Harris, C., & Schott, M. (2018). Unwarranted inferences from statistical mediation tests–An analysis of articles published in 2015. Journal of Experimental Social Psychology, 75, 95–102. Fiedler, K., Jung, J., Wänke, M., & Alexopoulos, T. (2012). On the relations between distinct aspects of psychological distance: An ecological basis of construal-level theory. Journal of Experimental Social Psychology, 48(5), 1014–1021. Forster, J., & Higgins, E. T. (2005). How global versus local perception fits regulatory focus. Psychological Science, 16, 631–636. Frederick, S. (2005). Cognitive reflection and decision making. The Journal of Economic Perspectives, 19(4), 25–42. Garcia-Retamero, R., Okan, Y., & Maldonado, A. (2015). The impact of depression on self–other discrepancies in decision making. Journal of Behavioral Decision Making, 28(1), 89–100. Gershoff, A. D., & Koehler, J. J. (2011). Safety first? The role of emotion in safety product betrayal aversion. Journal of Consumer Research, 38(1), 140–150. Gorlin, M., & Dhar, R. (2012). Bridging the gap between joint and individual decisions: Deconstructing preferences in relationships. Journal of Consumer Psychology, 22(3), 320–323. Green, J. D., Burnette, J. L., & Davis, J. L. (2008). Third-party forgiveness: (Not) forgiving your close other's betrayer. Personality and Social Psychology Bulletin, 34(3), 407–418. Greenstein, M. J., & Xu, X. (2015). My Mug Is Valuable, But My Partner's Is Even More So: Economic Decisions for Close Others. Basic and Applied Social Psychology, 37(3), 174–187. Hamilton, R. W., & Thompson, D. V. (2007). Is there a substitute for direct experience? Comparing consumers’ preferences after direct and indirect product experiences. Journal of Consumer Research, 34(4), 546–555. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press. Helgadóttir, A. (2015). Asymmetric dominance effect in choice for others (Unpublished doctoral dissertation). Iceland: Reykjavik University. Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational principle. Advances in Experimental Social Psychology, 30, 1–46. Higgins, E. T. (2005). Value from regulatory fit. Current Directions in Psychological Science, 14(4), 209–213. Higgins, E. T., Friedman, R. S., Harlow, R. E., Idson, L. C., Ayduk, O. N., & Taylor, A. (2001). Achievement orientations from subjective histories of success: Promotion pride versus prevention pride. European Journal of Social Psychology, 31(1), 3–23. Huang, N., Burtch, G., Hong, Y., & Polman, E. (2016). Effects of multiple psychological distances on construal and consumer evaluation: A field study of online reviews. Journal of Consumer Psychology, 26(4), 474–482. Iyengar, S. S., & Lepper, M. R. (2000). When choice is demotivating: Can one desire too much of a good thing? Journal of Personality and Social Psychology, 79(6), 995–1006. Jung, D., Sul, S., & Kim, H. (2013). Dissociable neural processes underlying risky decisions for self versus other. Frontiers in Neuroscience, 7, 1–12. Kennedy, K. A., & Ames, D. (2013). Telling others to be tough: How and why advisors urge actors to be assertive. Poster presented at the Society for Personality and Social Psychology Conference, New Orleans, LA. Kenny, D. A. (1994). Interpersonal perception: A social relations analysis. Guilford Press. Klayman, J. (1983). Analysis of predecisional information search patterns. Advances in Psychology, 14, 401–414. Kray, L. J. (2000). Contingent weighting in self–other decision making. Organizational Behavior and Human Decision Processes, 83, 82–106. Kray, L. J., & Gonzalez, R. (1999). Differential weighting in choice versus advice: I’ll do this, you do that. Journal of Behavioral Decision Making, 12, 207–218. Kuhn, K. M. (2015). Selecting the good vs. rejecting the bad: Regulatory focus effects on staffing decision making. Human Resource Management, 54(1), 131–150. Laran, J. (2010). Goal management in sequential choices: Consumer choices for others are more indulgent than personal choices. Journal of Consumer Research, 37(2), 304–314. Lee, H., Herr, P. M., Kardes, F. R., & Kim, C. (1999). Motivated search: Effects of choice accountability, issue involvement, and prior knowledge on information acquisition and use. Journal of Business Research, 45(1), 75–88.

Results We conducted a 2 (target: self/other) × 2 (information type: alternative/attribute) mixed-design ANOVA, which yielded a significant main effect of target, F(1, 135) = 6.48, p = .012, d = .44. Across both information types, participants making choices for others sought more information (M = 6.06, SD = 2.21) than participants making choices for themselves (M = 5.19, SD = 1.78). In addition, we found a significant interaction between target and information type which demonstrates the trade-off between alternative- and attribute-search, F(1, 135) = 27.48, p < .001, f = 0.45. Participants making choices for others considered more alternatives (M = 3.34, SD = 1.59) than attributes (M = 2.72, SD = 1.49), F(1, 135) = 12.97, p = .004, d = 0.40. In contrast, participants making choices for themselves pursued more attributes (M = 3.07, SD = 1.01) than alternatives (M = 2.12, SD = 1.17), F(1, 135) = 20.44, p < .001, d = 0.87. As in Studies 1a/b and 2, this behavior was not predicted by duration spent making choices, which did not differ by condition, t(135) = 1.14, p = .256. We also found that promotion focus was higher among participants making choices for others (M = 20.18, SD = 3.29) than for themselves (M = 18.74, SD = 2.76), F(1, 135) = 7.68, p = .006, d = 0.47. Moreover, prevention focus was higher among participants making choices for themselves (M = 19.39, SD = 2.11) than for others (M = 17.78, SD = 2.17), F(1, 135) = 19.43, p < .001, d = 0.75. In light of these results, we used bootstrapping (Hayes, 2013) to test whether regulatory focus mediated the link between target and alternative- and attribute-search. We carried out two bootstrapping procedures to determine (1) whether promotion focus mediates the relation between target and alternative-search, and (2) whether prevention focus mediates the relation between target and attribute-search. For these respective indirect effects, results indicated that the 95% confidence intervals (with 1000 samples) were [0.10, 0.49] and [0.24, 0.78]. Neither confidence interval contains zero, confirming that promotion and prevention focus acted as mediators (the direct effects were not significant, ps > .25). In a separate pair of bootstrapping analyses, we carried out the above tests with participants’ companion regulatory focus as a covariate. Which is to say, we tested (1) whether promotion focus mediates the relation between target and alternative-search while controlling for prevention focus, and (2) whether prevention focus mediates the relation between target and attribute-search while controlling for promotion focus. For these respective indirect effects, results indicated the same pattern as above – the 95% confidence intervals (with 1000 samples) were [0.11, 0.51] and [0.25, 0.73], and the direct effects were not significant, ps > .25. References Adam, H., & Galinsky, A. D. (2012). Enclothed cognition. Journal of Experimental Social Psychology, 48(4), 918–925. Amanatullah, E. T., & Morris, M. W. (2010). Negotiating gender roles: Gender differences in assertive negotiating are mediated by women’s fear of backlash and attenuated when negotiating on behalf of others. Journal of Personality and Social Psychology, 98(2), 256–267. Andersson, O., Holm, H. J., Tyran, J. R., & Wengström, E. (2014). Deciding for others reduces loss aversion. Management Science, 62(1), 29–36. Anik, L., & Norton, M. I. (2014). Matchmaking promotes happiness. Social Psychological and Personality Science, 5(6), 644–652. Bettman, J. R., & Park, C. W. (1980). Effects of prior knowledge and experience and phase of the choice process on consumer decision processes: A protocol analysis. Journal of Consumer Research, 7, 234–248. Blunden, H., & Gino, F. (2018). How the other half thinks: The psychology of advising. In E. L. MacGeorge, & L. M. Van Swol (Eds.). The oxford handbook of advice (pp. 43–68). Oxford University Press. Botti, S., & Iyengar, S. S. (2006). The dark side of choice: When choice impairs social welfare. Journal of Public Policy & Marketing, 25(1), 24–38. Chang, C. C., Chuang, S. C., Cheng, Y. H., & Huang, T. Y. (2012). The compromise effect in choosing for others. Journal of Behavioral Decision Making, 25(2), 109–122.

74

Organizational Behavior and Human Decision Processes 147 (2018) 65–75

Y. Liu et al.

perceived purchasing power is smaller for others than for the self. Journal of Consumer Research, 45(1), 109–125. Polman, E., & Emich, K. J. (2011). Decisions for others are more creative than decisions for the self. Personality and Social Psychology Bulletin, 37(4), 492–501. Polman, E., & Vohs, K. D. (2016). Decision fatigue, choosing for others, and self-construal. Social Psychological and Personality Science, 7(5), 471–487. Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36(4), 717–731. Pronin, E. (2008). How we see ourselves and how we see others. Science, 320(5880), 1177–1180. Pronin, E., Olivola, C. Y., & Kennedy, K. A. (2008). Doing unto future selves as you would do unto others: Psychological distance and decision making. Personality and Social Psychology Bulletin, 34(2), 224–236. Rusbult, C. E., Finkel, E. J., & Kumashiro, M. (2009). The Michelangelo phenomenon. Current Directions in Psychological Science, 18(6), 305–309. Semin, G. R., Higgins, T., de Montes, L. G., Estourget, Y., & Valencia, J. F. (2005). Linguistic signatures of regulatory focus: How abstraction fits promotion more than prevention. Journal of Personality and Social Psychology, 89(1), 36–45. Simon, H. A. (1959). Theories of decision-making in economics and behavioral science. American Economic Review, 49(3), 253–283. Sokol-Hessner, P., Hsu, M., Curley, N. G., Delgado, M. R., Camerer, C. F., & Phelps, E. A. (2009). Thinking like a trader selectively reduces individuals’ loss aversion. Proceedings of the National Academy of Sciences, 106(13), 5035–5040. Tetlock, P. E. (1992). The impact of accountability on judgment and choice: Toward a social contingency model. Advances in Experimental Social Psychology, 25(3), 331–376. Thomson, K. S., & Oppenheimer, D. M. (2016). Investigating an alternate form of the cognitive reflection test. Judgment and Decision Making, 11(1), 99–113. Tu, Y., Shaw, A., & Fishbach, A. (2016). The friendly taking effect: How interpersonal closeness leads to seemingly selfish yet jointly maximizing choice. Journal of Consumer Research, 42(5), 669–687. Tunney, R. J., & Ziegler, F. V. (2015). Toward a psychology of surrogate decision making. Perspectives on Psychological Science, 10(6), 880–885. White, D., Sutherland, C. A., & Burton, A. L. (2017). Choosing face: The curse of self in profile image selection. Cognitive Research: Principles and Implications. Wray, L. D., & Stone, E. R. (2005). The role of self-esteem and anxiety in decision making for self versus others in relationships. Journal of Behavioral Decision Making, 18(2), 125–144. Zhao, M., & Xie, J. (2011). Effects of social and temporal distance on consumers’ responses to peer recommendations. Journal of Marketing Research, 48(3), 486–496. Zhu, R., & Meyers-Levy, J. (2007). Exploring the cognitive mechanism that underlies regulatory focus effects. Journal of Consumer Research, 34(1), 89–96. Ziegler, F. V., & Tunney, R. J. (2012). Decisions for others become less impulsive the further away they are on the family tree. PloS ONE, 7(11), e49479. Zikmund-Fisher, B. J., Sarr, B., Fagerlin, A., & Ubel, P. A. (2006). A matter of perspective: Choosing for others differs from choosing for yourself in making treatment decisions. Journal of General Internal Medicine, 21(6), 618–622.

Liberman, N., Idson, L. C., Camacho, C. J., & Higgins, E. T. (1999). Promotion and prevention choices between stability and change. Journal of Personality and Social Psychology, 77(6), 1135–1145. Liberman, N., Trope, Y., & Stephan, E. (2007). Psychological distance. In E. T. Higgins, & A. Kruglanski (Eds.). Social psychology: Handbook of basic principles(2nd ed.). New York: Guilford. Liberman, N., Trope, Y., & Wakslak, C. (2007). Construal level theory and consumer behavior. Journal of Consumer Psychology, 17(2), 113–117. Liu, P. J., Campbell, T. H., Fitzsimons, G. J., & Fitzsimons, G. M. (2013). Matching choices to avoid offending stigmatized group members. Organizational Behavior and Human Decision Processes, 122(2), 291–304. Liu, H., Wang, L., Yao, M., Yang, H., & Wang, D. (2017). Self-other decision-making differences in loss aversion: A regulatory focus perspective. Journal of Applied Social Psychology, 47(2), 90–98. Lu, J., Liang, Y., & Duan, H. (2017). Justifying decisions: Making choices for others enhances preferences for impoverished options. Social Psychology, 48(2), 92–103. Lu, J., Liu, Z., & Fang, Z. (2016). Hedonic products for you, utilitarian products for me. Judgment and Decision Making, 11(4), 332–341. Lu, J., & Xie, X. (2014). To change or not to change: A matter of decision maker’s role. Organizational Behavior and Human Decision Processes, 124(1), 47–55. Lu, J., Xie, X., & Xu, J. (2013). Desirability or feasibility self-other decision-making differences. Personality and Social Psychology Bulletin, 39(2), 144–155. Maglio, S. J., Trope, Y., & Liberman, N. (2013). The common currency of psychological distance. Current Directions in Psychological Science, 22(4), 278–282. Mata, A., Fiedler, K., Ferreira, M. B., & Almeida, T. (2013). Reasoning about others' reasoning. Journal of Experimental Social Psychology, 49(3), 486–491. Parker, A. M., De Bruin, W. B., & Fischhoff, B. (2007). Maximizers versus satisficers: Decision-making styles, competence, and outcomes. Judgment and Decision making, 2(6), 342–350. Payne, J. W. (1976). Task complexity and contingent processing in decision making: An information search and protocol analysis. Organizational Behavior and Human Performance, 16, 366–387. Payne, J. W., Bettman, J. R., & Johnson, E. J. (1988). Adaptive strategy selection in decision making. Journal of Experimental Psychology: Learning, Memory, and Cognition, 14, 534–552. Payne, J. W., Bettman, J. R., & Johnson, E. J. (1993). The adaptive decision maker. New York: Cambridge University Press. Pham, M. T., & Higgins, T. E. (2005). Promotion and prevention in consumer decision making: State of the art and theoretical propositions. In S. Ratneshwar, & D. G. Mick (Eds.). Inside consumption: Frontiers of research on consumer motives, goals, and desires (pp. 8–43). London: Routledge. Polman, E. (2010). Information distortion in self-other decision making. Journal of Experimental Social Psychology, 46(2), 432–435. Polman, E. (2012b). Self–other decision making and loss aversion. Organizational Behavior and Human Decision Processes, 119(2), 141–150. Polman, E. (2012a). Effects of self–other decision making on regulatory focus and choice overload. Journal of Personality and Social Psychology, 102(5), 980–993. Polman, E., Effron, D. A., & Thomas, M. R. (2018). Other People’s Money: Money’s

75