Doing Worse but Feeling Better: Consequences of Collective Choice Nuno Jose Lopes, University of Navarra Elena Reutskaja, IESE Business School

ASSOCIATION FOR CONSUMER RESEARCH

Association for Consumer Research, University of Minnesota Duluth, 115 Chester Park, 31 West College Street Duluth, MN 55812

Doing Worse But Feeling Better: Consequences of Collective Choice Nuno Jose Lopes, University of Navarra Elena Reutskaja, IESE Business School

Two studies demonstrate that the decision process of a group of two people is essentially cooperative. This leads dyad members to sacrifice their favorite alternatives, making them choose an alternative with an inferior value comparatively to individuals. Additionally, and contrary to individuals, dyads’ decision process is dominated by positive emotion.

[to cite]: Nuno Jose Lopes and Elena Reutskaja (2018) ,"Doing Worse But Feeling Better: Consequences of Collective Choice", in NA - Advances in Consumer Research Volume 46, eds. Andrew Gershoff, Robert Kozinets, and Tiffany White, Duluth, MN : Association for Consumer Research, Pages: 690-391.

[url]: http://www.acrwebsite.org/volumes/2411731/volumes/v46/NA-46

[copyright notice]: This work is copyrighted by The Association for Consumer Research. For permission to copy or use this work in whole or in part, please contact the Copyright Clearance Center at http://www.copyright.com/. Doing Worse but Feeling Better: Consequences of Collective Choice Nuno Jose Lopes, University of Navarra, Spain Elena Reutskaja, IESE Business School, Spain

EXTENDED ABSTRACT taxing decision such as choosing from a large choice set. It had the participation of 42 individuals (36% women) and 44 dyads (dyad Conceptualization members knew each other; 51% women), between-subjects. Movies Consumer decisions are often made together with other people was the only category. Procedure: each participant alone started by (Davis 1976). Nonetheless, marketing scholars typically investigate evaluating every alternative on the same scale as study 1. Then, to set consumers as independent individuals who chose alone, leaving joint participants’ emotions to neutrality they watched a 4-minutes docu- decisions under-researched (Bagozzi 2012). However, choosing with mentary. Afterwards, from a choice set with 30 movies participants others differs substantially from choosing alone. This article com- had to select their favorite to watch briefly in the end. This was done pares the implications on decision outcome and emotions felt when either individually or in dyads. Then, a short questionnaire was filled. either an individual or a group of two people chose from a set with The selection and questionnaire tasks were repeated six times. Final- many alternatives. ly, one of the movies selected as favorite was watched for 5-minutes. Group decision is not the averaging of its members (Zajonc Wo- Along study 2, the emotions of each participant were recorded losin, and Wolosin 1972). When several people choose together, they every 35 milliseconds with an automated facial recognition software. frequently have different tastes and goals, which can lead them to This allowed1 to know the intensity (0 = minimum to 100 = maxi- disagreement. Despite this latent conflict, groups are typically still mum) of each of the six basic emotions: enjoyment, sadness, anger, able to reach a decision agreement. This is possible because collec- disgust, fear, and surprise (Ekman 1992). This information was then tive consumer decision is fundamentally cooperative in nature (Fish- aggregated in two-seconds’ intervals and the emotion with the higher, Grégoire, and Murray 2011), which implies that conflicting pref- est average intensity along the interval was considered the dominant erences are overcome through self-sacrifice and mutual concessions one. Then, the dominant valence was coded as positive if the stron- (Corfman and Lehaman 1987). Therefore, knowing the consequenc- gest emotion was enjoyment, or negative if it was anger, sadness, es of cooperation and self-sacrifice on the value of the alternative disgust, or fear. Surprise was out of analysis as the literature still selected constitutes one of the contributions of the current article. debates its valence. Iyengar, Wells, and Schwartz (2006) demonstrated that individuals who do better in taxing decisions can actually feel worse. Major Findings However, it is unclear whether these effects are also extended to col- Study 1 showed that dyads selected an item with an inferior lective choice. On one hand, research on individual decision-making value in comparison to individuals (Mdyad = 3.51, SDdyad = 1.26; Mind. 2 suggests that engaging in a difficult selection can lead to the experi- = 4.50, SDind. = 0.84; χ (1) = 63.85, p < .0001). This difference was ence of negative emotions (Bettman, Luce, and Payne 1998). Given the consequence of dyad members giving up her top-rated alterna- that group decision might imply contradictory preferences which can tive, which implied an average sacrifice of 1.10 value points (SDdyad degenerate into affective conflict (Amason 1996), groups might ex- = 1.01). Nonetheless, dyads were not less satisfied with the option perience even stronger negative emotions than individuals. Never- selected comparatively to individuals (Mdyad = 7.78, SDdyad = 1.76; 2 theless, groups’ cooperative mindset and the fact that human beings Mind. = 7.97, SDind. = 1.80; χ (1) = 0.27, p = .602). Moreover, 69% of have a tendency to enjoy sharing an activity with others (Baumeis- the dyads behaved very cooperatively as in at least 2/3 of their selec- ter and Leary 1995; Raghunathan and Corfman 2006) could elicit tions (a) both dyad members conceded the same or (b) they engaged positive emotions in dyadic decision. Which of these drivers has a in sequential reciprocity (“You win now, I win next”). stronger influence on dyadic emotions experienced during choice is Besides replicating study 1 findings, study 2 revealed that still unknown in the consumer literature and is the other goal of the whereas individuals’ selection process was dominated by negative current article. emotions (73% of the two-seconds intervals), dyadic selection, on the contrary, was dominated by positive emotion (45% of the inter- Method vals). Proportion tests confirmed the differences between individuals Study 1 aimed to compare the subjective value of the alternative and dyads both for the negative (z = 44.85, p < .0001) and the posi- selected by individuals and dyads, and assess the magnitude of sac- tive (z = -34.22, p < .0001) dominant emotions. Moreover, study 2 rifice in collective decisions. Participants were randomly assigned to showed that while choosing, dyad members experienced a level of the individual or collective condition, between-subjects. They were enjoyment almost five times stronger than that of individuals (Mdyad 2 60 individuals (63% women) and 41 dyads (61% women; dyad mem- members = 26.13, SDdyad members = 15.44; Mind. = 5.67, SDind. = 8.34; χ (1) = bers did not know each other). Every choice set had 30 alternatives of 46.61, p < .0001). Level of dyad familiarity had no effect. the same category (poster, chocolate, ice cream, and desk lamp; ran- This article demonstrated that although dyads do worse in terms dom assignation and order). Procedure: first, each participant alone of selection outcome, they do not report a lower level of satisfaction. had to evaluate every alternative in the choice set (-5 = “I don’t like Furthermore, collective consumers feel better as they enjoy the selec- it at all” to +5 = I like it very much”). Afterwards, participants ticked tion process much more. which alternatives they would consider from the choice set, followed by the selection of the favorite alternative. Finally, a short question- REFERENCES naire was filled. This procedure was done three times. In the collec- Amason, Allen C. (1996), “Distinguishing the Effects of Functional tive condition, both the consideration and the selection stages were and Dysfunctional Conflict on Strategic Decision Making: done by dyad members together. Resolving a Paradox for Top Management Teams,” Academy The goal of Study 2 was to compare the valence of the domi- of Management Journal, 39(1), 123-148. nant emotion experienced by individuals and dyads while making a

Bagozzi, Richard P. (2012), “Alternative Approaches for Thinking Fisher, Robert J., Yany Grégoire, and Kyle B. Murray (2011), About and Modeling Consumer Decisions in Relationships,” “The Limited Effects of Power on Satisfaction with Joint Journal of Consumer Psychology, 22(3), 315-319. Consumption Decisions,” Journal of Consumer Psychology, Baumeister, Roy F. and Mark R. Leary (1995), “The Need 21(3), 277-89. to Belong: Desire for Interpersonal Attachments as a Iyengar, Sheena S., Rachael E. Wells, and Barry Schwartz (2006), Fundamental Human Motivation,” Psychological Bulletin, “Doing Better but Feeling Worse. Looking for the “Best” Job 117(3), 497-529. Undermines Satisfaction,” Psychological Science, 17(2), 143- Bettman, James R., Mary Frances Luce, and John W. Payne (1998), 50. “Constructive Consumer Choice Processes,” Journal of Raghunathan, Rajagopal and Kim Corfman (2006), “Is Happiness Consumer Research, 25(3), 187-217. Shared Doubled and Sadness Shared Halved? Social Influence Corfman, Kim P. and Donald R. Lehmann (1987), “Models of on Enjoyment of Hedonic Experiences,” Journal of Marketing Cooperative Group Decision-Making and Relative Influence: Research, 43(3), 386-394. An Experimental Investigation of Family Purchase Secisions,” Zajonc, Robert B., Robert J. Wolosin, and Myrna A. Wolosin Journal of Consumer Research, 14(1), 1-13. (1972), “Group Risk-Taking Under Various Group Decision Davis, Harry L. (1976), “Decision Making Within the Household,” Schemes,” Journal of Experimental Social Psychology, 8(1), Journal of Consumer Research, 2(4), 241-260. 16-30. Ekman, Paul (1992), “An Argument for Basic Emotions,” Cognition & Emotion, 6(3-4), 169-200. Brand Movement Gregorio Fuschillo, Kedge Business School, France Jon Bertilsson, Lund University, Sweden Andrea Lucarelli, Lund University, Sweden

EXTENDED ABSTRACT Beppe Grillo is a textual brand as he is an Italian celebrity – a For three decades, consumer research has paid a great deal of humorist – with his community of fans gathered around his blog – attention to the role of brands in fostering social formations such beppegrillo.it. Founded officially in 2009, with the help of web strat- as subcultures, communities, and publics. Some of these formations egist Gianroberto Casaleggio, around Beppe Grillo and his followers’ manipulate elements of consumer culture for the sake of ideologi- proposals to adopt social networks (i.e. Meetup) to communicate and cal and cultural changes. In this regard, consumers organize them- coordinate local meetings, Five Star Movement is today the biggest selves in movements, namely anti-brand movements, which overtly political party in Italy (election 2018), with mayors of the party in criticize brands. Following these streams of research, this article large cities such as Rome and Turin. We collected data through the introduces the notion of brand movement to conceptualize current netnographic method (Kozinets, 2002) by analyzing online dis- phenomena whereby consumers-citizens turn to brands to produce a courses - TV shows, journal articles, press reviews, social networks, political change in society. In turn, brands have the power to chan- and consumer generated content. The amount of data totaled 148 nel, mobilize, and empower these consumers-citizens to gain more hours of video and about 200 pages of visual and textual data. political legitimation and success in social territory. We define brand Based on Melucci’s theory of new social movements, we iden- movement as a brand-related platform that propels the evolution and tify four main markers that constitute the empirical grounding for dissemination of political views, ideas, and actions. brand movement. These are “challenging administrative logic on Consumer-related social formations are a key phenomenon of symbolic grounds”, “acting political, changing the social”, “func- current consumer societies (Jenkins, 2006; Kozinets, and Handel- tioning as daily socio-technical networking” and “consciousness of man, 2004). Studies in consumer culture show the way in which con- living in a global (consumer) society”. sumption nourishes subcultures, communities, tribes, collaborative The conceptualization of brand movement offers an understand- networks, collectives, movements, activism, and publics of consumers ing of empirical phenomena under analysis in which brands allow (Arvidsson and Caliandro, 2016; Cova et al., 2007; Kozinets, 2001; citizens-consumer to participate in political processes; this is in order Scaraboto and Fischer, 2013; Schouten and McAlexander, 1995; to gain more space and power in the society, instead of contrasting/ Thompson et a., 2006; Weijo et al., 2018). Brands are at the core of resisting market logic or using/engaging with the market in order to these sociocultural phenomena (Luedicke et al., 2010; Muniz and gain socio-political power. In doing so, the article sheds new light on O’Guinn, 2001; Schau et al., 2009). More specifically, the strong the entanglement between brands and consumer activism (Kozinets connection of consumers to their brands leads to the rise of social 2002; Kozinets & Handelaman 2004). In fact, rather than emphasiz- groups of devotees turning the brand into a cult object (Belk and ing the role of consumer activism in resisting the market and/or its Tumbat, 2005; Muniz and Schau, 2005). However, consumers can global brands (Kozinets & Handelman 2004; Thompson and Arsel, also unite against a brand when they despise it (Izberk-Bilgin, 2012; 2004; Thompson et al., 2006; Weijo et al., 2018), this study illustrates Kozinets, and Handelman, 2004). While most of the research in con- that brand-related activism may be used to attain real, substantial sumer culture has focused especially on those brand-related social political and democratic ends. All in all, via brand movement, rather formations emerging in opposition to the society as a whole, as an than providing consumers with support for social belonging (Muñiz & alternative to it, or both (Kozinets, 2001; Muniz and O’Guinn, 2001; O’Guinn 2001; Schouten & McAlexander 1995), or a means of self- Schouten and McAlexander, 1995), we still know little about those promotion (Arvidsson & Caliandro 2016), brands may offer consum- brand-related social movements that gain momentum in their interac- ers the power to achieve actual political goals. This is because brand tion with the current society. movements are able to connect with, tap into, soak up and integrate In their pioneering study on brand-related social movements, values and ideologies from different worlds of worth into a novel- Kozinets and Handelman (2004) provide evidence of alternative pragmatic configuration of political ideas in order to challenge the social formations in consumer societies, specifically that they more existing political landscape. politically active than communal formations as tribes and publics (Arvidsson and Caliandro, 2016; Cova et al., 2007). However, their REFERENCES analysis focuses especially on anti-brand movements, telling little Arvidsson, A., & Caliandro, A. (2015). Brand public. Journal of about those emerging around consumers’ strong connection brands. Consumer Research, 42(5), 727-748. As noted in social movement theory, social movements arise espe- Belk, R., & Tumbat, G. (2005). The cult of cially from shared common beliefs, but also from non-traditional po- Macintosh. Consumption markets & culture, 8(3), 205-217. litical actions aimed at redressing the power gap between underdog Cova, B., Kozinets, R. V., & Shankar, A. (2007). Tribes, Inc.: the individuals and powerful elites (Melucci, 1989). In cultural studies, new world of tribalism. Consumer tribes, 3-26. these beliefs mostly manifest themselves in the form of a textual Izberk-Bilgin, E. (2012). Infidel brands: Unveiling alternative brand – e.g. celebrities – that consumers use to resist dominant ide- meanings of global brands at the nexus of globalization, ologies, develop a sense of belonging among them, and gain social consumer culture, and Islamism. Journal of Consumer territory (Fiske, 1989; Jenkins, 1992; Kozinets, 2001). Drawing from Research, 39(4), 663-687. this broader theoretical framework, this study aims at investigating Fiske, J. (1989). Understanding Popular Culture. New York: the role that textual brands play in political social movements in to- Routledge. day’s consumer societies. Jenkins, H. (1992). Textual Peachers. Nueva York: Routledge. The present study explores Beppe Grillo’s movement, which Jenkins, H. (2006). Fans, bloggers, and Gamers. New York, NY: became later known by its current name of the Five Star Movement. New York University Press.

Kozinets, R. V. (2001). Utopian enterprise: Articulating the Schouten, J. W., & McAlexander, J. H. (1995). Subcultures of meanings of Star Trek’s culture of consumption. Journal of consumption: An ethnography of the new bikers. Journal of consumer research, 28(1), 67-88. Consumer research, 22(1), 43-61. Kozinets, R. V. (2002). The field behind the screen: Using Schau, H. J., Muñiz Jr, A. M., & Arnould, E. J. (2009). How brand netnography for marketing research in online communities. community practices create value. Journal of marketing, Journal of marketing research, 39(1), 61-72. 73(5), 30-51. Kozinets, R. V., & Handelman, J. M. (2004). Adversaries of Thompson, C. J., & Arsel, Z. (2004). The Starbucks Brandscape consumption: Consumer movements, activism, and ideology. and Consumes’ (Anticorporate) Experiences of Glocalization. Journal of consumer research, 31(3), 691-704. Journal of Consumer Research, 31(3), 631-642. Luedicke, M. K., Thompson, C. J., & Giesler, M. (2009). Consumer Thompson, C. J., Rindfleisch, A., & Arsel, Z. (2006). Emotional identity work as moral protagonism: How myth and ideology branding and the strategic value of the doppelgänger brand animate a brand-mediated moral conflict.Journal of Consumer image. Journal of marketing, 70(1), 50-64. Research, 36(6), 1016-1032. Weijo, H. A., Martin, D. M., Arnould, E. J., & Ger, G. (2018). Melucci, A. (1982). L’invenzione del presente: movimenti, identità, Consumer Movements and Collective Creativity: The Case of bisogni individuali (Vol. 146). Bologna: Il Mulino. Restaurant Day. Journal of Consumer Research Muniz, A. M., & O’Guinn, T. C. (2001). Brand community. Journal of Consumer research, 27(4), 412-432. Scaraboto, D., & Fischer, E. (2012). Frustrated fatshionistas: An institutional theory perspective on consumer quests for greater choice in mainstream markets. Journal of Consumer Research, 39(6), 1234-1257. Cultivating a Network of Trust: Exploring the Trust Building Agency of Objects in Home Sharing Marian Makkar, Auckland University of Technology, New Zealand Drew Frankin, Auckland University of Technology, New Zealand

EXTENDED ABSTRACT tion between host and guest, but are also engendered in the materials With trust so central to the function and success of home shar- of the home such as mattress and duvet quality: ing, this research proposes that trust is best established through in- They can see our reviews. They can see what the people in terplay of trust entities. However, apart from literature detailing trust Bordeaux and Saint Sebastien said about us as guests whether we’re in platform security (Lu, Wang, and Hayes, 2012), in the seller (Ert, good, clean. They can say no to us if they didn’t want to have us. Fleischer, and Magen, 2016), or in reviews (Filieri, 2016), little is Without positive reviews I don’t think they can trust us. (Karen, understood about how trust is built in home sharing. In order to of- guest). fer a more refined understanding of trust, we explore the sources of Communication and transparency are important dimensions rational, affective and contextual trust and suggest the value in trac- of trust building in services and are enacted by agents, or boundary ing objects as a key property of trust building in home sharing. We spanners, during the service encounter (Anderson and Narus, 1990). review this emergence through the lens of consumer culture perspec- Yet in home sharing, these activities are manifest even before an tives (Arnould and Thompson, 2005) embedded in organizational agreement to stay, or encounter, takes place. trust literature (Korsgaard, Brower, and Lester, 2015) to assist in To co-create is to jointly-configure a customer solution and is providing a vocabulary that can serve as an organizing framework. more than simply transactional in nature. This activity results in a To understand how trust is built in home sharing networks (Airbnb), more customized solution and, by extension, trusting disposition be- this study adopts interpretivist research practices. This approach en- tween parties. Co-created value in home sharing is realized through ables an in-depth understanding of participants’ Airbnb experiences homely practices such as having dinner with the host: (shared with hosts or full unit rentals) specifically relating to trust in The key to her home rather than a hotel was important. You go engaging in the network. This study is based on a three-year multi- in and you use the key to open YOUR apartment and this is where sited ethnography of Airbnb consumption experiences that includes YOU’RE living. You go in and you feel like you can do the grocer- netnographic data (Kozinets, 2010) of Airbnb online communities in ies.” (Karen, guest). New Zealand, interviews with guests and hosts and field observation. Shared values also contribute to trust building in home sharing. Previous research suggests that attributions of trust are influ- By unraveling the Airbnb network, it became apparent that shared enced by rational, affective and contextual factors (Korsgaard et al., values are realized through familiarity with the sharing economy 2015). Johnson and Grayson (2005) propose rational trust as the con- world. Shared values may even be encoded in the home appliances fidence or willingness to rely on a service producer who is competent or design that creates closeness with the other: and reliable. Rational trust has also been posed to include evaluations I don’t need to meet the host but like to feel welcome and really of satisfaction (Baxter, 2012; Ganesan, 1994), communication (Do- it’s the little extras that count…I love quirky decor and don’t want to ney, Barry, and Abratt, 2007), integrity (Mayer, Davis, and Schoo- stay in someplace that lacks a soul (guest, online). rman, 1995), value co-creation (Franklin and Marshall, 2016) and Integrity and benevolence are other dimensions of trust that are transparency (Eggert and Helm, 2003). Affective trust represents the delivered through practices of unspoken rules of hospitality, respect non-calculative, emotional, or empathetic, element of trust within a of home and house rules and the needs of the other. Homely material relationship. This kind of relationship has been described as the mood objects, such as family photos on the wall, can also signify decency, felt by the parties and the strong mutual feeling, or shared values, sincerity and the familial care and protection by the other. Benev- upon which their relationship is based (Sekhon, Roy, Shergill, and olence, or acts of service outside of profit motive, are discovered Pritchard, 2013). This functions as a projection of the image that the through the act of gift-giving and the meanings of the gifts (Sherry other party will exercise benevolence toward the other if presented Jr, 1983) that adhere to hospitality culture: with an opportunity to do so (Sirdeshmukh, Singh, and Sabol, 2002). “I’ve got two necklaces from Chinese guests and one gave me Lewicki and Bunker (1996) describe contextual trust as manifesting a little book and one gave us a little sculpture, like a little carving, so within relationships that are embedded in a broader context, such as just really little things but it’s a nice gesture.” (Angie, host). organizational membership or shared community, as the actors share Within an increasingly networked and fluid society, where ac- certain characteristics, thus signaling trust. tors fluctuate between physical to digital platforms and devices, this This study predicts that building trust in home sharing may in- research demonstrates that objects feature as a key agent of trust volve processes of pre-objectification of material objects (the home) building in home sharing. through interactions with the digital platforms (home photos, reviews and profile photos) (Ferreira and Scaraboto, 2016), thus enhancing REFERENCES other sources of trust. Anderson, James C. and James A. Narus (1990), “A Model By considering the agency of objects in building trust in home of Distributor Firm and Manufacturer Firm Working sharing (Latour, 2004), we offer a broader view of seven well-es- Partnerships,” Journal of Marketing, 42-58. tablished dimensions of dyadic trust in the existing trust literature. Arnould, Eric and Craig J. Thompson (2005), “Consumer Culture These dimensions include competence, communication, transpar- Theory (CCT): Twenty Years of Research,” Journal of ency, value co-creation, shared values, integrity and benevolence. Consumer Research, 31(4), 868-882. Airbnb guests initially consider attributions of competence in Baxter, Roger (2012), “How Can Business Buyers Attract Sellers’ their Airbnb hosts through their digital profiles prior to making con- Resources?: Empirical Evidence For Preferred Customer tact. These include star ratings, reviews and badges such as Super- Treatment from Suppliers,” Industrial Marketing Management, host titles. Attributions of competence continue during communica- 41(8), 1249-1258.

Doney, Patricia M., James M. Barry, and Russell Abratt (2007), Kozinets, Robert V. (2010), “Netnography: Doing Ethnographic “Trust Determinants and Outcomes in Global B2B Services,” Research Online,” London: Sage Publications Ltd. European Journal of Marketing, 41(9/10), 1096-1116. Latour, Bruno (2004), “Nonhumans,” in Patterned Ground: Eggert, Andreas and Sabrina Helm (2003), “Exploring the Impact Entanglements of Nature and Culture, ed. Harrison S, Pile S of Relationship Transparency on Business Relationships: and Thrift N, London: Reakiton Books, 224–227. A Cross-Sectional Study Among Purchasing Managers in Lewicki, Roy J. and Barbara B. Bunker (1996), “Developing Germany,” Industrial Marketing Management, 32(2), 101-108. and Maintaining Trust in Work Relationships,” in Trust in Ert, Eyal, Aliza Fleischer, and Nathan Magen (2016), “Trust and Organizations: Frontiers of Theory and Research B2 - Trust in Reputation in the Sharing Economy: The Role Of Personal Organizations: Frontiers of Theory and Research. Thousand Photos in Airbnb,” Tourism Management, 55, 62-73. Oaks, CA: Sage. Ferreira, Marcia C. and Daiane Scaraboto (2016), ““My Plastic Lu, June, Luzhuang Wang, and Linda A. Hayes (2012), “How Dreams”: Towards an Extended Understanding of Materiality do Technology Readiness, Platform Functionality and Trust and the Shaping of Consumer Identities,” Journal of Business Influence C2C User Satisfaction?”Journal of Electronic Research 69(1):191–207. Commerce, 13(1), 50–69. Filieri, Raffaele (2016), “What Makes an Online Consumer Review Mayer, Roger C., James H. Davis, and F. David Schoorman (1995), Trustworthy?” Annals of Tourism Research, 58, 46-64. “An Integrative Model of Organizational Trust,” Academy of Franklin, Drew and Roger Marshall (2016), “Building Trust in Management Review, 20(3), 709-734. Business-to-Business Relationships: The Enabling Role of Co- Sekhon, Harjit, Sanjit Roy, Gurvinder Shergill, and Adrian creation and Integrity,” paper presented at Australia and New Pritchard (2013), “Modelling Trust in Service Relationships: Zealand Marketing Academy Conference. A Transnational Perspective,” Journal of Services Marketing, Ganesan, Shankar (1994), “Determinants of Long-Term Orientation 27(1), 76-86. in Buyer-Seller Relationships,” Journal of Marketing, 58(2), 1. Sherry Jr, John F. (1983), “Gift Giving in Anthropological Johnson, Devon and Kent Grayson (2005), “Cognitive and Perspective,” Journal of Consumer Research, 10(2), 157-168. Affective Trust in Service Relationships,”Journal of Business Sirdeshmukh, Deepak, Jagdip Singh, and Barry Sabol (2002), Research, 58(4), 500-507. “Consumer Trust, Value, and Loyalty in Relational Korsgaard, M. Audrey, Holly H. Brower, and Scott W. Lester Exchanges,” Journal of Marketing, 66(1), 15-37. (2015). “It isn’t Always Mutual: A Critical Review of Dyadic Trust,” Journal of Management, 41(1), 47-70. “Yes, I Can” or “No, I Can’t” – Effect of Extraneous Affirmation- and Negation-Evoking Contexts on Brand Recall Memory: The Role of Semantic Activations Sudipta Mandal, Indian Institute of Management, Ahmedabad, India Arvind Sahay, Indian Institute of Management, Ahmedabad, India Sanjeev Tripathi, Indian Institute of Management, Indore, India

EXTENDED ABSTRACT jective feelings (Adelmann and Zajonc 1989; Semin and Smith In the real world, consumers often draw up a shopping list by re- 2002). Thus, calling brand names. Lynch Jr and Srull (1982) classified such choice options as “memory-based”. If consumers cannot recall a brand, they Hypothesis 3 Mere association of any brand-related informa- are less likely to consider it, and therefore, purchase it. Consequently, tion, with an extraneous horizontal shaking of factors that affect memory-based choices are important practically and the head (embodied “No”) should impair the theoretically. Our research focuses on how (i.e., the phenomenon) and recall memory of that brand as compared to an why (i.e., the psychological mechanism) the recall of brands could be extraneous vertical nodding of the head (embod- impacted by extraneous brand placement contexts that evoke a nega- ied “Yes”). tion versus an affirmation (such as a “NO” versus a “YES”). Fundamentally, it is possible that the experience from a source Our research is the first to show that an entirely unrelated ne- concept increases the accessibility in memory of specific semantic gation (vs. affirmation) in a brand’s placement context leads to im- concepts that such experiences exemplify. These concepts may then paired brand recall memory. The principal contribution of our work spread over to others, based on associative network models (“spread- is the investigation of the underlying mechanism. We demonstrate ing semantic activation”, Collins and Loftus 1975). Consequently, that negations spontaneously generate negation-related semantic as- we hypothesize that, negation, results in spreading of concepts se- sociations, which through a spreading activation mechanism (Collins mantically related to negation. Such concepts trigger exclusion of and Loftus 1975) influences brand recall memory. Specifically, we the negated and its associated entities from an individual’s immedi- show that the nature of these negation-related semantically activated ate purview of consideration. Spreading semantic activation coupled concepts, decreases consumers’ “general judgments of importance” with the aforementioned act of exclusion, reduces the importance of toward discrete consumer issues, which in turn impairs brand recall not only the negated entity and anything associated with it, but also memory. induces generally lower judgments of importance in an individual’s Extant research has shown that negation is associated with cog- mind. Being cognitive misers, humans disregard the unimportant, nitive inhibition, and latencies (Carpenter and Just 1975; Giora et and forget under negation. Thus, al. 2007; Kaup and Zwaan 2003). A negation operator prevents the activation of the core concept or schema being negated (MacDonald Hypothesis 4 The accessibility of negation-related semantic and Just 1989). concepts should mediate the effect of negation Research has also shown the importance of context factors in on the judgment of importance. brand placement effects (Newell, Salmon, and Chang 2006; Van Reijmersdal, Smit, and Neijens 2010). For example, there is an evidence of a spill-over effect induced by the context towards conven- Hypothesis 5 Higher judgments of importance lead to higher tional advertising, as reflected through consumers’ emotions (e.g., brand recall memory compared to lower judg- involvement, liking, feelings) (Pavelchak, Antil, and Munch 1988; ments of importance. Moorman, Neijens, and Smit 2005). We hypothesize that concepts that are coexistent with an ex- We conducted five laboratory experiments to validate our hy- traneous negation are susceptible to spill-over inhibitory effects of potheses, all of which were supported. Experiment 1 shows that an negation. Thus, extraneous negation (vis-à-vis an extraneous affirmation) elicited by a brand’s placement context impairs subsequent brand recall memory Hypothesis 1 Mere association of any brand-related informa- (H1). Experiment 2 shows that consumers’ motivation to remember tion, with an extraneous context or episode that moderates the effect observed in H1 (H2). Experiment 3 provides elicits a negated response, impairs the recall of convergent evidence from an embodied cognition perspective (H3). that brand as compared to an extraneous context Experiment 4 shows that accessibility of negation-related semantic or episode that elicits an affirmative response. concepts mediate the effects of explicit negation on the judgment of importance (H4). Finally, Experiment 5 shows that higher judgments Often “motivation” determines how people react to their envi- of importance lead to higher brand recall compared to lower judg- ronment when processing information, and formulating attitudinal/ ments of importance (H5). behavioural responses (Darke, Chattopadhyay, and Ashworth 2006), The principal theoretical contribution of our work is in inves- we hypothesize that when consumers are motivated to remember, the tigating the underlying mechanism leading to extraneous negation- effect in H1 should disappear. Thus, induced forgetting. Moreover, it is possibly the first research that demonstrates, both semantically and through embodiment, that an Hypothesis 2 When consumers are motivated to remember, the extraneous affirmation/negation can impact memory. A fundamental brand recall should be almost similar under both practical implication for brand managers is to decide what type of extraneous negation and affirmation contexts. programs/contexts they would like to use to advertise/promote their brands. Managers need to appreciate that negation-inducing contexts Furthermore, evidence of brain mechanisms linking language could potentially adversely impact the brand recall of their brands and action, is indicative of “embodied cognition” (Niedenthal 2007). and consequently their revenues. One facet of embodiment is the fact that motor actions impact sub- Advances in Consumer Research 696 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 697 REFERENCES MacDonald, Maryellen C., and Marcel A. Just. 1989. “Changes in Adelmann, Pamela K., and Robert B. Zajonc. 1989. “Facial Activation Levels with Negation.” Journal of Experimental Efference and the Experience of Emotion.”Annual Review of Psychology: Learning, Memory, and Cognition 15 (4): 633. Psychology 40 (1): 249–280. Moorman, Marjolein, Peter C. Neijens, and Edith G. Smit. 2005. Carpenter, Patricia A., and Marcel A. Just. 1975. “Sentence “The Effects of Program Responses on the Processing of Comprehension: A Psycholinguistic Processing Model of Commercials Placed at Various Positions in the Program and Verification.”Psychological Review 82 (1): 45. the Block.” Journal of Advertising Research 45 (1): 49–59. Collins, Allan M., and Elizabeth F. Loftus. 1975. “A Spreading- Newell, Jay, Charles T. Salmon, and Susan Chang. 2006. Activation Theory of Semantic Processing.” Psychological “The Hidden History of Product Placement.” Journal of Review 82 (6): 407. Broadcasting & Electronic Media 50 (4): 575–594. Darke, Peter R., Amitava Chattopadhyay, and Laurence Ashworth. Niedenthal, Paula M. 2007. “Embodying Emotion.” Science 316 2006. “The Importance and Functional Significance of (5827): 1002–1005. Affective Cues in Consumer Choice.”Journal of Consumer Pavelchak, Mark A., John H. Antil, and James M. Munch. 1988. Research 33 (3): 322–328. “The Super Bowl: An Investigation into the Relationship Giora, Rachel, Ofer Fein, Keren Aschkenazi, and Inbar Alkabets- among Program Context, Emotional Experience, and Ad Zlozover. 2007. “Negation in Context: A Functional Approach Recall.” Journal of Consumer Research 15 (3): 360–367. to Suppression.” Discourse Processes 43 (2): 153–172. Semin, Gün R., and Eliot R. Smith. 2002. “Interfaces of Social Kaup, Barbara, and Rolf A. Zwaan. 2003. “Effects of Negation and Psychology with Situated and Embodied Cognition.” Situational Presence on the Accessibility of Text Information.” Cognitive Systems Research 3 (3): 385–396. Journal of Experimental Psychology: Learning, Memory, and Van Reijmersdal, Eva, Edith Smit, and Peter Neijens. 2010. Cognition 29 (3): 439. “How Media Factors Affect Audience Responses to Brand Lynch Jr, John G., and Thomas K. Srull. 1982. “Memory and Placement.” International Journal of Advertising 29 (2): Attentional Factors in Consumer Choice: Concepts and 279–301. Research Methods.” Journal of Consumer Research 9 (1): 18–37. Linguistic Antecedents of Anthropomorphism N. Alican Mecit, HEC Paris, France Tina M. Lowrey, HEC Paris, France L. J. Shrum, HEC Paris, France

EXTENDED ABSTRACT expected, participants in the treatment group who used pronouns like In consumer research, humanlike form has been regarded as a native French speaker anthropomorphized more than did the con- the prerequisite for anthropomorphism (Hur, Koo, and Hofmann, trol group who used all English third-person singular pronouns (he/ 2015; Kim, Peng Chen, and Zhang, 2016). Given that activation of she/it). knowledge structures about humans and nonhumans and their appli- In study 3, we tested whether native speakers of languages that cation to a nonhuman target determines the extent to which human- do not distinguish grammatically between humans and nonhumans like qualities are attributed to the target (Epley, Waytz, and Cacioppo implicitly consider linguistic cues such as the grammatical gender 2007), we argue that humanlike form is not essential for consumers as characteristics of the anthropomorphized entity. In French, each to anthropomorphize nonhuman agents, because even in the absence noun has a specific article that indicates the grammatical gender of of humanlike characteristics, people may continue to anthropomor- the noun: le (masculine) and la (feminine). To test our hypothesis, phize, as some languages make their speakers focus on the similari- we identified a nonhuman noun (cache), which can take either the ties, rather than the differences, between humans and nonhumans. feminine or the masculine article. We manipulated the name of a Recent empirical evidence also shows that whether the language of fashion brand in an ad as “le cache” or “la cache,” and measured a given country has gendered pronouns or not is a strong predic- brand attitudes of 242 French speakers. The expected gender x gen- tor of gender equality in that country (Prewitt-Freilino, Caswell, and dered pronoun was significant. Male participants in the masculine Laakso 2012). In English, not only are there different pronouns for grammatical gender condition evaluated the brand more favorably male (he) and female (she), but English also separates humans (he compared to male participants in the feminine grammatical gender and she) from non-humans (it). condition. Similarly, female participants in the feminine grammatical Although most languages distinguish grammatically between gender condition evaluated the brand more favourably compared to humans and nonhumans, some languages (e.g., French, Turkish), female participants in the masculine grammatical gender condition, do not make these distinctions. Given that language shapes thought although the difference was not significant. (Whorf 1952), we propose that (1) native speakers of these languages These results show that French speakers imbue nonhumans with are more likely to attribute human qualities to nonhuman agents (i.e., human characteristics even in the absence of anthropomorphizing vi- anthropomorphize) than those whose native language separates hu- sual cues. When reasoning about nonhumans, in this case the brand mans from nonhumans, and that (2) native speakers of languages “le/la cache,” native French speakers activate knowledge about hu- lacking differential treatment implicitly consider linguistic cues as mans and apply it to nonhumans, which results in an interplay that is characteristics of the anthropomorphized entity. similar to what one can observe between humans. To test our hypotheses, we first conducted a pilot study with Overall, we show that humanlike form is not essential for con- participants (n = 41) whose native language was either English or sumers to anthropomorphize nonhuman agents. Native speakers of Turkish. We administered the individual differences in anthropomor- languages that do not separate humans from nonhumans are more phism questionnaire (Waytz, Cacioppo, and Epley 2010) in English likely to anthropomorphize than those whose native language treats in both conditions. Consistent with our expectation, native Turkish humans and nonhumans differently. We further show that native speakers anthropomorphized more than did native English speakers. speakers of languages lacking differential treatment implicitly con- In study 1 (n = 225), to address concerns that answering ques- sider linguistic cues, such as the grammatical gender, as characteris- tions in a second language may affect responses (e.g., anxiety, (Ma- tics of the anthropomorphized entity and behave accordingly. cIntyre and Gardner 1994), we administered the same anthropomorphism scale to either native French-speaking participants or native REFERENCES English-speaking participants, and also manipulated whether native Epley, Nicholas, Adam Waytz, and John T. Cacioppo French-speaking participants answered in French or English. Data (2007), “On Seeing Human: A Three-Factor Theory of from 17 participants who failed an attention check were excluded Anthropomorphism,” Psychological Review, 114 (4), 864-886. from analyses. As expected, native French speakers scored higher on Hur, Julia D., Minjung Koo, and Wilhelm Hofmann (2015), the anthropomorphism scale than did English speakers independent “When Temptations Come Alive: How Anthropomorphism of the language of administration. French participants who answered Undermines Self-control,” Journal of Consumer Research, 42 the survey in English did not differ from French participants who (2), 340-358. answered the survey in French. Kim, Sara, Rocky Peng Chen, and Ke Zhang (2016), In study 2 we addressed issues of causality by manipulating the “Anthropomorphized Helpers Undermine Autonomy and pronoun categories of native English speakers on MTurk (n = 120) Enjoyment in Computer Games,” Journal of Consumer through a learning task (Roberson, et al. 2005). Data from 13 par- Research, 43 (2), 282-302. ticipants who failed to follow the instructions were excluded from MacIntyre, Peter D., and Robert C. Gardner (1994), “The Subtle analyses. In the learning task, participants rewrote 60 sentences by Effects of Language Anxiety on Cognitive Processing in the replacing the underlined noun with a pronoun. In the control group, Second Language,” Language Learning, 44 (2), 283-305. participants replaced the nouns with regular pronouns (he/she/it), but Prewitt-Freilino, Jennifer L., T. Andrew Caswell, and Emmi K. in the treatment we asked participants not to use “it.” Instead, like Laakso (2012), “The Gendering of Language: A Comparison French, we asked them to use either “he” or “she,” depending on of Gender Equality in Countries with Gendered, Natural the arbitrary gender indicated in parentheses next to the underlined Gender, and Genderless Languages,” Sex Roles, 66 (3), 268- noun. Participants then completed the anthropomorphism scale. As 281.

Roberson, Debi, Jules Davidoff, Ian R.L. Davies, and Laura R. Whorf, Benjamin Lee (1952), “Language, Mind, and Reality,” A Shapiro (2005), “Color categories: Evidence for the cultural review of general semantics 167-188. relativity hypothesis,” Cognitive Psychology, 50 (4), 378–411. Waytz, Adam, John Cacioppo, and Nicholas Epley (2010), “Who Sees Human? The Stability and Importance of Individual Differences in Anthropomorphism,”Perspectives on Psychological Science, 5 (3), 219-232. Preferences for Insight and Effort Differ across Domains and Audiences Gaetano “Nino” Miceli, University of Calabria, Italy Irene Scopelliti, City University of London, UK Maria Antonietta Raimondo, University of Calabria, Italy

EXTENDED ABSTRACT song. Participants evaluated the song more favorably and were will- A creative process is “the sequence of thoughts and actions that ing to pay more, when it was presented through insight-based (vs. lead to a novel, adaptive production” (Lubart 2001, 295). Narratives effort-based) narratives. In study 2B we measured intentions to fund of the creative process can enhance the perceived quality of a prod- a Kickstarter project based on scientific research that was presented uct and stimulate inferences on the creator’s ability to produce ad- through insight- vs. effort-based narratives of the creative process. ditional high-quality ideas and outputs (Baas et al. 2015; Mourgues When it was presented through effort-based narrative, the project et al. 2016), improve empathy between creators and their audiences was considered (marginally) more likely to reach the fund-raising (Davis 1983; Friestad and Wright 1994), and increase process trans- goal and participants allocated more funds to it than when it was parency (Buell and Norton 2011; Buell, Kim and Tsay 2017). These presented through an insight-based narrative. narratives can emphasize either the experience of insight (Schooler, Studies 3A and 3B replicated studies 2A and 2B but also tested Fallshore, and Fiore 1995) – the sudden, dream-like, spontaneous the proposed fit mechanism, as well as the role of expertise, in the and illuminating experience facilitating the emergence of new ideas– artistic and scientific domain, respectively. Participants in Study 3A or the application of effort (Lucas and Nordgren 2015) – the me- evaluated the song from Study 2A more favorably and were will- thodical, planned, and rational stage in which ideas are organized and ing to pay more for it when they read it was the outcome of insight transformed into a new product. than effort. The perceived fit between the creative process narrative People hold beliefs on how creative products are generated (e.g., and the image of the creator mediated both effects. In addition, an Sternberg 1985; Runco and Bahleda 1986; O’Connor, Nemeth and insight-based creative process produced more favorable evaluations Akutsu 2013), and these beliefs can influence product evaluations of an artistic product, but this effect was attenuated as recipients’ ex- due to reliance on stereotypical knowledge and heuristics (Stern- pertise increased. Participants in Study 3B evaluated the project from berg 1985; Levy, Stroessner, and Dweck 1998). We maintain that Study 2B as (marginally) more likely to reach the fund-raising goal an insight-based narrative of the creative process (Rothenberg 1970; and assigned more funds to it when it was described as effort-based Kasof 1995) has greater fit with artistic domains, whereas an effort- than as insight-based. Also in this case, the perceived fit between the based narrative of the creative process has greater fit with scientific type of creative process and the work mediated both effects. In line domains (Lucas and Nordgren 2015). Because lay beliefs influence with our prediction, expertise did not interact with the type of narra- evaluations of creative products (Baas et al. 2015; Sternberg 1985), tive of the creative process, as both experts and non-experts prefer ef- we expect artistic products to receive better evaluations when their fort-based narratives. However, the perceived fit between the nature creative process is described through insight-based narratives rather of the creative process and the image of the creator only mediated the than effort-based narratives and that the opposite pattern holds for effect when participants’ expertise was low to average. scientific products. In Study 4, participants read descriptions of a new product pre- We also expect experts to be more analytical in processing in- sented as either artistic or scientific, and as the outcome of either an formation and less sensitive to heuristics, to own to a higher extent insight- or an effort-based creative process. When the product was the meta-cognitive skills necessary to evaluate others’ level of abil- described as artistic, participants evaluated it more favorably and ity (Kruger and Dunning 1999), and to be more receptive towards were willing to pay more for it, if they read it was the outcome of effort-based narratives of the creative process (Ericsson, Krampe, insight rather than of effort. When the product was presented as sci- and Tesch-Romer 1993; Ericsson, Prietula, and Cokely 2007), thus entific product, the opposite pattern was observed. to respond relatively more favorably to information on central and This research clarifies that beliefs on the nature of the creative concrete – rather than ephemeral and transient – factors characteriz- process depend on the domain of the creative output, and that the ing the creative process. Six studies support the proposed conceptual most effective communication strategy regarding the creative pro- framework and the existence of optimal narratives for the promotion cess leading to a product depends critically on both the nature (sci- of artistic and scientific products to different audiences. entific or artistic) of the product, and the level of expertise of the Study 1 verified the existence of associations between anin- audience targeted. sight-based creative process and artistic domains, and between an effort-based creative process and scientific domains. Participants REFERENCES read the descriptions of twelve new successful artistic works and sci- Baas, Matthijs, Severine Koch, Bernard A. Nijstad Nijstad, and entific works and rated the extents to which they were the outcome of Carsten K. W. De Dreu, (2015). “Conceiving Creativity: The insight and effort. Each art work received higher insight ratings than Nature and Consequences of Laypeople’s Beliefs about the effort ratings, and each science work received higher effort ratings Realization of Creativity,” Psychology of Aesthetics, Creativity, than insight ratings. Furthermore, all art works received higher in- and the Arts, 9(3), 340–354. sight ratings than all science works, and all scientific works received Buell, Ryan W., and Michael I. Norton (2011). “The labor illusion: higher effort ratings than all art works. How operational transparency increases perceived value,” Studies 2A and 2B investigated whether describing a creative Management Science, 57(9), 1564–1579. process using insight- vs. effort-based narratives affects evaluations ______, Tami Kim, and Chia-Jung Tsay (2017). “Creating of new products in artistic and scientific domains, respectively. Par- reciprocal value through operational transparency,” ticipants in Study 2A read an interview in which a band described the Management Science, 63(6), 1673–1695. creative process leading to their last song either as insight-based or effort-based. Afterwards, they listened to a 45-second excerpt of the

Davis, Mark H. (1983). “Measuring individual differences in Lucas, Brian J., and Loran F. Nordgren (2015). “People empathy: Evidence for a multidimensional approach,” Journal underestimate the value of persistence for creative of Personality and Social Psychology, 44, 113–126. performance,” Journal of Personality and Social Psychology, Ericsson, K. Anders, Ralf T. Krampe, and Clemens Tesch-Römer 109(2), 232–243. (1993). “The role of deliberate practice in the acquisition of Mourgues, Catalina, Mei Tan, Sascha Hein, Julian G. Elliott, and expert performance,” Psychological Review, 100(3), 363–406. Elena L. Grigorenko (2016). “Using creativity to predict ______, Michael J. Prietula, and Edward T. Cokely (2007). future academic performance: An application of Aurora’s five “The making of an expert,” Harvard Business Review, 85(7/8), subtests for creativity,” Learning and Individual Differences, 114–121. 51, 378–386. Friestad, Marian, and Peter Wright (1994). “The persuasion O’Connor, Alexander J., Charlan J. Nemeth, and Satoshi Akutsu knowledge model: How people cope with persuasion (2013). “Consequences of beliefs about the malleability of attempts,” Journal of Consumer Research 21(1), 1–31. creativity,” Creativity Research Journal, 25(2), 155–162. Kasof, Joseph (1995). “Explaining creativity: The attributional Rothenberg, Albert (1970). Inspiration, insight and the creative perspective,” Creativity Research Journal, 8(4), 311–366. process in poetry. College English, 32(2), 172–183. Kruger, Justin, and David Dunning (1999). “Unskilled and unaware Runco, Mark A. and Michael D. Bahleda (1986). “Implicit Theories of it: how difficulties in recognizing one’s own incompetence of Artistic, Scientific, and Everyday Creativity,” The Journal lead to inflated self-assessments,”Journal of Personality and of Creative Behavior, 20(2), 93–98. Social Psychology, 77(6), 1121–1134. Schooler, Jonathan W., Marte Fallshore, and Stephen M. Fiore Levy, Sheri R., Steven J. Stroessner, and Carol S. Dweck (1998). (1995). “Epilogue: Putting insight into perspective,” In R.J. “Stereotype formation and endorsement: The role of implicit Sternberg and J. E. Davidson (ed.), The nature of insight. theories,” Journal of Personality and Social Psychology, Cambridge, MA: The MIT Press. 559–587. 74(6), 1421–1436. Sternberg, Robert J. (1985). “Implicit theories of Intelligence, Lubart, Todd I. (2001). “Models of the Creative Process: Past, Creativity, and Wisdom,” Journal of Personality and Social Present and Future,” Creativity Research Journal, 13(3–4), Psychology, 49(3), 607–627. 295–308. Believing in and Reacting to Rumors: The Role of Congruity and Nature of Existing Predisposition Satadruta Mookherjee, Binghamton University State University of New York, USA Subimal Chatterjee, Binghamton University State University of New York, USA

EXTENDED ABSTRACT about the restaurant/actor described the restaurant using healthy (fatty) ingredients in their food/the actor receiving a prestigious, humanitarian Introduction award (arrested for disorderly conduct). We measured how good (bad) The prevalence of rumors is a common phenomenon in today’s they thought the positive (negative) news to be (9-point scale captur- world and spreading rapidly as more and more consumers adapt to ing valence and magnitude of information). Finally, the participants the digital world of the internet and social media. Our research inves- in the rumor (fact) condition read that the news was a rumor, unveri- tigates conditions under which consumers treat rumors (unverified fied by major media outlets (fact, verified by the major news outlets). information) identical to facts (the same, but verified, information). Thereafter, we measured to what extent participants believe the news We investigate to what extent congruent/incongruent information (9-point scale), and how their attitude towards the focal restaurant/ac- (rumors/facts) about liked/disliked entities (both people and prod- tor (composite of affect, cognition, and behavior) has changed asa ucts) affect consumers’ belief in that information and the resulting result (9-point scale, with 5.0 or mid-point indicating no change). attitude change towards the focal entity. Analysis and Results Theory We created a congruity variable (congruent/incongruent infor- Research suggests that two things happen (almost automati- mation) where the congruent (incongruent) conditions matched posi- cally) when we expose a consumer to a piece of information (fact tive news with a liked (disliked) entity and negative information with or rumor) about a focal entity. First, she automatically compares a disliked (liked) entity. We ran two ANOVAs, the first for belief it to her baseline or prior knowledge (Higgins et al. 1977), and this and the second for attitude change. The predictors were information process determines the compatibility or congruity of the information exposure (fact/rumor), congruency (congruent/incongruent), and with what she knows and believes (e.g., to what extent the rumor predisposition towards the focal entity (like/dislike). We controlled fits with her pre-disposition towards a focal entity; McGuire, 1972). for (1) magnitude effects (for predisposition and information), (2) Second, she either accepts the information (fact or rumor) as true or the focal entity (person/product and (3) the participant’s gender, age, rejects it as false (Allport and Lepkin 1945), and is more likely to act tolerance for inconsistency, and gullibility. upon it only if she believes the information to be true (Rosnow, Yost, Our first ANOVA found that consumers believe (1) less in ru- and Esposito 1986). mor than fact (M’s of 5.44 and 6.35), and (2) less in incongruent than According to cognitive consistency theories, congruent infor- congruent information (M’s of 5.21 and 6.57). However, a signifi- mation about a focal entity elicits positive feelings (Festinger, 1957) cant information by congruency interaction shows that consumers and consumers will believe such information automatically (without discriminate between rumor and fact somewhat less for incongruent verification). This line of reasoning suggests that consumers will information (M’s of 4.89 and 5.52) than congruent information (M’s not distinguish between fact and rumor for congruent information. of 5.98 and 7.14). The results indicate that consumers do not accept However, and on the contrary, Borgstede et al. argue that consum- congruent information blindly (just because they fit with what they ers analyze congruent information about a focal entity systematically expect). They appear to carefully process the information to separate before acting on it. This is because consumers do not wish to get fact from rumor, and, in fact, do so more carefully relative to incon- their hopes up about the focal entity too much (e.g., following a posi- gruent information. tive rumor about a liked brand/person) only to see it dashed later on In our second ANOVA, once we controlled for belief, we found (rumor turns out to be false). attitude change towards the entity to be similar across rumor and fact Incongruent information about a focal entity, on the other hand, (M’s of 5.24 and 5.41), but more positive following exposure to con- creates surprise by evoking unexpected cognitive challenge (Alden gruent than incongruent information (M’s of 5.52 and 5.13). We also et al. 2000) and motivates the consumer to analyze and resolve the found a three-way interaction between the type of information, con- discrepancy (Schutzwohl and Borgstedt 2005). If the information is gruency, and predisposition. The results indicate that, in the case of just a rumor (e.g., negative rumors about a liked entity), then con- incongruent information, rumor/fact leads to identical attitude change sumers should have less reasons to be tense or worried, and the desire irrespective of how consumers feel about the entity (M’s of 5.24 and to lessen the tensions associated with incongruity should motivate 5.00 for liked entities and M’s of 4.94 and 5.35 for disliked entities). consumers to distinguish between rumor and fact and act accord- However, in the case of congruent information, attitude change for a ingly. liked entity is less positive following rumor than fact (M’s of 6.49 and Below, we report a study testing how exposure to congruent/ 7.25) but identical for a disliked entity (M’s of 4.03 and 3.98). incongruent information about liked/disliked entities may lead con- From a managerial perspective, the key lesson here is to identify sumers to treat rumors as equivalent to fact (e.g. believe the rumor when consumers treat a rumor as a fact (e.g., when are they more just as much as they believe the fact). susceptible to deceptions from the competition). The fact that attitude change following incongruent information about a liked entity Participants, Stimuli and Measures is identical for rumor and fact should alert managers to the neces- We asked 825 M-Turk workers to think about a fast food res- sity of inoculating their own consumers against negative competitive taurant/actor that they liked/disliked very much and indicate how messages. On the contrary, given that attitude change following con- much they liked/disliked the entity (9-point scale, capturing valence gruent information for a disliked entity is sensitive to the fact/rumor and magnitude of predisposition). Next, they read a positive/negative distinction suggests that the company has to be sure about the facts piece of news about the focal entity. The positive (negative) news before maligning a competitor’s brand to its own audience.

Advances in Consumer Research 702 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 703 REFERENCES Kanouse, D. E. “Hanson. LR (1971). Negativity in evaluations.” EE Alden, Dana L., Ashesh Mukherjee, and Wayne D. Hoyer. “The Jones et al., Attribution: Perceiving the causes of behavior: effects of incongruity, surprise and positive moderators 47-62. on perceived humor in television advertising.” Journal of Knapp, Robert H. “A psychology of rumor.” Public opinion Advertising 29.2 (2000): 1-15. quarterly 8.1 (1944): 22-37. Allport, Floyd H., and Milton Lepkin. “Wartime rumors of waste Lewandowsky, Stephan, et al. “Misinformation and its correction and special privilege: why some people believe them.” The continued influence and successful debiasing.” Psychological Journal of Abnormal and Social Psychology 40.1 (1945): 3. Science in the Public Interest 13.3 (2012): 106-131 Antoniadis, Sotirios, Iouliana Litou, and Vana Kalogeraki. “A McGuire, W. J. (1972). Attitude change: The information model for identifying misinformation in online social processing paradigm. In C. G. McClintock (Ed.), Experimental networks.” OTM Confederated International Conferences” On social psychology (pp. 108–141). New York, NY: Holt, the Move to Meaningful Internet Systems”. Springer, Cham, Rinehart, & Winston 2015. Rosnow, Ralph L., John H. Yost, and James L. Esposito. “Belief Feldman, Shel. “Motivational aspects of attitudinal elements and in rumor and likelihood of rumor transmission.” Language & their place in cognitive interaction.” Cognitive consistency. Communication 6.3 (1986): 189-194. 1966. 75-108. Schützwohl, Achim, and Kirsten Borgstedt. “The processing of Festinger, L. (1957). A theory of cognitive dissonance. Evanston, affectively valenced stimuli: The role of surprise.” Cognition IL: Row, Peterson. & Emotion 19.4 (2005): 583-600. Hayes, Andrew F. “Partial, conditional, and moderated von Borgstede, Chris, Maria Andersson, and André Hansla. moderated mediation: Quantification, inference, and “Value-congruent information processing: the role of issue interpretation.” Communication Monographs 85.1 (2018): involvement and argument strength.” Basic and Applied Social 4-40. Psychology 36.6 (2014): 461-477. Higgins, E. Tory, William S. Rholes, and Carl R. Jones. “Category accessibility and impression formation.” Journal of experimental social psychology 13.2 (1977): 141-154. Priming & Privacy: How Subtle Trust Cues Online Affect Consumer Disclosure and Purchase Intentions James A. Mourey, DePaul University, USA Ari Ezra Waldman, New York Law School, USA

EXTENDED ABSTRACT the studies that each rely on maintaining privacy in online contexts Consumers are quick to say that maintaining privacy in an in- as a difficult scenario. creasingly digital world is more and more difficult (Pilot). Yet recent research (Mourey) suggests that consumers are more sensitive Study 1 to subtle cues when they experience a choice as “feeling difficult.” Study 1 relies on verbal priming to cue trust, distrust, or nothing The purpose of the current project is to explore whether the increased (control). Participants (N = 122) completed word descrambling tasks feelings of difficulty in maintaining privacy online counterintuitively containing trust cue words, distrust cue words, or neutral words. Par- makes consumers more likely to disclose personal information when ticipants then drafted a post for their preferred social network and exposed to subtle cues that prime “trust.” Five studies show that asked to indicate which information from a list of twenty personal individuals are more likely to disclose highly personal information they would be willing to disclose. Participants also indicated how based on subtle cues of trust including verbal trust primes (Study 1), much they trusted the site and completed the 12-question Interpreta- social network size (Study 2), friends’ online engagement (Study 3), tion of Difficulty (IoD) scale regarding their perceptions of difficulty and the verbal and visual fluency of a website’s privacy policy (Study doing the task. Results revealed that trust primes led to more trust 4), all without their conscious awareness. and distrust primes to more distrust for people who interpreted difficulty to mean “impossible,” (the opposite was true for participants Literature Review who interpreted difficulty to mean “important”). Thus, we obtain ini- A growing body of research has shown that privacy disclosure tial support for our hypothesis. decisions are strongly influenced by contextual factors. Consider, for example, observing others sharing private information makes indi- Study 2 viduals more likely to share, an online “herding effect” (Acquisti et Study 2 relies on social network size and prior theorizing on al., 2012; John et al., 2011). These prior projects found that individu- “herding effects” in which social network size is a cue of trust (larger als engage in comparative behavior based on the signals of others = more trust). The study design was the same as Study 1 with the and suggests that such signals–e.g., social network size, the engage- verbal primes replaced by participants’ attention being drawn to the ment of others online–implicitly influence behavior. size of their social network. Results revealed that participants (N = However, recent research also suggests that the influence 203) were willing to disclose more private information the larger of subtle cues (a.k.a. primes) is rarely as straightforward as once their network if difficulty was interpreted as impossible, with trust thought. Indeed, the same studies involving subtle cues and priming mediating the effect. (Schnall, Benton, and Harvey, 2008) have been replicated (Arbesfeld et al., 2014), not replicated, or replicated with results in the opposite Study 3 direction (Johnson et al., 2014). To address these inconsistencies, Study 3 also looked at a practical application of this subtle cue recent research (Mourey, forthcoming) proposed a two-part model online influence by using a feature common to most social networks: to explain both when primes are likely to exert a stronger influence number of likes on an item. Greater engagement from friends has and in which direction those primes are likely to influence behavior. been shown to cue more engagement from individuals. Here, the First, he proposes that individuals are more likely to employ con- priming stimuli was an ad liked by 24 (or 2) friends, and the depen- textual cues when tasks “feel difficult.” Second, when difficulty is dent measure of interest was how likely participants indicated they interpreted to mean a task is “impossible” and individuals “give up” were to try the made-up restaurant featured in the ad. Trust was also mentally (i.e., not waste cognitive effort on an impossible task), a measured. Results showed that participants (N = 197) randomly as- prime exerts a prime-consistent effect. Alternatively, when difficulty signed to the “many friends” ad condition were more likely to try is interpreted to mean a task is “important,” individuals process more the restaurant if they interpreted the difficulty of the task as “impos- deliberately and, in doing so, exhibit prime-reactance (Ferguson and sible.” Effects, again, were mediated via trust, as friend engagement Bargh, 2004). influenced trust, which, in turn, influenced willingness to try an un- For the current project, participants should be more sensitive to known restaurant. trust/distrust primes when engaging in a difficult task online (e.g., maintaining privacy, assessing an unknown company), and the direc- Study 4 tion of that influence depends on what they interpret that difficulty The final study used privacy policy as the contextual stimulus, to mean. varying both the perceptual fluency (figure/ground contrast) and the verbal fluency (complexity of language). Participants (N = 214) first Pilot reviewed the privacy policy, then indicated the same disclosure deci- The purpose of the pilot was to show that participants do, in sions as done in the prior studies, as well as measures of difficulty fact, experience difficulty with respect to maintaining privacy in on- and trust. Results showed that easy-to-read (difficult-to-read) privacy line contexts. Participants (N = 107) found maintaining privacy dif- policies increased (diminished) trust which, in turn, increased (de- ficult online in general (M = 4.31, SD = 1.60; t(106) = 1.99, p < .05), creased) disclosure of private information. on social networks (M = 4.41, SD = 1.70; t(106) = 2.50, p < .01), and in the real world (M = 4.42, SD = 1.55; t(106) = 2.81, p < .01), and General Discussion that maintaining privacy was becoming more difficult (M = 5.56, SD Five studies support that 1) consumers find maintaining privacy = 1.18; t(106) = 13.74, p < .001). This first study provided support for increasingly difficult, and 2) in these difficult situations, subtle cues

Advances in Consumer Research 704 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 705 of trust/distrust exert greater influence on disclosing personal infor- Hayes, Andrew F. and Jörg Matthes (2009), “Computational mation. Counterintuitively, increasing difficulty maintaining privacy Procedures for Probing Interactions in OLS and Logistic online makes people more sensitive to cues that encourage more dis- Regression: SPSS and SAS Implementations,” Behavioral closure. Research Methods, 41, 924-936. Johnson, David, Felix Cheung, and Brent Donnellan (2014), REFERENCES “Cleanliness Primes Do Not Influence Moral Judgment,” Alter, Adam L., Daniel Oppenheimer, Nicholas Epley, and Rebecca Retrieved 13:00, August 15, 2016 from http://www. N. Eyre (2007), “Overcoming Intuition: Megacognitive PsychFileDrawer.org/replication.php?attempt=MTcy Difficulty Activates Analytic Reasoning,” Journal of Kurzban, Robert, Angela Duckworth, Joseph W. Kable, and Justus Experimental Psychology: General, 136(4), 569-576. Myers (2013), “An Opportunity Cost Model of Subjective Arbesfeld, Julia, Tricia Collins, Demetrius Baldwin, and Kimberly Effort and Task Performance,” Behavioral and Brain Sciences, Daubman (2014), “Clean Thoughts Lead to Less Severe Moral 36(6), 661-679. Judgment,” Retrieved 13:00, August 15, 2016 from http:// Li, Xitong and Wu, Lynn, Herding and Social Media Word-of- www.PsychFileDrawer.org/replication.php?attempt=MTc3 Mouth: Evidence from Groupon (September 25, 2014). Avnet, Tamar, Michel Tuan Pham, and Andrew T. Steven (2012), Mourey, James A. (forthcoming), “Prime Time: A Practical “Consumers’ Trust in Feelingsas Information,” Journal of Explanation of When and How Primes Influence Choice and Consumer Research, 39 (December), 1-16. Behavior.” Besman, Melissa, Caton Dubensky, Leanna Dunsmore, and Oyserman, Daphna, George Smith, and Kristen Elmore (2014), Kimberly Daubman (2013), “Cleanliness Primes Less “Identity-Based Motivation: Implications for Health and Severe Moral Judgments,” Retrieved 13:00, August 15, Health Disparities,” Journal of Social Issues, 70, 206-225. 2016 from http://www.PsychFileDrawer.org/replication. Schwarz, Norbert (2012), “Feelings-as-Information Theory,” in php?attempt=MTQ5. Handbook of Theories of Social Psychology, ed. P. Van Lange, Chartrand, Tanya L., Amy N. Dalton, and Gavan J. Fitzsimons A. Kruglanski, E.T. Higgins, Washington DC: Sage, 289-308. (2007), “Nonconscious relationship reactance: When Srull, Thomas K. and Robert S. Wyer, Jr., (1979), “The Role of Significant Others Prime Opposing Goals,” Journal of Category Accessibility in the Interpretation of Information Experimental Social Psychology, 719-726. About Persons: Some Determinants and Implications,” Journal Hayes, Andrew F. (2013), “Introduction to Mediation, Moderation, of Personality and Social Psychology, 37(10), 1660-1672. and Conditional Process Analysis: A Regression Based Approach,” The Guilford Press. A Salience Theory of Three Exposure Effects Kellen Mrkva, Columbia Business School, USA Leaf Van Boven, University of Colorado Boulder, USA

EXTENDED ABSTRACT Experiment 2 was a replication of Experiment 1. American We propose and support a salience theory of exposure effects adults who were recruited from Amazon Mechanical Turk (N = which posits that repeated exposure influences evaluations by in- 109) participated online in exchange for $1.00. Four participants creasing salience, the relative quality of standing out in relation to dropped out of the study prior to viewing all three slideshows and other stimuli in the environment. Across four experiments, we dem- were thus removed prior to analyses (resulting N = 105; 54 female, onstrate that exposure effects previously attributed to absolute (mere) Mage = 34.12). Repeated exposure again increased salience, t(30.45) exposure are actually explained by heightened relative salience. Ad- = 5.64, b = 0.44, p < .001, affective intensity, t(32.10) = 4.88, b = ditionally, we demonstrate that exposure, in addition to increasing 0.20, p < .001, and evaluative extremity, t(34.40) = 2.09, b = 0.03, liking (Zajonc 1968), increases affective intensity and makes evalu- p = .044. Unlike in Experiment 1, repeated exposure did not appre- ations more extreme. ciably increase liking, t(37.10) = 1.19, b = 0.03, p = .243. A multiple In Experiments 1-2, we replicated the procedure used in most mediation model suggested that the effects of repeated exposure on mere exposure research. In Experiment 1, 116 American adults who evaluative extremity and affective intensity were partially accounted were recruited from Amazon Mechanical Turk participated in ex- for by salience, whereas fluency and apprehension did not account change for $1.25. Sixteen dropped out of the study prior to complet- for these effects. In Experiment 3, we directly manipulated salience to complete a ing the key measures (resulting N = 100, 50 female, Mage = 36.39). Participants viewed eight stimuli, presented in a slideshow with ex- causal chain in which exposure increases salience (Experiments 1-2) posure manipulated: Two stimuli were presented 9 times, two were and salience increases evaluative extremity and affective intensity presented 3 times, two were presented 1 time, and two were not pre- (Experiment 3). Undergraduates (N = 71, 45 female, Mage = 19.34) sented (1.0 second per presentation). Following the slideshow, liking participated for course credit. We directly manipulated salience by was measured as in previous research: “How much do you like each presenting 12 stimuli on a screen simultaneously, with 1 oriented dif- image” (-3 = dislike, 0 = neutral, 3 = like), and we assessed affect ferently from the others (1 horizontal and 11 diagonal or 1 diagonal intensity (1 = not at all intense; 9 = extremely intense). We also mea- and 11 horizontal). This manipulates salience while equating expo- sured self-reported salience, fluency, and apprehension to test our sure duration and inherent differences between horizontally-oriented hypothesis that salience accounts for the exposure effect, whereas and diagonally-oriented stimuli. Salience increased evaluative ex- fluency and apprehension account for less if any of the exposure ef- tremity, t(43.84) = 4.70, b = 0.43, p < .001 and affective intensity, fect. Prevailing theories of mere exposure posit that increased flu- t(40.08) = 4.73, b = 0.91, p < .001, supporting our proposed causal ency (Bornstein and D’Agostino 1994; Reber, Winkielman, and chain. Schwarz 1998) or reduced apprehension (Harrison 1977; Zajonc Because salience is a relative quality of standing out in rela- 1968) account for the exposure effect. Because salience increases tion to other objects, we hypothesized in Experiment 4 that relative evaluative extremity (Downing, Judd, and Brauer, 1992; Sadler and exposure influences evaluations more than absolute exposure. Par- Tesser 1973), we also hypothesized that exposure would increase ex- ticipants (N = 400 from Mechanical Turk) were assigned to either tremity (measured as distance from the neutral midpoint of the liking the within-subjects condition (similar to Experiments 1-2) or one scale; Downing et al., 1992). This procedure was repeated for three of three between-subjects conditions in which they viewed stimuli stimulus sets commonly-used in mere exposure experiments (Chi- presented at one level of exposure (i.e., 9 times, 3 times, or 1 time). nese characters, paintings, and Turkish words). This was designed to isolate exposure from relative salience; we hy- We used linear mixed effects models to test each prediction. pothesized that exposure would increase liking and extremity when Each model treated Participant and Stimulus as random effects to confounded with relative salience (i.e., within-subjects condition) properly model variance associated with participants and stimuli but not when exposure was isolated from salience (i.e., between- (Judd, Westfall, and Kenny 2012). subjects conditions). As predicted, relative exposure increased liking As predicted, exposure increased salience, t(83.60) = 7.14, b = and evaluative extremity. In contrast, absolute exposure (between- 0.76, p < .001. Exposure also increased liking, t(70.08) = 6.50, b = subjects) did not. A difference-in-differences analysis revealed that 0.26, p < .001, replicating previous research (Zajonc 1968). Addi- the relative exposure effect was significantly larger than the absolute tionally, exposure made evaluations more extreme, t(24.85) = 3.42, b exposure effect. = 0.09, p = .002. Exposure also increased affective intensity,t (34.54) These results cohere with previous findings that relative com- = 4.48, b = 0.39, p < .001. parisons are more impactful than absolute states (e.g., Shafir, Dia- We also conducted multiple mediation analyses (Preacher and mond, and Tversky 1997). They also build on the emerging evidence Hayes 2008) to test the hypothesis that salience accounted for the that salience may account for several judgment and decision making effects of exposure on liking, evaluative extremity, and affective in- phenomena (Bordalo, Gennaioli, and Shleifer 2012; Mrkva and Van tensity. These analyses revealed that there was an indirect effect of Boven 2017; Yechiam and Hochman 2013). The findings have im- relative exposure on liking through salience as a mediator (ab = 0.10, portant theoretical implications as well, suggesting that effects previ- 95% CI [0.08, 0.13]). Mediation analyses were also consistent with ously attributed to absolute (mere) exposure and fluency are actually our hypotheses that salience accounted for the effects of exposure on explained by heightened relative salience. evaluative extremity and affective intensity. Fluency and apprehension accounted for smaller proportions of each effect (if any), con- REFERENCES trary to prevailing theories that these variables account for exposure Bordalo, Pedro, Nicola Gennaioli, and Andrei Shleifer (2012), effects. “Salience in Experimental Tests of the Endowment Effect.” The American Economic Review 102, 47-52.

Bornstein, Robert F., and Paul R. D’Agostino (1994), “The Preacher, K. J., & Andrew F. Hayes (2008), “Asymptotic and Attribution and Discounting of Perceptual Fluency: Resampling Strategies for Assessing and Comparing Indirect Preliminary Tests of a Perceptual Fluency/Attributional Model Effects in Multiple Mediator Models.” Behavior Research of the Mere Exposure Effect.” Social Cognition 12, 103-128. Methods, 40(3), 879-891. Downing, James W., Charles M. Judd, and Markus Brauer Reber, Rolf, Piotr Winkielman, and Norbert Schwarz (1992), “Effects of Repeated Expressions on Attitude (1998), “Effects of Perceptual Fluency on Affective Extremity.” Journal of Personality and Social Psychology, 63, Judgments.” Psychological Science, 45-48. 1, 17-29. Sadler, Orin, and Abraham Tesser (1973), “Some Effects of Harrison, Albert A (1977), “Mere Exposure.” Advances in Salience and Time Upon Interpersonal Hostility and Attraction Experimental Social Psychology, 10, 39-83 During Social Isolation.” Sociometry, 99-112. Judd, Charles M., Jacob Westfall, and David A. Kenny (2012), Shafir, Eldar, Peter Diamond, and Amos Tversky (1997), “Money “Treating Stimuli as a Random Factor in Social Psychology: illusion.” The Quarterly Journal of Economics 112, 341-374. A New and Comprehensive Solution to a Pervasive but Spencer, Steven J., Mark P. Zanna, and Geoffrey T. Fong (2005), Largely Ignored Problem.” Journal of Personality and Social “Establishing a Causal Chain: Why Experiments Are Often Psychology 103, 54-69. More Effective Than Mediational Analyses in Examining Montoya, Matthew, Robert S. Horton, Jack L. Vevea, Psychological Processes.” Journal of Personality and Social Martyna Citkowicz, and Elissa A. Lauber (2017), “A Re- Psychology, 89, 6, 845-851. examination of the Mere Exposure Effect: The Influence Yechiam, Eldad, and Guy Hochman (2013), “Losses as Modulators of Repeated Exposure on Recognition, Familiarity, and of Attention: Review and Analysis of the Unique Effects of Liking.” Psychological Bulletin, 143, 5 459-498. Losses over Gains.” Psychological Bulletin 139, 497-518. Mrkva, Kellen, and Leaf Van Boven (2017), “Attentional Zajonc, Robert B. (1968), Attitudinal Effects of Mere Accounting: Voluntary Spatial Attention Increases Budget Exposure. Journal of Personality and Social Psychology, 9, Category Prioritization.” Journal of Experimental Psychology: 1-27. General, 146, 9, 1296-1306. Non-Normative Influence of Self-Decided Prices on Product-Related Inferences Sudipta Mukherjee, Virginia Tech, USA Mario Pandelaere, Virginia Tech, USA

EXTENDED ABSTRACT Results Introduction and Theoretical Background Excluding 10 participants who exhibited response reversal/ Price is a key factor in consumer decision making (Adaval and response inconsistency (willingness to buy the product at a higher Monroe, 2002; De Langhe et al., 2014). From a purely economic per- price but not at a lower price), a one-way ANOVA showed that self- spective, consumers should be willing to buy a product if the value decided prices were lower (vs. higher) in the case of ascending (vs. that they expect from the product is higher than the product price descending) price presentation order (Mascending = $19.26, SD = 14.39;

– implying that consumers should determine the value of a prod- Mdescending = $35, SD = 25.85; F (1, 100) = 14.86; p < 0.0001; d = uct independent of the price. However, ample research shows that 0.75). More importantly, as hypothesized, a one-way ANOVA results consumers often use the marketer-provided prices to make product reveal that the ascending (vs. descending) price presentation resulted inferences (Zeithaml, 1988; Parasuraman et al., 1988; Brucks et al., in less favorable perceived product usefulness (Mascending = 3.54, SD

2000). This is rational assuming that marketers decide product-prices = 1.65; Mdescending = 4.18, SD = 1.65; F (1, 110) = 4.24; p = 0.042; d based on the market dynamics of supply and demand, which would = 0.38). Further, it was found that the effect of price presentation lead a product price to be closely aligned to its inherent value (i.e. order on product-related inference (perceived usefulness) was me- quality and usefulness). diated through self-decided prices (10000 sample 95% C.I.: (-0.27, While, inferring product value from marketer-provided prices -0.07)). Interestingly, the inverse direction of the indirect effect, the may be rational, this is not the case when consumers themselves gen- normatively expected influence of product-related inferences on self- erate or decide on product prices as in pay-as-you-wish/pay-what- provided prices, was not significant (10000 bootstrap 95% CI (-1.86, you-want pricing schemes (Kim et al., 2009), auctions (Greenleaf, 0.43). 2004) and price negotiations (Srivastava et al., 2000). Normatively, product value should influence self-decided prices rather than the re- Study 2 verse. Still, we argue that consumers may infer value from the price This study tested the robustness of the results obtained in the they decide to pay; this may lead to biased inferences when self- previous study and showed that the underlying effect was caused not decided prices are affected by irrelevant contextual factors. We argue by difference in the starting prices between ascending and descend- that due to sheer habit of inferring value from marketer-provided ing price order conditions, but rather by differences in self-decided prices (Plassmann et al., 2007, 2008; Rao and Monroe, 1989; Sci- prices. tovszky, 1944), consumers will over apply the price-quality heuristic Design and Procedure in the case of self-decided prices too. That is, higher the self-decided Using 170 (49% female; M = 20.91) undergraduate students, price, the more favorable would be the inferred product value. age this study utilized a 2 (Price presentation order: ascending vs. descending) x 2 (Price steps: $10 vs. $20) between-subject factorial Study 1 design. This study differed from the previous study in three ways. Study 1 tests the idea that an undue influence on self-decided First, prices ranged from $100 to $200. Second, price step was also prices affects subsequent judgments of product usefulness. To influ- manipulated by varying the prices in the self-decided price elicitation ence self-generated prices, we had participants indicate for a range procedure either in steps of $10 ($10 price step condition) or $20 of prices if they would consider buying a product at that price. Prices ($20 price step condition). We anticipated that because the number were presented from low to high or from high to low; based on the of different prices being presented is larger (vs. smaller) in the price phenomenon of anchoring (Kamins, Dreze, and Folkes, 2004; Ariely step $10 (10 different prices) (vs. price step $20 (5 different prices), and Simonson, 2003), we expected a higher willingness-to-pay in the self-decided prices will differ more between ascending and de- the latter condition, causing higher judgments of product usefulness. scending price presentation orders more when the price step is $10 as Design and Procedure compared to when the price step is $20. Further, as the starting price This study implemented a dichotomous-choice willingness-to- was the same for both price step $10 and price step $20, any sys- pay procedure in which respondents (111 undergraduate students, tematic differences between the effect of price presentation order on

67% female, Mage = 20.74) were provided with multiple prices for product-related inferences for price step $10 vs. price step $20 condi- the same product (Bluetooth speaker) and were asked to indicate tions would indicate that the underlying effect of price presentation whether they would be willing to buy the product at each of the order on product-related inferences is due to self-decided prices and presented prices. The prices (ranging from $10 to $100) were either not due to the starting price itself. Third, the product-related infer- systematically increasing (ascending price order condition) or de- ences was operationalized through measuring anticipated product creasing (descending price order condition). The final (vs. first) price satisfaction. that respondents were willing to buy the product at was calculated Results as their willingness-to-pay price in the ascending (vs. descending) For self-decided prices, we found a significant interaction be- price order condition. The willingness-to-pay price is conceptually tween price presentation order and price step (F(1, 156) = 4.55, p = identical to the self-decided price (Kim, Natter, and Spann 2009). 0.034) such that ascending (vs. descending) price presentation order Having obtained the self-decided price, the experimental procedure resulted in lower self-decided prices only when price step is $10 (M next required respondents to indicate how useful they thought the a- = $43.9, SD = 63.75; M = $94.15, SD = 80.65; F (1, 156) product would be. scending descending = 10.57; p = 0.001; d = 0.69), and not when price step is $20 (Mascending

= $50.24, SD = 61.34; Mdescending = $53.24, SD = 72.73; F (1, 156) = 0.036; p > 0.5).

Advances in Consumer Research 708 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 709 We also found a significant interaction between price presenta- Results tion order and price step for anticipated product satisfaction (F (1, As expected, the amount bid in the low bid amount condition

156) = 4.19, p = 0.042), such that while ascending (vs. descending) was lower than that in the high bid amount condition (Mbid amount low price presentation order resulted in lower anticipated product satis- = $18.25, SD = 4.19; Mbid amount high = $88.56, SD = 22.6; F (1, 316) faction only in the price step $10 condition (Mascending = 3.87, SD = = 1495.92, p < 0.0001, d = 4.33). A general linear model, which

1.11; Mdescending = 4.42, SD = 1.01; F (1, 156) = 3.4; p = 0.067; d = included the main effects of bid amount (low vs. high), perspective

0.52), and not in the price step $20 condition (Mascending = 4.01, SD = (self vs. other), and their interaction, revealed that while the main ef-

1.69; Mdescending = 3.7, SD = 1.45; F (1, 156) = 1.1; p = 0.3). fect of bid amount on expected product quality was significant (Mbid

Finally, we ran a moderated mediation model (Preacher and amount low = 4.53, SD = 1.44; Mbid amount high = 4.87, SD = 1.34; F (1, 314) Hayes (2008) PROCESS model 7) to test whether price step mod- = 4.82, p = 0.029, d = 0.24), the main effect of perspective (self vs. erates the indirect effect of price presentation order on anticipated other) (p=0.2) and the interaction between bid amount and perspec- product satisfaction through self-decided prices. A 95% C.I. (-0.53, tive (p=0.33) was not significant. -0.02) for a 10000 bootstrap sample indeed reveals that the index for moderated mediation (Index = -0.25, SE (boot) = 0.13) does not Study 4 contain 0. While the 95% C.I. for a 10000 bootstrap sample for the Design and Procedure indirect effect of price presentation order on anticipated product sat- Undergraduate students (N = 119; M = 20.87, 53% female) isfaction did not contain 0 for the $10 price step (-0.47, -0.10), it did age were randomly assigned to one cell of a 2 (bid amount: low vs. high) contain 0 for the $20 price step (-0.18, 0.15). This suggests price between-subject design. We used the game show paradigm as in steps indeed does moderate the mediating influence of self-decided study 3. In addition, this study also included a price-quality schema prices on the relationship between price presentation order and an- measure to test if the effect is stronger for those with stronger price- ticipated product satisfaction. More importantly, it provides further quality schema; this would indicate that the over-application of a evidence that people infer product value from self-decided prices. price-quality heuristic explains the observed effect. Study 3 Results Study 3 aims to replicate the previous findings using a different As expected, the amount bid in the low bid amount condition paradigm to influence self-decided prices. In addition, the present was lower than that in the high bid amount condition (Mbid amount low = study attempts to rule out cognitive dissonance (Festinger, 1962; Eli- $17.55, SD = 5.16; Mbid amount high = $89.05, SD = 21.62; F (1, 117) = iot and Devine, 1994) and self-perception (Bem, 1972; Calder and 620.51, p < 0.0001, d = 4.55). More importantly, the expected prod- Staw, 1975) as rival explanations. uct quality was higher in the high bid amount condition as compared to the low bid amount condition (M = 3.41, SD = 1.61; M Design and Procedure bid amount low bid = 4.06, SD = 1.55; F (1, 117) = 5.06, p = 0.026, d = 0.41). MTurk panel members (N= 318; 55% female; M = 36.14) amount high age Further, there was a significant interaction (F (1, 115) = 17.32, p = were randomly assigned to one cell of a 2 (bid amount: low vs. high) 0.0001) between experimental condition and price-quality schema in x 2 (perspective: self vs. other) between-subject factorial design. predicting expected product-quality, such that while the effect was This study utilized an interactive gameplay scenario to influence significant for those with strong price-quality schema (p < 0.0001) prices. Respondents were either required to take part in an online but not for those with weak price-quality schema (p = 0.22). interactive simulated game titled “Wheel of Fortune”. Respondents were asked to imagine that they were taking part in the last stage of Study 5 a multistage game during which, at each stage, they were required to bid a certain amount of money in order to win a product. Respon- Design and Procedure dents were required to bid the amount before knowing what product Undergraduate students (N = 239; Mage = 38.13, 64% female) they could win – thus ensuring that the bid amount was completely were randomly assigned to one cell of a 2 (amount left: $20 vs. $100) unconnected to the inherent product characteristics. Upon putting in x 2 (reminder about random nature of bid amount: present vs. absent) their bid amounts, respondents were informed about the product that between-subjects design. We used the same paradigm as in Study 3 they had won (the same product, sunglasses, for all respondents), and and 4. In addition to manipulating bid amount, we also manipulated were asked to indicate the expected product quality. whether participants were reminded that the bid amount was just the In order to manipulate the bid amount, respondents were told outcome of a random process, that is, just the amount they happened that they either had $20 left or $100 left to bid for the last round, to have left on the last round of gameplay. Half of the respondents and that this amount could not be converted into cash or credit and were asked to indicate if the amount was indeed random, whereas the would go to waste unless used for bidding on the product. Other other half did not get that cue. We expected that the over-application than manipulating the bid amount, this was also useful in ruling out of a price-quality heuristic is a mindless process. If so, any cue that the potential alternative explanation of cognitive dissonance (aris- makes people more mindful about the nature of the self-decided ing from potentially indicating a low expected quality after having price should attenuate or even eliminate the effect of bid amount of bid a high amount), by providing respondents a clear justification of perceived product value. Finally, we also measured individual dif- their bid amount. Further, gameplay perspective (self vs. other) was ferences in price-quality beliefs to replicate the moderation of the manipulated: Half of the participants had to imagine playing them- previous study. selves, whereas the other half had to imagine watching someone Results else participating. If similar results are obtained when participants As before, bid amount was significantly higher in the $100 imagine observing others as when they imagine playing themselves, left condition as compared to the $20 left condition (M = these effects cannot be attributed to self-perception processes (nor bid amount low $17.98, SD = 4.03; M = $82.68, SD = 27.48; F (1, 237) = cognitive dissonance). They would nevertheless support the idea that bid amount high 640.58, p < 0.0001, d = 3.29). More importantly, a general linear people infer product value, even when prices are clearly not reflec- model analysis, with bid amount and reminder intervention as the tive of the value the market assigns to the product. 710 / Non-Normative Influence of Self-Decided Prices on Product-Related Inferences fixed factors showed a significant interaction (F (1, 235) = 9.44, p De Langhe, Bart, Stijn MJ van Osselaer, Stefano Puntoni, and Ann = 0.002) between bid amount (low vs. high) and reminder interven- L. McGill (2014), “Fooled by Heteroscedastic Randomness: tion (present vs. absent) on expected quality. The expected product Local Consistency Breeds Extremity in Price-Based Quality quality was significantly higher in the high bid amount condition as Inferences,” Journal of Consumer Research, 41 (August), compared to the low bid amount condition (Mbid amount low = 3.95, SD 978-94.

= 1.46; Mbid amount high = 5, SD = 1.21; F (1, 235) = 17.16, p < 0.0001, Elliot, Andrew J., and Patricia G. Devine (1994), “On the d = 0.78) when reminder intervention was absent, but not when the Motivational Nature of Cognitive Dissonance: Dissonance as reminder intervention was present (F (1, 235) = 0.033, p > 0.5). Fur- Psychological Discomfort”, Journal of Personality and Social ther, showing the robustness of the finding from the previous study, Psychology, 67 (September), 382-94. there was a significant interaction (F (1, 114) = 7.63, p = 0.0067) in Festinger, Leon (1962), A Theory of Cognitive Dissonance. Vol. 2. the expected direction between price quality schema and bid amount Stanford University Press. in the conditions without reminder intervention. Greenleaf, E. A. (2004). “Reserves, Regret, and Rejoicing in Open English Auctions,” Journal of Consumer Research, 31(2), General Discussion, Theoretical Contributions, and 264-73. Managerial Implications Kamins, Michael A., Xavier Dreze, and Valerie S. Folkes (2004), The present research provides evidence for the non-normative “Effects of Seller-Supplied Prices on Buyers’ Product influence of self-decided prices on product-related inferences, and Evaluations: Reference Prices in an Internet Auction contributes to marketing literature in several important ways. First, Context,” Journal of Consumer Research, 30 (no. 4), 622-28. it contributes to the existing price-quality relationship literature by Kim, Ju-Young, Martin Natter, and Martin Spann (2009), showing that consumers infer product value not only from marketer- “Pay What You Want: A New Participative Pricing provided prices but also from self-decided prices. Second, it contrib- Mechanism,” Journal of Marketing, 73 (no. 1), 44-58. utes to the existing literature on price presentation order by showing Parasuraman, Arun, Valarie A. Zeithaml, and Leonard L. Berry that ascending and descending price presentation orders can differ in (1988), “Servqual: A Multiple-Item Scale for Measuring terms of their downstream influence on product-related inferences. Consumer Perceptions of Service Quality.” Journal of Given that price presentation order is an important commonplace Retailing, 64 (April), 12-40. contextual factor that is relevant to a variety of marketer-customer Plassmann, Hilke, John O’Doherty, and Antonio Rangel (2007), interactions, such as auctions, negotiations, salesperson-customer “Orbitofrontal Cortex Encodes Willingness to Pay in Everyday interactions, amongst others, the findings from this research thus Economic transactions.” Journal of Neuroscience, 27 provide both substantial theoretical and substantive contributions. (September), 9984-88. Plassmann, Hilke, John O’Doherty, Baba Shiv, and Antonio REFERENCES Rangel (2008), “Marketing Actions Can Modulate Neural Adaval, Rashmi, and Kent B. Monroe (2002), “Automatic Representations of Experienced Pleasantness,” Proceedings of Construction and Use of Contextual Information for Product the National Academy of Sciences, 105 (January), 1050-54. and Price Evaluations”, Journal of Consumer Research, 28 Rao, Akshay R., and Kent B. Monroe (1989), “The effect of (March), 572-88. price, brand name, and store name on buyers’ perceptions of Ariely, Dan, and Itamar Simonson (2003), “Buying, Bidding, product quality: An integrative review,” Journal of Marketing Playing, or Competing? Value Assessment and Decision Research, August, 351-57. Dynamics in Online Auctions”, Journal of Consumer Scitovszky, Tibor (1944), “Some Consequences of the Habit Psychology, 13 (January), 113-23. of Judging Quality by Price,” The Review of Economic Bem, Daryl J. (1972), “Self-Perception Theory” in Advances in Studies, 12 (January), 100-5. Experimental Social Psychology, vol. 6, Academic Press., Srivastava, Joydeep, Dipankar Chakravarti, and Amnon Rapoport 1-62. (2000), “Price and Margin Negotiations in Marketing Brucks, Merrie, Valarie A. Zeithaml, and Gillian Naylor (2000), Channels: An Experimental Study of Sequential Bargaining “Price and Brand Name as Indicators of Quality Dimensions Under One-Sided Uncertainty and Opportunity Cost of for Consumer Durables.” Journal of the Academy of Marketing Delay,” Marketing Science, 19 (May), 163-84. Science, 28 (July), 359-74. Zeithaml, Valarie A. (1988), “Consumer Perceptions of Price, Calder, Bobby J., and Barry M. Staw (1975), “Self-Perception of Quality, and Value: A Means-End Model and Synthesis of Intrinsic and Extrinsic Motivation,” Journal of Personality and Evidence,” The Journal of Marketing, July, 2-22. Social Psychology, 31 (April), 599. The “Upper Limit Framing” Effect: Upper Limit Framing of a Cost Estimate Influences Consumption Choices Sudipta Mukherjee, Virginia Tech, USA Frank May, Virginia Tech, USA

EXTENDED ABSTRACT choice scenarios (study 2: restaurant wait time; study 3: Uber wait time). In both studies, respondents were asked to imagine that they Introduction and Theoretical Background had to wait for a period of time, and were presented with wait time Consumers often make consumption choices on the basis of es- estimates either using less than or not more than framing. The wait timates, rather than actual temporal or monetary costs. Frequently, time estimates provided in study 2 (study 3) were: less than 30 minutes marketers interchangeably use “less than” vs. “not more than” frame to and not more than 30 minutes (less than 20 minutes and not more than present consumers with an estimated upper limit of the underlying cost. 19 minutes). The wait times in study 3 were chosen to rule out the For example, an Uber driver may tell you that he will arrive in “less than alternative explanation for the upper limit framing effect – potential 20 minutes,” or “not more than 20 minutes.” Although these frames are non-equivalence of the two frames. Results from both studies showed equivalent and interchangeably used, we show that these frames result that less than (vs. not more than) framing resulted in lower perceived in differential consumption choices – what we term the upper limit wait time and higher likelihood to wait. Further, perceived wait time framing effect. We show that less than (vs. not more than) framing of mediated the influence of upper limit frame on likelihood to wait. the upper limit of a temporal or monetary cost estimate can result in dif- Study 4 tested the influence of upper limit framing on estimated ferent consumer expectations regarding the underlying cost involved, monetary cost (getting a car repaired by a mechanic; less than $475 and systematically influence subsequent consumption choices. vs. not more than $475). It was found that less than (vs. not more We draw our hypotheses from research on negation theory. Ne- than) framing resulted in lower perceived cost of repairs, which me- gation theory literature suggests that verbal inferences that involve diated the likelihood of getting the car repaired by the mechanic. negations are more effortful to process than equivalent affirmations Studies 5 (house price: less than $200,000 vs. not more than (Dale and Duran, 2011; Nordemeyer and Frank, 2014; Khemlani, $200,000) and 6 (plumbing repair cost: less than $7,500 vs. not more Orenes, and Johnson-Laird, 2012; 2014), although under certain than $7,500) show that the upper limit framing effect reverses when circumstances they can be equally fluent to process (Johnson-Laird, the underlying cost is large—not more than framing results in lower 1970; Clark, 1974). In turn, the less fluent processing associated with perceived costs versus less than framing. negations leads individuals to neglect the negation. For example, Finally, study 7 showed that there is a significant interaction someone deemed “innocent” will appear more free of guilt than between upper limit frame (less than vs. not more than) and estimate someone deemed “not guilty.” Indeed, for cost estimates involving amount (car repair cost: $250 vs. $2500) in predicting perceived cost small amounts, our findings are consistent with this model: framing and likelihood to incur cost. Specifically, for the small estimated re- the upper limit as “less than” leads to contracted cost perceptions (vs. pair cost ($250), the less than (vs. not more than) frame resulted in “not more than” framing). significantly lower perceived repair cost and higher likelihood to get Interestingly, we are the first to our knowledge to find conditions the car repaired. However, the reverse was true for large estimated under which a negation frame is actually more fluent than an affirma- repair cost ($2500). Furthermore, fluency of processing the cost es- tion frame--when negation frames involve large cost estimates. Ex- timate was higher (lower) for less than (vs. not more than) frame for tant research has shown that large numbers require abstract construal small (large) amount, and mediated the interactive effect of frame to process (Hansen, Kutzner, and Wanke, 2012; Stanescu-Cosson et and amount on cost perception. al., 2000; Dehaene, et al., 2003), which matches with the abstract nature of negation framed assertions (Gilead, Liberman, and Maril, Contribution 2012). Consequently, the abstract mindset required to process large This research makes a number of substantial contributions. numbers matches the abstract nature of negations, leading to more First, we contribute to the negation literature by being the first to fluent processing (Lee and Aaker, 2004; Labroo, Dhar, and Schwarz, demonstrate situations in which negation frames are actually easier 2007). Thus, for cost estimates involving large amounts, the well- to process and have stronger meaning than affirmation frames (e.g., established effect of negations being more difficult to process should someone deemed “not guilty” can appear more free of guilt than reverse due to a greater match between the mindset primed by large someone deemed “innocent”). Additionally, we contribute to con- amounts and negation-framed assertions. Consequently, for large es- sumer research by showing how framing of cost estimates influence timates, framing the estimate as not more than results in contracted consumption choices and that equivalent frames for a cost estimate cost perceptions (vs. less than framing). can result in distinct consumption choices. We also contribute to the literature on temporal and monetary costs by showing how less than Methods and Results versus not more than framings can influence cost expectations. Fur- Study 1 utilized a single factor (upper limit framing: less than ther, this research contributes to the semantic framing literature by vs. not more than) between-subjects experimental design. Specifically, finding a new semantic framing effect that involving negation based respondents were presented with a real choice context – choosing to inference making. take part in a survey about the business school. Respondents in the less Our research also presents substantial managerial implications. than (vs. not more than) condition were told that the survey will take Given that marketers interchangeably use less than and not more less than (vs. not more than) 10 minutes. It was found that respondents than framing to express the upper limit of an estimate, we provide in the less than condition were significantly more likely to choose to an important insight that the framing of the upper limit can influence take part in the survey (vs. those in the not more than condition). perception of the underlying cost, and that this effect depends on the Studies 2 and 3 utilized single factor (upper limit framing: less estimated amount. than vs. not more than) experimental designs involving hypothetical

Advances in Consumer Research 711 Volume 46, ©2018 712 / The “Upper Limit Framing” Effect: Upper Limit Framing of a Cost Estimate Influences Consumption Choices REFERENCES Khemlani, S., Orenes, I., & Johnson-Laird, P. N. (2014). The Clark, H.H. (1974). Semantics and comprehension. In T.A. Sebeok negations of conjunctions, conditionals, and disjunctions. Acta (Ed.) Current trends in linguistics, Vol. 12: Linguistics and Psychologica, 151, 1-7. adjacent arts (pp. 1291-1428). The Hague: Mouton. Labroo, A. A., Dhar, R., & Schwarz, N. (2007). Of frog wines and Dale, R., & Duran, N. D. (2011). The cognitive dynamics of frowning watches: Semantic priming, perceptual fluency, and negated sentence verification. Cognitive science, 35(5), 983- brand evaluation. Journal of Consumer Research, 34(6), 819- 996. 831. Dehaene, S., Piazza, M., Pinel, P., & Cohen, L. (2003). Lee, A. Y., & Aaker, J. L. (2004). Bringing the frame into Three parietal circuits for number processing. Cognitive focus: the influence of regulatory fit on processing neuropsychology, 20(3-6), 487-506. fluency and persuasion. Journal of personality and social Gilead, M., Liberman, N., & Maril, A. (2012). Construing psychology, 86(2), 205. counterfactual worlds: The role of abstraction. European Nordemeyer, A., & Frank, M. (2014, January). A pragmatic account Journal of Social Psychology, 42(3), 391-397. of the processing of negative sentences. In Proceedings of the Hansen, J., Kutzner, F., & Wänke, M. (2012). Money and thinking: Cognitive Science Society (Vol. 36, No. 36). Reminders of money trigger abstract construal and shape Stanescu-Cosson, R., Pinel, P., van de Moortele, P. F., Le Bihan, D., consumer judgments. Journal of Consumer Research, 39(6), Cohen, L., & Dehaene, S. (2000). Understanding dissociations 1154-1166. in dyscalculia: a brain imaging study of the impact of number Johnson-Laird, P. N. (1970). The interpretation of quantified size on the cerebral networks for exact and approximate sentences. na. calculation. Brain, 123(11), 2240-2255. Khemlani, S., Orenes, I., & Johnson-Laird, P. N. (2012). Negation: A theory of its meaning, representation, and use. Journal of Cognitive Psychology, 24(5), 541-559. Approach and Loss Aversion: Consumer Responses to Approaching and Receding Stimuli in Advertising Lana Mulier, Ghent University, Belgium Iris Vermeir, Ghent University, Belgium Hendrik Slabbinck, Ghent University, Belgium

EXTENDED ABSTRACT Manikin (SAM) scales measuring valence and arousal. Three con- Consumers are increasingly confronted with dynamically mov- trol measures were randomly assessed: the Behavior Identification ing stimuli in commercials, video games and virtual reality. Although Form (BIF; Vallacher and Wegner 1989; α = .90), the Behavioral In- the effect of physical (near vs. far) and psychological (concrete vs. hibition and Behavioral Activation System Scales (BIS/BAS; Carver abstract) distance of stimuli has been broadly investigated (Codis- and White 1994; α = .83), and the Fight Flight Freeze Questionnaire poti and De Cesarei 2007; De Cesarei and Codispoti 2006; Loftus (FFFQ; Maack et al. 2015; α = .91). and Harley 2005), much less is known about the effect of movement Multilevel analyses, controlling for the covariates, showed a (approaching vs. receding) on advertising appeals and consumer de- (marginally) significant effect of condition on valence (F(4, 229) = cisions. Designing effective dynamic marketing campaigns requires 2.40, p = .051). Participants indicated more negative valence toward better understanding of consumers’ reactions to stimuli either mov- approaching neutral stimuli compared to static-middle neutral stim- ing away from or toward them. Building on the approach and loss uli, partly confirming H1a. However, there was no significant effect aversion biases, we predict that people would show more negative of condition on arousal (F(4, 229) = .11, p = .978). H1b could not (positive) emotions and attitudes on advertising effectiveness toward be confirmed, as receding neutral stimuli did not evoke significantly approaching (receding) stimuli. more positive feelings than non-receding neutral stimuli. People have an innate tendency of approach aversion, because In Study 2, testing H1a-b and H2a-b, participants (MTurk; N approaching stimuli can be potentially more harmful and danger- = 123, Mage = 38.24, 60% men) were randomly assigned to 1 of 5 ous than receding stimuli. Approach aversion means that people feel conditions (similar to study 1) in a between-subjects design. One ob- more negatively toward a stimulus that is approaching rather than ject from study 1 served as stimulus for a fictive advertisement on receding or static. This overgeneralized bias arises for a priori nega- second-hand products. Stimulus movement was manipulated as in tive, nonnegative and ambivalent stimuli, and for stimuli moving in study 1. Attitude toward the ad (Aad), attitude toward the brand (Ab), space, time, or probability (Davis et al. 2011; Hsee et al. 2014; Müh- and purchase intention (PI), were used as measures of advertising ef- lberger et al. 2008). For instance, Hsee et al. (2014) reported more fectiveness. Participants indicated their feelings toward the product, negative feelings toward neutral, negative and positive approaching on SAM scales measuring valence and arousal. Aad and Ab were stimuli than toward non-approaching stimuli. In line with these stud- measured on three-item, 7-point semantic differential scales (MacK- ies, we hypothesize that people will express more negative feelings enzie and Lutz 1989; αAad = .97, αAb = .97). PI was measured on a toward approaching neutral stimuli versus non-approaching neutral three-item, 7-point semantic differential scale (Pornpitakpan 2004; α stimuli (H1a). = .96). Three control variables were randomly measured: the BIF (α However, it is still unclear whether there would be a parallel = .92), the BIS/BAS scales (α = .82), and the GREEN scale, measur- effect for receding stimuli, in the sense that these stimuli would ing environmentally friendly consumption values (Haws et al. 2014; evoke more positive feelings than approaching stimuli (Hsee et al. α = .95). 2014). Specifically, loss aversion can manifest itself in people’s reac- ANCOVA analyses firstly showed a significant effect of conditions toward moving stimuli, because receding stimuli could induce tion on valence (F(4, 112) = 3.40, p = .012). Participants indicated a sense of loss. Loss aversion means that losses and disadvantages more negative valence toward approaching neutral stimuli compared have greater impact on people’s preferences than gains and advan- to non-approaching neutral stimuli, confirming H1a. Contrastingly, tages (Tversky and Kahneman 1991). Since people find it worse to participants indicated more positive valence toward receding neu- lose something than to gain something, and thus attach more value tral stimuli than toward approaching neutral stimuli, partly confirm- to losses than gains, receding stimuli could evoke more positive ing H1b. Secondly, condition could not significantly predict arousal feelings because the perceived value seems to increase (Hsee et al. (F(4, 112) = .64, p = .633). Thirdly, there were (marginally) signifi- 2014). As such, we predict that people will express more positive cant effects of condition on Aad (F(4, 112) = 2.93, p = .024), Ab feelings toward receding neutral stimuli versus non-receding neutral (F(4, 112) = 3.15, p = .017), and PI (F(4, 112) = 2.25, p = .068). Par- stimuli (H1b). ticipants expressed more negative Aad, Ab, and PI toward approach- Moreover, as emotions affect advertisements’ evaluations (Poels ing neutral stimuli than toward non-approaching stimuli, confirming & Dewitte, 2006), we hypothesize that people will evaluate an adver- H2a. In contrast, participants expressed more positive Aad, Ab, and tisement more negatively when the ad contains approaching neutral PI toward receding neutral stimuli than toward approaching stimuli, stimuli versus non-approaching neutral stimuli (H2a). Contrastingly, partly confirming H2b. people will evaluate an advertisement more positively when the ad This research revealed that consumers react more negatively to contains receding neutral stimuli versus non-receding neutral stimuli approaching stimuli, supporting approach aversion, and more posi- (H2b). tively to receding stimuli, providing preliminary evidence for loss aversion as underlying mechanism. Moreover, advertisements using In Study 1, testing H1a-b, participants (MTurk; N = 242, Mage = 37.45, 49% men) were randomly assigned to 1 of 5 conditions approaching products decrease ad evaluations, while advertisements (approaching, receding, static-middle, static-near, or static-far stim- using receding products increase ad evaluations. This study is the uli) in a between-subjects design. Twelve neutral objects served as first to demonstrate the importance of understanding how consumers experimental stimuli, for which five videos were created. Stimulus react to approaching and receding stimuli in advertising. Using the movement was manipulated by a change in size. Participants indi- right kind of movement can strengthen consumers’ attitudes toward cated their feelings toward each object, on 9-point Self-Assessment the products being advertised.

Advances in Consumer Research 713 Volume 46, ©2018 714 / Approach and Loss Aversion: Consumer Responses to Approaching and Receding Stimuli in Advertising REFERENCES Maack, Danielle J., Buchanan, Erin and Young, John (2015), Carver, Charles S. and White, Teri L. (1994), “Behavioral “Development and psychometric investigation of an inventory inhibition, behavioral activation, and affective responses to to assess fight, flight, and freeze tendencies: The fight, flight, impending reward and punishment: The BIS/BAS scales,” freeze questionnaire,” Cognitive Behaviour Therapy, 44 (2), Journal of Personality and Social Psychology, 67 (2), 319-33. 117-27. Codispoti, Maurizio and De Cesarei, Andrea (2007), “Arousal MacKenzie, Scott B. and Lutz, Richard J. (1989), “An empirical and attention: Picture size and emotional reactions,” examination of the structural antecedents of attitude toward the Psychophysiology, 44 (5), 680-86. ad in an advertising pretesting context,” Journal of Marketing, Davis, Joshua Ian, Gross, James J. and Ochsner, Kevin N. (2011), 53 (2), 48-65. “Psychological distance and emotional experience: What you Mühlberger, Andreas, Neumann, Roland, Wieser, Matthias J. and see is what you get,” Emotion, 11 (2), 438-44. Pauli, Paul (2008), “The impact of changes in spatial distance De Cesarei, Andrea and Codispoti, Maurizio (2006), “When on emotional responses,” Emotion, 8 (2), 192-98. does size not matter? Effects of stimulus size on affective Poels, Karolien and Dewitte, Siegfried (2006), “How to capture modulation,” Psychophysiology, 43 (2), 207-15. the heart? Reviewing 20 years of emotion measurement in Haws, Kelly L., Winterich, Karen Page and Naylor, Rebecca advertising,” Journal of Advertising Research, 46 (1), 18-37. Walker (2014), “Seeing the world through GREEN-tinted Pornpitakpan, Chanthika (2004), “The effect of celebrity endorsers’ glasses: Green consumption values and responses to perceived credibility on product purchase intention,” Journal environmentally friendly products,” Journal of Consumer of International Consumer Marketing, 16 (2), 55-74. Psychology, 24 (3), 336-54. Tversky, Amos and Kahneman, Daniel (1991), “Loss aversion in Hsee, Christopher K., Tu, Yanping, Lu, Zoe Y. and Ruan, Bowen riskless choice: A reference-dependent model,” Quarterly (2014), “Approach aversion: Negative hedonic reactions Journal of Economics, 106 (4), 1039-61. toward approaching stimuli,” Journal of Personality and Social Vallacher, Robin R. and Wegner, Daniel M. (1989), “Levels of Psychology, 106 (5), 699-712. personal agency: Individual variation in action identification,” Loftus, Geoffrey R. and Harley, Erin M. (2005), “Why is it easier to Journal of Personality and Social Psychology, 57 (4), 660-71. identify someone close than far away?” Psychonomic Bulletin & Review, 12 (1), 43-65. More than Just a Number: The Negative Effect of 100% Claims Nira Munichor, Bar-Ilan University, Israel Liat Levontin, Technion, Israel

EXTENDED ABSTRACT fect, we used a foreign product–a Russian jam, such that participants Many product labels include numerical claims. Consumers in- could only understand the numbers, but not the text, written on the terpret these claims in different ways, beyond their numerical value. label. Put differently, the text on the label was exaggeratedly unin- Non-round numbers, for example, are considered more informative formative for participants. A MANOVA revealed significant differ- and scientific than round numbers (Guang-Xie and Kronrod 2012). ences in product evaluations (F(2, 150)=4.30, p=.015) and percep- Correspondingly, evaluations of products associated with non-round tions of the product’s consumers (F(2, 150)=4.77, p=.01). Product numbers are based more on cognitions, whereas evaluations of prod- evaluations were lower with a 100% claim (M=3.42, SD=1.16) than ucts associated with round numbers lean more on feelings (Wadhwa with either a 99% claim (M=4.06, SD=1.26, p=.02) or a non-numeric and Zhang 2015). Round numbers, on the other hand, symbolize (non-understandable) claim (M=3.89, SD=.97, p=.11). Perceptions completion, and consumers are therefore more willing to accept of- of the jam’s consumers were also less favorable with a 100% claim fers with round prices (Yan and Pena-Marin 2017). (M=3.02, SD=1.06) than with either a 99% claim (M=3.51, SD=1.16, In the current research we focus on a common yet underex- p=.06) or a non-numeric claim (M=3.62, SD=.93, p=.01). plored numerical claim, a 100% claim (e.g., “100% natural”). More If the effect of a pseudo-informative 100% claim is indeed trig- than any other round number, 100% mathematically denotes com- gered by decreased consideration of the numerical value of 100%, pleteness, and accordingly, in everyday language, the term “100%” and enhanced consideration of its symbolism, then our effect should stands for completeness, fullness or perfection (Lee 2014). Thus, it emerge also when a 100% claim is compared to other, superior, claims is not surprising that marketers intuitively believe that a 100% claim involving percentages that are symbolic rather than numerically infor- can make products more appealing, an intuition reflected in their ex- mative. Thus, Study 3 used a 101%-satisfaction claim. Note that while tensive use of such claims. Indeed, Canadian consumers perceive a 101% has no numerical logic, a 101%-satisfaction claim may imply “100% Canadian milk” claim on milk and ice cream as an indication extra effort to satisfy consumers, which consumers may find more con- of product quality (Forbes-Brown et al. 2016). vincing. Participants (n=81) reviewed a hotel picture that contained We suggest that this intuition may backfire when the 100% either a 100% or 101% “satisfaction guarantee” claim. As expected, claim is pseudo-informative, namely, does not convey meaningful evaluations of the hotel were lower with a 100% claim (M=4.94, information regarding the associated product—for example, a claim SD=0.91) than with a 101% claim (M=5.36, SD=0.84; F(1,79)=4.53, stating that a drink contains “100% juice”, which does not specify p=.04), and perceptions of the hotel’s consumers were less favorable percentages of fruit content or other similar information. In these (M=3.92, SD=1.10 vs. M=4.46, SD=1.17, F(1,79)=4.61, p=.035). cases, consumers may rely less on the numerical information con- Study 4 aimed to provide more direct evidence for our hypoth- veyed by the claim, and refer more to its symbolism. In addition, esis that consumers rely more heavily on the symbolism, rather than because the perfection symbolized by the term “100%” might be per- the numerical value, of a pseudo-informative 100% claim. Thus, ceived as “cheap talk” that cannot actually be measured, consumer’s Study 4 tested whether measurability perceptions mediate the claim product evaluations decrease. effect on behavioral intentions (the mean of willingness to taste and Four experiments show that participants evaluate products less willingness to buy the product, Cronbach’s α=.75). Participants favorably when those products carry a pseudo-informative 100% (n=200) reviewed a picture of ice-bars that included one of three claim, as compared with a 99%, non-numerical, and even 101% uninformative claims: “Made with [100%/99%/no number men- claim (that serves as a symbol). We further show that the effect of a tioned] juice blend and other added ingredients.” The results of a pseudo-informative 100% claim is triggered by perceptions of low mediation analysis (PROCESS Model 4, multicategorial, with 5000 measurability of the information embedded in the claim, suggesting resamples; Hayes 2013) show that both the 99% claim and the non- an enhanced focus on 100% as a symbol. numeric claim were perceived as more measurable than the 100% Study 1 tested the effect of a pseudo-informative 100% claim claim (B=.63, SE=.28, p=.03, 95% CI [.07,1.18] and B=.63, SE=.27, on evaluations. Participants (n=99) reviewed a picture of a beverage p=.025, 95% CI [.08,1.18], respectively). Importantly, the relative with a label that contained either a “100% juice” or a “99% juice” indirect effects of the difference between the 100% claim and both claim. They provided product evaluations by indicating on 7-point the 99% claim and the non-numeric claim on behavioral intentions scales how appealing, tasty, healthy, fresh, natural, good versus through measurability were significant (B=.13, SE(boot)=.07, 95% bad, of high versus low quality, and close to perfection this product CI [.02, .31] and B=.13, SE(boot)=.07, 95% CI [.02,.30], respective- seemed to them (Cronbach’s α=.89). Participants also reported their ly), suggesting that the negative effect of a 100% claim on willing- perceptions of the product’s consumers: how successful, powerful, ness to adopt the product ensues from perceptions of lower measur- and high versus low in status they perceived them to be (Cronbach’s ability of that claim. α=.89). As expected, when the label contained a 100% claim, prod- Our results suggest that using 100% claims for products may uct evaluations were lower (M=4.36, SD =1.17) than when the label sometimes have unintuitive effects on consumers, which may yield contained a 99% claim (M=4.81, SD=1.07; F(1, 97)=3.86, p=.05), negative effects. and perceptions of the product’s consumers were less favorable (M=3.76, SD=1.16 vs. M=4.19, SD=.97; F(1, 97)=3.93, p=.05). REFERENCES Thus, a product label containing a 100% claim can sometimes harm Forbes-Brown, Shelicia, Eric T. Micheels, and Jill E. Hobbs product evaluations. (2016), “Consumer Willingness to Pay for Dairy Products Study 2 (n=153) compared the effect of a 100% claim not only With the 100% Canadian Milk Label: A Discrete Choice to that of a 99% claim, but also to that of a non-numeric claim. To Experiment,” Journal of International Food & Agribusiness make sure the (uninformative) numbers on the label triggers our ef- Marketing, 28 (3), 203-24.

Guang-Xin, Xie and Ann Kronrod (2012), “Is the Devil in the Wadhwa, Monica, and Kuangjie Zhang (2015), “This Number Just Details? The Signaling Effect of Numerical Precision Feels Right: The Impact of Roundedness of Price Numbers on in Environmental Advertising Claims,” Journal of Product Evaluations,” Journal of Consumer Research, 41 (5), Advertising, 41 (4), 103-17. 1172-85. Hayes, Andrew F. (2013), Introduction to Mediation, Moderation, Yan, Dengfeng, and Jorge Pena-Marin (2017), “Round Off the and Conditional Process Analysis. New York, NY: The Bargaining: The Effects of Offer Roundness on Willingness to Guilford Press. Accept,” Journal of Consumer Research, 44 (2), 381-95. Lee, Christopher T. (2014), “Consumer Linguistics: A Markedness Approach to Numerical Perceptions,” unpublished dissertation, Marketing Department, University of Oregon, Eugene, OR. Spreading of Alternatives Without a Perception of Choice Kurt P. Munz, New York University, USA Vicki G. Morwitz, New York University, USA

EXTENDED ABSTRACT assigned to advocate for one brand versus a competitor in a role- Choosing a product changes a consumer’s attitudes toward it. playing “sales game.” At time 1, we measured attitudes toward vari- Brehm (1956) classically demonstrated this by asking consumers to ous retailer brands. At time 2 two weeks later, we randomly selected first rate several products, and then choose between two they had two brands each participant had rated equally, and then randomly as- rated about equally. He found that attitudes toward chosen products signed one as “assigned” and the other as “competitor.” Participants improved and rejected products declined upon subsequent measure- believed they were to enter a chat room to attempt to convince an- ment. He explained this “post-choice spreading of alternatives” other participant to shop at their assigned retail brand. They would be (Brehm, 1956) by cognitive dissonance (Festinger, 1957). Specifi- in competition against a third participant trying to sell the competitor cally, the negative features of the chosen option became inconsistent brand. The chat room never occurred. Participants never actually ad- with the knowledge of having chosen them. The resulting unpleasant vocated for their assigned brand, nor was any additional information feeling of dissonance motivated choosers to downplay the impor- provided about the brands. Immediately after assignment to condi- tance of those features, allowing them to be happier with their choice tion, we re-measured attitudes toward the retailers. As predicted, we (Simon, Greenberg, & Brehm, 1995). observed a significant interaction F(1,47)=5.7, p =.021. Attitudes Researchers took Brehm’s “free-choice paradigm” (1956) as toward assigned brands improved (M=2.3, SD=2.1 from M=1.3, evidence that choosing causally changes attitudes (Festinger, 1957, SD=2.5), F(1,47)=6.7, p =.013, but attitudes toward competitors did 1964). Recently however, Chen and Risen (2010) argued that the not (M=1.1, SD=2.6 from M=1.2, SD=2.5), F<1. paradigm contains a methodological flaw. To demonstrate that choice Study 2 demonstrated that outcome assignment led to chang- causally affects attitudes, they argued that one would need to ensure es in feature importance. Modeled after hotwire.com, participants the choice outcome could not be driven by existing preference (2010; (N=175) in a longitudinal experiment booked a hypothetical ho- Risen & Chen, 2010). tel room through a web site where they chose only the neighbor- In response, several researchers demonstrated post-choice hood of the hotel; the web site assigned them a specific hotel in that spreading while randomly assigning outcomes rather than allowing neighborhood without further input. At time 1, we measured base- participants to make actual choices (Egan, Bloom, & Santos, 2010; line hotel feature importance. At time 2 we manipulated whether an Johansson, Hall, Tarning, Sikstrom, & Chater, 2014; Sharot, Velas- assigned hotel performed well on two secondary features: service quez, & Dolan, 2010). For example, Sharot et al. asked participants and amenities. We expected that self-reports of the importance of to make a choice based on information ostensibly presented sublimi- the manipulated features should reflect the level assigned. As pre- nally. The “choice” outcome was actually randomly assigned. When dicted, on a 100-point bipolar tradeoff scale we observed significant participants believed they had made a choice, their attitudes toward spreading in relative feature importance to match hotel assignments chosen options improved, but when they believed a computer had F(1,173)=7.97, p=.005. On independent 7-point scales, poorly per- chosen, there was no change in attitude (Sharot et al., 2010). The forming features were trivialized F(1,173)=7.22, p=.008. Specifi- authors concluded that a “sense of agency” was critical to observ- cally, poorly performing attributes decreased in importance (p’s<.05) ing spreading. Similarly, following a procedure involving altering but well-performing attributes remained unchanged (n.s.). choices by sleight of hand, others have concluded that “it is not the Study 3 showed moderated mediation. Two weeks after measur- choice per se that drives the preference change but rather the belief ing baseline attitudes and feature importance, participants (N=410) that a certain choice has been made” (Johansson et al., 2014, p. 288). were told their employer had booked a hotel room for them for a These authors interpreted their results as consistent with self-percep- business trip. Replicating study 1, attitudes spread to favor assigned tion theory (Bem, 1972), where people infer their own preferences hotels F(1,408)=17.67, p<.001. Replicating study 2, on a 100-point from the choices they have made. bipolar tradeoff scale we observed significant spreading in relative In contrast, we argue that neither choice nor the self-perception feature importance to match hotel assignments F(1,408)=23.80, of having a made a choice is required to observe spreading of al- p<.001. On independent 7-point scales poorly performing features ternatives. Instead, spreading occurs whenever a consumer accepts were trivialized F(1,408)=52.61, p<.001, also replicating study 2. an outcome. Consumers normally accept the outcomes of their own We additionally measured acceptance of the outcome by agreement choices, but they can also accept outcomes without believing they with four measures. Acceptance moderated the amount of spreading chose them personally. For example, when assigned to sell a product in feature importance F(1,406)=14.11, p<.001, such that spreading (study 1), people must accept that single course of action in order to was more pronounced as acceptance rose. Changes in importance act effectively (Harmon-Jones, Harmon-Jones, & Levy, 2015). Con- mediated changes in attitude, a significant moderated mediation 95 sumers may also accept outcomes when they use a decision process CI: [-.60,-.14]. that does not allow agency over the final outcome (study 2) or when Across three studies, we found that spreading of alternatives they accept an employer’s prerogative to assign them a product for does not require a self-perception of choice. Rather, spreading de- use in their work (study 3). We show that, consistent with the process pends on accepting an outcome. Consistent with the logic of post- presumed for post-choice spreading, consumers trivialize the impor- choice spreading, changes in feature importance mediated changes in tance of the negative features of the assigned outcomes (studies 2 and attitude. Mere acceptance seems to yield the same attitude reapprais- 3). These changes in importance mediate changes in attitude (study als currently attributed to choice. 3). Finally, individual levels of acceptance moderate the intensity of spreading (study 3). REFERENCES Study 1 demonstrated spreading (attitude improvement) fa- Bem, D. J. (1972). Self-perception theory. Advances in voring an assigned outcome. Participants (N=48) were randomly Experimental Social Psychology, 6, 1–62.

Brehm, J. W. (1956). Postdecision changes in the desirability of Johansson, P., Hall, L., Tarning, B., Sikstrom, S., & Chater, N. alternatives. The Journal of Abnormal and Social Psychology, (2014). Choice blindness and preference change: You will like 52(3), 384–389. https://doi.org/10.1037/h0041006 this paper better if you (Believe You) chose to read it! Journal Chen, M. K., & Risen, J. L. (2010). How choice affects and reflects of Behavioral Decision Making, 27(3), 281–289. https://doi. preferences: Revisiting the free-choice paradigm. Journal of org/10.1002/bdm.1807 Personality and Social Psychology, 99(4), 573–594. https:// Risen, J. L., & Chen, M. K. (2010). How to Study Choice-Induced doi.org/10.1037/a0020217 Attitude Change: Strategies for Fixing the Free-Choice Egan, L. C., Bloom, P., & Santos, L. R. (2010). Choice-induced Paradigm. Social and Personality Psychology Compass, 4(12), preferences in the absence of choice: Evidence from a blind 1151–1164. https://doi.org/10.1111/j.1751-9004.2010.00323.x two choice paradigm with young children and capuchin Sharot, T., Velasquez, C. M., & Dolan, R. J. (2010). Do monkeys. Journal of Experimental Social Psychology, 46(1), decisions shape preference? Evidence from blind choice. 204–207. https://doi.org/10.1016/j.jesp.2009.08.014 Psychological Science, 21(9), 1231–1235. https://doi. Festinger, L. (1957). A theory of cognitive dissonance. Row, org/10.1177/0956797610379235 Peterson, and Company. Simon, L., Greenberg, J., & Brehm, J. W. (1995). Trivialization: Festinger, L. (1964). Conflict, Decision, and Dissonance (1967th the forgotten mode of dissonance reduction. Journal of ed.). Stanford: Stanford University Press. Personality and Social Psychology, 68(2), 247–260. https:// Harmon-Jones, E., Harmon-Jones, C., & Levy, N. (2015). An doi.org/10.1037/0022-3514.68.2.247 action-based model of cognitive-dissonance processes. Current Directions in Psychological Science, 24(3), 184–189. https://doi.org/10.1177/0963721414566449 When Novices have more Influence than Experts: Empirical Evidence from Online Peer Reviews Peter Nguyen, Ivey Business School, Canada Xin (Shane) Wang, Western University, Canada Xi Li, City University of Hong Kong, China June Cotte, Ivey Business School, Canada

EXTENDED ABSTRACT both levels. Is it simply the case that expert reviewers always have Over the past decade, online peer review platforms have played greater influence on consumers than their novice counterparts? If not, a major role in shaping consumer choice. Eliciting and displaying in what case might this not be true, or perhaps even reversed? Why? online reviews has become a priority for many firms as part of their marketing efforts to influence consumer decisions and stimulate firm Hypotheses sales. Just as firms have been interacting with expert professionals Consistent with extant research on expertise (Alba and Hutchin- (Biswas, Biswas, and Das 2006; Wang 2005), firms are now also son 1987), because of their abundant practice in generating prior re- engaging with expert, or elite, users on peer review platforms (Chae views, expert reviewers are more skillful than their novice counter- et al. 2016; Stone 2014). Businesses invite expert users to try their parts (Solomon 1990). Therefore, we expect that at the individual products/services, with the objective of having expert users create level of reviews, reviews by expert users will be more influential, as and share high-quality reviews. Given the increasing engagement be- judged by readers, than reviews by novice users (H1). tween businesses and elite users, we are interested in the relationship However, at the aggregate level of reviews (volume and va- between user expertise and influence of user-generated reviews on lence), we suspect the opposite to be true, where novice users have consumers. more influence than their expert counterparts. With regard to the To get a more descriptive sense of how consumers are influenced volume measure, because of the Pareto distribution (Arnold 2015), by online peer review platforms, let’s consider a situation where a where there are substantially more novice users than expert users, we consumer accesses a review website to make a choice on a service predict that for any given service provider, under any (recent) time provider. Perhaps the consumer is at a conference in a new city, and frame, there will be more reviews generated by novice users than by decides to use a peer review website, such as tripadvisor.com, to find expert users. This would suggest that novice users are the main driv- a restaurant for the evening. Consider the navigation process from ers of increasing the volume measure (H2). accessing the user-generated review website to selecting a restaurant In the case of the valence measure, because people tend to for the evening. She might begin by first narrowing down the options simplify complex information in a categorical/dichotomous manner with a number of filters, including price range (cheap eats, mid- (e.g., good versus bad; Billig and Tajfel 1973; Gutman 1982) and range, fine dining), meal type (breakfast, lunch, brunch, dinner), and experiences tend to lead to a more nuanced perspective (Galinsky regions in the city which are located nearby. Even after the initial fil- and Moskowitz 2000), we hypothesize that for any given experience, ters, she is likely presented with tens, if not, hundreds of restaurants. novice users assign more polarizing ratings (i.e., ratings that are more In order to narrow the list of restaurants for her consideration set, she distant from the midpoint of the rating scale) than expert users (H4). might use aggregate-level measures, such as the volume of reviews As a result, we expect that novice users will have a greater tendency and the user rating average; considering only restaurants that have in shifting the existing average valence, especially when the average lots of reviews and are highly rated. Finally, from the handful of res- valence is close to the midpoint of the rating scale (H3). taurants left in consideration, the consumer may read several reviews of each in order to help finalize a choice of restaurant for the evening. Hypothesis 1 At the individual level of reviews, reviews by This vignette highlights an important feature – a multi-stage expert (vs. novice) users are more influential on filtering approach, from aggregate-level to individual – in the con- consumers. sumer decision-making journey (Boatwright and Nunes 2001). This filtering approach explains why, although there are hundreds of thou- Hypothesis 2 Novice (vs. expert) users play a greater role in sands, if not millions, of peer reviews on a review platform for res- driving up the volume measure. taurants within a given city, consumers do not read even a modest portion of them all. Aggregate level measures, such as the number of recent reviews and their average valence, affect the upstream rank- Hypothesis 3 Novice (vs. expert) users play a greater role in ing algorithm (Ghose, Ipeirotis, and Li 2012), which determines the shifting the valence measure up/down. rank of each service provider and the page on which they appear. Aggregate level measures are also used by consumers (Dai et al. Hypothesis 4 Novice (expert) users adopt a more polarizing 2017; Fisher, Newman, and Dhar 2018), which can narrow down (gradient) rating approach relative to their their consideration set. Finally, a handful of reviews from each of the counterparts. considered service providers are read to finalize consumer choice. Extant research on the impact of online reviews has primarily only considered reviews at either the aggregate (Babić Rosario et al. Hypothesis 5 For service providers that generally provide 2016; Chevalier and Mayzlin 2006; Duan, Gu, and Whinston 2008) moderately high [moderately low] quality ser- or the individual levels (Moore 2015; Mudambi and Schuff 2010; vice, novices assign significantly higher [lower] Packard and Berger 2017). In this research, we examine online peer ratings than experts. reviews in terms of reviewer influence on consumers at the aggregate and the individual levels, and investigate the relationship between reviewer expertise and influence of reviews on consumers at

Advances in Consumer Research 719 Volume 46, ©2018 720 / When Novices have more Influence than Experts: Empirical Evidence from Online Peer Reviews Study 1: Qunar (Field Data) they have visited [electronic products they have purchased] over the The purpose of Study 1 is to test whether expert or novice re- past year (note that the dependent measure is specific to restaurants). viewers are more influential on consumers at the individual and ag- Participants were then presented with a description of a moderately gregate levels of online peer reviews. We scraped and analyzed over positive or moderately negative experience at a restaurant and then 130,000 reviews on 60 hotels, spanning 4 cities from Qunar.com, a asked to assign a star rating for the experience. major Chinese-based reputation platform. A two-way analysis of variance revealed a significant interac- Our main independent variable of interest is user expertise, tion between contributor level and experience valence on assigned which is defined as the number of reviews the user has generated on star rating (F(2,187) = 5.68; p < .05). For the moderately positive the platform prior to posting the current review. Our dependent vari- experience description, participants primed as expert raters assigned able of influence at the individual level is review favorability. Our marginally lower ratings (M = 4.35, SD = 0.64) than those primed as dependent variables of influence at the aggregate levels are review novice raters (M = 4.57, SD = 0.50; t(1,94) = 1.89, p = .06). For the volume and rating-average distance. We also assess differences in moderately negative experience description, there was no significant rating pattern with the rating polarity measure. For the operational difference in ratings between expert and novice raters (MNovice = definitions of the dependent variables. 1.61, SD = 0.57 vs. MExpert = 1.84, SD = 0.75; p = .13). Most impor- To test the relationship between user expertise and the different tantly, consistent with H4, we find that participants primed as expert dependent variables, we conducted an analysis of covariance (AN- raters assign ratings that were less polarizing (M = 1.29 average units COVA) for each. Across our tests, we include a number of control from the midpoint of a five-point scale,SD = 0.62) than those primed variables, including hotel identification, purpose of travel, and how as novice raters (M = 1.47, SD = 0.54, t = 2.12, p < .05). long ago the review was posted from the date the review data was Conclusion scraped. Using an experiment, we showed that priming rating expertise, We first find that, consistent with H1, users with more review- specifically, rating familiarity with a focal domain, reduces the ex- writing experience generate reviews that readers find more favor- tremity of ratings. This reflects the more gradient rating approach able (MExperts = 0.40 vs. MNovices = 0.32 average ‘like’ votes per favored by users with high rating expertise in the earlier field data. review; F(1, 132 979) = 14.76, p < .001). Next, we find that there is a The parallel findings between our field data in Study 1 and our exper- significant negative relationship between user expertise and volume iment data in Study 2 allows us confidently draw the causal relation- of reviews posted within the most recent six months (β = -846.7, t ship between more extensive rating experiences and less polarizing = -2.32, p < .05). Results are robust for reviews posted within most ratings, as well as rule out unobserved reviewer characteristics as recent one month, two months, and one year (all p’s < .05). Novices possible confounds of the effect. do indeed drive the volume of reviews (H2). Consistent with H3, we find that users with less review-writing experience on average assign Study 3: TripAdvisor (Field Data) ratings that are further from the average, for hotels with moderate The purpose of Study 3 is twofold. First, we test whether our averages (M = 0.88 vs. M = 0.82 units from user rating Novices Experts findings from Study 1 from the Chinese-based platform Qunar.com existing average valence; F(1, 22 819) = 29.85, p < .001). Finally, replicate in a North American-based platform, TripAdvisor.com. as per H4, we find that novice users assign more polarizing ratings Second, we test whether the more polarizing rating approach by their expert counterparts (M = 1.64 vs. M = 1.55 than Novices Experts novice users is systematic, such that for certain service providers, average units from the midpoint of a five-point scale; F(1, 132 979) novice reviewers systematically assign higher and lower ratings than = 772.69, p < .001). expert reviewers. We scraped and analyzed over 100,000 reviews Conclusions on 60 hotels, spanning 6 cities, on tripadvisor.com, a major online In Study 1, using Qunar hotel review data, we demonstrate that North-American-based travel review platform. The same variables at the individual review level, expert (vs. novice) users have greater are collected as those in Study 1. influence, as judged by reader assessment. However, at the aggre- First, for replication of findings from Study 1. Second, we test gate level, we find that opposite effect, novice (vs. expert) users have whether, for certain service providers, novice reviewers systemati- greater influence, in terms of driving up the volume of reviews and cally assign higher and lower ratings than expert reviewers. We find shifting the average valence. We demonstrate that the shifting of the that for hotels that generally provide moderately high quality service average valence can be explained by novice users’ tendency to assign (average valence of 4.4-4.6), novice users systematically assigned more polarizing ratings relative to their expert counterparts. higher ratings (M = 4.53) than expert users (M = 4.43; β = -0.056, t = -7.16, p < .001), and therefore are systematically pushing up the Study 2: Priming Experienced Raters (Experiment) average valence. However, for hotels that generally provide moder- The purpose of Study 2 to test the effects of user expertise on ately low quality service (average valence of 3.5-3.7), novice users patterns of assigned star ratings in an experiment, by priming par- systematically assigned lower ratings (M = 3.27) than expert users ticipants as either expert or novice raters. User expertise (expert vs. (M = 3.40; β = 0.060, t = 2.50, p = .01), and therefore are systemati- novice) was manipulated by having participants recall and evaluate cally pulling down the average valence. memories of experiences either related or unrelated to the focal ser- Conclusions vice, prior to assigning a rating to the focal experience. The design In Study 3, using hotel review data from a North American- of the experiment is a 2 rating expertise (high vs. low) x 2 descrip- based review platform, TripAdvisor.com, we replicate the main re- tion valence (positive vs. negative) between-subjects design. The sults from Studies 1 and 2. More importantly, we show that the more outcome measure in the experiment is the assigned star rating, along polarizing rating approach by novice users is not random, but sys- a 5-point scale from 1-Terrible to 5-Excellent. tematic, such that for service providers that generally provide mod- Online participants (N = 190, %female = 56.3%, MAge = 35.0, erately high [moderately low] quality service, novice users assign SDAge = 11.1) on Amazon Mechanical Turk were randomly assigned significantly higher [lower] ratings than expert users, and therefore, to the beginner or veteran contributor level. Participants assigned to the expert (novice) condition were asked to think about restaurants Advances in Consumer Research (Volume 46) / 721 are systematically pushing up [pulling down] the user rating aver- Boatwright, Peter, and Joseph C. Nunes (2001), “Reducing ages. Assortment: An Attribute-Based Approach,” Journal of Marketing, 65 (3), 50-63. General Discussion Billig, Michael, and Henri Tajfel (1973), “Social Categorization In our research, we examined the influence of expert and novice and Similarity in Intergroup Behaviour,” European Journal of users on online peer review platforms at the individual and aggregate Social Psychology, 3 (1), 27-52. review levels. In Study 1 (Qunar), we show that at the individual Biswas, Dipayan, Abhijit Biswas, and Neel Das (2006), “The review level, expert users have greater influence, as judged by read- Differential Effects of Celebrity and Expert Endorsements ers’ assessments. on Consumer Risk Perceptions. The Role of Consumer However at the aggregate review level, novice users have great- Knowledge, Perceived Congruency, and Product Technology er influence, in terms of driving the volume of reviews and shifting Orientation,” Journal of Advertising, 35 (2), (2006): 17-31. the average valence. Further, we demonstrate that novice users’ shift- Chae, Inyoung, Andrew T. Stephen, Yakov Bart, and Dai Yao ing of the average valence can in part be explained by their more (2016), “Spillover Effects in Seeded Word-of-Mouth polarizing rating approach relative to their expert counterparts. In Marketing Campaigns,” Marketing Science, 36 (1), 89-104. Study 2, testing the idea that novice users adopt a more polarizing Chevalier, Judith A., and Dina Mayzlin (2006), “The Effect of approach, we validate the causal relationship between lack of rat- Word of Mouth on Sales: Online Book Reviews,” Journal of ing experience and rating polarity, and rule out unobserved reviewer Marketing Research, 43 (3), 345-54. characteristics as possible confounds. In Study 3 (TripAdvisor), we Dai, Weijia (Daisy), Ginger Zhe Jin, Jungmin Lee, and Michael replicate the findings from Study 1. Also, we show that for moder- Luca (2017), “Aggregation of Consumer Ratings: An ately high [moderately low] service providers, novice users assign Application to Yelp.com,” Quantitative Marketing and higher [lower] ratings than experts. Economics. Although it is generally thought that experts play a more domi- Duan, Wenjing, Bin Gu, and Andrew B. Whinston (2008), “Do nant role in shaping consumer opinion (Alba & Hutchinson 1987), Online Reviews Matter? An Empirical Investigation of Panel the present research demonstrates this is not always the case. With Data,” Decision Support Systems, 45 (4), 1007-16. the flourishing of user-generated content, consumers are less likely Fisher, Matthew, George E. Newman, Ravi Dhar, and Stijn van to access all the available content. Instead, they rely more and more Osselaer (2018), “Seeing Stars: How the Binary Bias Distorts on aggregate level measures that summarize the abundance of con- the Interpretation of Customer Ratings,” Journal of Consumer tent to guide their consumption choice (Dai et al. 2017; Fisher et al. Research. 2018). Our results show that novices have more impact at the aggre- Galinsky, Adam D., and Gordon B. Moskowitz (2000), gate level measures in terms of driving the volume of reviews and “Perspective-Taking: Decreasing Stereotype Expression, shifting the average valence. Our results suggest that firms should be Stereotype Accessibility, and In-group Favoritism,” Journal of cautious, and perhaps even hesitant, when focusing on having expert Personality and Social Psychology, 78 (4), 708-24. users generate reviews about their product/service. Further, firms Ghose, Anindya, Panagiotis G. Ipeirotis, and Beibei Li (2012), that generally provide moderately high quality service may consider “Designing Ranking Systems for Hotels on Travel Search completely avoiding expert users, as experts assign significantly Engines by Mining User-Generated and Crowdsourced lower ratings than novices, and therefore, relative to their novices, Content,” Marketing Science, 31 (3), 493-520. are systematically pulling down the average valence. Gutman, Jonathan (1982), “A Means-End Chain Model Based To conclude, this research provides evidence, in the context of on Consumer Categorization Processes,” The Journal of user-generated content platforms, for when and how novices have Marketing, 46 (2) 60–72. greater influence on consumers and the systematically different rat- Moore, Sarah G. (2015), “Attitude Predictability and Helpfulness ing approaches adopted by expert and novice users. The findings in Online Reviews: The Role of Explained Actions and from our study are important to service providers, particularly as Reactions,” Journal of Consumer Research, 42 (1), 30-44. consumers move away from traditional offline media and towards Mudambi, Susan M., and David Schuff (2010), “What Makes a online digital media where user-generated content plays an increas- Helpful Review? A Study of Customer Reviews on Amazon. ingly larger role in shaping consumer choice. com,” MIS Quarterly, 34 (1), 185-200. Packard, Grant, and Jonah Berger (2017), “How Language Shapes REFERENCES Word of Mouth’s Impact,” Journal of Marketing Research, 54 Alba, Joseph W., and J. Wesley Hutchinson (1987), “Dimensions of (4), 572-88. Consumer Expertise,” Journal of Consumer Research, 13 (4), Stone 2014, “Elite Yelpers Hold Immense Power, and They Get 411-54. Treated Like Kings by Bars and Restaurants Trying to Curry Arnold, Barry C. (2015), Pareto Distribution. John Wiley & Sons, Favor”. Business Insider. http://www.businessinsider.com/ Ltd. how- to-become-yelp-elite-2014-8 Babić Rosario, Ana, Francesca Sotgiu, Kristine De Valck, and Wang, Alex (2005), “The Effects of Expert and Consumer Tammo HA Bijmolt (2016), “The Effect of Electronic Word Endorsements on Audience Response,” Journal of Advertising of Mouth on Sales: A Meta-Analytic Review of Platform, Research, 45 (4), 402-12. Product, and Metric Factors,” Journal of Marketing Research, 53 (3), 297-318. Does It Pay to Be Virtuous? Examining Whether and Why Firms Benefit From Their CSR Initiatives Dionne Nickerson, Georgia Institute of Technology, USA Michael Lowe, Georgia Institute of Technology, USA Adithya Pattabhiramaiah, Georgia Institute of Technology, USA

EXTENDED ABSTRACT uct category that did not announce CSR initiatives. We thus use the Formally defined as discretionary business practices and contri- synthetic control method to examine this effect (Abadie, Diamond, butions of corporate resources intended to improve societal well-be- and Hainmueller 2010). The intuition behind this approach is that it ing (Korschun, Bhattacharya, and Swain 2014); corporate social re- creates a “synthetic control unit,” computed as a weighted combina- sponsibility (CSR) is increasingly present in consumer consciousness. tion of all control units, to compare against the treated unit. In fact, we often see consumers seeking out and supporting brands that We find that, on average, corrective and compensating CSR ac- they perceive as contributing to the greater societal good, incentiviz- tions are associated with a sales increase, whereas cultivating CSR ing brands to engage in CSR (Hughes 2016). actions are associated with a sales decrease. We show that this ef- While CSR is increasingly important to brand strategy, specific fect differs according to consumers’ prior knowledge of the brand, guidance on the nature of activities that a brand should undertake, measured as brand reputation. The results suggest that, for brands and a documentation of the ensuing returns, are lacking. Detecting with lower reputations, corrective CSR leads to an overall increase in potential differences in consumer response to CSR activities might brand sales. The pattern of increases in brand sales is similar for low- have been challenging in prior research for three important reasons. er reputation brands that undertake compensating CSR. In contrast, First, prior work has typically not distinguished between particular both higher and lower reputation brands experience sales decreases types of CSR investment (Jayachandran, Kalaignanam, and Eilert after engaging in cultivating CSR. 2013). Second, research has centered around financial performance We explore the mechanism behind these sales effects in a con- measures such as firm market value and stock response - which are trolled laboratory setting. An ANOVA (N=148) examining the im- more separated from direct consumer involvement (Luo and Bhat- pact of the CSR on brand attitudes using brand reputation (higher tacharya 2006; Mishra and Modi 2016). Third, other work has fo- vs. lower) and CSR type (corrective, compensating, or cultivating), cused on documenting the influence of CSR campaigns by looking at shows that the significance of the impact of a CSR activity depends measures of consumers’ intentions and attitudes (Chernev and Blair on the a priori reputation of the brand involved. Consistent with the 2015). To our knowledge, our paper forms the first attempt at lever- model, the most positive consumer response was for “corrective” ac- aging field data to offer direct empirical support for the existence of tions undertaken by a lower reputation brand, a response that was benefits to firms from engaging in different types of CSR, because of significantly higher than an identical corrective action undertaken by direct consumer response. a higher reputation brand (Mlower/corrective = 1.52, Mhigher/corrective = .474; We propose a new framework for managing CSR that is rooted F(1,142) = 14.53, p <.001). This was also true for compensating ac- in the perspective of consumer-brand relationships, by drawing on the tions (Mlower/compensating = 1.30, Mhigher/compensating = .464; F(1,142) = 8.63, concepts of forgiveness in relationships. We distinguish between three p = .004). There was no such difference between lower and higher types of CSR engagement: correcting for the negative societal impact reputation brands for “cultivating” CSR actions (Mlower/cultivating = .827, of a brand’s business operations, compensating for the negative soci- Mhigher/cultivating = .617; F(1,142) = .494, p = .483). etal impact of a brand’s business operations without making changes to We used a bootstrap analysis to test for the mediating effect of operations, and cultivating consumer goodwill through prosocial acts consumers’ perceptions of the sincerity of brands’ CSR actions. An unrelated to the negative societal impact of a brand’s business opera- ANOVA with CSR type and brand reputation predicting perceived tions. Conceptualizing CSR in this manner can be advantageous for sincerity had a marginal interaction (F(2, 146) = 2.68, p = .072), and brand managers, due to its direct focus on the influence of CSR on significant main effects of CSR type (F(1, 147) = 6.75, p = .002) and consumer actions. We argue that the three different types of CSR en- brand reputation (F( 1, 147) = 50.37, p < .0001). Overall, perceived gagement can potentially engender different consumer attributions of sincerity significantly mediated the moderated effect of CSR type on sincerity in a brand’s CSR efforts, and that these differences can be brand attitudes for both corrective-type actions (corrective: a1 x b1 important for understanding any potentially heterogeneous impact that = -.1989, 95% C.I. = -.5322 to -.0333, p < .05), and compensating each type of engagement may have on brand sales. CSR actions (compensating: a2 x b1 = -.1988, 95% C.I. = -.5355 to We take a mixed methods approach to understand these rela- -.0183, p < .05). These results reveal that consumer perceptions of tionships. We first leverage observational data to examine theex- brand sincerity significantly mediate the relationship between CSR istence of a causal impact of firms’ CSR announcements on brand type and brand attitude. sales. We then examine the underlying psychological mechanisms These results show that the payoffs from CSR vary depending that may be influencing changes in consumer purchase behavior in a on the type of CSR engagement and that this effect can be materially controlled experimental setting. significant. These findings contribute to the literature by investigat- To investigate the effect of CSR announcements on brand sales, ing CSR as a multidimensional construct and by empirically demon- we collect CSR press releases from prominent CPG (consumer pack- strating how firms’ CSR initiatives affect consumer responsiveness. aged goods) brands as well as detailed brand sales data. Our analysis dataset contains 33 brands that announced CSR initiatives and 100 REFERENCES brands that did not, spanning 15 CPG product categories. We com- Abadie, Alberto, Alexis Diamond and Jens Hainmueller (2010), pare differences in brand sales before and after CSR announcements “Synthetic control methods for Comparative Case Studies: for brands that announced CSR initiatives (“treated” brands) and Estimating the Effect of California’s Tobacco Control compare the analogous difference in brand sales for a closely compa- Program,” Journal of the American Statistical Association, rable set of brands (“control” brands) within the focal brand’s prod- 105(June), 493-505.

Chernev, Alexander and Sean Blair (2015), “Doing Well by Korschun, Daniel, C.B. Bhattacharya and Scott D. Swain (2014), Doing Good: The Benevolent Halo of Corporate Social “Corporate Social Responsibility, Customer Orientation, and Responsibility,” Journal of Consumer Research, 41(April), the Job Performance of Frontline Employees,” Journal of 1412-1425. Marketing, 78(May), 20-37. Cone Communications (2017), “2017 Cone Communications Luo, Xueming and C. B. Bhattacharya (2006), “Corporate Social CSR Study” http://www.conecomm.com/2017-cone- Responsibility, Customer Satisfaction, and Market Value,” communications-csr-study-pdf/. Journal of Marketing, 70(October), 1-18. Hughes, B. (2016), “Why Corporate Social Responsibility is Mishra, Saurabh and Sachin B. Modi (2016), “Corporate Social Essential for Brand Strategy,” Huffington Post. https://www. Responsibility and Shareholder Wealth: The Role of Marketing huffingtonpost.com/brian-hughes/why-corporate-social-resp b Capability,” Journal of Marketing, 80(January), 26-46. 9282246.html. Jayachandran, Satish, Kartik Kalaignanam and Meike Eilert (2013), “Product and Environmental Social Performance: Varying Effect on Firm Performance,” Strategic Management Journal, 34(October), 1255-1264. All We Need is Love: Examining Differences in Time and Money Donations between Dyads and Individuals Hristina Nikolova, Boston College, USA

EXTENDED ABSTRACT There was no difference in the monetary donations of dyads and in- Donations of both time and money are critical for societal well- dividuals, p=.36. being: in 2013, the dollar estimate of the annual volunteered time Study 2 replicated these findings using real monetary and time in the U.S. was $167 billion and the total monetary donations were donations. Participants (n=135) completed the study for $6 and were $335 billion. Time and money donations have important benefits assigned either to the individuals (n=55) or dyads condition (n=40 for consumers’ personal well-being as well: consumers who donate pairs). Participants read information about a charity called The World monetary/time resources are happier, have higher life satisfaction, Needs More Love Letters (www.moreloveletters.com). Participants self-esteem, better physical and mental health (Thoits & Hewitt, were asked to write an encouragement letter to a couple Jon and Jas- 2001). mine, who were both battling cancer. The number of words in the In the present research, we focus on the joint pro-social behav- letters participants wrote represented the time donation measure. Fi- ior of pairs of consumers (dyads) and compare it to that of individual nally, participants were given their compensation and a donation en- consumers. Specifically, we examine whether there are differences velope in which they could donate some of it to Jon and Jasmin. Re- between dyads and individuals in their donations of time and mon- sults showed that dyads (M=35.38, SD=29.76) wrote encouragement ey – two fundamental resources in consumers’ lives (Liu & Aaker, letters that contained less words than those of individuals (M=48.22, η 2 2008). SD=51.67), F(1, 90)=6.82, p=.01, p =.03, controlling for amount We propose that while dyads and individuals will make similar of time spent on the task (b=.37, t(90)=12.76, p<.0001). There was monetary donations, dyads will donate less time than individuals. no difference in the monetary donations of dyads and individuals, p We base this prediction in the functionalist model of donations pro- = .22. posed by Clary and colleagues (1998, 1999), who argue that donating In Study 3 (n=119), dyads (M=1.70 hours, SD=.96) were will- time serves six personal/social functions: ego protection (cope with ing to donate less hours to help an organization called HopeLab with life difficulties), value expression (express altruistic values), career a fundraising event than individuals (M=2.43 hours, SD=1.85), F(1, η 2 growth (improve career prospects), knowledge (gain knowledge/ 74)=4.90, p=.03, p =.06. There was no difference in the monetary skills), self-enhancement (enhance self-growth), and, most impor- donations of dyads and individuals, p=.42. Importantly, we find that tantly, social motives (develop/strengthen social ties). To the extent dyads (M=4.27, SD=1.46) perceived lower happiness of giving time that these needs are active and unmet, people will be more likely to than individuals (M=4.76, SD=1.42), F(1, 117)=2.99, p=.09. This donate time finding suggests that dyads might have a proximal satisfaction of On the other hand, donating money serves a more limited set their social needs (the other person), which reduces the value of out- of personal (not social) goals. Specifically, money donations have side opportunities for social utility. been most strongly linked to value expression, self-enhancement or In Study 4, we provide direct evidence of this mechanism us- identity-related motives (Aaker & Akutsu, 2009; Meier, 2007). Un- ing married couples. Participants (n=307) were recruited on MTurk like donating time, donating money cannot serve any social functions and asked to indicate their relationship status. Participants who were such as building social ties. single (n=131) represented the group of individuals and reported Thus, we argue that time donation decisions are rooted in the how many hours/what percent of income they donated each month. degree to which social needs are salient at the time of the choice. Participants who were married (n=176) represented the group of dy- Importantly, when individuals are making decisions in dyads, this ads; they reported the amount of time/money they donated jointly as joint decision context itself offers a proximal means of satisfying the a couple each month. All participants indicated the extent to which social needs (Nikolova et al., 2018) that time donations may other- they felt that their needs for social connections were satisfied. To ac- wise fulfill. In other words, in joint decision-making contexts dyads count for excess number of zeros on the time donation DV (78.4%), should be more likely to perceive that their needs for social connec- a Zero-Inflated Poisson regression was run: couples donated less tions are satisfied. As a result, the dyadic context temporarily lowers time each month than single participants (b=-.28, Wald χ2(1)=10.79, the importance of using the time donation decision to satisfy social p=.001). There was no difference in the percent of income donated needs. We predict that this should lead to lower time donations of by couples and individuals, p=.87. Importantly, couples (M=4.81, dyads relative to individuals. However, the goals fulfilled by money SD=1.47) believed that their social connection needs were satisfied donations, being less social, do not change when individuals decide to a larger extent than single participants (M=4.27, SD=1.81), F(1, alone or in dyads: individuals always desire to express their values 287)=7.98, p=.005, and the satisfaction of social connection needs and enhance the self, whether or not they are in a pair (Ariely et al., mediated the effect of decision-maker on time donation (b=.21, 2009). Thus, we do not anticipate any differences in the monetary SE=.12, 95%CI: .050, .555). donations of dyads and individuals. We test these predictions in four Taken together, this research contributes to the pro-social be- studies. havior literature by being the first to compare the joint pro-social In Study 1, participants (n=248) were either paired to make decisions of dyads and individuals, as well as offers important impli- joint decisions (n=98 pairs) or made the same decisions individually cations for organizations which rely heavily on volunteering couples (n=52). Pairs/individuals imagined that they were shopping together/ (e.g., schools counting on parents’ time donations). alone when they noticed a donation campaign for helping prema- ture babies. They indicated the amount of money ($)/time (hours) REFERENCES they were willing to donate to the campaign together/alone. Dyads Aaker, J. L. & Satoshi Akutsu (2009). Why do people give? The (M=1.49 hours, SD=1.69) were willing to donate less time than indi- role of identity in giving. Journal of Consumer Psychology, 19, η 2 267-270. viduals (M=2.65 hours, SD=2.58), F(1, 144)=10.66, p=.001, p =.07. Advances in Consumer Research 724 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 725

Ariely, D., Bracha, A., & Meier, S. (2009). Doing good or doing Meier, S. (2007). A survey on economic theories and field evidence well? Image motivation and monetary incentives in behaving on pro-social behavior. In Economics and Psychology: A prosocially. American Economic Review, 99(1), 544-555. Promising New Cross-Disciplinary Field, ed. Bruno S. Frey Clary, E. G. & Snyder, M. (1999). The motivations to volunteer: and Alois Stutzer, 51-88. Cambridge: MIT Press. Theoretical and practical considerations. Current Directions in Nikolova, H., Lamberton, C. & Coleman, N. V. (2018). Stranger Psychological Science, 8, 156-159. danger: when and why pairs of consumers make sub-ethical Clary, E. G., Ridge, R. D., Stukas, A. A., Snyder, M., Copeland, J., decisions. Journal of Consumer Research, 45(1), 90-108. Haugen, J., & Miene, P. (1998). Understanding and assessing Thoits, Peggy A., & Hewitt, L. N. (2001). Volunteer work and well- the motivations of volunteers: A functional approach. Journal being. Journal of Health and Social Behavior, 42(2), 115-131. of Personality and Social Psychology, 74(6), 1516-1530. Liu, W., & Aaker, J. (2008). The happiness of giving: The time‐ask effect.Journal of Consumer Research, 35(3), 543-557. The Psychology of Fun: Inquiry into Consumers’ Fun Experiences Travis Tae Oh, Columbia University, USA Michel Tuan Pham, Columbia University, USA

EXTENDED ABSTRACT uct offerings or situational properties were unfavorable to “let loose” One prominent pleasure construct that plays a vital role in peo- of oneself. ple’s consumption is fun. The experience of “having fun” is ubiqui- Given the importance of liberation as a major determinant in the tous among today’s consumers, who spend considerable amounts of fun construct and due to space limitations, we present a more detailed time and money on fun activities such as a week-long cruise. More- analysis only for this dimension below. over, firms frequently promote “fun” as an attractive promise in their Liberation advertising messages and strategically position themselves (e.g., Vir- The concept of liberation can be found similarly in the Epicu- gin Airlines) as fun brands. rean ideology of “ataraxia” which represents freedom from mental Despite its growing presence as a unique concept in real-world discomfort as the highest order of pleasure (O’Keefe 2010). While circles, fun has yet to receive adequate academic attention as a psy- meaningful consumption (i.e. “eudaimonia”) plays a significant role chological construct on its own right. Most of what is considered in people’s lives, we argue that fun as a modern day epithet of at- research on “fun,” has utilized the word as a proxy of experiential araxia is also critical to understanding well-being in consumer psy- utility (Babin et al., 1994) with little regard for its psychological con- chology. In particular, we identify three sub-categories of liberation tents (Alba & Williams, 2013; Pham 2013). Thus, the main goal of as (1) Freedom from worries, (2) Carefree mentality, and (3) Release this paper is to propose a theoretical framework on the psychology from tension. For instance, a 27 year old teaching assistant narrates of “having fun” in consumers’ lives, and present fun as a unique con- about her family trip on the Disney Cruise as, “Disney- I feel like it’s struct, differentiated from related concepts such as happiness. not just for kids… but just to see adults sing Frozen songs, and just To reach this objective, we adopt a multi-method approach with be… Carefree! Just to see people carefree, not caring what anybody an inductive focus, based on data consisting of 21 depth interviews, thinks, singing all the songs, getting on all these rides, I find that 335 narrative experiences, and 1,600 selfies. fun.”

METHODOLOGY FUN AS A DISTINCT PLEASURE CONSTRUCT To capture the full range of the phenomenon, we conducted We primarily focus on the differences between fun and happi- three types of studies in different modalities (oral, written, and picto- ness, as these two constructs are closely related, and have substantial rial), broadening the scope of our understanding of fun in multiple implications to consumption experiences (e.g., Nicolao et al. 2009). contexts. The results below are primarily from the narrative experiences data Depth Interviews. 21 interviews were conducted in the phe- and we leave out results from selfies due to space limitations. nomenological tradition (Thompson et al. 1989). Participants ranged from ages 18 to 73, were balanced in gender, and came from diverse Results socioeconomic backgrounds (e.g., homeless to retired). Participants in the happy condition reported the highest level Narrative Experiences. We recruited participants from an on- of meaningfulness (p < .01). People in the fun condition reported line panel and randomly assigned them to three conditions: Fun, that time progressed faster and was more fun than the happy condi- Happy, and Personally Interesting. After participants wrote about tion (all ps < .05). More importantly, participants in the fun condi- their respective experiences, they filled out various process measures tion reported significantly higher feelings of liberation (p < .001) and that were expected to be more or less representative of fun. connectedness (p = .08) than the happy condition. LIWC analysis Prompted Selfies. We partnered with a mobile market research showed that between conditions, “friends” was higher in the fun con- company that specializes in collecting selfie data. Through its app, dition (p = .06) and “family” was higher in the happy condition (p we recruited participants and randomly assigned them to one of four < .01). conditions (Fun, Happy, Proud, and Relaxed) to upload a selfie re- Discussion garding that particular moment. We find evidence that consumers distinguish fun experiences as In the following sections, we present (1) a psychological frame- more fun, less meaningful, and progressing faster than happy experi- work of consumer fun, and (2) how fun is distinct from related posi- ences. These results suggest that consumers readily differentiate the tive constructs. two constructs. Interestingly, people felt happier with their family and more fun with their friends, implying that the content of connect- EMERGING FRAMEWORK OF CONSUMER FUN edness matter, over and beyond social intimacy or belongingness. As shown in Figure 1, the cross-data analyses reveal six major psychological underpinnings of fun: (a) Novelty, (b) Spontaneity, DISCUSSION (c) Connectedness, (d) Boundedness, (e) Engagement, (f) Libera- This research presents a novel framework of fun, as a unique tion, and two affective layers (not shown in Figure): (g) Increasing construct. We also show that consumers readily distinguish fun as a Positive Arousal and (h) Decreasing Negative Arousal. In particular, unique construct, different from related concepts as happiness. among the six psychological concepts that contribute to fun, we find Based on these findings, we define fun as an affective pleasure that “Liberation” is the most crucial element in defining the fun con- construct identified by a strong sense of engaged liberation, which struct. This feeling of liberation is central to both the motivational often arises from bounded, novel experiences that involve moments and experiential aspects of fun- that is, consumers were unwilling to of spontaneity and connectedness with others. spend their resources on fun consumption if current circumstances Our framework and investigation contributes to the growing lit- were too restrictive (e.g., lack of money, exam week) or if the prod- erature on hedonic consumption by providing a content-rich insight into a highly relevant pleasure construct. Moreover, the multi-meth-

Advances in Consumer Research 726 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 727 od approach expands the epistemological lens traditionally used in Nicolao, Leonardo, Julie Irwin, and Joseph Goodman (2009), consumer research. Future studies could operationalize and manipu- “Happiness for Sale: Do Experiential Purchases Make late the underlying mechanisms to further validate the framework, Consumers Happier than Material Purchases?” Journal of and distinguish from other related concepts as humor and play. Consumer Research, 36 (August), 188–98. O’Keefe. (2010), Epicureanism, Berkley, CA: University of REFERENCES California Press. Alba, J. W., & Williams, E. F. (2013). Pleasure principles: A review Pham, M. T. (2013). The seven sins of consumer psychology. of research on hedonic consumption. Journal of Consumer Journal of Consumer Psychology, 23(4), 411–423. Psychology, 23(1), 2–18. Thompson, Craig J., William B. Locander, and Howard R. Pollio Babin, B. J., Darden, W. R., & Griffin, M. (1994). Work and/or Fun: (1989), “Putting Consumer Experience Back into Consumer Measuring Hedonic and Utilitarian Shopping Value. Journal of Research: The Philosophy and Method of Existential- Consumer Research, 20(4), 644. Phenomenology,” Journal of Consumer Research, 16 (2), 133–46. Do Altruistic Individuals “Share” More Contents on Social Media? Travis Tae Oh, Columbia University, USA Keith Wilcox, Columbia University, USA

EXTENDED ABSTRACT to present themselves favorably and accurately by “sharing” video The power of video contents to “go viral” has sparked much contents on social media. interest in managerial and academic circles. Despite the prevalence of the “Share” function in almost all online social network services STUDY 1 and its important implications in viral marketing, the psychological We investigate how altruism specifically relates to sharing on processes and motivational factors underlying the behavior of “shar- social media by measuring people’s actual behaviors on Facebook. ing” contents on social media have been understudied. We recruited 267 online participants to login to their Facebook Word-of-mouth literature shows that altruism is a viable pre- accounts and browse their “Activity Log” to report the number of ac- dictor of sharing information in consumer contexts (Sundaram et al. tivities (e.g., “Like”) conducted during the previous month. We also 1998), but it is unclear whether altruistic individuals would share measured altruism (Philippe Rushton et al. 1981), self-monitoring more contents on social media. Purely altruistic acts typically incur (Snyder 1974), and cooperativeness (Cloninger et al. 1993). Only a significant “cost” to the benefactor, but sharing on social media re- altruism positively predicted “sharing” on Facebook (p < .05), while quires minimal costs while exposing the self to the scrutiny of many other traits showed negative or null effects. others. Thus, even if altruism increases sharing on social media as The results of study 1 is particularly reliable given that we es- in other word-of-mouth contexts, the primary motive is more likely tablish an effect by measuring actual past behaviors. to be self-oriented (Berger 2014) than other-focused (Dichter 1966). Indeed, Baumeister (1982) theorizes that the self-presentational mo- STUDY 2 & 3 tive of gaining favorable evaluations from others often accounts for If self-presentational concerns motivate altruistic individuals to altruism and that heightening such concern will promote prosocial “Share,” we posit that triggering sensitivity to others’ judgments on behaviors. Yet, this relationship has not been directly tested. the self should also increase likelihood of sharing on social media In this research, we examine the relationship between altruism for non-altruistic individuals, and cancel out the positive correlation and self-presentation motives in the context of sharing contents on of altruism on sharing. In contrast, if self-presentational motives are social media. We propose that self-presentation concerns lead altru- the main mechanism, priming other-focused motives (i.e., for the istic individuals to share more contents on social media, because they enjoyment of others) should not moderate the positive relationship view sharing as a means to gain favorable evaluation from others. As between altruism and sharing online. In the following studies, we suggested by Bausmeister (1982), we expect that altruistic individu- test this hypothesis by directly manipulating the sensitivity to others’ als will be motivated to gain favorable evaluations from others, and judgment, and also measure positive feelings of the self (e.g., proud) share contents to reap positive rewards of feeling good about the self. as a mediator. Through a series of 4 studies, we provide novel evidence that in- All participants engaged in a hypothetical task of watching a deed altruistic people are more likely to share contents on social me- 15 second video (pretested for moderate likability) and were asked dia because of self-presentational concerns and rewards of positive to imagine sharing the video on Facebook. We primed individuals to feelings about the self. Interestingly, low altruistic individuals will think about others’ passing judgment on them in one condition or to also share more video contents when self-motivational cues are trig- think about how others would enjoy the video (study 3), while no ad- gered, and this concern for others’ evaluations is dormant for such in- ditional instructions were given in the control condition (study 2). In dividuals in neutral states on social media. The findings contribute to study 3, we measured positive feelings to test whether self-oriented our understanding of the relationship between self-oriented motives rewards were mediating sharing contents on social media. and altruism, and offer insight into the psychology of sharing video As expected, the main effect of altruism was positive and sig- contents on social media, an important and pervasive phenomenon nificant in all studies (p<.01). More importantly, the interactions be- that underlies viral marketing. tween altruism and the judgment manipulation were significant in both studies. The conditional effects of the judgment manipulation PILOT STUDY increased the likelihood of sharing for the low altruistic people, but In this study, we test whether individuals with high altruism are had no effect in the high altruistic individuals. Furthermore, positive more likely to have (1) higher self-presentational concerns and (2) feelings showed mediated moderation patterns (p <.05). belief that others are judging them, especially in social media. We recruit 233 people and measure the extent of self-presenta- CONCLUSION tional concerns when thinking about sharing video contents on Face- Through a series of five studies, we present a robust finding book. Additionally, participants imagine that they had just shared a that altruistic individuals are more likely to share contents on social video, and report their belief on others’ judging them on social me- media. We find support that self-presentational motivations – gain- dia. Altruism was positively correlated with self-presentational con- ing a favorable evaluation from others – account for this peculiar cerns (p <.01). When participants were grouped by levels of altruism, relationship. Our evidence shows that altruistic individuals are more only the high altruistic individuals showed a significantly greater sensitive to others’ judgments on social media, and that positive self- sensitivity to judgment by others on social media. Furthermore, self- oriented feelings mediate their intention to share contents online. presentational motives mediated the positive relationship between One limitation of the current research is that only one type of altruism and belief of others’ judgment when sharing on Facebook. contents (video) is used, limiting the generalizability of the findings. Thus, a stronger belief of others’ judgment on the self was pres- Although we show a positive relationship between sharing and altru- ent in high altruistic individuals, who in turn were more motivated ism with actual past Facebook activities data, future research should address this issue by testing additional types of contents.

Advances in Consumer Research 728 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 729 REFERENCES Philippe Rushton, J., Chrisjohn, R. D., & Cynthia Fekken, G. Baumeister, R. F. (1982). A self-presentational view of social (1981). The altruistic personality and the self-report altruism phenomena. Psychological Bulletin, 91(1), 3–26. scale. Personality and Individual Differences, 2(4), 293–302. Berger, J. (2014). Word of mouth and interpersonal communication: Snyder, M. (1974). Self-monitoring of expressive behavior. Journal A review and directions for future research. Journal of of Personality and Social Psychology, 30(4), 526–537. Consumer Psychology, 24(4), 586–607. Sundaram, D. S., Mitra, K., & Webster, C. (1998). Word-of-mouth Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). communications: A motivational analysis. Advances in A psychobiological model of temperament and character. Consumer Research, 25, 527-531 Archives of General Psychiatry, 50(12), 975–90. Wilcox, K., & Stephen, A. T. (2013). Are Close Friends the Enemy? Dichter, E. (1966). How word of mouth advertising works. Harvard Online Social Networks, Self-Esteem, and Self-Control. Business Review, 44, 147-166. Journal of Consumer Research, 40(1), 90–103. An Idea Opposed to Another Idea is Always the Same Idea: Reconsidering the Materialistic Aspects of Voluntary Simplicity Mathieu Alemany-Oliver, Toulouse Business School, France Justyna Kramarczyk, Adam Mickiewicz University in Poznań, Poland

EXTENDED ABSTRACT While voluntary simplicity is generally associated with a rejec- In consumer research literature, voluntary simplicity is gener- tion of or indifference to materialism, the widely accepted definitions ally understood as a non-materialistic or an anti-materialistic life- of materialism by Belk (1984), Richins and Dawson (1992), Richins style. Our study of Polish simplifiers stands in contrast to this view and Rudmin (1994), and Shrum et al. (2013) confirm that our par- as it reveals the importance that possessions can have in the lives of ticipants are materialistic and that there is, consequently, a form of our participants. The participants’ relations to objects reveal a form voluntary simplicity that is materialistic. While the Polish simplifiers of materialism that we describe as negative (vs. positive) and that we observed and questioned proclaim themselves to be anti-materi- emphasizes non-accumulation while leaving possessions a central alists, they unknowingly reinforce the materialistic consumer culture place in the lives of consumers. by continuing to attach great importance to what they possess and Voluntary simplicity can be defined as “the choice out of free reject. These results support Shaw and Newholm (2002) as well as will […] to limit expenditures on consumer goods and services, and Shaw and Moraes (2009), with the idea that voluntary simplicity oc- to cultivate non-materialistic sources of satisfaction and meaning” curs within, not outside, a consumerist market system. These results (Etzioni 1998, 620). It implies regaining control over one’s own life also reinforce the observations by Nepomuceno and Laroche (2015) and searching for meaning while avoiding mass consumption and that materialists themselves can sometimes resist consumption. accumulation (Elgin 1981; Zavestoski 2002). Voluntary simplic- Finally, this research is an opportunity to rethink materialism. ity is often associated with environmental values, ethical practices, The results emphasize the critical role that possessions play in the material simplicity, distance from money, freedom, self-sufficiency, lives of participants, but by their absence or scarcity and without the self-knowledge, and the spiritual world (Cherrier 2009; Cherrier and simplifiers intending to accumulate more. Therefore, we encourage Murray 2007; Elgin and Mitchell 1977; Etzioni 1998; Gregg 1936; researchers in our field to move away from the idea of a materialism Huneke 2005; Leonard-Barton 1981; Moisander and Pesonen 2002; that would only or mainly be reduced to the accumulation of goods. Shama 1981, 1985; Shaw and Newholm 2002). With such a concep- Instead, we propose to differentiate between positive materialism, tion of voluntary simplicity, it is not surprising to find close links which encourages accumulation, and negative materialism, which between voluntary simplicity and materialism –or, specifically, its favors non-accumulation while leaving possessions a central place rejection– in academic works and other discourses on simplifiers in the lives of consumers. We end this extended abstract by saying (Etzioni 1998; Craig-Lees and Hill 2002; Nepomuceno and Laroche that negative materialism is potentially a durable solution and could 2015, 2017). Unfortunately, this emphasis on a negative correlation respond to Scott, Martin and Schouten’s (2014) call for new relation- between voluntary simplicity and materialism leads to a wrong as- ships with materiality. sumption, namely that voluntary simplicity separates consumers This research was supported by a grant from the National Sci- from the material world. And yet, it would not be counterintuitive ence Centre in Poland (no. 2014/13/N/HS6/02686) awarded to Justy- to think that the fewer possessions we have, the more we care about na Kramarczyk. them and develop special relationships with them (Hill and Stamey 1990). Belk (1988) also points out that possessions are not mere ob- REFERENCES jects but imbued with meanings that extend the self. Even the ab- Alexander, Samuel and Simon Ussher (2012), “The Voluntary sence or refusal of objects can constitute a way of shaping the self Simplicity Movement: A Multi-National Survey Analysis in (Lastovicka and Fernandez 2005; Hogg, Banister, and Stephenson Theoretical Context,” Journal of Consumer Culture, 12 (1), 2009). 66-86. The results of a two-year ethnography of Polish simplifiers Belk, Russell W. (1984), “Three Scales to Measure Constructs highlight materialistic aspects of voluntary simplicity that had not Related to Materialism: Reliability, Validity, and Relationships previously been underlined or had been emphasized very little. As in to Measures of Happiness,” in NA - Advances in Consumer previous studies, we have observed that participants are motivated to Research Volume 11, ed. Thomas C. Kinnear, Provo, UT: live with fewer possessions without necessarily spending less (mate- Association for Consumer Research, 291-297. rial simplicity), to take control of their lives (self-determination), and Belk, Russell W. (1988), “Possessions and the Extended to better explore their inner lives (personal growth). It is interesting Self,” Journal of Consumer Research, 15 (2), 139-68. to note that the vast majority of participants pay little attention to Cherrier, Hélène (2009), “Disposal and Simple Living: Exploring ecological concerns or ethical principles, even though other studies the Circulation of Goods and the Development of Sacred have identified these concerns as motivating factors for simplifiers Consumption,” Journal of Consumer Behaviour, 8 (6), 327-39. (Leonard-Barton 1981; Alexander and Ussher 2012; Shaw and Ne- Cherrier, Hélène and Jeff B. Murray (2007), “Reflexive wholm 2002). One possible explanation for these differences may Dispossession and the Self: Constructing a Processual Theory come from Poland’s socialist past, which is more conducive to the of Identity,” Consumption Markets & Culture 10 (1), 1-29. development of a less ideological and mainly self-oriented version of Craig‐Lees, Margaret and Constance Hill (2002), “Understanding voluntary simplicity. This would coincide with the results that Kala, Voluntary Simplifiers,” Psychology & Marketing, 19 (2), 187- Galčanová, and Pelikán (2017) obtained in the Czech Republic, an- 210. other formerly socialist country. Last but not least, the results show Elgin, Duane (1981), Voluntary Simplicity: Toward a Way of Life how participants can attach great importance to their possessions and that is Outwardly Simple, Inwardly Rich, New York, NY: show relationships with objects that are very similar to those devel- Morrow. oped by materialistic consumers.

Elgin, Duane and Arnold Mitchell (1977), “Voluntary Nepomuceno, Marcelo Vinhal and Michel Laroche (2017), “When Simplicity,” Planning Review, 5 (6), 13-15. Materialists Intend to Resist Consumption: The Moderating Etzioni, Amitai (1998), “Voluntary Simplicity: Characterization, Role of Self-Control and Long-Term Orientation,” Journal of Select Psychological Implications, and Societal Business Ethics, 143 (3), 467-83. Consequences,” Journal of Economic Psychology, 19 (5), Richins, Marsha L. and Floyd W. Rudmin (1994), “Materialism and 619-43. Economic Psychology,” Journal of Economic Psychology, 15 Gregg, Richard (1936), “Simplicity,” Visva-Bharati Quarterly, (2), 217-31. reprinted in Manas, 27 (36), 1-13 (September 4, 1974). Richins, Marsha L. and Scott Dawson (1992), “A Consumer Values Hill, Ronald Paul and Mark Stamey (1990), “The Homeless in Orientation for Materialism and its Measurement: Scale America: An Examination of Possessions and Consumption Development and Validation,” Journal of Consumer Research, Behaviors,” Journal of Consumer Research, 17 (3), 303-21. 19 (3), 303-16. Hogg, Margaret K., Emma N. Banister, and Christopher Scott, Kristin, Diane M. Martin, and John W. Schouten (2014), A. Stephenson (2009), “Mapping Symbolic (anti-) “Marketing and the New Materialism,” Journal of Consumption,” Journal of Business Research, 62 (2), 148-59. Macromarketing, 34 (3), 282-90. Huneke, Mary E. (2005), “The Face of the Un‐Consumer: An Shama, Avraham (1981), “Coping with Staglation: Voluntary Empirical Examination of the Practice of Voluntary Simplicity Simplicity,” Journal of Marketing, 120-34. in the United States,” Psychology & Marketing, 22 (7), 527- Shama, Avraham (1985), “The Voluntary Simplicity Consumer,” 50. Journal of Consumer Marketing, 2 (4), 57-63. Kala, Lukáš, Lucie Galčanová, and Vojtěch Pelikán (2017), Shaw, Deirdre and Caroline Moraes (2009), “Voluntary Simplicity: “Narratives and Practices of Voluntary Simplicity in the An Exploration of Market Interactions,” International Journal Czech Post-Socialist Context,” Sociologický Časopis/Czech of Consumer Studies, 33 (2), 215-23. Sociological Review, 53 (6), 833-55. Shaw, Deirdre and Terry Newholm (2002), “Voluntary Simplicity Lastovicka, John L., and Karen V. Fernandez (2005), “Three Paths and the Ethics of Consumption,” Psychology & Marketing, 19 to Disposition: The Movement of Meaningful Possessions to (2), 167-85. Strangers,” Journal of Consumer Research, 31 (4), 813-23. Shrum, Lawrence J., Nancy Wong, ... and Jill Sundie (2013), Leonard-Barton, Dorothy (1981), “Voluntary Simplicity Lifestyles “Reconceptualizing Materialism as Identity Goal Pursuits: and Energy Conservation,” Journal of Consumer Research, 8 Functions, Processes, and Consequences,” Journal of Business (3), 243-52. Research, 66 (August), 1179-85. Moisander, Johanna and Sinikka Pesonen (2002), “Narratives of Zavestoski, Stephen (2002), “The Social–Psychological Bases of Sustainable Ways of Living: Constructing the Self and the Anti-Consumption Attitudes,” Psychology & Marketing, 19 Other as a Green Consumer,” Management Decision, 40 (4), (2), 149-65. 329-42. Nepomuceno, Marcelo Vinhal and Michel Laroche (2015), “The Impact of Materialism and Anti-Consumption Lifestyles on Personal Debt and Account Balances,” Journal of Business Research, 68 (3), 654-64. When Sharing Isn’t Caring: The Influence of Seeking the Best on Sharing Favorable Word of Mouth about Unsatisfactory Purchases Nicholas J. Olson, Texas A&M University, USA Rohini Ahluwalia, University of Minnesota, USA

EXTENDED ABSTRACT expectations, when the purchases were completed, participants in the Past word of mouth (WOM) research typically finds that, if maximizing condition were more likely to share favorable WOM (M consumers share WOM with others, it tends to be consistent with = 4.36) than those in the non-maximizing condition (M = 3.68; p < actual perceptions (Berger 2014). Thus, a negative consumption ex- .05). However, this effect was unexpectedly reversed when the pur- perience, such as an unsatisfactory purchase, tends to elicit negative chases were merely considered, as participants in the maximizing (vs. positive) WOM. In our work, we consider when the opposite condition were marginally less likely to share favorable WOM (M may be true. Specifically, we examine the possibility that a goal of = 3.04) than those in the non-maximizing condition (M = 3.71; p = attaining the best, or maximizing, enhances consumers’ tendency to .07). share favorable WOM about unsatisfactory purchases. Study 3 was similar to study 1, but after listing a close other Prior research demonstrates that when consumers have a goal and indicating their likelihood of saying good things and encourag- of attaining the best, they are concerned with not only getting the ing the other to make their same unsatisfying purchase (favorable best, but also being best relative to others (Weaver et al. 2015). WOM), they also reported their concern with enhancing the relative When those seeking the best, or maximizers, make an unsatisfying standing of their purchase, and how much better they expected to purchase, it signals that they have failed to attain an objectively op- feel about their own purchase if the other made the same purchase. timal outcome, which elicits strong negative feelings (e.g., regret; As predicted, maximizing promoted the tendency to share favorable Schwartz et al. 2002). However, their objectively poor decision can WOM about unsatisfying purchases (β = .11, t = 1.94, p < .05), and still be relatively good if others make the same poor decision, which this effect was mediated by maximizers’ heightened concern with en- can be encouraged by speaking favorably about the bad purchase hanced relative standing, which led them to believe they would feel (Berger 2014). Extending this notion, we propose that maximizers better about their purchase if a close other made the same purchase (vs. non-maximizers) will be more likely to share favorable WOM (indirect effect = .019, 95% CI = .0015 to .0436). about their unsatisfactory purchase, because they want others to Study 4 replicated past studies in an actual choice context, and make the same poor decision to enhance the relative standing of and considered the downstream consequences of maximizers’ behavior. subsequent feelings toward their own outcome. Participants (students) reported chronic maximizing tendencies, then We further posit that maximizers will only exhibit this behav- completed a pen task in which they chose a pen from two options, ior when sharing WOM with interpersonally close (vs. distant) oth- learned their chosen pen was supposedly inferior (unsatisfying), and ers, as close others are a more consequential source of comparison then anonymously shared WOM (how favorable their own pen was) (Festinger 1954), making their outcomes especially important in de- with a purported fellow student, who would later choose from the termining the relative standing of consumers’ own choices. Finally, same assortment, and whose choice would be revealed. Participants we consider the downstream consequences of maximizers’ conduct, also indicated their expected decision regret if the fellow student proposing that when they successfully persuade others to make their chose the same pen. Supporting predictions, maximizing enhanced same poor purchase, they will feel better about their own purchase, the tendency to share favorable WOM (β = .33, t = 2.52, p < .05), but will also experience heightened feelings of guilt because of their because maximizers expected their decision regret would be reduced deceptive behavior (Hoffman 1982). We tested and found support for if the fellow student chose their pen (indirect effect = .095, 95% CI our predictions in five studies. = .0026 to .2830). Following a 10-minute delay, participants learned In study 1, participants recalled a past unsatisfactory purchase, the fellow student had supposedly chosen their pen, and they indi- then listed either a close friend (close other) or acquaintance (dis- cated their actual choice regret, as well as current feelings of guilt. tant other). All participants then reported their likelihood of recom- While maximizers’ choice regret was diminished following this in- mending their purchase (sharing favorable WOM) to the other. They formation (β = -.36, t = -2.75, p < .01), their guilt was enhanced (β = also indicated how much better they expected to feel about their own .28, t = 2.07, p < .05). purchase if the other made the same purchase. Finally, after a filler Study 5 was identical to study 4, except it manipulated whether task, participants reported chronic maximizing tendencies (Diab et the purported fellow student would choose between participants’ al. 2008). pen and a superior or an inferior alternative. In the former condition, Supporting our predictions, maximizing had no effect on word maximizing’s positive effect on WOM was replicated (β = .37, t = of mouth in the distant other condition (β = .06, t = .53, p = NS). 3.06, p < .01), but in the latter condition, it was attenuated (β = .007, However, in the close other condition, it enhanced the tendency to t = .05, p = NS). share favorable word of mouth about unsatisfying purchases (β = .34, t = 2.83, p < .01). Moreover, this effect was mediated by maximizers’ REFERENCES heightened expectations that they would feel better about their own Berger, Jonah (2014), “Word of Mouth and Interpersonal purchase if close (but not if distant) others made the same purchase Communication: A Review and Directions for Future (95% CI = .0087 to .0649). Research,” Journal of Consumer Psychology, 24 (4), 586-607. Study 2 manipulated maximizing via an essay task, then asked Diab, Dalia L., Michael A. Gillespie, and Scott Highhouse (2008), participants to imagine three scenarios in which an unsatisfactory “Are Maximizers Really Unhappy? The Measurement of purchase was either completed or merely considered. After imagin- Maximizing Tendency,” Judgment and Decision Making, 3 (5), ing each scenario, participants listed a close other and reported their 364-70. likelihood of recommending and saying good things about their pur- Festinger, Leon (1954), “A Theory of Social Comparison chase (favorable WOM) to that other (7-pt scales). Consistent with Processes,” Human Relations, 7 (2), 117-40.

Hoffman, Martin L. (1982), “Development of Prosocial Motivation: Weaver, Kimberlee, Kim Daniloski, Norbert Schwarz, and Empathy and Guilt,” in The Development of Prosocial Keenan Cottone (2015), “The Role of Social Comparison for Behavior, ed. Nancy Eisenberg, NY: Academic Press, 281-313. Maximizers and Satisficers: Wanting the Best or Wanting to Be Schwartz, Barry, Andrew Ward, John Monterosso, Sonja the Best?” Journal of Consumer Psychology, 25 (3), 372-88. Lyubomirsky, Katherine White, and Darrin R. Lehman (2002), “Maximizing Versus Satisficing: Happiness Is a Matter of Choice,” Journal of Personality and Social Psychology, 83 (5), 1178-197. How Awe Might Be Awesome: The Role of Awe in Consumers’ Food Consumption and Perceptions of Misshapen Produce Begum Kaplan-Oz, University of Massachusetts Amherst, USA Elizabeth G. Miller, University of Massachusetts Amherst, USA

EXTENDED ABSTRACT tions also differed with respect to happiness and peace (p’s<. 05), Consumers face numerous food choices that impact the well- these emotions did not predict the DVs when we included them as being of themselves, the environment, and society. One such issue covariates in the main model (p>.4). Participants indicated their which is starting to receive greater attention is that of food waste. likelihood of choosing virtuous and vice food options (a healthy Nearly one-third of food produced is wasted globally in developed (virtue-framed) chocolate, tasty (vice-framed) chocolate, and six countries (Loebnitz et al 2015). Thus, understanding how consumer snack options (3 virtues, 3 vices)). Consistent with H1, participants choices and perceptions contribute to food waste and identifying in the awe condition were marginally more likely than those in the interventions that reduce waste have numerous benefits. In this re- neutral condition to prefer the virtue-framed chocolate (Mawe=4.91, search, we explore the role of awe in consumers’ food consumption. Mneutral=4.37; F(1,136)=3.03, p=.08) and the virtue food options Awe is a “moral, spiritual, and aesthetic” emotion (Keltner and Haidt (Mawe=4.51, Mneutral=4.25; F(1,136)=2.92, p=.09). A bootstrap 2003, p.297) that has been found to play a role in consumers’ pro- analysis revealed a significant mediation by small self (95% CIlow- social behavior (Piff et al 2015). Further, it has a number of features er-bound=.11, CIupper-bound=.64), supporting H2. that suggest it may influence consumer food choices relevant to food Study 2 shows that awe increases participants’ positive percep- waste. tions and purchase intentions towards misshapen produce. Partici- Two factors that contribute to food waste are leftovers (Dewey pants (N=504 mTurkers) watched either an awe-eliciting or neutral 2017), particularly of vices which are more likely than virtues to be video (Kaplan-Oz and Miller 2017). Results confirmed the effec- wasted (Kim and Kim 2016), and misshapen produce (Loebnitz et al tiveness of the manipulation (Mawe=4.92, Mneutral=3.53, p<.001). The 2015). In regards to leftovers, engaging in virtue food consumption awe and neutral conditions also differed with respect to happiness is regarded as normal (Rook 1987) whereas engaging in vices elic- and peace (p’s<. 05) but when we included them as covariates in the its guilt (Ramanathan and Williams 2007). Recent research suggests model, neither predicted the DVs (p>.5). Participants rated their per- that since consumers feel guiltier when they consume vices, they are ceptions (organic/natural) and likelihood of purchase for three dif- more likely to leave leftover vices than leftover virtues (Kim and ferent foods (apple, carrot, eggplant; Loebnitz et al 2015) of varying Kim 2016) which contributes to food waste. Related to misshapen abnormality. Finally, participants completed food vanity measures. produce, each year, 30-40% of produce is discarded, regardless of A repeated measures analysis tested the impact of food shape abnor- whether it is healthy to eat (USDA 2016). Thus, this situation also mality, emotion and food vanity on buying intentions for the three contributes to food waste (Loebnitz et al 2015). food items and revealed a significant main effect for food abnormal- In this research, we show that awe influences both consumers’ ity (F(2, 492)=105.575, p<0.001). Participants were much less likely consumption of virtuous foods and their perceptions of misshapen to purchase the extremely abnormal foods (M=2.95) compared to produce. Awe has been shown to shift one’s focus from the self to normal (M=5.23, p<.001) and moderately abnormal foods (M=4.36, others resulting in a diminished self (small self) and to make people p<.001). The main effect for emotion was not significant (p > .4), feel more connected to nature and other people around them (Keltner even though the means were in the expected direction (Mawe=4.21; and Haidt 2003; Shiota et al 2007). Accordingly, this feature might Mneutral=4.14). increase consumers’ desire to be healthy and live longer by engaging As expected, the results revealed a significant food vanity x in virtuous food consumption (H1). Additionally, feelings of small abnormality interaction F(2,492)=16.887, p<.001), with those high self have been shown to increase consumers’ well-being and life in food vanity less likely than those low in food vanity to purchase satisfaction (Rudd et al. 2012; Piff et al. 2015). Thus, we propose moderately (Mhigh=4.05, Mlow=4.67) and extremely abnormal that awe’s influence on small self may increase consumers’ desire produce (Mhigh=2.09, Mlow=3.80). However, as predicted in H4, for virtuous foods (H2). Further, feelings of awe can elicit percep- this interaction was qualified by a marginally significant emotion x tions of aesthetics and beauty (Keltner and Haidt 2003). Therefore, abnormality x food vanity interaction (F(2,492)=2.49, p=.08). The awe’s ability to shift perceptions towards beauty and to enable a negative effects of food vanity on purchase intentions of misshapen greater desire for nature (Piff et al 2015) could increase consumers’ produce were reversed for those in the awe condition for extremely purchase intentions (H3a) and perceptions (organic/natural) towards abnormal produce (Mawe-highV=2.28, Mneutral-highV=1.90). misshapen produce (H3b). However, such effects may be affected by Similar results were obtained for perceptions of the food as the importance consumers place on appearance. Previous research organic/natural. Participants were less likely to perceive extremely shows that perceptions of beauty impact consumers’ behaviors (e.g. abnormal produce as organic/natural (M=4.23) compared to nor- Eagly et al 1991; Lee et al 2015) which is referred as consumer van- mal (M=5.23, p<.001) and moderately abnormal foods (M=5.07, ity (Netemeyer et al 1995). Building on that, we define food vanity p<.001; F(2,492)=23.708, p<.001). The main effect of emotion was as the importance an individual attaches to the physical appearance not significant (p>.4). There was a significant food vanity x abnor- of food. We propose that awe’s effects on purchase intentions and mality interaction (F(2,492)=9.871, p<.001), with those high in food perceptions of misshapen produce will be strongest for those who vanity less likely than those low in food vanity to purchase moder- care about how their food looks (high level of food vanity; H4a-b). ately (Mhigh=4.85, Mlow=5.31) and extremely abnormal produce Study 1 shows that awe increases participants’ desires towards (Mhigh=3.64, Mlow=4.82). However, the 3-way interaction with virtuous foods. Participants from Amazon Mechanical Turk (N=137) emotion failed to reach significance (p=.2), although the means fol- were randomly assigned to view an emotion-inducing video (awe, lowed the same pattern as for purchase intentions. neutral). Results confirmed the effectiveness of the manipulation Thus, across two studies, we find that awe influences consumers’ food purchasing behavior and perceptions. Awe increased con- (Mawe=4.8, Mneutral=3.69, p<.002). While the awe and neutral condi- Advances in Consumer Research 734 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 735 sumers’ preferences for virtue foods and increased their perceptions Keltner, Dacher, and Jonathan Haidt (2003), “Approaching awe, a and likelihood of purchasing misshapen produce (for those high in moral, spiritual, and aesthetic emotion.” Cognition & emotion food vanity). These findings have implications for alleviating food 17 (2), 297-314. waste. Kim, J. C., & Kim, J. (2016), “Waste Aversion For Virtue Versus Food waste is a major concern of our society. While some re- Vice”, ACR North American Advances. search has explored the effect of negative emotions (e.g., shame, re- Lascu, D. N. (1991), “Consumer guilt: examining the potential of a gret, guilt; Sirieix et al 2017) on consumers’ attitudes and behaviors new marketing construct”, ACR North American Advances. related to leftovers, the role of positive emotions (such as awe) has Loebnitz, N., Schuitema, G., & Grunert, K. G. (2015), “Who not been studied nor has the impact of emotion on perceptions of ab- buys oddly shaped food and why? Impacts of food shape normally shaped food. Our research highlights the impact of awe as abnormality and organic labeling on purchase intentions”, a potential intervention for reducing food waste caused by leftovers Psychology &Marketing, 32(4), 408-421. and misshapen produce. This research contributes to an increased Ludwig, D. S., Peterson, K. E., & Gortmaker, S. L. (2001), understanding of the role of awe in consumer behavior. Additionally, “Relation between consumption of sugar-sweetened drinks and this research contributes to consumers’ food consumption literature childhood obesity: a prospective, observational analysis”, The by presenting the positive role of a unique emotion in increasing con- Lancet, 357(9255), 505-508. sumers’ healthy food consumption which accordingly contributes to Maddock, J. (2004), “The relationship between obesity and the their wellbeing. prevalence of fast food restaurants: state-level analysis”, American journal of health promotion, 19(2), 137-143. REFERENCES Netemeyer, R. G., Burton, S., & Lichtenstein, D. R. (1995). Trait Bloom, J. (2010). American wasteland: How America throws away aspects of vanity: Measurement and relevance to consumer nearly half of its food (and what we can do about it). Da Capo behavior. Journal of consumer research, 21(4), 612-626. Lifelong Books. Piff, Paul K., Pia Dietze, Matthew Feinberg, Daniel M. Stancato, Chandler, A. (2016), “Why Americans Lead the World in and Dacher Keltner (2015), “Awe, the small self, and prosocial Food Waste”, https://www.theatlantic.com/business/ behavior.” Journal of personality and social psychology 108 archive/2016/07/american-food-waste/491513/ [accessed on (6), 883. 03-13-18] Ramanathan, Suresh and Patti Williams (2007), “Immediate Dewey, C. (2017), “Why Americans have stopped eating leftovers”, and Delayed Emotional Consequences of Indulgence: https://www.washingtonpost.com/news/wonk/wp/2017/10/31/ The Moderating Influence of Personality Type on Mixed why-americans-have-stoppedeating-leftovers/?utm_ Emotions,” Journal of Consumer Research, 34(2), 212–23. term=.4c8f32366912 [accessed on 03-17-18] Rook, Dennis W. (1987), “The Buying Impulse,” Journal of Greenberg, J. (2017), “No, food waste is not the third biggest Consumer Research, 14(2), 189–99. source of greenhouse gases”, http://www.politifact.com/ Rudd, Melanie, Kathleen D. Vohs, and Jennifer Aaker. (2012), global-news/statements/2017/feb/14/world-food-program/no- “Awe expands people’s perception of time, alters decision foodwaste-not-third-biggest-source-greenhouse-/ [accessed on making, and enhances well-being.” Psychological science 03-15-18] 23(10), 1130-1136. Gunders, D. (2012), “Wasted: How America is losing up to 40 Shiota, Michelle N., Dacher Keltner, and Amanda Mossman. percent of its food from farm to fork to landfill”, Natural (2007), “The nature of awe: Elicitors, appraisals, and effects Resources Defense Council, 26. on self-concept.” Cognition and emotion 21 (5), 944-963. Harvey, C. (2016), “Our wasted food is a huge environmental Sirieix, L., Lála, J., & Kocmanová, K. (2017), “Understanding the problem – and it’s only getting worse”, https://www. antecedents of consumers’ attitudes towards doggy bags in washingtonpost.com/news/energy environment/ restaurants: Concern about food waste, culture, norms and wp/2016/04/07/howwasting- less-food-can-help-save-the emotion”, Journal of Retailing and Consumer Services, 34, planet/?utm_term=.5179850ce4e5 [accessed on 03-15-18] 153-158. Harvey, C (2016). “The enormous carbon footprint of food that USDA. (2016), “Does the U.S. have a food loss and waste we never even eat”, https://www.washingtonpost.com/news/ reduction goal?”, https://www.usda.gov/oce/foodwaste/faqs. energy-environment/wp/2016/03/28/the-enormouscarbon- htm [accessed on 03-12-18] footprint-of-the-food-we-never-eat/?utm_term=.7aa772d25b3e Wang, G. J., Volkow, N. D., Telang, F., Jayne, M., Ma, Y., [accessed on 03-15-18] Pradhan, K., ... & Biegon, A. (2009), “Evidence of gender Izard, Carroll E. (1977), Human Emotions, New York: Plenum differences in the ability to inhibit brain activation elicited by Press. food stimulation”, Proceedings of the National Academy of Kaplan-Oz, B. & Elizabeth G. Miller (2017), “Emotional drivers Sciences, 106(4), 1249-1254. of materialism: The Role of Awe in Materialistic Tendencies”, Williams, P., Coleman, N. V., Morales, A. C., & Cesareo, L. (2018), Proceedings of the 2017 Marketing and Public Policy “Connections to Brands that Help Others versus Help the Self: Conference, Washington DC, p.70 The Impact of Awe and Pride on Consumer Relationships with Social-Benefit and Luxury Brands”, Journal of the Association for Consumer Research, 3(2). With or Without You: How Second Person Pronouns Engage Listeners Grant Packard, Wilfrid Laurier University, Canada Jonah Berger, University of Pennsylvania, USA

EXTENDED ABSTRACT ing (e.g., “Do the lyrics make you think about someone you know?”). What makes cultural products popular? Could products that lit- Lastly, we collected measures for alternative explanations. We ob- erally speak to “you” be more engaging? Academics have long been tained the lyrics for each of the participant-reported songs to generate interested in whether the use of second person pronouns (e.g., you, “you” pronoun statistics for each song. Independent judges coded your, and yours) shapes audience involvement (Pennebaker 2011). each “you” use as more consistent with self- or other-referencing. While some have argued that this small category of words seems to Regression analysis revealed that “you” pronouns increase song enhance engagement (e.g., increasing likes of online posts, Cruz et liking (B = .05, t = 2.56, p = .01). This relationship was mediated by al. 2017), other work has found little impact of second person pro- the song’s ability to evoke another person in the participant’s own nouns on listeners (Packard et al. 2018). life (indirect effect = .02, 95% CI [.008, .046]). Neither a perceived We predict that “you” pronouns do enhance engagement with connection with the singer nor seeing into the singer’s personal life listeners. Importantly, however, rather than being narratively trans- mediated the relationship, casting doubt on these alternatives. ported (Escalas 2007) as the driver of a story or by making them Replicating the field data results, the impact of “you” pronouns the focal subject or actor (e.g., “You’re driving down the street…”) was driven by the number of use cases judged to be other-referencing as described in research on self-referencing (Burnkrant and Unnava (B = .03, t = 2.50, p = .01), but not by those judged as self-referenc- 1995; Meyers-Levy and Peracchio 1996), we propose that “you” pro- ing (B = -.01, t = -1.02, p = .31). nouns engage audiences by inducing a feeling of social attachment Study 3 sought to replicate the main effect of “you” pronouns by (Hazan and Shaver 1990). In short, “you” pronouns could work by manipulating pronoun use in experimenter-controlled lyrics, and test evoking a “you” other than the listener: a person to whom the lis- for moderation of the effect by the extent to which the song’s content tener could imagine singing or speaking the words. We examine this evokes a salient other. proposition—that other-referencing underlies the positive impact of Participants (N = 226) who were currently involved in a ro- “you” pronouns—in three ways. mantic relationship were primed to think about their partner using a First, from a linguistic perspective, other-referencing “you” use thought listing exercise, and then indicated their attitude towards this should typically occur when this pronoun is presented as the object person. Participants were then presented with one of three versions of a protagonist’s thoughts or actions (“I love you”) rather than as a of a positively valenced love song featuring one of either “you,” “s/ self-referencing protagonist (“You love me”). he” or “it/that” pronouns (e.g. “Nothing can take [your, her (or his), Second, if other-referencing drives the engagement benefit of its] place”). “you” pronouns, this relationship will be mediated by second person Corroborating prior results, the version of the song with “you” pronouns’ ability to evoke an other to whom the listener is socially pronouns was liked significantly more than the “s/he” version (B = attached (Hazan and Shaver 1990). .43, t = 3.40, p < .001) and the “it/that” version (B = .31, t = 2.34, p Third, we expect that the effect of “you” use should be enhanced = .02). Consistent with the notion that the song was connecting them when second person pronoun references appear in a positively va- to others in their own life, the positive effect of “you” pronouns on lenced manner. People are motivated to hold positive perceptions of attitudes towards the positive love song was stronger among people the self, and these perceptions are maintained in part through the who had more positive attitudes towards their relationship partner belief that others regard them in a positive light (Baumeister 1982, (+1SD, B = -.53, t = -.377, p < .001), but had a weaker effect among 1998; Leary 1996). people who had less positive attitudes towards that partner (-1SD; B Three studies using field data and lab participants test these pre- = .11, t = .56, p = .58). dictions. This research demonstrates that “you” pronouns have a signifi- Study 1 collected a dataset of 4,200 song rankings from the cant impact on audience engagement. Textual analysis of thousands Billboard charts across 1,879 unique songs. We then captured each of songs and two lab studies suggest that second person pronouns song’s lyrics and a variety of covariates (e.g., artist name, radio air- engage the listener by evoking a personal other. This work deepens play). The lyrics were processed to identify all instances of second our understanding of when and why second person pronouns may person pronouns. Independent judges coded whether each “you” ap- succeed (or fail) in engaging the listener, offering new insights for peared in an other-referencing or self-referencing role, as well as the both consumer and linguistic psychologists valence of each “you” usage. Results indicate that the more “you” pronouns that appear in a REFERENCES song, the more successful that song is (reverse-coded chart rank B = Baumeister, Roy. F. (1982), “A self-presentational view of social .19, t = 3.35, p < .001, model 1). This result was robust to a variety phenomena,” Psychological Bulletin, 91(1), 3–26. of controls (model 2). Baumeister, Roy F. (1998), “The self,” The Handbook of Social Songs were more successful the more they mentioned “you” as Psychology (4th ed.), ed. Daniel Gilbert, Susan Fiske, and the focus of someone’s attention (other-referencing; B = .1, t = 3.08, Gardner Linndzey, New York, NY: McGraw-Hill, 680–740. p = .002). The number of “you” pronouns that invoked the audience Burnkrant, Robert E. and H. Rao Unnava (1995), “Effects of Self- as the actor themselves (self-referencing) or that were unassigned, Referencing on Persuasion,” Journal of Consumer Research, however, had no effect (ts < 1.48, ps > .14; models 3 and 4). Songs 22 (June), 17-26. were also more successful if the “you” uses were judged to be more Cruz, Ryan E., James M. Leonhardt, and Todd Pezzuti (2017), positive (B = .58, t = 2.73, p = .006) “Second Person Pronouns Enhance Consumer Involvement Study 2 asked 200 participants to report any song they had and Brand Attitude,” Journal of Interactive Marketing, 39, heard, and their attitude towards it. We then measured other-referenc- 04-116.

Escalas, Jennifer E. (2007), “Self-Referencing and Persuasion: Meyers-Levy, Joan and Laura A. Peracchio (1996), “Moderators Narrative Transportation versus Analytical Elaboration,” of the Impact of Self-Referenence on Persuasion,” Journal of Journal of Consumer Research, 33 (March), 421-429. Consumer Research, 22 (March), 408-423. Hazan, Cindy and Philip R. Shaver (1990), “Love and work: An Packard, Grant, Sarah G. Moore, and Brent McFerran (2018), attachment theoretical perspective,” Journal of Personality “(I’m) Happy to Help (You): The Impact of Personal Pronoun and Social Psychology, 59(2), 270-280. Use in Customer-Firm Interactions,” Journal of Marketing Leary, Mark R. (1996) Self-presentation: Impression management Research, 55(4), 541-555. and interpersonal behavior. Boulder, CO: Westview. Pennebaker, James W. (2011). The secret life of pronouns: What our McGrath, Lisa and Maria Kuteeva (2012), “Stance and engagement words say about us. New York, NY: Bloomsbury Press. in pure mathematics research articles: Linking discourse features to disciplinary practices,” English for Specific Purposes, 31, 161-173. The Neutral Face of Blue: How Color Can Make Consumers Stay Sensitive Sung-Hee W. Paik, University of Oregon, USA Jiao Zhang, University of Oregon, USA Aparna Sundar, University of Oregon, USA

EXTENDED ABSTRACT high versus low (F(1, 134) = 4.59, p < .05). When the red logo was According to The Largest U.S. Charities, released by Forbes presented, the difference was not significantn.s. ( ). (Forbes 2017), more than half of the top 50 charities in the US use In Study 2, we used a different charitable cause (child abuse) red, blue, or both in their logos. Despite this widespread use of red and added a black logo as a control. Participants were recruited on- and blue, limited research has investigated the use of these colors in line (N=263) to take part in a 2 (victim number: low vs. high) 3 donation appeals. A logo’s color, especially its hue, is known to influ- (logo color: blue vs. red vs. black) between-subject experiment. A 2 ence marketing outcomes such as perceived benefits or brand person- 3 ANOVA controlling for participants’ demographics yielded a sig- ality (Bottomley and Doyle 2006; Labrecque and Milne 2012). The nificant victim-number logo-color interaction (F(2, 254) = 3.60, p < color of a logo may also be associated with meaningful concepts and .05). As in Study 1, only when the blue logo was presented, donation even unrelated consumer inferences about the entity that the color intention was significantly higher when the number of victims was represents (e.g. ethicality of a practice; Sundar and Kellaris 2017). high vs. low (F(1, 254) = 7.60, p < .01). The difference between Past research on color and donation appeals studied how background victim-number conditions was not significant when a red or black colors used in donation appeals influenced donation (Choi, Singh, logo was presented (n.s.). and Rangan 2011; Mehta et al. 2011). However, the influence of logo In Study 3, we examined how logo color influences prospective color, and hues in particular, on prosocial intentions is less under- donors’ sensitivity to victims’ social distance (e.g., nationality). We stood. tested the prediction that consumers would donate more money to The current research investigates how the color of a charity logo help socially close victims than to help socially distant victims if the influences consumers’ sensitivity to the charitable cause and to dona- logo of a charitable organization was blue vs. other colors. Partici- tion. One aspect of this is sensitivity to the number of victims. As pants (N=510) living in the US were randomly assigned to one of six charitable causes often respond to natural disaster or disease and the conditions in a 3 (logo color: blue, red, green) 2 (victim national- number of donation recipients can be massive, examining sensitivity ity: US vs. Africa) between-subject experiment. They were told that to victim numbers can be crucial. Another important criteria when children in the US (vs. Africa) were suffering from a rare disease and interpreting the charitable cause is the social distance between donor asked to indicate how much money on a sliding scale ($0–$30) they and victims. Past research has shown that people indicate a greater would be willing to donate to the charity. A 3 2 ANOVA yielded a willingness to help when the social distance between donor and vic- significant logo-color victim-nationality two-way interaction (F(2, tim is close (Loewenstein and Small 2007; Small and Simonsohn 504) = 3.30, p = .038). Consistent with our prediction, when the blue 2007). Given that many global charities ask donors to give to victims logo was presented, participants donated significantly more money in other countries, the question of whether peripheral cues such as when the victims were in the US than when they were in Africa (F(1, color can change the perception of victims is important. 504) = 10.02, p = .002). By contrast, when the red (F(1, 504) = .00, Studies have shown that blue and red have distinct effects on n.s.) or green (F(1, 504) = .01, n.s.) logo was presented, participants psychological functioning and human behavior. Mehta and Zhu donated a similar amount of money, regardless of the victims’ na- (2009) showed that red versus blue induces different types of moti- tionality. vation (i.e., avoidance vs. approach) in cognitive tasks. In the domain Overall, our studies show that the hue used in a charity logo of a of advertising, Puzakova et al. (2016) showed that blue (vs. red) can donation appeal influences consumers’ sensitivity to victim numbers broaden cognitive scope and make individuals more diagnostic in and to the social distance to victims, which in turn impacts consum- overall evaluation of information. Based on previous research on ers’ donation. Considering the choice of hues in charity logos, our color, we examine whether the hue of a charity’s logo influences findings have practical implications. Furthermore, our studies con- potential donors’ sensitivity to important objective dimensions of tribute to the literature on color and donation by demonstrating that charitable causes (e.g., number of victims, social distance to vic- proper use of color can promote sensitivity to the charitable cause tims). Specifically, we tested the prediction that, as blue can make and can increase donation with regard to victim numbers and social donors more diagnostic about the objective information presented in distance. an appeal, potential donors’ willingness to donate will show greater sensitivity to the number of victims and the social distance to victims REFERENCES if the charity’s logo color is blue rather than red. Bottomley, Paul A. and John R. Doyle (2006), “The Interactive In Study 1, we tested whether consumers’ donation intention Effects of Colors and Products on Perceptions of Brand Logo was more sensitive to the number of victims when a blue (vs. red) Appropriateness,” Marketing Theory, 6 (1), 63–83. logo was presented. Participants (N=141) were recruited online to Choi, Jungsil, Surendra N. Singh, and Priyamvadha Rangan (2011), take part in a 2 (victim number: low vs. high) 2 (logo color: blue vs. “Unanticipated Marketing Effects of Color on the Efficacy red) between-subject experiment. Participants were asked to read the of Charitable Appeals,” In Robert Mittelstaedt Doctoral donation appeal of a fictitious charity supporting a rare disease and Symposium, 199–206. indicate their donation intention. Victim number (30 vs. 3000) and Forbes (2017), “The 100 Largest U.S. Charities,” Accessed March logo color (blue vs. red) were manipulated. A 2 2 ANOVA control- 10, 2018. https://www.forbes.com/top-charities/list/#tab:rank. ling for participants’ gender, age, and income yielded a marginally Labrecque, Lauren I. and George R. Milne (2012), “Exciting Red significant interaction (F(1, 134) = 3.56, p = .06). The results show and Competent Blue: The Importance of Color in Marketing,” that, when the blue logo was presented, participants indicated a sig- Journal of the Academy of Marketing Science, 40 (5), nificantly higher donation intention when the number of victims was 711–727.

Loewenstein, George, and Deborah A. Small (2007), “The Puzakova, M., Kwak, H., Ramanathan, S. and Rocereto, J. F. Scarecrow and the Tin Man: The Vicissitudes of Human (2016), “Painting Your Point: The Role of Color in Firms’ Sympathy and Caring,” Review of General Psychology, 11 (2), Strategic Responses to Product Failures via Advertising and 112–126. Marketing Communications,” Journal of Advertising, 45 (4), Mehta, Ravi, Boyoun Grace Chae, Rui Juliet Zhu, and Dilip Soman 1–12. (2011), “Warm Or Cool Color?: Exploring the Effects of Color Small, Deborah A., and Uri Simonsohn (2007), “Friends of on Donation Behavior,” NA-Advances in Consumer Research, Victims: Personal Experience and Prosocial Behavior,” Journal Volume (39), 190–191. of Consumer Research, 35 (3), 532–542. Mehta, Ravi, and Rui Juliet Zhu (2009), “Blue or Red? Exploring Sundar, Aparna, and James J. Kellaris (2017), “How Logo Colors the Effect of Color on Cognitive Task Performances,” Science, Influence Shoppers’ Judgments of Retailer Ethicality: The 323 (5918), 1226–1229. Mediating Role of Perceived Eco-friendliness,” Journal of Business Ethics, 146 (3), 685–701. When Disadvantage is an Advantage: Benevolent Partiality in Consumer Donations Gabriele Paolacci, Erasmus University Rotterdam, The Netherlands Gizem Yalcin, Erasmus University Rotterdam, The Netherlands

EXTENDED ABSTRACT that a sizable number of participants (about 40% in our studies) give How should an altruistic consumer choose whom to help? The money to the project in the most disadvantaged country rather than to classical utilitarian answer, championed by the Effective Altruism the one allowing to save more lives (Study 2A), a result that cannot movement, urges people to choose the charitable alternatives that be explained by beliefs that either project will be underfunded (Study produce the greatest welfare benefits (e.g., MacAskill 2015, Sing- 2B), and that replicate when people can allocate their donation across er 2015), maximizing the size of the benefit and the number of re- projects (Study 2C). cipients who benefit irrespective of their characteristics (Pellegrino Further studies addressed several predictors of benevolent par- 2017). The size criterion posits that one should give towards the tiality. In Study 3, we manipulated whether or not participants were causes that bring more benefit, which typically and aptly prescribes asked to weigh the importance of the choice criteria prior to making that, all else being equal, people should help the most disadvantaged. a choice, mimicking a typically recommended step in prescriptive Because of the diminishing marginal returns of wealth, one should, models of decision-making. We found that this intervention signifi- for instance, donate $100 to a poorer family rather than to a slightly cantly reduced benevolent partiality, making participants more likely less poor family, because $100 produces a more sizable benefit for to choose the project saving more lives. In Study 4, we found that the former (MacAskill 2015, Singer 2015). donors are particularly sensitive to disadvantage when this is caused The effectiveness of donating to the disadvantaged, however, is by uncontrollable factors (e.g., natural soil type) compared to con- contingent on donations directly alleviating the disadvantage; when trollable factors (e.g., overharvesting). Finally, consistent with our economic disadvantage (or any other form of disadvantage) is unaf- moral framework, we found that benevolent partiality is less preva- fected by one’s donation, it does not increase the size of the potential lent among people who have more permissive attitudes towards in- benefit and it is therefore irrelevant for one’s donation decision. For strumental harm, a core dimension of utilitarian thinking which cap- instance, although economic disadvantage is more severe in Chad tures people’s beliefs that sacrifices are acceptable if they produce a than in Nigeria, one should donate polio vaccines to hospitals in Ni- larger benefit (Kahane et al. 2017). geria (where polio has not been eradicated yet) rather than in Chad Altogether, our findings join recent research documenting how (where polio has been eradicated). When disadvantage is orthogonal donation decisions are often sensitive to factors other than effective- to the cause, therefore, utilitarianism prescribes that donors should ness (e.g., Berman et al. 2018; Cryder et al. 2017; Evangelidis and only attend to the number criterion: because “everybody to count Van den Bergh 2013; Fetherstonhaugh et al. 1997; Small, Loew- for one, nobody for more than one” (Bentham 1789/1983), people enstein, and Slovic 2007). Importantly, whereas this literature has should give towards the causes that allows benefiting more people typically documented instances of “heuristic ineffectiveness” that as opposed to fewer. is often corrected when people jointly evaluate differently effective In contrast to utilitarian prescriptions, we document a phenom- charitable options (e.g., Kogut and Ritov 2005), we documented enon that we call benevolent partiality. Across seven hypothetical benevolent partiality under the most conservative conditions—with donation studies, we find that many consumers weigh the well-being donors evaluating options jointly and deliberately choosing to save of members of disadvantaged groups more than what the disadvan- fewer lives. Practically, this suggests that highlighting the socioeco- tage implies from a utilitarian standpoint, choosing to save the lives nomic disadvantage of donation recipients may increase prosocial of members of more disadvantaged groups even if the same donation behavior, even within the context of donations that address needs allows saving more lives. that are unaffected by disadvantage and do not affect disadvantage. In Study 1, participants decided whether to give $10 towards shipping polio vaccines to a country or mosquito nets to another REFERENCES country—both being fictional third-world countries with a poor Bentham, Jeremy (1789), An Introduction to the Principles of socioeconomic outlook. In a between-participants design, we ma- Morals and Legislation, ed. Burns and HLA Hart London: nipulated orthogonally the relative number of beneficiaries across Methuen. projects and the relative socioeconomic disadvantage of the two Berman, Jonathan Z., Alixandra Barasch, Emma E. Levine and countries (e.g., average income, literacy rate). Participants responded Deborah A. Small (2018), “Impediments to Effective Altruism: much more strongly to the socioeconomic disadvantage in the recipi- The Role of Subjective Preferences in Charitable Giving”, ents’ country than to the number of people who would benefit from Psychological Science, forthcoming. the donation. Cryder, Cindy, Simona Botti and Yvetta Simonyan (2017), “The In studies 2A-2B-2C, we provide individual-level evidence of Charity Beauty Premium: Satisfying Donors’ “Want” Versus benevolent partiality by eliciting participants’ direct preferences for “Should” Desires”, Journal of Marketing Research, 54 (4), saving more people vs. saving people of more disadvantaged groups. 605-618. Importantly, we took several precautions to make economic disad- Evangelidis, Ioannis and Bram Van den Bergh (2013), “The vantage irrelevant from a utilitarian standpoint: Donations impacted Number of Fatalities Drives Disaster Aid: Increasing lives saved but not quality of life (i.e., donations do not alleviate Sensitivity to People in Need”, Psychological Science, 24 (11), disadvantage), and were described as the only means of help avail- 2226-2234. able to recipients (i.e., disadvantage cannot be used as a proxy for the Fetherstonhaugh, David, Paul Slovic, Stephen Johnson and James inaccessibility of other solutions). We also repeatedly emphasized Friedrich (1997), “Insensitivity to the Value of Human Life: the restrictedness aspect of the donations, and analyzed our data with A Study of Psychophysical Numbing”, Journal of Risk and and without participants who failed to report in a comprehension Uncertainty, 14 (3), 283-300. check that money would be exclusively used to save lives. We found

Kahane, Guy, Jim A. Everett, Brian D. Earp, Lucius Caviola, Pellegrino, Gianfranco (2017), “Effective Altruism and the Nadira S. Faber, Molly J. Crockett and Julian Savulescu Altruistic Repugnant Conclusion”, Essays in Philosophy, 18 (2017), “Beyond Sacrificial Harm: A Two-Dimensional Model (1), 4. of Utilitarian Psychology”, Psychological Review. Singer, Peter (2015), The Most Good You Can Do: How Effective Kogut, Tehila and Ilana Ritov (2005), “The Singularity Effect Altruism is Changing Ideas About Living Ethically, Text of Identified Victims in Separate and Joint Evaluations”, Publishing. Organizational Behavior and Human Decision Processes, 97 Small, Deborah A., George Loewenstein and Paul Slovic (2007), (2), 106-116. “Sympathy and Callousness: The Impact of Deliberative MacAskill, William (2015), Doing Good Better: Effective Altruism Thought on Donations to Identifiable and Statistical Victims”, and a Radical New Way to Make a Difference, Guardian Faber Organizational Behavior and Human Decision Processes, 102 Publishing. (2), 143-153. How Passive Form Messages in CSR Advertisement Improve Consumer Reaction to the Campaign Taehoon Park, University of South Carolina, USA Anastasiya Pocheptsova Ghosh, University of Arizona, USA Elise Chandon Ince, University of South Carolina, USA

EXTENDED ABSTRACT Extending the prior research on active and passive form sen- Corporate Social Responsibility (CSR) literature has docu- tences to the context of CSR advertising, we examine how consum- mented that running a pro-social campaign provides indirect ben- ers react to a pro-social campaign depending on the sentence form efits to companies such as higher purchase likelihood and better per- of advertising messages and congruence between CSR campaign ceived product performance (Sen and Bhattacharya 2001; Chernev and the company. Specifically, we argue that active sentence form and Blair 2015). However, these positive benefits are observed only in a CSR advertisement will focus consumers more on the agent (the when consumers believe that the company has sincere motive for company running CSR campaign) highlighting congruency between engaging in pro-social campaign (Drumwright 1996; Ellen, Webb, the company and the cause. By contrast, passive form sentences and Mohr 2006). should de-emphasize the role of the agent (the company running One factor that affects perceptions of company’s sincere mo- CSR campaign) attenuating the impact that company-cause congru- tive for running CSR is company-cause congruence. Company-cause ence has on CSR evaluation. congruence refers to the extent to which the supporting company’s As a consequence, we expect that, when high congruence sig- core business is associated with the supported charity (Simmons and nals self-benefit, consumer reaction to such CSR campaign will be Becker-Olsen 2006). Some work in CSR and in sponsorship domains improved when messages in advertisement are written in passive has shown that a high congruent campaign is more favored by con- (e.g., People in need are helped by Company A) as opposed to active sumers as it fulfills people’s expectations (Pracejus and Olsen 2004; (e.g., Company A helps people in need) form. This happens because, Rifon et al. 2004). By contrast, when there is a low company-cause by shifting focus from the agent (company) to the acted-upon agent congruence, consumers are more likely to elaborate on the campaign (CSR campaign), passive form CSR messages increase the com- due to its unexpectedness and often end up with having more nega- pany’s perceived commitment to the campaign. However, since the tive thoughts about it (Simmons and Becker-Olsen 2006; Koscha- relationship between congruence and self-benefit does not hold for te-Fischer, Huber, and Hoyer 2016). low congruence, we should not observe these effects. Yet another line of research has suggested that high company- We test our propositions in four studies. Study 1a and 1b offer cause congruence can signal a greater self-benefit salience (Fore- evidence of the impact of sentence form on company’s evaluation hand and Grier 2003; Yoon, Gürhan-Canli, and Schwarz 2006). For through two different measures of consumer reaction. Study 2fo- example, Yoon et al. (2006) showed that consumers reported high cuses on the impact of sentence form on commitment to the cam- self-benefit salience when a tabacco company was supporting cancer paign, while study 3 reports the complete path through a significant research than when it was supporting environmental charity. Simi- moderated mediation. larly, Forehand and Grier (2003), find that consumers see higher self- Study 1a employed a 2 (congruence: low vs. high) by 2 (sen- benefit salience when a computer software company supports a fight tence form: active vs. passive) between-subject design. Undergradu- against computer illiteracy versus general illiteracy among adults. ate participants (n = 173) read about a description of petroleum com- When consumers infer that a company runs its campaign with an pany, whose core business is responsible for greenhouse effect. Next, ulterior motive, their reaction to the company and CSR campaign is participants were asked to examine the company’s CSR campaign generally less favorable (Ellen et al. 2006). Therefore, when a high advertisement. For the high congruence condition, the campaign was congruence campaign implies more of a self-centered benefit, posi- about supporting a charity to conserve the natural environment. For tive effect of high congruence campaign will be attenuated. the low congruence condition, the campaign was about supporting Linguistics literature has documented that semantically same cancer research. In addition, we manipulated sentence form. For the content can be perceived differently depending on whether it is de- high congruence condition, active form read “Exxon supports the scribed in an active or passive form (Turner and Rommetveit 1968; World Wide Fund for Nature,” while the passive form read “The Johnson-Laird 1968; Tannenbaum and Williams 1968). For example, World Wild Fund for Nature is supported by Exxon.” For the low pro-social behavior of a company can be described either in an active congruence condition, the active form read “Exxon supports the Na- form (e.g., Company A supports charity B) or a passive form mes- tional Cancer Institute,” while the passive form read “The National sage (Charity B is supported by company A). Tannenbaum and Wil- Cancer Institute is supported by Exxon.” After reading one of four liams (1968) documented that sentence form has a functional differ- CSR advertisements depending on experimental condition, partici- ence in focus, with active sentence having a more conceptual focus pants evaluated the company using two seven-point scales (1 = Very on the actor agent, whereas passive form places more attention on the negative, 7 = Very positive; 1 = Very bad, 7 = Very good). The two acted-upon agent. For example, after reading active vs. passive form items were averaged to form a composite index ( = .95). An ANOVA sentences, participants recalled actor agents more correctly for active revealed a significant interaction between congruence and sentence sentences, whereas acted-upon agents were recalled better from the form on evaluation of the company (F (1, 169) = 5.19, p = .024). passive sentences (Turner and Rommetveit 1968). Difference in con- As expected, for high congruence campaign, participants rated the ceptual focus also impacts readers’ attitudes toward the behavioral company more favorably when the CSR advertisement used passive agent of an event. Using context of negative events, recent research than active form messages ( = 4.37, = 3.73; F (1, 169) = 6.11, p = has shown that people blame the behavioral agent more when the .014). No difference was revealed for the low congruence condition ( situation is described in active sentences compared to passive sen- = 4.03, = 4.23; F (1, 169) = .57, n.s.). tences (Fausey and Boroditsky 2010). A 2 (congruence: low vs. high) by 2 (sentence form: active vs. passive) between-subject study was conducted with online panel

Advances in Consumer Research 742 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 743 participants (n = 171). The target company was a car manufacturer, Study 3 directly manipulated consumer skepticism to examine which was recently involved in a scandal about environmental pollu- the relationship between suspicion of self-centered benefit, commit- tion. Similar to study 1a, participants read about the target company, ment to CSR and company’s evaluation. Prior work demonstrates and then were exposed to either one of four CSR advertisements that suspicion of an ulterior motive decreases evaluation of the agent depending on experimental condition. The CSR campaign was either (Campbell and Kirmani 2000). Holding the campaign constant, we supporting natural environment protection (high congruence) or sup- manipulated the level of skepticism toward advertisement in general, porting art and culture education (low congruence). Sentence form and examined whether passive CSR messages improve consumer re- was manipulated as in study 1a. For the high congruence condition, action through greater perceived commitment. active form read “Volkswagen supports protection of the natural en- A 2 (skepticism: low vs. high) by 2 (sentence form: active vs. vironment around the world,” while the passive form read “The natu- passive) between-subject design study was conducted with online ral environment protection around the world is supported by Volk- panel participants (n = 136). Skepticism toward advertisement was swagen.” For the low congruence condition, the active form read manipulated by asking participants to evaluate an irrelevant product “Volkswagen supports art and culture education,” while the passive advertisement before main stimuli. In the high skepticism condition, form read “Art and culture education is supported by Volkswagen.” the target product was claimed to be better based on the company’s After reading the advertisement, participants reported intention to own research, whereas the research was done by independent or- participate in CSR campaign on two seven-point scale items. One ganization in the low skepticism condition. Next, all participants item was about intention to share the advertisement with their friends read the same CSR advertisement of a chocolate company that helps on social media (1 = Definitely not share the ad, 7 = Definitely share homeless people, as in study 2 low congruence condition. Then they the ad), and the other one was to report their willingness to partici- rated perceived commitment of the company to CRS campaign with pate in the campaign by donating money (1 = Not at all, 7 = Very two items ( = .89). Finally, the participants evaluated the company much). The two items were averaged to form a composite index ( with two items as in study 1a ( = .94). Replicating study 2 effect, = .80). Replicating the results of study 1a, an ANOVA revealed a an ANOVA revealed a significant interaction between skepticism significant interaction between congruence and sentence form on and sentence form on perceived commitment (F (1, 132) = 3.94, p = participation intentions index (F (1, 167) = 6.39, p = .012). In the .049), such that the company seemed more committed with passive high congruence condition, participants expressed higher participa- (vs. active) messages in high skepticism condition ( = 4.84, = 4.19; tion intention with the passive (vs. active) form message ( = 3.31, = F (1, 132) = 3.35, p = .07), but not in low skepticism condition ( = 2.39; F (1, 167) = 4.86, p = .029). No difference was observed for the 4.46, = 4.80; F (1, 132) = .95, n.s.). Replicating studies 1a and 1b low congruence condition ( = 2.41, = 2.98; F (1, 167) = 1.85, n.s.). effect, ANOVA showed a marginally significant interaction on com- First two studies demonstrated the predicted effect of sentence pany evaluation (F (1, 132) = 3.45, p = .066), such that the company form on improving evaluation of CSR and the company running it, was favored more with passive (vs. active) messages in high skepti- when the pro-social campaign is highly congruent with the compa- cism condition ( = 4.77, = 4.14; F (1, 132) = 2.89, p = .091), but not ny’s core business. One potential limitation of previous studies is in low skepticism condition ( = 4.82, = 5.15; F (1, 132) = .85, n.s.). that the company was also involved in negative consequences related A moderated mediation analysis (model 7, 5000 bootstraps) demon- to its core business. To address this limitation, in study 2, partici- strated that perceived commitment mediated the interaction effect on pants were directly exposed to a CSR advertisement without reading company evaluation only for the high skepticism condition (.422, company’s information first to verify that the passive form effect still 90% CI: .015, .855). holds without any prior knowledge about the company. With four studies, we show that passive form messages in a Further, study 2 measured whether passive form sentence im- CSR advertisement improve consumer reaction to the company proves perceived company commitment for a high congruent CRS and its CSR campaign when there is high congruence between the campaign. Given that the passive (vs. active) form shifts focus from campaign and its core business. We argue that this happens because the company to the CSR campaign, we expect that it will highlight passive CSR messages reduce suspicion about self-centered benefits how committed the company is to its CSR campaign. But this effect from the campaign by shifting focus from the agent (company) to the will only emerge when the campaign is highly congruent with the acted-upon agent (CSR campaign). cause. This is because consumers attribute self-centric benefit to the This research contributes to a CSR literature by extending prior company solely when level of company-cause congruence is high. findings about company-cause congruence. By showing how pas- No such attribution occurs when the congruency level is low. sive CSR messages improve high congruent campaign evaluation, Online panel participants (n = 162) read a CSR advertisement we highlight a new way to enhance the success of high fit CSR cam- of a chocolate company. The target campaign for high congruence paign. We also demonstrate the positive effect of high congruence condition was helping hungry people in need, whereas supporting can be improved by signaling the company’s commitment. In ad- building houses for homeless was the campaign in the low condition, our findings provide evidence that sentence form influences gruence condition. After reading the advertisement, participants agent evaluation in the context of CSR campaigns. For managers, reported the extent to which they agreed with a statement that the this research suggests an easy-to-implement tool that maximizes the company is truly committed to its CSR campaign on a seven-point positive effect of high fit CSR campaign. Compared to other factors scale (1=Strongly disagree, 7=Strongly agree). An ANOVA revealed such as length of campaign or donation amount, changing sentence a significant congruence by sentence interaction on perceived com- structure is a cost efficient way for companies to improve evaluation mitment (F (1, 158) = 3.70, p = .056). As expected, for the high con- and success of their CSR campaigns. gruence condition, participants in the passive form condition (vs. active) reported that the company was more committed to its campaign REFERENCES ( = 4.90, = 4.22; F (1, 158) = 4.10, p = .045), whereas no difference Campbell, Margaret C. and Amna Kirmani (2000), “Consumers’ was observed for the low congruence condition ( = 4.40, = 4.63; F Use of Persuasion Knowledge: The Effects of Accessibility (1, 158) = .49, n.s.). and Cognitive Capacity on Perceptions of an Influence Agent,” Journal of Consumer Research, 27(1), 69–83. 744 / How Passive Form Messages in CSR Advertisement Improve Consumer Reaction to the Campaign

Chernev, Alexander and Sean Blair (2015), “Doing Well by Pracejus, John W. and G. Douglas Olsen (2004), “The Role Doing Good: The Benevolent Halo of Corporate Social of Brand/cause Fit in the Effectiveness of Cause-Related Responsibility,” Journal of Consumer Research, 41(6), Marketing Campaigns,” Journal of Business Research, 57(6), 1412–25. 635–40. Drumwright, Minette E. (1996), “Company Advertising with Rifon, Nora J., Sejung Marina Choi, Carrie S. Trimble, and Hairong a Social Dimension: The Role of Noneconomic Criteria,” Li (2004), “Congruence Effects in Sponsorship: The Mediating Journal of Marketing, 60(4), 71–87. Role of Sponsor Credibility and Consumer Attributions of Ellen, Pam Scholder, Deborah J. Webb, and Lois A. Mohr (2006), Sponsor Motive,” Journal of Advertising, 33(1), 29–42. “Building Corporate Associations: Consumer Attributions for Sen, Sankar and C.B. Bhattacharya (2001), “Does Doing Good Corporate Socially Responsible Programs,” Journal of the Always Lead to Doing Better? Consumer Reactions to Academy of Marketing Science, 34(2), 147–57. Corporate Social Responsibility,” Journal of Marketing Fausey, Caitlin M. and Lera Boroditsky (2010), “Subtle Linguistic Research, 225–43. Cues Influence Perceived Blame and Financial Liability,” Simmons, Carolyn J. and Karen L. Becker-Olsen (2006), Psychonomic Bulletin & Review, 17(5), 644–50. “Achieving Marketing Objectives Through Social Forehand, Mark R. and Sonya Grier (2003), “When Is Honesty Sponsorships,” Journal of Marketing, 70(4), 154–69. the Best Policy? The Effect of Stated Company Intent on Tannenbaum, Percy and Frederick Williams (1968), “Generation Consumer Skepticism,” Journal of Consumer Psychology, of Active and Passive Sentences as a Function of Subject 13(3), 349–56. of Object Focus,” Journal of Verbal Learning and Verbal Johnson-Laird, Philip N. (1968), “Shorter Articles and Notes the Behavior, 7(1), 246–50. Interpretation of the Passive Voice,” Quartely Journal of Turner, Elizabeth Ann and Ragnar Rommetveit (1968), “Focus of Experimental Psychology, 20(1), 69–73. Attention in Recall of Active and Passive Sentences,” Journal Koschate-Fischer, Nicole, Isabel V. Huber, and Wayne D. Hoyer of Verbal Learning and Verbal Behavior, 7(2), 543–48. (2016), “When Will Price Increases Associated with Company Yoon, Yeosun, Zeynep Gürhan-Canli, and Norbert Schwarz Donations to Charity Be Perceived as Fair?,” Journal of the (2006), “The Effect of Corporate Social Responsibility (CSR) Academy of Marketing Science, 44(5), 608–26. Activities on Companies With Bad Reputations,” Journal of Consumer Psychology, 16(4), 377–90. To Touch or Not to Touch?: How Touch Influences Decision Confidence Sang Kyu Park, University of Florida, USA Yang Yang, University of Florida, USA

EXTENDED ABSTRACT situations where touching noticeably provides no information. How- An unfortunate reality of online shopping is that people do not ever, because touching actually provides no useful information in have the opportunity to touch the actual products. When choosing such cases, the positive expectation is shattered when given the op- between different options online, people often wish that they could portunity to touch, and it thus ironically leads to decreased decision touch the options in person and believe that doing so would result confidence. We further demonstrate the downstream consequences in an easier choice (Childers, Carr, Peck, and Carson 2001; Zhou, of diminished confidence including lower willingness-to-pay (e.g., Dai, and Zhang 2007). In most cases, touching provides useful in- Chaxel 2016) and lower choice satisfaction (e.g., Iyengar and Lepper formation by allowing people to examine the haptic characteristics 2000; Roster and Richins 2009). The proposed effect of touch is first of a target including texture and shape, which in turn widely affect shown in a controlled, non-product context for a cleaner test, and attitudes and behaviors (see Krishna 2012 for a review). However, later generalized to product choice contexts. little is known for cases where touching does not provide any useful information. Do people still desire the opportunity to touch and ex- Studies 1A & 1B pect touching to be helpful? For instance, do people naïvely believe Study 1A first establishes that people generally mispredict that that being able to touch a bottle of wine would help, even though nondiagnostic haptic information would enhance one’s decision con- touching the bottle itself offers no informative advantage? If so, does fidence. Participants were asked to imagine two people, A andB, touching in those cases actually help? participating in a lucky draw for a free chocolate bar individually. Participants were given an image of two identical opaque envelopes. Theoretical Background They were told that each person had to pick from two envelopes Building on the primordial nature of touch and the ubiquitous where one has a winning card with the letter X on it, and the other use of haptic information, we first posit that people rely heavily on has a non-winning card with the letter Y on it. They were further in- touch. Touch is indeed the very first sense humans develop (Atkin- formed that the instructor had person A touch both envelopes before son and Braddick, 1982; Bernhardt 1987), and is also the most fun- making a decision and had person B make a choice by pointing to damental way we explore and interact with the world (Gallace and one of the envelopes without touching. Without learning about the Spence 2009; Lederman and Klatzky 1987). Touching is ever-more outcome, participants were asked to predict each person’s decision important in consumption contexts as it not only provides important confidence and the likelihood that they picked the winning envelope. information about the product (Peck and Childers 2003), but also Despite the fact that merely touching the two identical envelopes because doing so itself may induce positive affect (Peck and Wiggins does not provide any useful information, participants predicted that 2006). Previous research has further shown that people also tend to doing so would increase confidence for the decision maker. The pat- incorporate non-diagnostic, irrelevant haptic cues into judgment and tern favoring those who touched was replicated when the scenario decision making (e.g., Ackerman, Nocera, and Bargh 2010; Kirshna was administered between-subjects (i.e., description of person A pro- and Morrin 2008). vided in one condition, and of person B provided in the other), and Given the fruitful evidence, we propose that people have long when scenario was altered to adopt a first-person perspective (i.e., been accustomed to gathering information through touch, and thus asking participants in which case they would feel more confident, expect that touching would invariably prove to be useful. Similar to rather than making judgments about person A or B). how internalized habits and heuristics continue to function in situ- The results were reversed in Study 1B where we had participants ations where they serve no purpose (Arkes and Ayton 1999; Baron actually go through the lucky draw procedure described in Study 1A. 2000; Amir and Ariely 2007), we contend that people overgeneralize As in Study 1A, participants were told that one envelope had the their belief in effectiveness of touch-for-information so as to predict winning card X while the other had non-winning card Y inside, and that touching would aid their decisions even in situations where it that they would win a free chocolate bar if they pick the envelope explicitly provides no informative value. Our conceptualization is in with the winning card. In the no-touch condition, the instructor held line with the recent stream of research showing that people overgen- up both envelopes and participants made their choice by pointing to eralize various kinds of beliefs. (Hsee, Yang, and Ruan 2015; Pey- one envelope. In the touch condition, the instructor handed the two sakhovich and Rand 2015). envelopes to participants so that they could touch the envelopes and While people overgeneralize and expect that touching a tar- make their decision. Participants’ confidence and perceived likeli- get would help their decision even in cases where it shouldn’t, we hood of having picked the winning card was measured before the show that touching decreases decision confidence when people are results were revealed. Counter to people’s prediction, when given actually given the opportunity to touch and realize that it gives them an opportunity to touch, experiencers felt less confident about their no valid information. That is, when touching fails to acquire useful decision, and responded that they were less likely to have picked the information, expectation disconfirmation and thwarted information envelope with the winning card. search effort renders individuals frustrated (Diehl and Poynor 2010; Raghubir and Krishna 1999; Strebel, O’Donnell, and Myers 2004; Studies 2A & 2B Sun and Spears, 2012) and less confident in their decision (Berkow- We replicate this pair of results with an altered design to rule itz 1989; Roseman 1984). Tinted by the inherent penchant towards out an alternative explanation: touching increases attachment to both touch, however, people fail to anticipate the dismay; only after they envelopes and makes it harder to forgo either option (Carmon and experience the frustration do they realize it. Wertenbroch 2003; Peck and Shu 2009), causing decreased confi- In sum, we propose that consumers hold an overgeneralized na- dence. Only one envelope was involved in the modified procedure. ïve belief that touching would somehow help their decision even for People were told that there were 20 envelopes in total of which 10

Advances in Consumer Research 745 Volume 46, ©2018 746 / To Touch or Not to Touch?: How Touch Influences Decision Confidence had a card with letter X and the other 10 had a card with letter Y Specifically, participants in the no-touch condition were pro- written on it. The instructor would randomly pick one envelope and vided with a picture of two different flavors of coffee pods from the the lucky draw entrant’s job was to guess which letter was inside the same brand priced equally. They were told that one of them received envelope. Participants in Study 2A (the predictors) expected the per- a higher rating for taste and that their job was to pick the better coffee son with the opportunity to touch the envelope to have a higher con- pod. In the touch condition, the same information was given except fidence. However, participants in Study 2B (the experiencers) felt that the two coffee pods were provided in a cup next to the monitor. less confident when they actually touched the envelope compared Participants in both conditions were told “Please take a moment and to those who did not touch the envelope. We also measured involve- examine the two coffee pods,” and had to wait at least 10 seconds ment and ownership, but there were no differences between touch vs. before they could proceed. Then, participants indicated their choice, no touch conditions in Study 2B. followed by the two confidence measures, satisfaction for choice, and willingness to pay (slider scale from $0 to $2) for the chosen Studies 3A and 3B option. The results showed that participants who touched the two Studies 3A (the predictors) and 3B (the experiencers) general- coffee pods felt less confident about their decision, and thought that ize the effect to product choice contexts. The predictors were asked they were less likely to have picked the better one. Also, participants to imagine two people, person A and person B trying to choose be- were less satisfied with their choice when they touched the two cof- tween two bottles of wine separately. Participants were shown two fee pods, mediated by the average of confidence and likelihood mea- bottles of wines (both 2014 Napa Valley Cabernet Sauvignons) of sure (5,000 bootstrap samples, 95% CI [068. 491]). Those in touch equal price which they could enlarge to read the information on the condition also reported significantly lower willingness to pay, again labels. They were further told that one of the wines received a high- mediated by the two confidence measures (5,000 bootstrap samples, er rating, and that each person was trying to choose the wine that 95% CI [005. 056]). The significance of mediation analyses did not received the higher rating (the two wines actually received similar change when confidence and likelihood measures were used sepa- ratings in reality). In the scenario, it was explicitly stated that both rately for both WTP and satisfaction. person A and person B were fully informed about the descriptions We also measured expectation disconfirmation for those in the provided on the label of each wine, but they were not informed of the touch condition (“How informative was the experience of touching ratings. Note that we used wines as touching the two identical bottles the two coffee pods? 1 = much less informative than I expected, 4 = does not provide any useful information. In line with the previous equally informative as I expected, 7 = much more informative than I results, the predictors judged the person who had the opportunity to expected, Gurhan-Canli and Maheswaran 2000; Swan and Trawick touch to be more likely to have picked the better wine. 1981). In line with our conceptualization, participants reported that For experiencers (Study 3B), we actually showed the two bot- the touch experience was significantly less informative than expect- tles of wine mentioned above and provided information about the ed (M = 3.05, vs. midpoint: t(125) = -7.75, p < .001). two wines as in Study 3A (i.e., origin and price). They were further told that one of the wines received a higher rating, and were asked to General Discussion choose the wine they thought received the higher rating, mirroring a The results shed light on the optimistic bias people hold for typical purchase decision. We told participants that they would get touching, and how such belief may negatively influence their deci- a free candy if they pick the wine with the higher rating to motivate sion confidence. While people expect touching to increase decision decision. In the no-touch condition, we held up the two bottles of confidence even in situations where touching should not generate wines so that the participants can see the front label. We asked “Can any useful information, providing the opportunity to touch actually you tell us which one you think is the wine that received the higher reduces confidence. The findings here extends our understanding of rating?” and participants responded by either pointing to one of the the role of touch in consumption contexts by providing pioneering wines, or by saying the name of the wine. In the touch condition, we evidence for when and how touching may have an undesirable ef- handed over the two bottles of wine so that people can hold them fect on decision confidence. Further, our research shows that people and asked the same question mentioned above. The back label of the hold a strong, positive naïve optimism towards touch, and overgen- wine was removed to hold information constant across touch and eralize its effectiveness. Thus, this paper joins the emerging stream no-touch conditions. Prior to disclosing which wine was better, par- of research examining how people may readily apply deeply-rooted ticipants’ perceived likelihood of having picked the better wine was beliefs and habits in irrelevant contexts (Hsee, Yang, and Ruan 2015; measured. In stark contrast to the pattern of predictors, people who Peysakhovich and Rand 2015). actually touched both bottles of wine felt that they were less likely to On the managerial side, the findings suggest that providing the have chosen the better wine (i.e., less confident) in comparison to the opportunity to touch for consumers may not be a panacea. Specifi- counterparts who made the decision without touching. cally, for products for which haptic cues are irrelevant for decision making, touching may in fact decrease choice satisfaction and will- Study 4 ingness-to-pay. The results imply that it may not be in the best inter- Study 4 (experiencers) replicates our findings using a differ- est of managers to facilitate touch for those products, even though ent product (i.e., coffee pods), and utilizing a less obtrusive proce- consumers may seek to do so. Non-touch channels including online dure. One could argue that 1) explicitly instructing participants to and catalogue retailers may also want to implement strategies that ‘touch’ the objects in the touch condition may be driving the effect prevent people from automatically overgeneralizing their belief in (confound and/or demand characteristics) or 2) the way in which touch. researchers interact with participants in the touch conditions could have led to decreased confidence. To address these concerns, we REFERENCES conducted a computer-mediated study in the lab to minimize experi- Ackerman, Joshua M., Christopher C. Nocera, and John A. menter effects. Also, we held the instructions constant across touch Bargh (2010), “Incidental haptic sensations influence social vs. no-touch conditions without using the word ‘touch’ nor ‘feel’. judgments and decisions,” Science, 328(5986), 1712-1715. Advances in Consumer Research (Volume 46) / 747

Amir, On, and Dan Ariely (2007), “Decisions by rules: The case Krishna, Aradhna, and Maureen Morrin (2007), “Does touch affect of unwillingness to pay for beneficial delays,” Journal of taste? The perceptual transfer of product container haptic Marketing Research, 44(1), 142-152. cues,” Journal of Consumer Research, 34(6), 807-818. Arkes, Hal R., and Peter Ayton (1999), “The sunk cost and Lederman, Susan J., and Roberta L. Klatzky (1987), Concorde effects: Are humans less rational than lower “Hand movements: A window into haptic object animals?” Psychological bulletin, 125(5), 591. recognition,” Cognitive psychology, 19(3), 342-368. Atkinson, J., and O. Braddick (1982), “Sensory and perceptual Peck, Joann, and Terry L. Childers (2009), “To have and to capacities of the neonate,” Psychobiology of the human hold: The influence of haptic information on product newborn, 191-220. judgments,” Journal of Marketing, 67(2), 35-48. Berkowitz, Leonard (1989), “Frustration-aggression hypothesis: Peck, Joann, and Suzanne B. Shu (2009), “The effect of mere examination and reformulation,” Psychological touch on perceived ownership,” Journal of consumer bulletin, 106(1), 59. Research, 36(3), 434-447. Bernhardt, Janice (1987), “Sensory capabilities of the fetus,” MCN: Peck, Joann, and Jennifer Wiggins (2006), “It just feels good: The American Journal of Maternal/Child Nursing, 12(1), 44- Customers’ affective response to touch and its influence on 47. persuasion,” Journal of Marketing, 70(4), 56-69. Carmon, Ziv, Klaus Wertenbroch, and Marcel Zeelenberg (2003), Peysakhovich, Alexander, and David G. Rand (2015), “Habits of “Option attachment: When deliberating makes choosing feel virtue: Creating norms of cooperation and defection in the like losing,” Journal of Consumer Research 30(1), 15-29. laboratory,” Management Science, 62(3), 631-647. Chaxel, Anne-Sophie (2016), “Why, When, and How Raghubir, Priya, and Aradhna Krishna (1999), “Vital dimensions in Personal Control Impacts Information Processing: A volume perception: Can the eye fool the stomach?,” Journal of Framework,” Journal of Consumer Research, 43(1), 179-197. Marketing Research, 36(3), 313-326. Childers, Terry L., Christopher L. Carr, Joann Peck, and Stephen Roseman, Ira J. (1984), “Cognitive determinants of emotion: A Carson (2001), “Hedonic and utilitarian motivations for online structural theory,” Review of personality & social psychology. retail shopping behavior,” Journal of Retailing, 77(4), 511- Roster, Catherine A., and Marsha L. Richins (2009), “Ambivalence 535. and attitudes in consumer replacement decisions,” Journal of Diehl, Kristin, and Cait Poynor (2010), “Great expectations?! Consumer Psychology, 19(1), 48-61. Assortment size, expectations, and satisfaction,” Journal of Strebel, Judi, Kathleen O’Donnell, and John G. Myers Marketing Research, 47(2), 312-322. (2004), “Exploring the connection between frustration Gallace, Alberto, and Charles Spence (2009), “The cognitive and consumer choice behavior in a dynamic decision and neural correlates of tactile memory,” Psychological environment,” Psychology & Marketing, 21(12), 1059-1076. bulletin, 135(3), 380. Sun, Qin, and Nancy Spears (2012), “Frustration and Consumer Gürhan-Canli, Zeynep, and Durairaj Maheswaran (200), “Cultural Evaluation of Search Advertising and Search Engine variations in country of origin effects,” Journal of Marketing Effectiveness: The Case of Hedonic Versus Utilitarian Research, 37(3), 309-317. Product,” Journal of Electronic Commerce Research, 13(2), Hsee, Christopher K., Yang Yang, and Bowen Ruan (2015), “The 122-134. mere-reaction effect: Even nonpositive and noninformative Swan, John E., and I. Frederick Trawick (1981), “Disconfirmation reactions can reinforce actions,” Journal of Consumer of expectations and satisfaction with a retail service,” Journal Research, 42(3), 420-434. of Retailing, 57(3), 49-67. Iyengar, Sheena S., and Mark R. Lepper (2000), “When choice Zhou, Lina, Liwei Dai, and Dongsong Zhang (2007), “Online is demotivating: Can one desire too much of a good shopping acceptance model-A critical survey of consumer thing?,” Journal of Personality and Social Psychology, 79(6), factors in online shopping,” Journal of Electronic Commerce 995-1006. Research, 8(1), 41-62. Krishna, Aradhna (2012), “An integrative review of sensory marketing: Engaging the senses to affect perception, judgment and behavior,” Journal of Consumer Psychology, 22(3), 332- 351. Beyond Subjectivity: Competing Governance Regimes and the Socio-Material Construction of Rational Consumer Action Léna Pellandini-Simányi, Università della Svizzera italiana, Switzerland

EXTENDED ABSTRACT Veresiu’s (2014) notion of ’capabilization’, to enable consumers to Rationality – originally assumed to be perfect, then to be more act on a subjectivity developed by discourses. bounded – is the core tenet of economic theories of action. Con- Our model does not deny that organizational discourses and de- sumer behavior provides a more nuanced picture with an entire sub- vices shape consumer action through subjectivity. However, it sug- discipline dedicated to the different modalities in which consumers gests that it is only one of the modalities through which action is process information and make decisions (Mowen 1988; Novak and shaped; and stresses the importance of the alternative route when Hoffman 2009; Olshavsky and Granbois 1989; Solomon et al. 2006; organizational discourses and devices act directly on consumer ac- Yoon, Sarial-Abi, and Gürhan-Canli 2012); only some of which, as tion without being mediated by subjectivity. an exception rather than a rule, resemble the rational decision-mak- Second, building on recent calls (Askegaard and Linnet 2011), ing model of economics. and prior research (Giesler and Veresiu 2014; Peñaloza 2000; Thomp- In recent years, CCT research took a step back from the ques- son and Tian 2008) to study how organizations shape meanings and tion of how consumers make decisions and inquired instead into how consumer subjectivity, our model extends the focus on consumers to consumers evolve into subjects capable and willing to make certain the producers of discourses and devices. Existing studies, drawing decisions: how they become particular consumer subjects. Research on a diverse body of theories, including governmentality (Foucault on the formation of consumer subjects in CCT (Giesler and Vere- 1997; Knights 1997) and performativity (MacKenzie, Muniesa, and siu 2014; Karababa and Ger 2011), sociology of governmentality Siu 2007) show how a specific organization produces discourses (Langley 2007; Miller and Rose 1990) and history of consumption and devices that not only reflect its ideas of the ‘consumer’ but also (Trentmann 2006) suggests that consumer subjectivity arises through contribute to the emergence of the consumer fitting these ideas. Our particular, historically-specific discourses. Building on a Foucauld- model introduces a comparative aspect to this analysis. It shows that ian tradition, many focused on how discourses assume and call forth different types of organizations hold different ideas of how consum- particular consumer subjects, such as responsible consumers (Giesler ers are and how they should be. Organizational ideas are products of and Veresiu 2014), active consumers (Karababa and Ger 2011) or different institutional frames, theories and interests prevalent in the ‘compliant patients’ (Thompson 2003). given organization. Extending these studies, this paper focuses on rational choice These diverse ideas of the consumer materialize in institution- and proposes a model of the construction of rational consumer ac- specific discourses and devices – all potentially capable of perform- tion. We develop our arguments through the case study of mortgage ing the consumer that they assume. The acknowledgement of this di- borrowing in post-socialist Hungary, based on 50 interviews with versity leads to our last and key contribution. While existing research borrowers, 49 expert interviews with bankers and regulators and ex- traces how a specific discourse or device shapes consumer subjec- tensive archive data. The model conceptualizes consumer choice as tivity, our model stresses that consumers are simultaneously subject an outcome of ‘performation struggles’ (Callon, Millo, and Muniesa to multiple, competing discourses and devices that are conducive 2007) between devices and discourses produced by different organi- to different consumer actions; in our case, to different degrees and zations, which are shaped by organization-specific information on forms of rational choice. Using Callon’s (2007, 343) notion of ‘per- consumers, organizational interpretative frames and organizational formation struggle’, we theorize actual consumer action as a tempo- aims. rary ‘win’ of one, or of a combination of, these competing devices We make three contributions. First, existing models imply that and discourses. In this respect our model differs from Giesler and a particular consumer action is driven by a specific subjectivity. Veresiu’s (2014) P.A.C.T. (personalization, authorization, capabiliza- Hence, in these models, discourses and practices shape consumer action, and transformation) model, which suggests that competing dis- tion through subjectivity. We propose an alternative route in which courses settle their battle first (through the first two stages), and then discourses and practices shape consumer action without being neces- the winner agenda gets to create its corresponding consumer subject sarily accompanied by a shift in subjectivity. To develop this point, (till the battle is re-opened). Instead, it chimes with Thompson’s we draw on the work of Michel Callon (Callon 1998; Callon et al. (Thompson and Tian 2008; Thompson 2003) approach that stresses 2007; Callon and Muniesa 2005). Callon and his followers argue that power struggles among different discourses in shaping meanings and for rational action to take place, rational subjectivity is not necessary. consumer subjects. Replacing personal agency with a distributed, sociomaterial view of In our model, competing discourses and devices simultaneously agency, Callon suggests that rational action is a product of socio-ma- act and struggle over the consumer choice. It is their relative power terial arrangements that enable people to act rationally, even if they that matters in formatting behavior. While studies of power follow- do not fully possess the subjectivity of the homo economicus (Cal- ing a Foucauldian line of analysis highlighted the subtle ways in lon 1998). For example, by using an on-line credit calculator, people which power operates through instilling self-governing mechanisms can make complex rational choices without possessing sophisticated (referenced as ‘governmentality’), our model follows Latour’s work calculative capacities. Drawing on this point, the literature on market (Latour 2005; Latour and Venn 2002) to add the very material, coer- devices documents how the material set-up of consumer choices and cive power dimension of devices shaping consumer choice. uses format consumer action, which may or may not be accompanied by a concomitant change in subjectivity (Callon et al. 2007; Cochoy REFERENCES 2008; Cochoy, Deville, and McFall 2017). This process differs from Askegaard, Søren and Jeppe Trolle Linnet (2011), “Towards an existing models of consumer subjectivity, which incorporates devic- Epistemology of Consumer Culture Theory: Phenomenology es either as means of developing a subjectivity or, as in Giesler and and the Context of Context,” Marketing Theory, 11 (4), 381–404.

Callon, Michel (1998), The Laws of the Markets, Oxford; Malden: MacKenzie, Donald, Fabian Muniesa, and Lucia Siu, eds. (2007), Blackwell. Do Economists Make Markets? On the Performativity of Callon, Michel, Y. Millo, and F Muniesa, eds. (2007), Market Economics, Princeton: Princeton University Press. Devices, Malden, MA ; Oxford: Blackwell Pub./The Miller, Peter and Nikolas Rose (1990), “Governing Economic Sociological Review. Life,” Economy and Society, 19 (1), 1-31. Callon, Michel and F. Muniesa (2005), “Economic Markets as Mowen, John C. (1988), “Beyond Consumer Decision Making,” Calculative Collective Devices,” Organization Studies, 26 (8), Journal of Consumer Marketing, 5 (1), 15-25. 1229-50. Novak, Thomas P. and Donna L. Hoffman (2009), “The Fit of Cochoy, F. (2008), “Calculation, Qualculation, Calqulation: Thinking Style and Situation: New Measures of Situation- Shopping Cart Arithmetic, Equipped Cognition and the Specific Experiential and Rational Cognition,”Journal of Clustered Consumer,” Marketing Theory, 8 (1), 15-44. Consumer Research, 36, 56-72. Cochoy, Franck, Joe Deville, and Liz McFall, eds. (2017), Markets Olshavsky, Richard W. and Donald H. Granbois (1989), “Consumer and the Arts of Attachment, London: Routledge. Decision Making - Fact or Fiction,” Journal of Consumer Foucault, Michel (1997), “The Ethics of the Concern for the Self Research, 6, 93-100. as a Practice of Freedom,” in The Essential Works of Michel Peñaloza, Lisa (2000), “The Commodification of the American Foucault. Volume I. Ethics: Subjectivity and Truth, ed. Paul West: Marketers’ Production of Cultural Meanings at the Trade Rabinow, New York: New Press, 281-301. Show,” Journal of Marketing, 64 (4), 82-109. Giesler, Markus and Ela Veresiu (2014), “Creating the Responsible Solomon, Michael R., Gary Bamossy, Soren Askegaard, and Consumer: Moralistic Governance Regimes and Consumer Margaret K. Hogg (2006), Consumer Behaviour: A European Subjectivity,” Journal of Consumer Research, 41 (3), 840-57. Perspective, Harlow; New York: Financial Times/Prentice Karababa, Emİnegül and Gülİz Ger (2011), “Early Modern Hall. Ottoman Coffeehouse Culture and the Formation of the Thompson, Craig J. (2003), “Natural Health Discourses and Consumer Subject “ Journal of Consumer Research, 37 (5), the Therapeutic Production of Consumer Resistance “ The 737-60. Sociological Quarterly, 44 (1), 81-107. Knights, D. (1997), “Governmentality and Financial Services: Thompson, Craig and Kelly Tian (2008), “Reconstructing the Welfare Crises and the Financially Self-Disciplined Subject,” South: How Commercial Myths Compete for Identity Value in Regulation and Deregulation in European Financial through the Ideological Shaping of Popular Memories and Services, ed. G. Morgan and D. Knights, Basingstoke: Countermemories “ Journal of Consumer Research, 34 (5), Palgrave Macmillan, 216-36. 595-613. Langley, Paul (2007), “Uncertain Subjects of Anglo-American Trentmann, Frank (2006), “The Modern Genealogy of the Financialization,” Cultural Critique, 65 (Fall), 67-91. Consumer: Meanings, Identities and Political Synapses,” Latour, Bruno (2005), Reassembling the Social: An Introduction to in Consuming Cultures, Global Perspectives: Historical Actor-Network-Theory, Oxford ; New York: Oxford University Trajectories, Transnational Exchanges, ed. John Brewer and Press. Frank Trentmann, Oxford, UK; New York, NY: Berg, 19-70. Latour, Bruno and Couze Venn (2002), “Morality and Technology: Yoon, Yeosun, Gülen Sarial-Abi, and Zeynep Gürhan-Canli The End of the Means,” Theory, Culture & Society, 19 (5/6), (2012), “Effect of Regulatory Focus on Selective Information 247-60. Processing,” Journal of Consumer Research, 39 (1), 93-110. Vicarious Pride: When Gift Customization Increases Recipients’ Appreciation of the Gift Marta Pizzetti, Università della Svizzera italiana, Switzerland Michael Gibbert, Università della Svizzera italiana, Switzerland

EXTENDED ABSTRACT toolkits vs. gift selection was made by surfing different websites) and Traditionally, customization researchers have investigated process length. We found that recipients appreciated the gift more whether and why consumers appreciate customization (e.g. Fiore, when it was customized, and the vicarious pride was greater in the Lee, and Kunz 2004; Franke, Keinz, and Steger 2009). Consumers customized gift condition. Vicarious pride mediated the relationship often customize products for someone else as a gift and value cus- between gift and appreciation. These findings provide further sup- tomization more highly if it is intended as a gift (Moreau, Bonney, port to the idea that recipients appreciate a gift more when it is cus- and Herd 2011). However, it hasn’t been examined the recipients’ tomized thanks to vicarious pride. This effect appears to be indepen- appreciation of customized gifts. The current paper focuses on how dent from time and energy spent on selecting or customizing the gift. customization affects the gift appreciation. Study 3 further examines the phenomenon and includes a rela- We contend that gift recipients appreciate customized gifts be- tional factor, since gift exchanges have a strong effect on relation- cause they experience vicarious pride. Pride is a pivotal consequence ships (Ruth, Brunel, and Otnes 2004) and vicarious experiences are of customization: Customizers refer pride in the customized product, affected by relational factors (Hatfield, Cacioppo, and Rapson 1994). which increases the value placed on the product (‘I designed it my- Here, we included relational anxiety because of its relevance in the self’ effect; Franke, Schreier, and Kaiser 2010). We propose that a gift-exchange literature (e.g., Nguyen and Munch 2014; Ward and similar mechanism may occur when a recipient takes the perspective Broniarczyk 2011). Specifically, we expect that that relational anxi- of a giver customizing a product: The recipient experiences pride ety is an obstacle to vicarious pride, because it prevents emotional vicariously. This proposition builds on simulation theory: Individu- connection with others (Bartz and Lydon 2004). Thus, when the reals react to others’ mental states and actions by mentally replicating lationship is filled with anxiety, the vicarious pride is not elicited them, and this internal simulation elicits the same psychological ef- and consequently recipients do not appreciate more the customized fects as the actual performance of the action (e.g. Decety and Som- gift. A 2 (gift: customized vs. non-customized) x 2 (anxiety: no vs. merville 2008; Rizzolatti and Craighero 2004). We propose a psy- high) between-subjects design involved 184 participants (58.7% chological transfer between giver and recipient: The feeling of pride male, Mage = 32.73). Relational anxiety was manipulated as difficulty generated by self-designing a product (Franke et al. 2010) translates (vs. ease) of being emotionally close to someone (Bartz and Lydon from the customizer to the final user of the product (i.e., the recipi- 2004). Then, the scenario about the gift (watch customization vs. ent). To test this hypothesis, we conducted three experiments. watch selection between standard watches) was presented, followed The aim of Study 1 is to demonstrate that recipients appreci- by items on gift appreciation and vicarious pride. Analyses showed ate a gift more when it is customized (vs. selected between a set of that vicarious pride was greater in the no-anxiety than in the high- standard options) because of vicarious pride. We recruited partici- anxiety condition and, in the no-anxiety condition, customized gifts pants through the snowball-sample procedure. The study involved elicited greater vicarious pride. Customized gifts were significantly more appreciated in the no-anxiety condition. A mediation analysis real pairs of friends. 74 participants (41.9% male, Mage = 25.45) were involved in a single factor between-subjects design (gift: customized on the no-anxiety condition confirmed that vicarious pride mediated vs. not-customized). The gift was exactly the same in both condi- the relationship between gift customization and gift appreciation. tions (a clothing look), but differently described. In the customized Three studies show that customization is a valuable source of gift condition, the scenario presented a selection process carried out gifts because of the activation of vicarious pride: Recipients are vi- step by step via adding look attributes (vs. the look was selected be- cariously proud of the customized product. The parallelism in terms tween a set of predefined looks). Then, we measured vicarious pride of pride between the giver and recipient also affects the gift evalua- (Franke et al. 2010) and allowed participants to change the attributes tion: Customized gifts are more highly appreciated when givers and of the look they did not like as a proxy for gift appreciation – the low- recipients are emotionally connected. Conversely, when the relation- er the changes, the higher the appreciation. The amount of changes ship is filled with anxiety, vicarious pride is prevented. The effect ranged from 0 (= no changes) to 4 (= all items were changed). The seems to be strong: we used two different measures of gift appre- results showed that recipients experienced vicarious pride and were ciation, visual and textual descriptions of the customization process, less willing to modify a gift when it was customized. Moreover, vi- and different products. Future research should investigate whether carious pride mediated the relationship between gift customization and how other characteristics of the gift, such as the fit with the giv- and appreciation. er’s desire, amplify or reduce the effect of vicarious pride. In Study 2 we replicated the experimental design but with a different measure of gift appreciation (two items from Ward and REFERENCES Broniarczyk 2011) and including a second factor. Indeed, an alter- Bartz, Jennifer A., and John E. Lydon (2004), “Close Relationships native explanation might be that customization is perceived by gift and the Working Self-Concept: Implicit and Explicit Effects of recipients as a time- and energy-consuming activity for the giver. Priming Attachment on Agency and Communion”, Personality Such time and energy efforts might be a driver of gift appreciation, and Social Psychology Bulletin, 30(11), 1389-1400. since recipients prefer gifts that require a high amount of time and Decety, Jean and Jessica A. Sommerville (2008), “Action energy to the giver (Robben and Verhallen 1994). 109 participants Representation as the Bedrock of Social Cognition: A (42.4% male; Mage = 25) participated in a 2(gift: customized vs. Developmental Neuroscience Perspective”. In The Oxford not-customized) x 2(time and energy: high vs. low) between-subjects Handbook of Human Action, eds. Ezequiel Morsella, John A design. The gift was exactly the same between the two conditions Bargh and Peter M. Gollwitzer, New York: Oxford University (a T-shirt), but we showed two different videos to manipulate the Press, 250–75. gift-selection process (customization process undertaken via online

Fiore, Ann M., Seung-Eun Lee, and Grace Kunz (2004), Rizzolatti, Giacomo and Laila Craighero (2004), “The Mirror- “Individual Differences, Motivations, and Willingness to Use a Neuron System”, Annual Review of Neuroscience, 27, Mass Customization Option for Fashion Products”, European 169–192. Journal of Marketing, 38(7), 835-849. Robben, Henry S. J. and Theo M. M. Verhallen (1994), “Behavioral Franke, Nikolaus, Peter Keinz and Christoph J. Steger (2009), Costs as Determinants of Cost Perception and Preference “Testing the Value of Customization: When Do Customers Formation for Gifts to Receive and Gifts to Give”, Journal of Really Prefer Products Tailored to Their Preferences?” Journal Economic Psychology, 15(2), 333-350. of Marketing, 73(5), 103-121. Ruth, Julie A., Frédéric F. Brunel and Cele C. Otnes (2004), “An Franke, Nikolaus, Martin Schreier and Ulrike Kaiser (2010), Investigation of the Power of Emotions in Relationship “The “I Designed it Myself” Effect in Mass Customization”, Realignment: The Gift Recipient’s Perspective”, Psychology & Management Science, 56(1), 125-140. Marketing, 21(1), 29-52. Hatfield, Elaine, John T. Cacioppo and Richard L. Rapson (1994), Ward, Morgan K. Susan M. Broniarczy (2011), “It’s not Me, it’s Emotional Contagion. New York: Cambridge University Press. You: How Gift Giving Creates Giver Identity Threat as a Moreau, C. Page, Leff Bonney and Kelly B. Herd (2011), “It’s Function of Social Closeness”, Journal of Consumer Research, the Thought (and Effort) that Counts: How Customizing for 38(1), 164-181. Others Differs from Customizing for Oneself”,Journal of Marketing, 75(5), 120-123. Nguyen, Hieu P. and James M. Munch (2014), “The Moderating Role of Gift Recipients’ Attachment Orientations on Givers’ Gift-giving”, Journal of Consumer Behavior, 13(5), 373-382. The Viciousness and Caring of Sharing: Conflicts and Motivations of Online Shamers Chen Pundak, Tel-Aviv University, Israel Yael Steinhart, Tel-Aviv University, Israel Jacob Goldenberg, IDC Herzliya, Israel

EXTENDED ABSTRACT As expected, the mean-centered interaction between moral con- Public shaming is defined as informally punishing individuals cern and wrongdoer-identification (0=Low-identifiability, 1=High- who have deviated socially or morally, by informing the public about identifiability) was significant (b=-.65, p=.0033). their conduct (Petley 2013). Our research focuses on individuals who While in the high-identifiability wrongdoer condition, we found engage in public shaming on social media (e.g., Twitter, Facebook) a significant positive association between moral concern and shar- against a wrongdoer who has acted immorally. While public shaming ing likelihood (b = .69, SE = .15, t(274) = 4.53, p < .001), under has long been a tool of education, its embrace by the digital world the low-identifiability wrongdoer condition, the association was not enabling users to easily and effectively spread information about the significant (b = .03, SE = .16, t(274) = .22, p = .83). misconduct (Goldman 2015; Jacquet 2015). In a replication study, participants (n=374) were presented with The decision to engage in public shaming consists of two con- a post that described a manager of a restaurant who had ignored a tradictory motivations: informal enforcement against deviant behav- complaint about possible food poisoning. Half the participants saw a ior (thereby preventing offenses such as animal abuse or discrimina- photo of the restaurant’s manager, and the other half did not. Again, tory customer service) and the violation of privacy rights and dignity the interaction between moral concern and wrongdoer-identification (exposing the personal details of the wrongdoer). condition had significant effect on participants’ likelihood of sharing We suggest that a person considering participating in online the post. shaming may try to balance these opposing considerations. Specifi- In our fourth study (n=393), we attenuated the negative out- cally, when a wrongdoer identifiability is high (e.g., when a post comes of shaming by presenting more than one target as being describing the behavior is accompanied by a clear image or a full shamed, thereby reducing the likelihood that a given wrongdoer name), shaming that individual can give rise to both potential conse- would be hurt: Participants were presented with a post describing quences described above. When the wrongdoer identifiability is low either a spa receptionist or a group of receptionists who had ignored (e.g., a vague image is shown, or only a first name is presented), an elderly customer. A photo of the receptionist or receptionists was shaming is more likely to give rise to the potential positive conse- presented either from the back (low-identifiability) or from the front quences than to trigger the potential negative consequences. This (high-identifiability). Also, we manipulated moral concern instead of tradeoff may lead prospective shamers to experience a moral dilem- measuring them. An ANOVA found a three-way interaction: Among ma, since moral concerns involve the desire to protect others but also participants under the high-moral-concern manipulation, those in the restrain them from doing harm (Graham et al. 2011; Janoff-Bulman low-identifiable-single-receptionist condition were more likely to and Carnes 2013, Crocket, 2017). Therefore, users with high levels share the post than were participants in the high-identifiable-recep- of moral concern are more likely to join in on public shaming when tionist condition (Mmanipulation_low-identifiable_receptionist=4.55 vs. Mmanipulation_high- the wrongdoer’s identifiability level is low rather than high. identifiable_receptionist= 3.46, t(385)=2.59, p=0.01). A preliminary study (n=100) confirmed the assumption that us- In our last study we focused on the underlying process of our in- ers acknowledge both positive (preventing similar cases and educat- teraction, as well as the moderating role of group relatedness. When ing others) and negative (causing harm to the wrongdoer) sides of a person is being shamed in his own community, the negative out- sharing shaming information. It further showed that users’ evalua- comes and the positive outcomes can be particularly strong. Partici- tions of these consequences vary according to the level of wrongdoer pants (n=305) were presented with a shaming post about an airline identifiability. passenger who had misbehaved. The passenger’s photo was either In our first study we demonstrated how moral concern interact blurry (low-identifiable) or sharp (high-identifiable). Also, the pas- with identifiability levels of the wrongdoer in driving actual choices senger was described either as part of the participants’ ingroup (same regarding participation in public shaming. Participants (n=141) were nationality) or as an outgroup member (different nationality). Higher presented with several posts and were asked to click “like” on one moral concern were positively correlated with sharing likelihood of the posts through their social media accounts. After making their when the wrongdoer was low-identifiable and an ingroup member choices, participants completed a moral concern scale. Logistic re- (b=.25, t(297)=2.04, p=.04). Surprisingly, we also found a significant gression found that level of moral concern significantly predicted the positive effect for moral concern when the wrongdoer was high-iden- likelihood to choose to “like” a shaming post in which the wrongdoer tifiable and an outgroup member (b=.28, t(297)=2.83, p=.005). The identifiability was low rather than high (χ(1)2=4.87, p=.027). Specifi- effect of moral concern on sharing likelihood was mediated by ex- cally, the exp(B) value indicated that an increase of one unit in the pected positive outcomes of sharing the post (b=-.21; 95% CI: -.43, average moral concern score was associated with an increase of 62% -.04) but not by the expected negative outcomes. in the odds ratio of choosing a post that include low-level-identifiable Taken together, our studies suggest that while people acknowl- wrongdoer. We replicated these results in a study (n=209) that con- edge the dual outcomes of shaming, morality and the extent to which sidered #METOO campaign posts. the wrongdoer is identifiable affect the decision to participate in pub- In our second study, participants (n=278) were presented with lic shaming. While negative outcomes are acknowledged, the expect- a shaming post that included a photo of a woman (the wrongdoer) ed positive consequences are the ones that drive potential shamers’ dragging a swan out of a lake. We manipulated the identifiability of decisions regarding whether or not to participate in public shaming. the wrongdoer by presenting different angles of the image: half the participants saw a high-level-identifiable wrongdoer, and half saw a low-level-identifiable wrongdoer. Participants were asked to rate the likelihood they would share this post.

Advances in Consumer Research 752 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 753 REFERENCES Janoff-Bulman, R., & Carnes, N. C. (2013). Surveying the moral Crockett, M.J. (2017). Moral outrage in the digital age. Nature landscape: Moral motives and group-based moralities. Human Behaviour, 1, 769–771. Doi: 10.1038/s41562-017- Personality and Social Psychology Review. 17(3) 219–236, 0213-3 Doi: 10.1177/1088868313480274 Graham, J., Nosek, B. A., Haidt, J., Iyer, R., Koleva, S., & Ditto, P. Kohm, S. A. (2009). Naming, shaming and criminal justice: Mass- H. (2011). Mapping the moral domain. Journal of Personality mediated humiliation as entertainment and punishment. Crime and Social Psychology, 101(2), 366-385. Doi: 10.1037/ Media Culture, 5, 188-205. Doi: 10.1177/1741659009335724 a0021847. Massaro, T. M. (1997). The meanings of shame: Implications for Hayes, A. F. (2015). An index and test of linear moderated legal reform. Psychology, Public Policy and Law, 3(4), 645- mediation. Multivariate Behavioral Research, 50, 1-22. Doi: 704. Doi: 10.1037/1076-8971.3.4.645 10.1080/00273171.2014.962683 Petley, J. (Ed.) (2013). Media and public shaming: Drawing the Jacquet, J. (2015). Is shame necessary? New uses for an old tool. boundaries of disclosure. London: I. B. Tauris & Co. New York: Vintage. Names Are the Mirrors of the Soul: The Role of Possessive Brand Names in Brand Evaluations Marina Puzakova, Lehigh University, USA Mansur Khamitov, Nanyang Technological University, Singapore

EXTENDED ABSTRACT identity marker (Sweet Tatiana): Mpossessive=5.45 vs. Mnon-possessive=5.02; Marketing practitioners frequently rely on brand anthropomor- F(1,115) = 3.690, p=.057. Next, study 2 (n=186, MTurk partici- phism (i.e., endowing brands with humanlike features (Puzakova and pants) replicates the effect in a different service context (a hair-salon Kwak 2017) as an effective strategy that increases affective reactions brand) and examines whether the positive effect of possessiveness and product evaluations (Aggarwal and McGill 2012). One of the holds only for brand names with an identity marker (e.g., Elegant popular means that brand strategists rely on to encourage anthropo- Charlie’s vs. Charlie’s). The results of this study revealed that when morphism is endowing a brand with a human name. For example, an identity marker was present, the possessive (vs. non-possessive)

Lily takes photos of consumers’ adventures, Dave assists customers brand name led to more favorable brand evaluations (Mpossessive=6.14 with banking fees, and a cup of fresh coffee is served at Tim Hor- vs. Mnon-possessive=5.68, F(1,182)=5.652, p=.018). Conversely, in the ton’s. absence of identity marker, brand evaluations did not differ between

Furthermore, a more nuanced analysis of these humanlike names the possessive and the non-possessive brand name conditions (Mposses- points out to subtle semantic variations in a way these brand names sive=5.88 vs. Mnon-possessive=5.78, p=.594, Figure 1). Similar results were are formed. For example, marketers use a possessive form in Tim obtained for the word-of-mouth and purchase intentions. Overall, in Horton’s and Trader Joe’s brand names, whereas a non-possessive line with our theorizing the effect of brand name possessiveness is version is used in the Lily and Dave brand names. Does this subtle mitigated when no identity marker is present. difference matter, and if yes, in what way? In this regard, prior re- Study 3 (n=186, MTurk participants) further examines the me- search in psychology and linguistics indicates that possessive forms diating role of consumers’ attributions of identity-relevant brand have crucial affective consequences (Shi et al. 2011). Despite the attributes and considers potential alternative explanations for our widespread prevalence of humanlike brand names and theoretically results. We tested our effects with advertisements for a brand of ca- important consequences of possessive semantic units, to the best of tering: Artful Mark’s vs. Artful Mark. Replicating results of stud- our knowledge, no prior studies have examined the effectiveness of ies 1 and 2, the findings reveal that a possessive brand name of a using possessiveness as a linguistic cue in anthropomorphized brand catering business that included an identity marker (Artful Mark’s) names. In response, we attempt to fill this gap. Our key contribution led to more favorable brand evaluations than a non-possessive brand lies in demonstrating an important role of brand name possessiveness name with an identity marker (Artful Mark): Mpossessive=6.03 vs. Mnon- in consumers’ anthropomorphized brand inferences and evaluations. possessive=5.74; F(1,184)=4.405, p=.037. Next, the results of the model As a theoretical basis, we draw from both psychological ownership testing the serial relationships (brand name possessiveness → per- theory and linguistics research that demonstrate that perceptions ceived brand ownership → attributions of identity-relevant brand of an individual’s ownership toward an object is easily formed by attributes → brand evaluations; model 6; Hayes 2013) revealed a presenting people with possessive pronouns preceding the target of significant overall mediation effect (Bbrandevaluations=.08, SE=.03, 95% ownership (Pierce, Kostova, and Dirks 2003; Shi et al. 2011). Thus, CI[.03, .16]), That is, a possessive (vs. non-possessive) form of a we propose that the use of a possessive form in brand names can lead brand name led to greater inferences of brand ownership which in to greater consumer inferences that a brand is owned by a specific turn enhanced participants’ attributions of identity-relevant brand at- external entity (implied in the brand name). tributes, thus, resulting in more favorable brand evaluations. Finally, Our second contribution lies in identifying a novel process of our results further ruled out several alternative accounts: responsibil- this effect. Psychologically owned items are seen as extensions of ity for service outcomes and inferred service quality. people who own them (Belk 1988) and can help people communi- Overall, this research identifies important consequences of in- cate important aspects of their identity to others (White and Argo ferring others’ (or anthropomorphized brands’) ownership states. In 2011). Because people frequently experience feelings of psychologi- particular, studies 1 through 3 demonstrate that the use of a posses- cal ownership and signal their identity through owned items to oth- sive (vs. non-possessive) form in anthropomorphized brand names ers (Pierce et al. 2003), we propose that individuals are also likely with an identity marker leads to more favorable brand evaluations, to view others’ owned items as signaling an owner’s identity. As word-of-mouth, and purchase intentions. Study 3 sheds light on the such, consumers are likely to perceive that an inferred brand owner’s underlying process through greater attributions of identity-relevant identity spills over to the brand, subsequently resulting in greater brand attributes, while also ruling out alternative explanations. inferences of identity-relevant brand attributes. Importantly, we ex- Our findings add a new dimension to work on anthropomor- pect this effect to occur only for brand names that include an identity phism (Aggarwal and McGill 2012; Puzakova and Kwak 2017). marker (i.e., a qualifier that provides consumers with the information Although the use of both possessive and non-possessive humanlike about an inferred owner’s identity). Because an identity marker pro- names is a widely prevalent tactic that enhances brand anthropomor- vides unique information about an inferred brand owner’s identity, phism, scant research has systematically investigated nuances of this consumers will view a possessive (vs. non-possessive) brand name branding strategy. In response, our work highlights the impact of an with an identity marker as performing better on identity-relevant important linguistic cue that market strategists can rely on to enhance brand attributes, thus, resulting in more favorable brand evaluations. new customers’ brand perceptions. We test these ideas in three studies. The results of study 1 (n=117, students at a U.S. East Coast university) reveals that a possessive brand name of a café (introduced via an advertisement) joint that includes an identity marker (Sweet Tatiana’s) leads to more favorable brand evaluations than a non-possessive brand name with an

Advances in Consumer Research 754 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 755 REFERENCES Puzakova, Marina and Hyokjin Kwak (2017), “Should Aggarwal, Pankaj and Ann L. McGill (2012), “When Brands Seem Anthropomorphized Brands Engage Customers? The Impact Human, Do Humans Act Like Brands? Automatic Behavioral of Social Crowding on Brand Preferences,” Journal of Priming Effects of Brand Anthropomorphism,”Journal of Marketing, 81 (6), 99-115. Consumer Research, 39 (2), 307-23. Shi, Zhan, Aibao Zhou, Wei Han, and Peiru Liu (2011), “Effects Belk, Russell W. (1988), “Possessions and the Extended Self,” of Ownership Expressed by the First-Person Possessive Journal of Consumer Research, 15 (2), 139-68. Pronoun,” Consciousness and Cognition, 20 (3), 951-55. Hayes, Andrew F. (2013), Introduction to Mediation, Moderation, White, Katherine and Jennifer J. Argo (2011), “When Imitation and Conditional Process Analysis: A Regression-Based Doesn’t Flatter: The Role of Consumer Distinctiveness in Approach, New York, NY: Guilford Press. Responses to Mimicry,” Journal of Consumer Research, 38 Pierce, Jon L., Tatiana Kostova, and Kurt T. Dirks (2003), “The (4), 667-80. State of Psychological Ownership: Integrating and Extending a Century of Research,” Review of General Psychology, 7 (1), 84-107. The Impact of Anthropomorphized Cute Brands on Consumer Preferences for Distinctive and Majority-Endorsed Products Marina A. Puzakova, Lehigh University, USA Nevena T. Koukova, Lehigh University, USA

EXTENDED ABSTRACT To summarize, we provide important insights regarding the Marketers strategically use anthropomorphism (i.e., imbu- downstream effects of consumer interactions with cute anthropomor- ing nonhuman objects with humanlike characteristics, motivations, phized brands on subsequent preferences for majority-endorsed and intentions or behaviors; Puzakova, Kwak and Rocereto 2013) as a distinctive brands. To our knowledge this is the first investigation brand positioning strategy. Prior work identifies downstream conse- on how exposure to anthropomorphized brands systematically affects quences of using this strategy including positive brand evaluations consumers’ preferences for other brands during a subsequent con- and emotional reactions towards the brand (Aggarwal and McGill sumption experience. Specifically, we argue that men associate cute 2007; Delbaere, McQuarrie and Phillips 2011). However, previous anthropomorphized brands with femininity, which in turn threatens research has not looked at how anthropomorphizing a brand can their gender identity. Hence, men’s exposure to anthropomorphized systematically affect consumer preferences for a different brand in cute products prior to choice in a different product category de- a consumption sequence. An understanding of this phenomenon is creases their preferences for distinctive brands and enhances their both theoretically and managerially important because consumers attraction toward majority-endorsed brands. We further demonstrate frequently shop for products in predictable patterns (e.g., looking for the underlying process: allowing male consumers to reaffirm their coffee creamers after selecting a pack of coffee). Importantly, mar- gender identity before the second choice task attenuates the nega- keters carefully place products in particular sequences in retail stores tive impact of cute anthropomorphized brands on their preferences and effectively utilize cross-selling tactics to increase the purchase of for distinctive brands. While prior work sheds light on the impact of a full product line (Knott, Hayes and Neslin 2002). anthropomorphism on consumer preferences of an entity being an- In this paper we examine a specific brand positioning strategy— thropomorphized (Aggarwal and McGill 2007; Puzakova and Kwak cute brand anthropomorphism (imbuing a humanized brand with 2017), our work establishes the novel downstream effects of interact- specific configuration of infantile features; Wang and Mukhopad- ing with cute anthropomorphized brands on subsequent consumption hyay 2015)—and demonstrate that consumer exposure to or inter- of other brands positioned as distinctive or majority-endorsed. On action with a cute anthropomorphized (vs. nonanthropomorphized) the practical front, our research highlights a very interesting interac- brand has significant downstream consequences for preferences for tion effect between gender and brand anthropomorphism in the con- products or brands positioned to be either distinctive or majority- text of a subsequent consumption task. The role of gender identity endorsed (popular). We further establish that these unique effects of in consumer exposure to cute anthropomorphized brands is an inter- cute brand anthropomorphism hold only for male consumers. Spe- esting juxtaposition of factors that marketers may wish to consider cifically, we demonstrate that men’s exposure to anthropomorphized when developing brand strategies while taking into consideration cute products prior to exposure to a distinctive brand in a different sequences of consumption choices. product category can have a negative effect on the preferences for the distinctive brand. In contrast, exposure to anthropomorphized REFERENCES cute products enhances male consumers’ preferences for majority- Aggarwal, Pankaj and Ann L. McGill (2007), “Is This Car endorsed brands. The theoretical rationale is that anthropomorphized Smiling at Me? Schema Congruity as Basis for Evaluating cute products threaten men’s gender identity and activate self-protec- Anthropomorphized Products,” Journal of Consumer tion responses. This activation of a self-protection response further Research, 34 (December), 468-79. facilitates an evolutionary response of “going with the group is safe” Bianchi, S. M., Robinson, J. P., & Milkie, M. A. (2006), “Changing because standing out from the crowd limits one’s chances of survival Rhythms of American Family Life,” Russell Sage Foundation, (Griskevicius et al. 2009). Thus, our work demonstrates that cute an- New York: NY. thropomorphized products increase conformity in men; these prod- Brosch, Tobias, David Sander, and Klaus R. Scherer (2007), “That ucts reduce attraction to distinctively-positioned brands and increase Baby Caught My Eye … Attention Capture by Infant Faces,“ attraction to brands positioned as majority-endorsed. Emotions, 7 (3), 685-89. We conducted three studies to test our hypotheses. Studies 1 Brough, Aaron R., James E. B. Wilkie, Jingjing Ma, Mathew S. and 2 demonstrate that exposure to cute anthropomorphized prod- Isaac, and David Gal (2016), “Is Eco-Friendly Unmanly? ucts reduces men’s attraction to distinctive brands in a subsequent The Green-Feminine Stereotype and Its Effect on Sustainable choice task, whereas this positioning strategy has no effect for wom- Consumption,” Journal of Consumer Research, 43, 567-82. en. Study 3 extends the findings of studies 1 and 2 by providing in- Delbaere, Marjorie, Edward F. McQuarrie, and Barbara J. sights on the underlying process. Our conceptualization predicts that Phillips (2011), “Personification in Advertising: Using a men’s exposure or interaction with cute anthropomorphized brands Visual Metaphor to Trigger Anthropomorphism,” Journal of threatens their masculine identity. If this is the case, then allowing Advertising, 40 (Spring), 121–30. consumers to reaffirm their identity before the second choice task Griskevicius, Vladas, Noah J. Goldstein, Chad R. Mortensen, Jill should attenuate the negative impact of cute anthropomorphized M. Sundie, Robert B. Cialdini, and Douglas T. Kenrick (2009), brands on preferences for brands positioned to be distinctive. Study 3 “Fear and Loving in Las Vegas: Evolution, Emotion, and highlights the underlying process using a moderation approach, and Persuasion,” Journal of Marketing Research, 46 (3), 384–95. shows that the negative impact of cute anthropomorphized brands on preferences for brands positioned to be distinctive can be attenuated by allowing consumers to reaffirm their identity before evaluating the distinctive brand.

Griskevicius, Vladas, Noah J. Goldstein, Chad R. Mortensen, Puzakova, Marina and Hyokjin Kwak (2017), “Should Robert B. Cialdini, and Douglas T. Kenrick (2006), “Going Anthropomorphized Brands Engage Customers? The Impact Along versus Going Alone: When Fundamental Motives of Social Crowding on Brand Preferences,” Journal of Facilitate Strategic (Non)Conformity,” Journal of Personality Marketing, 81 (6), 99-115. and Social Psychology, 91 (2), 281-94. Puzakova, Marina, Hyokjin Kwak, and Joseph F. Rocereto (2013), Huang, Xun (Irene), Ping Dong, and Anirban Mukhopadhyay “When Humanizing Brands Goes Wrong: The Detrimental (2014), “Proud to Belong or Proudly Different? Lay Theories Effect of Brand Anthropomorphization Amid Product Determine Contrasting Effects of Incidental Pride on Wrongdoings,” Journal of Marketing, 77 (3), 81-100. Uniqueness Seeking,” Journal of Consumer Research, 41 (3), Wan, Echo Wen, Jing Xu, and Ying Ding (2014), “To Be or Not 697-712. to Be Unique? The Effect of Social Exclusion on Consumer Knott, Aaron, Andrew Hayes, and Scott A. Neslin (2002), “Next- Choice,” Journal of Consumer Research, 40 (6), 1109-22. Product-to-Buy Models for Cross-Selling Applications,” Wang, Tingting and Anirban Mukhopadhyay (2015), “How Journal of Interactive Marketing, 16 (3), 59-75. Consumers Respond to Cute Products,” in The Psychology Kurt, Didem, J. Jeffrey Inman, and Jennifer J. Argo (2011), “The of Design: Creating Consumer Appeal, eds. Rajeev Batra Influence of Friends on Consumer Spending: The Role of and Colleen Seifert and Diann Brei. New York and London: Agency-communion Orientation and Self-Monitoring,” Routledge, 149-67. Journal of Marketing Research, 48 (August), 741-54. Zebrowitz, Leslie A. and Joann M. Montepare (2005), “Appearance Nenkov, Gergana Y. and Maura L. Scott (2014), “So Cute I could DOES Matter,” Science, 308 (June), 1565-66. Eat It Up: Priming Effects of Cute Products on Indulgent Consumption,” Journal of Consumer Research, 41, 326-41. Trust, But Verify: A Multi-level Examination of Online Reviews and Persuasion Knowledge Martin A. Pyle, Ryerson University, Canada Andrew Smith, Suffolk University, USA Yanina Chevtchouk, University of Glasgow, UK

EXTENDED ABSTRACT branded websites) or past experience where content was largely Stories of companies (and pranksters) manipulating online re- sponsored (e.g., certain blogs). Third-party sites tended to receive views are abound in the news. These actions potentially affect con- greater trust, but in a qualified form; our informants’ PK schemata sumer trust not only at the review-level, but also their perceptions of at the platform-level included known barriers for posting content, specific websites (platform-level) and online reviews as a whole (sys- and inferences about how that impacts the pervasiveness of decep- tem-level), as consumers struggle to discern real reviews from fake. tive reviews. Even when perceptions about a platform did not prompt The Persuasion Knowledge Model (PKM, Friestad & Wright, avoidance, they still tended to heighten PK or lead to other coping 1994) offers a useful framework for understanding how consumers mechanisms. As purchase importance increased, informants tended identify and respond to persuasion attempts. The basic idea behind to cope by engaging in cross-platform comparisons. this framework is that consumers develop contextualized persuasion At the system-level, informants continued to apply coping knowledge (PK) about persuasion tactics, agents, and appropriate mechanisms intended to corroborate the information. While one responses to allow them to navigate the marketplace. Researchers informant practiced complete avoidance of online reviews, the re- have explored PK in the context of direct persuasion situations, such maining informants tended to complement the information from on- as sales or advertising (Campbell & Kirmani, 2000; Kirmani & Zhu, line reviews with offline sources (trusted people) or through directly 2007), but the relationship between PK and online reviews is less visiting the retail outlet (webrooming). This verification and lack of clear, largely due to the limited research on the topic. reliance on online reviews increased in relation to the perceived im- From a broad perspective, online reviews can take many forms; portance of the purchase. from individual consumer reviews (e.g., Amazon, ePinions) to blogs The consistent activation of PK at every level, and the resulting and vlogs. Research suggests that subtler forms of sponsored con- coping mechanisms are simultaneously troublesome and indicative tent, such as branded editorials, do not always activate skepticism of opportunities from a managerial perspective. Considering the role and recognition of persuasive intent (Boerman, Willemsen, & Van of online reviews—reducing risk (Hennig-Thurau & Walsh, 2003) Der Aa, 2017). Consumer reviews may be particularly challenging, and simplifying the decision making process, the extra steps consum- as those that are sponsored may be nearly identical to genuine per- ers take to verify the information could lead to more people bypass- spectives. One study noted a wide variation in consumers’ applica- ing this information. However, this also suggests that efforts to high- tion of PK in this context (Bambauer-Sachse & Mangold, 2013), light verified reviews (as opposed to eliminating fake ones) could be though the prevalence of fake reviews has received more attention in well-received by consumers. the media since this study, perhaps leading to greater PK activation. Thus, we investigate: to what extent does PK activation occur in the REFERENCES context of online reviews? Bambauer-Sachse, S., & Mangold, S. (2013). Do consumers still Based on two available studies, avoidance was the most promi- believe what is said in online product reviews? A persuasion nent coping behavior when PK was activated in an online consumer knowledge approach. Journal of Retailing and Consumer review context (Bambauer-Sachse & Mangold, 2013; Reimer & Services, 20(4), 373-381. Benkenstein, 2016). However, the focus was at the review level, and Boerman, S. C., Willemsen, L. M., & Van Der Aa, E. P. (2017). did not consider the effects on consumer perceptions of the platform “This Post Is Sponsored”: Effects of Sponsorship Disclosure or system-wide, of reviews in general. Therefore, we investigate a on Persuasion Knowledge and Electronic Word of Mouth in second set of questions: how does PK manifest at the system- and the Context of Facebook. Journal of Interactive Marketing, 38, platform-levels? What subsequent coping mechanisms do consum- 82-92. ers employ? Campbell, M. C., & Kirmani, A. (2000). Consumers’ use of To investigate these questions, we use data from 22 semi-struc- persuasion knowledge: The effects of accessibility and tured depth interviews (McCracken, 1988) that involved general cognitive capacity on perceptions of an influence agent. questions as well as real online product searches, where informants Journal of consumer research, 27(1), 69-83. narrated their thoughts, and responded to prompting questions. The Friestad, M., & Wright, P. (1994). The persuasion knowledge data was analyzed through the PKM independently by two of the model: How people cope with persuasion attempts. Journal of authors, who also met for cross-informant comparison. consumer research, 21(1), 1-31. The data indicates high levels of skepticism towards reviews Hennig-Thurau, T., & Walsh, G. (2003). Electronic word-of- from all informants, suggesting the development of PK surrounding mouth: Motives for and consequences of reading customer this context. The PK schemata tended to include typical company articulations on the Internet. International journal of electronic approaches for influencing reviews (e.g., ‘click-farms’, sponsorship), commerce, 8(2), 51-74. and as expected, led to several reviews being avoided or counter- Kirmani, A., & Zhu, R. (2007). Vigilant against manipulation: The argued because informants felt they were fake. However, other cop- effect of regulatory focus on the use of persuasion knowledge. ing behaviors were also revealed, such as using site-level heuristics Journal of Marketing Research, 44(4), 688-701. to confirm consensus opinions. Notably, analysis revealed that many McCracken, G. (1988). The long interview (Vol. 13). Sage. informants did not differentiate between sponsored and biased re- Reimer, T., & Benkenstein, M. (2016). When good WOM hurts and views, categorizing both as fake. bad WOM gains: The effect of untrustworthy online reviews. At the platform-level, certain sites were simply avoided because Journal of Business Research, 69(12), 5993-6001. of preconceived notions of active filtering of negative reviews (e.g.,

Advances in Consumer Research 758 Volume 46, ©2018 Bundle Variety and Preference: A Neuromarketing Study Using Event-related Potentials Ruyi Qiu, Tsinghua University, China Xiaoang Wan, Tsinghua University, China

EXTENDED ABSTRACT p = .10. We also performed a 2 (Bundle Variety: lower or higher) × Bundling has raised many interests for decades (e.g., Hansen & 4 (Electrode Site: Fz, FCz, CPz, or Pz) repeated-measure ANOVA Martin, 1987; Janiszewski & Cunha Jr, 2004), but very few studies on the N2 amplitude. The results revealed no significant main or in- have been conducted to systematically examine the relationship be- teraction effects, all Fs < 1.66, ps > .21. That said, planned pairwise tween bundle variety and bundle preference. With the development comparisons showed that only the N2 amplitude in electrode Pz was of consumer neuroscience, three event-related-potential components, significantly smaller for bundles with higher variety than those with namely the P1, P2 and N2, have been linked to consumer preference lower variety, t (19) = 2.60, p < 0.05, Cohen’s d = 0.61; whereas there for single products (Goto et al., 2017; Guo et al., 2016; Micklebor- was no such effect in other electrodes, allts < 0.45, ps > .65. ough et al., 2014; Telpaz et al., 2015; Zhang et al., 2011). However, Furthermore, the 2 (Bundle Preference: most- or least-pre- it remains unclear whether these early components could reveal con- ferred) × 2 (Electrode Site: P7/P8 or PO7/PO8) × 2 (Hemisphere: sumer preference for product bundles, and if so, how the component left or right) repeated-measure ANOVA on the P1 amplitude revealed variety of bundles might modulate this effect. The present study ad- a marginal significant main effect of Bundle Preference, F (1, 19) = 2 dressed this issue to shed some light on the neural underpinnings of 3.94, p = 0.06, ηp = 0.17. However, the 2 (Bundle Variety: lower or consumer’s responses to product bundles. higher) × 2 (Electrode Site: P7/P8 or PO7/PO8) × 2 (Hemisphere: In the present study, we chose bundles consisted of three cups left or right) repeated-measure ANOVA on the P1 amplitude showed of yogurt in fruit flavors as experimental stimuli, because variety no such effect of Bundle Variety, F (1, 19) = 1.87, p = 0.19. By con- seeking behavior was found to be more likely to occur with hedonic trast, the analogous ANOVA on the P2 amplitude revealed a signifi- 2 products (Baltas et al., 2017; Inman, 2001). We first run an online cant main effect of Bundle Variety, F (1, 19) = 7.82, p < 0.05, ηp = survey on 79 naïve participants to determine which yogurt flavors 0.29; whereas the 2 (Bundle Preference: most- or least-preferred) × to use in the main experiment. Based on these results, we decided to 2 (Electrode Site: P7/P8 or PO7/PO8) × 2 (Hemisphere: left or right) use strawberry and yellow peach as two popular flavors (PFs), sea- repeated-measure ANOVA on the P2 amplitude showed no such ef- berry and water chestnut as two unpopular flavors (UFs), as well as fect of Bundle Preference, F (1, 19) = 0.001, p = 0.98. coconut, mulberry, and cherry as three flavors with moderate popu- In conclusion, the present study provides the first physiologi- larity (MFs). We used four types of bundles, including the 3PF, 3UF, cal evidence investigating the relationship between bundle variety 2P1U, and 3MF bundles. Importantly, the 3PF and 3UF bundles were and bundle preference. The results revealed both associations and identical bundles, whereas the 2P1U and 3MF bundles were coded as dissociations between consumer preference and variety seeking. Our bundles with lower and higher variety, respectively. findings suggest that the frontal area is more involved in inhibiting Twenty young healthy participants took part in the main ex- less preferred bundles (Aron et al., 2004, 2014), whereas the parietal periment which consisted of a pre- and a post-test survey, a wanting area is linked to the categorization of bundles according to its vari- test, and a selection test. At the beginning of the experiment, the par- ety level (Freedman & Assad, 2006, 2011; Levine & Schwarzbach, ticipants chose which one of the two popular flavors they preferred 2017). Furthermore, our findings suggest that bundle preference is more, and which one of the two unpopular flavors they preferred associated with early configural processing, whereas later featural less. The chosen flavors were then presented in the following tests processing is important for bundle variety (Boutsen et al., 2006; Mer- for each participant. In both of the pre- and post-test surveys, par- cure et al., 2008; Wang et al., 2016). These findings not only have ticipants were asked to indicate their liking of each bundle. The EEG implications in helping researchers and practitioners better under- signals were recorded during the wanting test in which participants stand consumer preference and variety seeking, but also indicate the viewed the picture of each bundle and used the mouse to pick a num- promising future of using ERP method in neuromarketing practice. ber from 0 to 100 to indicate their desire for it. Each participant completed two blocks of 120 trials each, while equal numbers of different REFERENCES types of bundles were mixed and presented in a random order. After Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition the wanting test, participants were presented with pairs of bundles in and the right inferior frontal cortex. Trends in Cognitive the selection test, and they were asked to choose the preferred ones. Sciences, 8(4), 170-177. The rating results revealed some significant positive correla- Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2014). Inhibition tions among the wanting scores, preference scores (the total number and the right inferior frontal cortex: one decade on. Trends in of times each bundle was chosen by each participant), and liking Cognitive Sciences, 18(4), 177-185. scores, all rs > 0.72, ps < 0.001, suggesting the validity of want- Baltas, G., Kokkinaki, F., & Loukopoulou, A. (2017). Does variety ing and preference scores. The results also revealed some signifi- seeking vary between hedonic and utilitarian products? The cant positive correlations between the N2 amplitude for each bundle role of attribute type. Journal of Consumer Behavior, 16, e1- and the wanting/preference scores, both rs > 0.26, ps < 0.05. Next, e12. we conducted a 2 (Bundle Preference: most- or least-preferred) × 4 Boutsen, L., Humphreys, G. W., Praamstra, P., & Warbrick, (Electrode Site: Fz, FCz, CPz, or Pz) repeated-measure ANOVA on T. (2006). Comparing neural correlates of configural the N2 amplitude. The results revealed a significant main effect of processing in faces and objects: an ERP study of the Thatcher 2 illusion. NeuroImage, 32(1), 352-367. Bundle Preference, F (1, 19) = 6.69, p < 0.05, ηp = 0.26. Planned pairwise comparison showed that in electrodes Fz, FCz, and CPz, Freedman, D. J., & Assad, J. A. (2006). Experience-dependent the N2 amplitude was significantly smaller for the most-preferred representation of visual categories in parietal cortex. Nature, bundles than for the least-preferred ones, all ts > 2.17, ps < 0.05, 443(7107), 85-88. whereas no such effect was significant in electrode Pz, t (19) = 1.71,

Freedman, D. J., & Assad, J. A. (2011). A proposed common neural Levine, S. M., & Schwarzbach, J. (2017). Decoding of auditory and mechanism for categorization and perceptual decisions. Nature tactile perceptual decisions in parietal cortex. NeuroImage, Neuroscience, 14(2), 143-146. 162, 297-305. Goto, N., Mushtaq, F., Shee, D., Lim, X. L., Mortazavi, M., Mercure, E., Dick, F., & Johnson, M. H. (2008). Featural and Watabe, M., & Schaefer, A. (2017). Neural signals of selective configural face processing differentially modulate ERP attention are modulated by subjective preferences and buying components. Brain Research, 1239, 162-170. decisions in a virtual shopping task. Biological Psychology, Mickleborough, M. J., Chapman, C. M., Toma, A. S., & Handy, T. 128, 11-20. C. (2014). Cognitive processing of visual images in migraine Guo, F., Ding, Y., Wang, T., Liu, W., & Jin, H. (2016). Applying populations in between headache attacks. Brain Research, event related potentials to evaluate user preferences toward 1582, 167-175. smartphone form design. International Journal of Industrial Telpaz, A., Webb, R., & Levy, D. J. (2015). Using EEG to predict Ergonomics, 54, 57-64. consumers’ future choices. Journal of Marketing Research, Hansen, W., & Martin, R. P. (1987). Optimal bundle pricing. 52(4), 511-529. Management Science, 36, 155-174. Wang, H., Guo, S., & Fu, S. (2016). Double dissociation Inman, J. J. (2001). The role of sensory-specific satiety in attribute- of configural and featural face processing on level variety seeking. Journal of Consumer Research, 28(1), P1 and P2 components as a function of spatial 105-120. attention. Psychophysiology, 53(8), 1165-1173. Janiszewski, C., & Cunha, M. (2004). The influence of price Zhang, Y., Kong, F., Chen, H., Jackson, T., Han, L., Meng, discount framing on the evaluation of a product bundle. J., & Najam ul Hasan, A (2011). Identifying cognitive Journal of Consumer Research, 30, 534-546. preferences for attractive female faces: An event-related potential experiment using a study-test paradigm. Journal of Neuroscience Research, 89(11), 1997-1893. A Rational Model to Predict Consumers’ Irrational Behavior Vahid Rahmani, Rowan University, USA

EXTENDED ABSTRACT βi = P(sqh| qh and rql; sql | ql and rqh); 1- βi = P(sql| qh and rql; The current research developed a mathematical model that could sqh | ql and rqh) predict consumers’ price-quality perceptions based on their ability 0 < βi < βc and motivation to process the product information accurately. This Given the above conditions, the consumption utility of high- model effectively demonstrates how the price can influence consum- quality products [i.e., m(qh|r)] and the consumption utility of lowers’ judgments of quality after they purchased and used the product. quality products [i.e., m(ql|r)] can be conceptualized as below: Furthermore, this model sheds light on the underlying reasons that price has different effects on consumers’ perceptions of quality un- (4) m(qh|r) = ql + αh × (qh – ql) ; αh = [(βc - βi) × α + βi] der different product categories. Finally, this model offers a powerful statistical tool that could be utilized to find the best price-points for (5) m(ql|r) = ql + αl × (qh – ql); αl = [1+ α (βc - βi) – βc] products in dynamic environments. The developed model is based on the reference-dependent util- Therefore, the gain loss utility of the model [n (q|r)] could be ity model proposed by Kőszegi and Rabin (2006). They argued that calculated as: expected utility is the sum of gain/loss utility [n (q|r), consistent with the predictions of prospect theory (Kahneman and Tversky 1979)], (6) n(q|r) = {ƞ[m(q|r) – m(r)] | αh >α or αl > α; ƞλ[m(q|r) – m(r)]| and reference-dependent consumption utility [m (q|r)], where m(q) αh < α or αl < α } and λ > 1 denotes the experience of quality and m(r) denotes the expectation of quality: Based on the gain loss utility function, the total utility functions of high-quality [U(qh|r)] and low-quality products [U(ql|r)] could be (1) U(q|r) = m (q|r) + n (q|r) conceptualized as below:

In this model, when m(q) – m(r) > 0, consumers will experience (7) U(qh|r) = {ql + αh (qh – ql) + ƞ (αh – α) (qh – ql) | αh > α; qL a gain utility equal to ƞ[m(q|r) – m(r)]. In contrast, when m(q) – m(r) + αh (qh – ql) < 0, consumers will experience a loss utility of ƞλ[m(q|r) – m(r)]. λ is always greater than one because as proposed by Kahneman and + ƞλ (αh – α) (qh – ql) | αh < α} Tversky (1979) losses have a greater influence on consumers’ experienced utility (a strong negative effect) than gains. Furthermore, (8) U(ql|r) = {qL + αl (qh - ql) + ƞ (αl – α) (qh - ql) | αl > α; ql + Kőszegi and Rabin (2006) argued that the consumption utility itself αl (qh - qL) + ƞλ (αl – α) (qh - qL) | αl < α} would depend on a reference point. As an example, driving a C-Class Note that if α < βi/(1+ βi- βc) then αh > α. Since 0 < α <1 and Mercedes-Benz would create a higher consumption utility when the 0 < [βi; βc; βi/(1+ βi- βc)]<1, we can define the P(αh > α) = βi/(1+ reference product of the driver is a Toyota Camry than when it is βi- βc) = γ. Similarly, if α < (1-βc )/(1+βi- βc) then αl > α; 0 < (1-βc a Rolls-Royce. This notion is consistent with the placebo effect of )/(1+βi- βc)<1; P(αl > α) = (1-βc )/(1+βi- βc) = 1-γ. Given the cal- the price that is reported in the literature (Shiv, Carmon, and Ariely culated values of γ, we can rewrite the expected utility models as: 2005). To further develop the model, the arguments put forward by (9) SQAh= ql + [αh + (ƞλ + ƞγ - ƞλγ) (αh – α)] (qh – ql) Gneezy, Gneezy, and Lauga (2014) were followed. Based on their study, Alpha was defined as the probability that the product was a high-quality product. Furthermore, as supported by numerous stud- (10) SQAl = ql + [αl + (ƞ + ƞλγ – ƞγ) (αl – α)] (qh – ql) ies in the pricing literature, it was assumed that Alpha was a function of price: (11) ɖSQAh/ɖα = [(βc - βi) + (ƞλ + ƞγ - ƞλγ) ((βc - βi) – 1)] (qh – (2) α = P(rqh) = ƒ (price) ql)

(3) m(r) = ql + α (qh - ql) (12) ɖSQAl/ɖα = [(βc - βi) + (ƞ + ƞλγ – ƞγ) ((βc - βi) –1)] (qh – ql)

After using the products, consumers acquire some additional The predictions of the model were tested using several samples information about the real quality of the product. For high-quality including 519,200 product/day observations from Amazon.com. The products, if consumers’ expectations are consistent with the product findings showed that the predictions of the developed model closely quality, most would judge the quality as high (i.e., sqh). When con- matched the observed relationships in the real-world data. sumers’ expectations of quality are inconsistent with the actual qual- The findings of this paper could present companies with a pow- ity of the products, a sizable percentage of them will make the wrong erful tool that they could use to find the best price points for their judgment about the quality of the product. Based on these arguments, products based on consumers’ product knowledge and thinking style. βc and βi are defined as the probabilities that consumers’ will ac- In practice, companies could conduct a small study to estimate the curately judge the quality when their expectations are consistent and parameters of the model, and then use the model to find the ideal inconsistent with the real quality, respectively. Therefore: price points for their products. Furthermore, this article makes signif- βc = P(sqh| qh and rqh; sql | ql and rql); 1- βc = P(sql| qh and icant theoretical contributions. Extant literature in pricing suggests rqh; sqh | ql and rql) that the effect of price on consumers’ perception of quality is not homogenous across different market segments. This article is the first to offer a mathematical model that can not only explain the previous Advances in Consumer Research 761 Volume 46, ©2018 762 / A Rational Model to Predict Consumers’ Irrational Behavior findings in the literature but predict consumers’ price-quality percep- Kahneman, Daniel, and Amos Tversky (1979), “Prospect Theory: tions based on consumer and product characteristics. An Analysis of Decision Under Risk,” Econometrica: Journal of the Econometric Society, 263-291. REFERENCES Kőszegi, Botond, and Matthew Rabin (2006), “A Model of Gneezy, Ayelet, Uri Gneezy, and Dominique Olié Lauga (2014), Reference-Dependent Preferences,” The Quarterly Journal of “A Reference-Dependent Model of the Price-Quality Economics, 121(4), 1133-1165. Heuristic,” Journal of Marketing Research, 51(2), 153-164. Shiv, Baba, Ziv Carmon, and Dan Ariely (2005), “Placebo Effects of Marketing Actions: Consumers May Get What They Pay For.” Journal of Marketing Research, 42 (4), 383-393. Pangs from Persuasion: When Recommendations Undermine Consumers’ Social Worth Suzanne Rath, Queen’s University, Canada Laurence Ashworth, Queen’s University, Canada Nicole Robitaille, Queen’s University, Canada

EXTENDED ABSTRACT Consistent with existing work, we start with the proposition Research on the Persuasion Knowledge Model (PKM) (Friestad that consumers often assume that persuasion is driven by agents’ and Wright 1994) has documented many negative responses to per- self-interest. In marketing contexts, this is typically a motive to sell. suasion attempts (Isaac and Grayson 2017), including lower attitudes However, we also argue that this is typically well understood by con- towards salespeople (Campbell and Kirmani 2000; Decarlo 2005; sumers. The critical feature of a persuasion attempt is that it com- Main et al. 2007; Reinhard et al. 2006), advertisements (Campbell municates to consumers that the agent is attempting to benefit from 1995; Campbell and Keller 2003; Darke and Ritchie 2007), and that particular consumer. Regardless of whether the agent actually brands (Ahluwalia and Burnkrant 2004; Wei et al. 2008). To explain benefits, this is important because it violates a norm of interpersonal these responses, several mechanisms have been implicated having conduct, to not take advantage of others. We argue that this conveys to do with perceived selling motives, such as suspicion of ulterior a lack of respect and value for the consumer, and that this is an im- motives, inferences of manipulative intent, and consumer skepticism portant reason for consumers’ subsequent reactions (derogating the (Kirmani and Zhu 2007; Isaac and Grayson 2016). However, it is not agent and leaving the interaction). completely clear why selling motives should necessarily incite these In this paper, four studies test various aspects of our theorizing. responses because goals to sell or make profits are well-known to First, we tested whether perceiving a communication as persuasion consumers before exposure to a communication (Babin et al. 1995; is sufficient to convey socially threatening information. Importantly, Boush et al. 1994) and are often top of mind (Thompson 1972; Bhat- we found that a persuasion (vs. neutral) communication lowered con- tacharjee and Dana 2017). An alternative possibility is that consum- sumers’ perceived social worth (8-item, seven-point Likert scale i.e., ers may see attempts to persuade as deceptive. Although this may ex- the salesperson thought highly of you, did not think much of you plain some empirical findings (Forehand and Grier 2003), perceived (r); valued you as a customer, etc…). It also augmented advantage- deception is unlikely to fully account for consumers’ responses since taking beliefs, and it impaired all downstream consumer responses. they respond in the same way without deception (Campbell and Kir- We found results consistent with simple and serial mediation. This mani 2000; Main et al. 2007; Ahluwalia and Burnkrant 2004; Rein- provides initial evidence of the basic effect. In the second study, we hard et al. 2006). provide conceptual replication and extend the findings by varying The current work investigates a different reaction consumers the extent to which the salesperson benefits from the advocated ac- may have to perceived persuasion, which we suggest is responsible, tion – something that should moderate the basic effect. We found in part, for their negative reactions. We suggest that consumers react that the effect of being a target of persuasion on social worth and to the interpersonal implications of being the target of a persuasion advantage-taking beliefs was reduced when the salesperson would attempt. We argue that labelling a communication as “persuasion” benefit less from the tactic compared to when they would benefit activates consumers’ beliefs that the agent is attempting to pursue more. Additionally, we found serial moderated mediation, replicating their own interests instead of (and perhaps at the expense of) con- the basic effect. Our third study provides a conceptual replication of sumers’ interests. This violates an important norm of interpersonal moderation-of-process by varying the diagnosticity of the persuasive conduct – not taking advantage of others – conveying a lack of re- communication. We found that when features of the advocated ac- spect and value for the consumer. In short, we suggest that perceiving tion were less diagnostic of marketers’ beliefs, the effect of being a communication as an attempt to persuade may convey threatening a target of persuasion on social worth and advantage-taking beliefs information about consumers’ social worth. was attenuated. Our final study attempts to isolate social worth, dem- Inferences of social worth (i.e. the extent to which an individual onstrating this is a critical element in consumers’ overall reaction feels respected and valued by others) is a critical part of psychologi- to persuasion. We showed that advantage-taking beliefs affect infer- cal and social functioning. They are an important part of self-esteem ences of social worth when consumers view attempts to profit from (Leary and Baumeister 2000; Leary 2005; Leary and Batts Allen others as a moral violation. 2011) and have a powerful influence over how individuals judge These four experiments provide evidence that persuasion at- and respond to others. In particular, we tend to dislike, derogate, and tempts violate a norm of interpersonal conduct – not taking advan- withdraw from people we perceive hold us with low regard (Buckley tage of others – which conveys a lack of respect and value for the et al. 2004; Leary, Twenge, and Quinlivan 2006; Miller 2001). consumer, and leads consumers to derogate the agent and withdraw People can infer social worth from a wide variety of social cues. from the interaction. These findings suggest that researchers and Sometimes this information is direct, for example, when others di- marketers must be wary of the social consequences persuasion at- rectly state their opinion (Smart Richman and Leary 2009). Other tempts can have on consumers’ reactions and wellbeing. times, individuals infer (a lack of) social worth from actions that can be seen as exclusionary or are perceived as rejection (Buckley REFERENCES et al. 2004), even indirectly. For instance, a large literature in or- Ahluwalia, Rohini. and Robert E. Burnkrant (2004), “Answering ganizational psychology has found that individuals frequently infer Questions about Questions: A Persuasion Knowledge social worth from organizations’ processes and procedures (Blader Perspective for Understanding the Effects of Rhetorical and Tyler 2015), and work in consumer psychology has found that Questions,” Journal of Consumer Research, 31 (1), 26-42. consumers interpret unfavorable prices and poor service in this way too (Ashworth and McShane 2012; Liao 2007). In a similar vein, we suggest that consumers may infer social worth from the assumptions they make about why an agent is attempting to persuade them.

Ashworth, Laurence. and Lindsay McShane (2012), “Why Do Isaac, Mathew S. and Kent Grayson (2017), “Beyond Skepticism: We Care What Others Pay? The Effect of Other Consumers’ Can Accessing Persuasion Knowledge Bolster Credibility,” Prices on Inferences of Seller (Dis)Respect and Perceptions Journal of Consumer Research, 43 (6), 895-912. of Deservingness Violation,” Journal of Retailing, 88 (1), Kirmani, Anma. and Rui Zhu (2007), “Vigilant Against 145-155. Manipulation: The Effect of Regulatory Focus on the Use of Babin, Barry J., James S. Boles, and William R. Darden (1995), Persuasion Knowledge,” Journal of Marketing Research, 44 “Salesperson Stereotypes, Consumer Emotions, and Their (4), 688-701. Impact on Information Processsing,” Journal of the Academy Leary, Mark R. (2005), “Sociometer Theory and the Pursuit of Marketing Science, 23 (2), 94-105. of Relational Value: Getting to the Root of Self-Esteem,” Bhattacharjee, Amit., Jason Dana, and Jonathan Baron (2017), European Review of Social Psychology, 16 (1), 75-111. “Anti-Profit Beliefs: How People Neglect the Societal Benefit Leary, Mark R. and Ashley Batts Allen (2011), “Belonging of Profits,” Journal of Personality and Social Psychology, 113 Motivation: Establishing, Maintaining, and Repairing (5), 671- 696. Relational Value,” In D. Dunning (Ed.), Social Motivation (pp. Blader, Steven L. and Tom R. Tyler (2015), “Relational Models of 37-55). New York: Psychology Press Taylor & Francis Group. Procedural Justice,” In Russell S. Cropanzano and Maureen Leary, Mark R. and Roy F. Baumeister (2000), “The Nature and L. Ambrose (Eds.), The Oxford Handbook of Justice in Function of Self-Esteem: Sociometer Theory,” In M. P. Zanna the Workplace (pp. 351-370). New York, NY, US: Oxford (Ed.), Advances in Experimental Social Psychology (Vol. 32, University Press. pp. 1-62), San Diego, CA, US: Academic Press. Boush, David M., Marian Friestad, and Gregory M. Rose Leary, Mark. R., Ellen S. Tambor, Sonja K. Terdal, and Deborah (1994), “Adolescent Skepticism toward TV Advertising Downs (1995), “Self-Esteem as an Interpersonal Monitor: The and Knowledge of Advertisr Tactics,” Journal of Consumer Sociometer Hypothesis,” Journal of Personality and Social Research, 21 (1), 165-175. Psychology, 68 (3), 518-530. Buckley, Katherine E., Rachel E. Winkel, and Mary R. Leary Leary, Mark R., Jean M. Twenge, and Erin Quinlivan (2006), (2004), “Reactions to Acceptance and Rejection: Effects of “Interpersonal Rejection as a Determinant of Anger and Level and Sequence of Relational Evaluation. Journal of Aggression,” Personality and Social Psychology Review, 10 Experimental Social Psychology, 40 (1), 14-28. (2), 111-132. Cambell, Margaret C. (1995), “When Attention-Getting Advertising Liao, Hui. (2007), “Do It Right This Time: The Role of Employee Tactics Elicit Consumer Inferences of Manipulative Intent: Service Recovery Performance in Customer-Perceived Justice The Importance of Balancing Benefits and Investments,” and Customer Loyalty After Service Failures,” Journal of Journal of Consumer Psychology, 4 (3), 225-254. Applied Psychology, 92 (2), 475-489. Campbell, Margaret C. and Kevin L. Keller (2002), “Brand Main, Kelley J., Darren W. Dahl, and Peter R. Darke (2007), Familiarity and Advertising Repetition Effects,”Journal of “Deliberative and Automatic Bases of Suspicion: Empirical Consumer Research, 30 (2), 292-304. Evidence of the Sinister Attribution Error,” Journal of Campbell, Margaret C. and Amna Kirmani (2000), “Consumers’ Consumer Psychology, 17 (1), 59-69. Use of Persuasion Knowledge: The Effects of Accessibility Miller, Dale T. (2001), “Disrespect and the Experience of and Cognitive Capacity on Perceptions of an Influence Agent,” Injustice,” Annual Review of Psychology, 52 (1), 527-553. Journal of Consumer Research, 27 (1), 69-83. Reinhard, Marc-Andre., Matthias Messner, and Siegfried Darke, Peter R. and Robin J. B. Richie (2007), “The Defensive Ludwig Sporer (2006), “Explicit Persuasive Intent and its Consumer: Advertising, Deception, Defensive Processing, and Impact on Success at Persuasion: The Determining Roles Distrust,” Journal of Marketing Research, 44 (1), 114-127. of Attractiveness and Likeableness,” Journal of Consumer DeCarlo, Thomas E. (2005), “The Effects of Sales Message and Psychology, 16 (3), 249-259. Suspicion of Ulterior Motives on Salesperson Evaluation,” Smart Richman, Laura. and Mark R. Leary (2009), “Reactions to Journal of Consumer Psychology, 15 (3), 238-249. Discrimination, Stigmatization, Ostracism, and Other Forms of Forehand, Mark R. and Sonya Grier (2003), ”When Is Honesty Interpersonal Rejection: A Multimotive Model,” Psychological the Best Policy? The Effect of Stated Company Intent on Review, 116 (2), 365-383. Consumer Skepticism,” Journal of Consumer Psychology, 13 Thompson, Donald L. (1972), “Stereotype of the Salesman,” (3), 349-356. Harvard Business Review, 50 (January-February), 20-25. Friestad, Marian. and Peter Wright (1994), “The Persuasion Wei, Mei-Ling., Eillen Fischer, and Kelley J. Main (2008), Knowledge Model: How People Cope with Persuasion “An Examination of the Effects of Activating Persuasion Attempts, “Journal of Consumer Research, 21(1), 1-31. Knowledge on Consumer Response to Brands Engaging in Hayes, Andrew F. (2018), “Introduction to Mediation, Moderation, Covert Marketing,” Journal of Public Policy & Marketing, 27 and Conditional Process Analysis: A Regression-Based (1), 34-44. Approach,” 2nd Edition, New York, NY: Guildford. The Self-Bolstering Effects of Repeated Affirmations Over Time Alejandra Rodriguez, Oklahoma State University, USA Ted Matherly, Oklahoma State University, USA

EXTENDED ABSTRACT According to self-affirmation theory, it is not necessary to ad- Should you listen again to that song you have been listening to dress a self-threat directly in order to maintain the perception of ade- on repeat, or should you listen to different song? We make consump- quacy; it is also possible to do so by affirming one’s overall self-con- tion choices like this on a daily basis without really knowing if more cept (Steele 1988). For example, if a student’s self-esteem has been (or less) variety is better for us—not just in terms of enjoyment (Gal- threatened after failing an exam, they can restore their perceptions of ak, Redden, and Kruger 2009; Ratner, Kahn, and Kahneman 1999), adequacy by behaving in ways that signal other aspects they value but also psychological wellbeing. This is not an insignificant choice, about themselves, such as through the consumption of products that however, since we regularly use consumption to help us cope when remind them of their athletic skills or social ties. Product choices our perceptions of adequacy are called into question. that convey higher status have been shown to provide affirmational The abundance of and chronic nature of threats to one’s ad- benefits to those whose self has been threatened (Sivanathan and Pet- equacy can cause lasting negative effects on a person’s physical, tit 2010). In addition, a recent study by Townsend and Sood (2012) emotional, and mental wellbeing (Cook et al. 2012; Creswell et al. showed that participants who affirmed by writing a values essay were 2005; Jonas et al. 2014; Sherman and Hartson 2011). By encourag- less likely to choose high design options (vs high functionality). In ing reflection on valued aspects of the self, self-affirmation through another study by the same authors, participants who chose the aes- consumption can be an effective way to prevent these harmful con- thetic product option were more open to arguments, similar to those sequences since possessions and experiences can be closely tied to who had affirmed by writing about their values. one’s sense of self (Belk 1988; Carter and Gilovich 2012; Sherman Single affirmations like those in the previous examples have and Cohen 2006; Steele 1988). A considerable body of research has been shown to have lasting effects. Nonetheless, considering the confirmed the effectiveness of affirmations in restoring the self (Man- wide range of sources from which threats may originate, a single af- del et al. 2017), as demonstrated by higher levels of self-esteem, im- firmation may not be strong enough to counteract all the threats one proved performance, increased optimism, and reduced stress levels might encounter on a daily basis (Cohen and Sherman 2014; Cook et (Sherman and Cohen 2006). However, this research has primarily al. 2012; Jonas et al. 2014; Mandel et al. 2017; Stinson et al. 2011). If investigated these outcomes in isolation, without accounting for the an affirmation attempt is not successful, a person may no longer see fact the individuals may be faced with a choice to either use a variety themselves as adequate, and will need to find another way to affirm of affirmations, or to repeatedly use a single one to defend against the (Galinsky, Stone, and Cooper 2000). myriad threats they may face every day (Cohen and Sherman 2014; It is common to provide participants in studies with a list of Cook et al. 2012). The purpose of this research is to investigate how values which can be affirming (McQueen and Klein 2006), but which the use of varied or repeated affirmations increases their effective- aspects of the self are affirmed naturally is usually determined by ness. what is available and mentally accessible at the time for the individu- Prior work on satiation might suggest that repeatedly using an al (Steele 1988). For instance, in many cases a person’s occupation or affirmation would make it less effective (Brickman and Campbell, social affiliations are especially valued; for others their personal pos- 1971; McSweeney and Swindell, 1999). In this case, variety in af- sessions, such as clothing and accessories, represent many aspects of firmations should be beneficial to their effectiveness. Conversely, an individual’s self and as a result can become chronically salient. repetitive consumption can also be reinforcing, and thus beneficial, Often people lack awareness of who they are, but can infer qualities since it can increase enjoyment, perceived personal control, and and characteristics from their own behavior (Bem 1972; Schooler, agency (Frederick and Loewenstein 1999; Peterson 1999), enabling Ariely, and Loewenstein 2003), and consuming can provide cru- individuals to cope adaptively with threats (Jonas et al. 2014). Thus, cial feedback—signaling to an individual that they are an adequate we predict that there is a trade-off between variety and time between person (Bodner and Prelec 2003). Furthermore, because individual consumption, such that variety will be more effective in counteract- identities are mentally interconnected, the values and aspects that are ing self-threats over longer periods; but as time between each affir- central to a person’s self-concept can become activated and interact mation decreases, its threat-reducing effect will attenuate as a rein- with other identities during consumption through a spreading activa- forcement process bolsters the affirmations. tion process (Anderson 1983; Reed and Forehand 2016; Swann and We believe the research put forth in this paper helps shed light Bosson 2010; Verplanken and Holland 2002). Therefore, if an indi- on these issues and helps inform current work in multiple areas. First, vidual consumes more varied products that are personally meaning- this research contributes to the longitudinal perspective of compensa- ful and representative of what they value, the ability to affirm should tory consumption by exploring the effects of repetition and variety on increase since more aspects of the self are likely to be activated in affirming behaviors. Second, we extend theories on reinforcement the mind of the consumer. In this case, each time a consumer contem- by examining them within the context of affirming consumption. plates the painting they have hanging on their living room wall, new Lastly, understanding how marketers may help maintain the effec- self-aspects can be reflected on and more meaning ascribed to the tiveness of an affirmation can go a long way in ensuring consumers’ experience—reaffirming how they exemplify the value they place on sustained enjoyment and wellbeing. This can in turn, strengthen con- art and aesthetic appreciation (Ahuvia 2005). Thus, we predict that sumers’ loyalty to the brands that facilitate the opportunity to self- variety will be beneficial to affirmations, such that: affirm through meaningful consumption. In the rest of the paper, we discuss why the effectiveness of an affirmation changes over time Hypothesis 1 Relative to repeating an affirmation, using a va- depending on how often one consumes the same product, and what riety of affirmations will have a positive impact role variety plays in the process. on self-esteem following repeated threats.

However, during shorter consumption periods, there may be a X 2 (time delay: short, long) between subjects design. In an initial positive effect of using the same affirmation, due to potential rein- task completed several weeks prior to the main study, participants forcement effects. Repetition may lead to sensitization (Frederick provided a list of personally meaningful songs (adapted from Galak and Loewenstein 1999; McSweeney and Swindell 1999), since pat- et al. 2009). In each round of the main study, participants first com- tern recognition and repetitive consumption have been shown to in- pleted a threat task where they wrote a short essay on their personal crease enjoyment, as well as provide a sense of stability, agency, and insecurities. Then, they listened to either their top rated song (same control over an individual’s surroundings and self-concept (Freder- affirmation condition), their top four rated songs (different affirma- ick and Loewenstein 1999; Norton and Gino 2014; Peterson 1999; tion condition). Lastly, participants completed the same measure of Whitson and Galinsky 2008). Furthermore, the familiar cues provid- self-esteem used in Study 1. The time gap between affirmations was ed by the repetitive consumption can enable an approach motivation manipulated by having participants wait either 30-60 minutes (short in dealing with future threats (Jonas et al. 2014). Reinforcement of gap) or 1-2 days (long gap) between rounds. valued aspects of the self could also help explain why people satiate Replicating the findings of the first study, comparing the differ- slower to products that have more sentimental value associated with ent and same affirmation conditions, in the long gap condition we them, since sentimental value can be linked to memories of our rela- observed a significant positive effect of time (β = .130, t = 3.48, p < tionships and other valued aspects of the self (Yang and Galak 2015). .01), qualified by a significant interaction effect (β = - .116, t = -2.36, However, because these reinforcement effects require that consum- p < .05). However, in the short gaps condition, although there was ers reflect consistently on a similar affirmation experience, we expect a positive effect of time (β = .126, t = 2.78, p < .01), the interaction that they will only occur during relatively shorter time periods, when effects were attenuated (β = .05, t = .87 p = .39), indicating that using individuals can recall the thoughts and feelings surrounding previous either varied or repeated affirmations had equally positive effects on uses of the affirmation. In other words, variety will not have as strong self-esteem. an effect in the short term due to the reinforcing effects of repetition. While consumption can provide many options and opportunities to affirm, there is also potential for repetition to reduce the effec- Hypothesis 2 The positive relative effect of affirmational vari- tiveness of an affirmation. Our results provide initial evidence which ety, compared to repeating an affirmation, will suggests that this not necessarily the case. Although variety seems to be attenuated in shorter time periods. increase effectiveness of an affirmation over longer periods (hypothesis 1), repetition can be beneficial in the short term. Together, the We tested these predictions in two longitudinal experiments, in results of two longitudinal studies support our prediction that using which participants completed a threatening task, and then affirmed a variety of affirmations is more effective than repeatedly consuming before completing the dependent measure assessing their self-per- an affirmation. However, over shorter periods, the reinforcing effect ceptions. Because there were multiple observations for each partici- of using an affirmation repeatedly eliminates this difference. Thus, pant, we employed models with individual fixed effects to capture to counteract self-threats over time, it may be best to vary one’s af- these within subject variances. firmations, unless there is a possibility of reinforcement of these ef- The first study sought to provide an initial test of our predic- fects. tions while employing a values essay task most commonly used in By considering how using a particular affirmation repeatedly self-affirmation research (McQueen and Klein 2006). Ninety partici- may influence its effectiveness in bolstering the self, the research pants were recruited from the Amazon Mechanical Turk panel and presented in this paper provides several theoretical implications. asked to complete four rounds of the study, one to two days apart. First, in contrast to extant research on self-affirmation which has Only participants who completed at least three rounds in total were mainly looked at the effects of single affirmation interventions, our included in the analyses (n = 54). The study used a two-level (af- research delves deeper into understanding the effects of multiple affirmations: same, different) between subjects design. The threat task firmations. Second, our work extends the literature on product vari- was designed to threaten their competence and consisted of respond- ety by exploring the novel context of affirming consumption. Third, ing to difficult-to-impossible CAPTCHAs, a commonly used test this research also adds support for the effectiveness of indirect forms used to distinguish human users from automated responses. Using of compensatory consumption (Mandel et al. 2017). Our results a procedure adapted from Cohen, Aronson, and Steele 2000, after show that it is not necessary to affirm within the same domain as the first trial, participants ranked a list of values, and wrote an es- a threat to one’s self in order to address its negative psychological say about their highest ranked value (same affirmation condition) or effects. Consumption provides many avenues for affirming the self several values (different affirmation condition) after each round of that do not call more attention to the threat by addressing it directly CAPTCHAs. Finally, participants rated their self-esteem using two (Lisjak et al. 2015). items adapted from Rosenberg 1979. These findings also help inform how marketers can tailor their Results revealed a simple effect of time (β = .242, t = 2.71, p < offerings to help consumers sustain their products’ affirmational ben- .01), qualified by a negative interaction effect of time and a dummy- efits—encouraging them to keep consuming their favorite products. coded same affirmation (β = -.288, t = -2.05, p < .05). This indicates For example, products can be offered only during limited times, in that, compared to those who used different affirmations, participants which case variety would be beneficial (Sevilla and Redden 2014). who used the same affirmation repeatedly in the face of threats had The amount of variety provided, such as with video and streaming less positive self-perceptions, suggesting that variety improves an services, can also be modified depending on consumers’ previous affirmation’s effectiveness and supporting H1. behavior. Controlling consumers’ sense of variety (versus actual va- The goal of the second study was two-fold. First, we sought to riety in options) may also help regulate their pace of consumption replicate these results using actual consumption. Second, we wanted (Redden and Galak 2013), and thus improve affirmational benefits. to demonstrate the boundary condition for the effectiveness of va- Finally, this research provides insight into enhancing consumer riety in affirmations proposed in our H2. Three hundred and thirty wellbeing, by highlighting ways in which individuals can adaptively participants were recruited from MTurk to complete the study over cope with (and possibly prevent) self-threats through the intrinsical- four rounds. The study used a 3 (affirmation: same, different, control) ly meaningful use of consumption (Kim and Rucker 2012; Schimel Advances in Consumer Research (Volume 46) / 767 et al. 2004). Enabling consumers to develop and sustain psycho- Galak, Jeff, Joseph P. Redden, and Justin Kruger (2009), “Variety logical resources can help in the short run to deal with experienced Amnesia: Recalling Past Variety Can Accelerate Recover from threats, and may also help in the long run to prepare for threats such Satiation,” Journal of Consumer Research, 36 (4), Galak, J., as personal shortcomings and health recommendations (Ehret and Redden, J.P. and Kruger, J., 2009. Variety amnesia: Recalling Sherman 2014; Schmeichel and Vohs 2009; Taylor and Broffman past variety can accelerate recovery from satiation. Journal of 2011). By further exploring the affirmational benefits of consump- Consumer Research, 36(4), 575-84. tion, we provide support for more natural and spontaneous ways to Galinsky, Adam D., Jeff Stone, and Joel Cooper (2000), “The self-affirm (compared to writing values essays), which may possibly Reinstatement of Dissonance and Psychological Discomfort be used more often and proactively by consumers (Emanuel et al. Following Failed Affirmations,” European Journal of Social 2016; Kim and Rucker 2012; Toma and Hancock 2013). Psychology, 30 (1), 123-47. Jonas, Eva, Ian McGregor, Johannes Klackl, Dmitrij Agroskin, REFERENCES Immo Fritsche, Colin Holbrook, Kyle Nash, Travis Proulx, and Ahuvia, Aaron C. (2005), “Beyond the Extended Self: Loved Markus Quirin (2014), “Threat and Defense: From Anxiety to Objects and Consumers’ Identity Narratives,” Journal of Approach,” Advances in Experimental Social Psychology, 49, Consumer Research, 32 (1), 171-84. 219-86. Anderson, John R (1983), “A Spreading Activation Theory of Kim, Soo and Derek D. Rucker (2012), “Bracing for the Memory,” Journal of Verbal Learning and Verbal Behavior, Psychological Storm: Proactive Versus Reactive 22 (3), 261-95. Compensatory Consumption,” Journal of Consumer Research, Belk, Russell W. (1988), “Possessions and the Extended Self,” 39 (4), 815-30. Journal of Consumer Research, 15 (2), 139–68. Lisjak, Monika, Andrea Bonezzi, Soo Kim, and Derek D. Rucker Bem, Daryl J. (1972), “Self-Perception Theory,” Advances in (2015), “Perils of Compensatory Consumption: Within- Experimental Social Psychology, 6, 1-62. Domain Compensation Undermines Subsequent Self- Bodner, Ronit and Drazen Prelec (2003), “Self-Signaling and Regulation,” Journal of Consumer Research, 41 (5), 1186-203. Diagnostic Utility in Everyday Decision Making,” in The Mandel, Naomi, Derek D. Rucker, Jonathan Levav, and Adam D. Psychology of Economic Decisions, Vol. 1, ed. Isabelle Brocas Galinsky (2017), “The Compensatory Consumer Behavior and Juan D. Carrillo, New York: Oxford University Press, Model: How Self-Discrepancies Drive Consumer Behavior,” 105-26. Journal of Consumer Psychology, 27 (1), 133-46. Carter, Travis J and Thomas Gilovich (2012), “I Am What I Do, McQueen, Amy and William M.P. Klein (2006), “Experimental Not What I Have: The Differential Centrality of Experiential Manipulations of Self-Affirmation: A Systematic Review,” Self and Material Purchases to the Self,” Journal of Personality and Identity, 5 (4), 289-354. and Social Psychology, 102 (6), 1304. McSweeney, Frances K. and Samantha Swindell (1999), “General- Cohen, Geoffrey L, Joshua Aronson, and Claude M Steele (2000), Process Theories of Motivation Revisited: The Role of “When Beliefs Yield to Evidence: Reducing Biased Evaluation Habituation,” Psychological Bulletin, 125 (4), 437-57. by Affirming the Self,”Personality and Social Psychology Norton, Michael I. and Francesca Gino (2014), “Rituals Alleviate Bulletin, 26 (9), 1151-64. Grieving for Loved Ones, Lovers, and Lotteries,” Journal of Cohen, Geoffrey L. and David K. Sherman (2014), “The Experimental Psychology: General, 143 (1), 266-72. Psychology of Change: Self-Affirmation and Social Peterson, Christopher (1999), “Personal Control and Well-Being,” Psychological Intervention,” Annual Review of Psychology, in Well-Being: Foundations of Hedonic Psychology, ed. Daniel 65, 333-71. Kahneman, Ed Diener and Norbert Schwarz, New York: Sage, Cook, Jonathan E., Valerie Purdie-Vaughns, Julio Garcia, and 288-301. Geoffrey L. Cohen (2012), “Chronic Threat and Contingent Ratner, Rebecca K., Barbara E. Kahn, Daniel Kahneman (1999), Belonging: Protective Benefits of Values Affirmation on “Choosing Less-Preferred Experiences for the Sake of Identity Development,” Journal of Personality and Social Variety,” Journal of Consumer Research, 26 (1), 1-15. Psychology, 102 (3), 479-96. Redden, Joseph P and Jeff Galak (2013), “The Subjective Sense Creswell, J. David, William T. Welch, Shelley E. Taylor, David of Feeling Satiated,” Journal of Experimental Psychology: K. Sherman, Tara L. Gruenewald, and Traci Mann (2005), General, 142 (1), 209. “Affirmation of Personal Values Buffers Neuroendocrine and Reed, Americus and Mark R. Forehand (2016), “The Ebb and Flow Psychological Stress Responses,” Psychological Science, 16 of Consumer Identities: The Role of Memory, Emotions and (11), 846-51. Threats,” Current Opinion in Psychology, 10, 94-100. Ehret, Phillip J. and David K. Sherman (2014), “Public Policy Rosenberg, Morris (1979), Conceiving the Self, New York: Basic and Health,” Policy Insights from the Behavioral and Brain Books. Sciences, 1 (1), 222-30. Schimel, Jeff, Jamie Arndt, Katherine M. Banko, and Alison Cook Emanuel, Amber S., Jennifer L. Howell, Jennifer M. Taber, (2004), “Not All Self-Affirmations Were Created Equal: The Rebecca A. Ferrer, William M.P. Klein, and Peter R. Harris Cognitive and Social Benefits of Affirming the Intrinsic (Vs. (2016), “Spontaneous Self-Affirmation Is Associated with Extrinsic) Self,” Social Cognition, 22 (1), 75-99. Psychological Well-Being: Evidence from a Us National Adult Schmeichel, Brandon J. and Kathleen Vohs (2009), “Self- Survey Sample,” Journal of Health Psychology, 1-8. Affirmation and Self-Control: Affirming Core Values Frederick, Shane and George Loewenstein (1999), “Hedonic Counteracts Ego Depletion,” Journal of Personality and Social Adaptation,” in Well-Being: Foundations of Hedonic Psychology, 96 (4), 770-82. Psychology, ed. Daniel Kahneman, Ed Diener and Norbert Schwarz, New York: Sage, 302-29. 768 / The Self-Bolstering Effects of Repeated Affirmations Over Time

Schooler, Jonathan W., Dan Ariely, and George Loewenstein Swann, William B. and Jennifer K. Bosson (2010), “Self and (2003), “The Pursuit and Assessment of Happiness Can Be Identity,” in Handbook of Social Psychology, Vol. 1, ed. Susan Self-Defeating,” in The Psychology of Economic Decisions, T. Fiske, Daniel T. Gilbert and Gardner Lindzey, New Jersey: Vol. 1, ed. Isabelle Brocas and Juan D. Carrillo, New York: Wiley, 589-628. Oxford University Press, 41-70. Taylor, Shelley E. and Joelle I. Broffman (2011), “Psychosocial Sevilla, Julio and Joseph P. Redden (2014), “Limited Availability Resources: Functions, Origins, and Links to Mental Reduces the Rate of Satiation,” Journal of Marketing and Physical Health,” Advances in Experimental Social Research, 51 (2), 205-17. Psychology, 44, 1-57. Sherman, David K. and Geoffrey L. Cohen (2006), “The Toma, Catalina L. and Jeffrey T. Hancock (2013), “Self-Affirmation Psychology of Self-Defense: Self-Affirmation Theory,” Underlies Facebook Use,” Personality and Social Psychology Advances in Experimental Social Psychology, 38, 183-242. Bulletin, 39 (3), 321-31. Sherman, David K. and Kimberly A. Hartson (2011), “Reconciling Townsend, Claudia and Sanjay Sood (2012), “Self-Affirmation Self-Protection with Self-Improvement,” in Handbook of through the Choice of Highly Aesthetic Products,” Journal of Self-Enhancement and Self-Protection, ed. Mark D. Alicke and Consumer Research, 39 (2), 415-28. Constantine Sedikides, New York: Guilford Press, 128-51. Verplanken, Bas and Rob W. Holland (2002), “Motivated Decision Sivanathan, Niro and Nathan C. Pettit (2010), “Protecting the Making: Effects of Activation and Self-Centrality of Values Self through Consumption: Status Goods as Affirmational on Choices and Behavior,” Journal of Personality and Social Commodities,” Journal of Experimental Social Psychology, 46 Psychology, 82 (3), 434-47. (3), 564-70. Whitson, Jennifer A. and Adam D. Galinsky (2008), “Lacking Steele, Claude M. (1988), “The Psychology of Self-Affirmation: Control Increases Illusory Pattern Perception,” Science, 322 Sustaining the Integrity of the Self,” Advances in Experimental (5898), 115-17. Social Psychology, 21, 261-302. Yang, Yang and Jeff Galak (2015), “Sentimental Value and Its Stinson, Danu Anthony, Christine Logel, Steven Shepherd, and Influence on Hedonic Adaptation,”Journal of Personality and Mark P. Zanna (2011), “Rewriting the Self-Fulfilling Prophecy Social Psychology, 109 (5), 767-90. of Social Rejection: Self-Affirmation Improves Relational Security and Social Behavior up to 2 Months Later,” Psychological Science, 22 (9), 1145-49. Consumers’ Journey into Access-Based Consumption: A Narrative Analysis Approach Swapnil Saravade, University of Texas Rio Grande Valley, USA Lorena Garcia-Ramon, University of Texas Rio Grande Valley, USA Jacob Almaguer, University of Texas Rio Grande Valley, USA Mohammadali Zolfagharian, University of Texas Rio Grande Valley, USA Hazel H. Dadanlar, University of Texas Rio Grande Valley, USA

EXTENDED ABSTRACT the “why” of ABC, with little emphasis on the “how” question. For Consumers are increasingly shifting from traditional ownership- example, how do consumers first encounter ABC products and ser- based consumption to relatively novel access-based consumption vices? How do they proceed after early encounters? How do different (hereafter, ABC). Access involves transactions that are market medi- events transpire in consumer ABC journeys? Answering these ques- ated but without transfer of ownership rights (Bardhi and Eckhardt tions is important because emerging research has identified sharing 2012 881). Typically, ABC manifests in the form of renting, sharing, economy as an economic opportunity allowing a sustainable form of subscribing, or borrowing. This manifestation could be in the context consumption leading to a decentralized, equitable, and sustainable of public services such as parks, museums, and libraries or could be economy (Martin 2016). In addition, it is speculated that the econom- in the context of private services such as Netflix, Spotify, and Airbnb. ic impacts of the sharing economy could grow from $15bn in 2013 to The primary focus of this study, however, is consumer engagement $335bn by 2025 (PwC Consumer Intelligence Series Report 2015). with private services. Specifically, the current research focuses on Despite such positive outcomes, Schor et al. (2016) find that con- business-to-consumer situations with internet as a platform that fa- sumers engaged in some sharing economy practices actually employ cilitates access of products/services to consumers. Examples of such distinction practices that deter others, who lack cultural capital, from private services include companies like Rent the Runway, and Bag partaking in the exchanges resulting in an inequitable marketplace. Borrow or Steal, which enable consumers to access fashion products The mixed findings in ABC research warrants an in-depth account such as handbags and jewelry. Moreover, music streaming services of consumers’ engagement with ABC. Therefore, a narrative textual such as Spotify and Pandora allow consumers to store music in cell approach was employed. A narrative analysis approach allows entry phones as opposed to buying stacks of music CDs. Broadly, ABC is into participants’ minds through in-depth interviews, enabling the a part of the sharing economy, wherein consumers place underuti- researcher to witness their stories and journeys (Reissman 1993). lized assets as market offerings to make a profit (Pricewaterhouse Using iterative analysis of consumer stories pertaining to their ABC Coopers (PWC) Consumer Intelligence Series Report 2015). As per experiences, this paper sheds light on the key “turns in the road” that the PwC report (2015), 19% of all US consumers participate in shar- consumers experience as they navigate ABC in the context of their ing economy transactions; 72% plan to participate in the near future; daily lives (Cortazzi 1994; Reissman 1993). and 57% view access as the new ownership. These numbers paint an optimistic picture of ABC’s promising growth in the future. THEORETICAL BACKGROUND Scholars have examined ABC from multiple perspectives. Since a wide range of activities pertain to access, there is a conceptual ABC Enablers: Internet and Apps overlap in the literature regarding the semantics of ABC. As a re- Sharing and accessing as an Internet phenomenon started with sult, there is certain degree of fragmentation and confusion regarding Napster and other peer-to-peer (P2P) file sharing networks, where us- different terms such as sharing, collaborative consumption, renting, ers were able to share and access digital files, often illegally (Giesler and access (Habibi, Davidson, and Laroche 2017; Habibi, Kim, and 2006; Henning-Thurau, Henning, and Sattler 2007).The Internet Laroche 2016). Belk (2007) defines sharing as “the act and process of served as the springboard for the creation of these access systems; distributing what is ours to others for their use and/or the act and pro- however, it is the rising role of smartphones and their capability to cess of receiving or taking something from others for our use” (126). access mobile applications that has truly enabled the adoption of In collaborative consumption, people coordinate the acquisition and access systems (Botsman and Rogers 2010; Zolfagharian and Yaz- distribution of a resource for a fee or other compensation (Belk 2014) danparast 2017). Research has identified mobile applications as key while engaging in joint activities with additional consumers (Felson drivers of access system adoptions such as car sharing and streaming and Spaeth 1978). Furthermore, renting goods usually involves an services (Hamari, Sjöklint, and Ukkonen 2016; Molhmann 2015). agreement between two parties whereby the renter obtains the right to use the rented object and relevant benefits for a specified period Liquid Consumption of time and fee (Moeller and Wittowski 2010). In contrast to owner- The current research draws on the logic of liquid consumption ship, consumers in ABC, derive benefits from access to objects or which is derived from the theory of liquid modernity. Liquid moder- networks which they could not afford to own due to spatial or en- nity “is a social condition where social structures are no longer stable vironmental constraints (Bardhi and Ekhardt 2012). In the current or long-term” (Bauman 2007, 1). Flexibility, mobility, and openness research, we focus on ABC and follow the definition by Bardhi and to change are key components of liquid modernity (Bardhi and Eck- Ekhardt (2012) as “transactions that may be market mediated in hardt 2017; Featherstone 1995; Hannerz 1996). Recently, Bardhi and which no transfer of ownership takes place” (881). Specifically, we Eckhardt (2017) applied the theory of liquid modernity to the domain narrow our focus to consumption through internet enabled transac- of consumption and conceptualized the logic of liquid consumption tions with emphasis on online applications which facilitate the ABC. defined as access-based, ephemeral, and dematerialized. The access- Extant research focused on ABC characteristics (Bardhi and based characteristic of liquid consumption is central to the current Eckhardt 2017), ABC benefits (Lambert and Rose 2012), and motiva- research. The primary focus of this study then is to understand con- tions for participation in ABC (Bardhi and Eckhardt 2012; Edbring, sumer engagement with private access services. Consumer micro- Lehner, and Mont 2016; Gruen 2016; Lawson et al., 2016). Overall, narratives of ABC were viewed through the lens of liquid consump- extant literature has mainly endeavored to understand the “what” and tion and synthesized to understand consumers’ journey with ABC.

Advances in Consumer Research 769 Volume 46, ©2018 770 / Consumers’ Journey into Access-Based Consumption: A Narrative Analysis Approach METHOD uncovered by the researchers along the spectrum of the complete Given the research purposes, a qualitative approach using in- text (Reissman 1993). The iterative process was complete when con- depth interviews was employed to extract consumer ABC narratives. sensus across researchers was reached (Schensul et al. 1999). Narratives serve as an instrument for individuals to make sense of Throughout the generation and analysis of the text, the re- the events that happen in the world (Reissman 1993). Narrative anal- searchers took appropriate steps in order to ensure credibility, de- ysis is considered to be an entry into the mind of an individual or, if pendability, confirmability, and integrity (Wallendorf and Belk analyzing a group of storytellers, into their culture (Cortazzi 1993). 1989). Credibility was established by debriefing participants to ver- Researchers interested in the lived experiences of consumers have ify an appropriate representation of their narrated reality. Depend- employed narrative analysis to interpret and make sense of these ex- ability was established by incorporating member checks in over half periences (Braun, Zolfagharian, and Belk 2016; Stern, Thompson, of the interviews where at least two members of the research team and Arnould 1998). By employing narrative analysis, the current re- were present. This also allowed for the triangulation of participants search analyzes consumers’ interpretations of their experiences with and researchers. Confirmability was established by audio recording ABC and important events as they occurred in the context of their each interview and making the digital transcriptions available to all lived experience. members of the research team. Lastly, integrity was established by Text was generated through semi-structured in-depth interviews creating rapport among participants and researchers and by assuring that allowed researchers to delve deep into consumer narratives (Mc- participants of the confidentiality of the research. Cracken 1988). All researchers were involved in the data collection process, resulting in a total of 19 interviews conducted in the South- FINDINGS west United States. The purpose of the interviews was to extract the The macro-narrative is visually presented in Figure 1 in the experiences and stories of the participants as they transitioned (or form of a pinwheel, with each of its components representing specif- not) into ABC. For this reason, a purposive sample of ABC users ic characteristics of it. Moreover, the study identifies specific drivers was recruited using the researchers’ social network. Thus, the decid- of ABC that act as the “wind” making the pinwheel turn. The resulting criterion for recruitment was participants’ prior experience with ing macro-narrative describes the key milestones consumers un- ABC. Participants’ age ranged from 20 to 65, with an average age of dergo through their experiences with ABC. More specifically, eight 27 (see appendix A for respondent profile). Participants represented sequential “turns in the road,” four major and four deviant, emerged a variety of occupations such as retired professionals, graduate stu- from the analysis of micro-narratives. The sequence of these “turns dents, and working professionals. Additionally, the extent of experi- in the road” is: initiation/rejection, experimentation/renunciation, ence with ABC services also ranged from participants who were very contamination/alternation, and devotion/reversion. It should be used to ABC and those who had not had much experience with it. noted that not all participants’ narratives described such journey, as The researchers encouraged respondents to tell their personal nar- some of them deviated from the journey at different points in time ef- ratives and probed for specific experiences contingently during the fectively abandoning ABC, while some remained in a specific stage dialogue. The interviews generated 196 single-spaced pages of text. never reaching the devotion stage. Transcripts generated from interviews are among the most popular While initiation, experimentation, contamination, and devotion forms of text used by consumer researchers (Arnold and Fischer represent key stages in consumers’ engagement with ABC; rejection, 1994; Braun et al. 2016; Stern et al. 1998). renunciation, alternation and reversion demonstrate alternate points Interviews started with general questions about participants’ ex- from which consumers deviate from ABC. The unique feature of periences with ABC that enabled them to provide as rich a descrip- the emergent macro-narrative is that, it not only captures consum- tion of their emotions and opinions as possible. A research protocol ers’ journey of becoming a devout user of ABC, but also captures was designed to ensure that the conversation did not deviate from its the deviant “turns in the road” from which consumers leave their intended direction and to remind researchers of important probes and journey with ABC. follow-ups to enrich consumer narratives (Rubin and Rubin 2005). Consumers’ journey with ABC begins with “initiation” which is Moreover, the use of the interview protocol enabled coverage of in- characterized by consumers’ random or purposeful internet search- tended domains across all interviews as well as consistency across es, confirmation to recommendations from trusted relationships, interviewers (McCracken 1988). The researchers began the inter- or through product or service referrals. Alternatively, this journey views by asking participants if they had any experience with ABC. is ceased when consumers’ outright “reject” the idea of engaging In some cases, participants were not familiar with the term ABC so in ABC for various reasons. Even after transitioning from initiation the researchers provided examples to bring consumers’ experiences to “experimentation” wherein consumers “test the waters” of ABC, with ABC (if any) to the fore. some consumers become ambivalent. They either continue and con- taminate their consumption patterns with ABC or decide to renounce ANALYSIS use of ABC. Consumers who continue with their occasional experi- Text generated from the interviews was subjected to an inter- mentation eventually experience “contamination” of their consump- pretive analysis where researchers moved from the text to the etic tion patterns with ABC. This “turn in the road” represents an im- understanding in an iterative manner. This resulted in the discov- portant milestone for ABC consumers because it is from this point ery of stages emerging from the text, informed by the researchers’ onwards that consumers start considering access-based products or framework (McCracken 1998). Each interview was analyzed indi- services as feasible options in their consumption decisions. Howev- vidually to identify the emergence of a micro-narrative and then con- er, for some consumers this contamination is not compelling enough. sidered in the “big picture” with the rest of the interviews to form an Specifically, with “alternation”, they continue their traditional con- overarching macro-narrative. Once all interviews were transcribed, sumption practices and integrate ABC whenever convenient. This the researchers analyzed the text, searching for the underlying macro alternation does not reflect the complete embrace of ABC, as tra- narrative that not only constituted the different plots of the overall ditional consumption is still present. While some ABC consumers story, but also identified the most plausible sequence of events. The may not move beyond the alternation milestone resulting from their macro-narrative materialized from the similarities and differences lack of devotion to such services/products. Devotion marks the pin- Advances in Consumer Research (Volume 46) / 771 nacle of consumers’ engagement with ABC, characterized by strong Braun, Jakob, Mohammadali Zolfagharian, and Russell W. Belk emotions and bonding behaviors toward access-based products or (2016), “How Does a Product Gain the Status of a Necessity? services. However, at the pinnacle of engagement, some ABC con- An Analysis of Necessitation Narratives,” Psychology & sumers experience inconsistency with the ideology, stance or over- Marketing, 33 (October), 209-22. all functioning of the access-based products or services. It is at this Cortazzi, Martin (1994), “Narrative analysis”, Language teaching, point, that consumers who previously showed devotion to ABC, re- 27 (July) 157-70. vert back to their previous consumption practices. This progression Edbring, Emma Gullstrand, Matthias Lehner and Oksana Mont of consumers’ journey with ABC through each of the milestones is (2016), “Exploring Consumer Attitudes to Alternative Models powered by the driving forces which are characterized by flexibility, of Consumption: Motivations and Barriers,” Journal of status seeking, temporary nature of use, and low price of engaging Cleaner Production, 123 (June), 5-15. with ABC (Edbring et al. 2016). Featherstone, Mike (1995), Undoing Culture: Globalization, Postmodernism and Identity, Thousand Oaks, CA: Sage. DISCUSSION Felson, Marcus and Joe L. Spaeth (1978), “Community Structure Our findings can inform and equip marketing managers with and Collaborative Consumption,” American Behavioral key insights in various areas such as market-driven strategies, cus- Scientist, 21 (March–April), 614–24. tomer satisfaction, market-driving practices, service recovery, prod- Giesler, Markus (2006), “Consumer Gift System: Netnographic uct and/or service sales and consequently, firm profitability. As per Insights from Napster,” Journal of Consumer Research, 33 the findings, some consumers acknowledged the benefits of ABC, (September), 283–90. while some cited excuses pertaining to physical safety, information Habibi, Mohammad R., Alexander Davidson, and Michel Laroche security, negative experiences, and liability of newness of the service (2017), “What Managers Should Know About the Sharing or products for distancing themselves with ABC. As prior literature Economy”, Business Horizons, 60 (January), 113-21. has shown, alternative methods of consumption do not always result Habibi, Mohammad R., Andrea Kim, and Michel Laroche (2016), in the positive outcomes they had originally been intended to provide “From Sharing to Exchange: An Extended Framework of (Schor et al. 2016). Marketers can tap into the reservations consum- Dual Modes of Collaborative Non-Ownership Consumption”, ers have for not engaging in ABC by providing safety and security Journal of the Association for Consumer Research, 1 (April), assurance or employee training as a potential remedy for negative 277-94. consumer experiences, among others. Hamari, Juho, Mimmi Sjöklint, and Antti Ukkonen (2016), “The The current research makes several contributions to the liquid Sharing Economy: Why People Participate in Collaborative consumption literature. First, it not only identifies the unique mile- Consumption”, Journal of the Association for Information stones involved in ABC, but also demonstrates that each of these Science and Technology, 67 (September), 2047-59. milestones has an alternative component that forms an integral part Hannerz, Ulf (1996), Transnational Connections: Culture, People, of the macro-narrative. Second, by analyzing consumer stories, Places. London: Routledge. the findings provide a nuanced understanding of liquid consump- Hennig-Thurau, Thorsten, Victor Henning, and Henrik Sattler tion as manifested through ABC. Each of the unique milestones that (2007), “Consumer File Sharing of Motion Pictures”, Journal emerged in the macro-narrative delves deeper into the experiences of of Marketing, 71 (October), 1-18. consumers. This is important, because consumer research has shown Lawson, Stephanie J., Mark R. Gleim, Rebeca Perren, and Jiyoung that under certain conditions, experiences make consumers happier Hwang (2016), “Freedom from Ownership: An Exploration of compared to material possessions (van Boven and Gilovich 2003). Access-Based Consumption,” Journal of Business Research, Lastly, this research contributes to the theory of liquid modernity 69 (August), 2615–23. which posits that social structures are no longer stable or long-term Lovelock, Christopher, and Evert Gummesson (2004), “Whither (Bauman 2000) by adding more shades to the theory of liquid mo- Services Marketing? In Search of a New Paradigm and Fresh dernity and explaining how consumers progress through different Perspectives,” Journal of Service Research, 7 (August), 20–41. milestones in their ABC journey. Martin, Chris J. (2016), “The Sharing Economy: A Pathway to Sustainability or a Nightmarish Form of Neoliberal REFERENCES Capitalism?”, Ecological Economics, 121 (January), 149-59. Bardhi, Fleura, and Giana M. Eckhardt (2012), “Access-Based McCracken, Grant (1988), The Long Interview, Beverly Hills, CA: Consumption: The Case of car Sharing.” Journal of Consumer Sage. Research, 39 (December), 881-98. Möhlmann, Mareike (2015), “Collaborative Consumption: ——— (2017), “Liquid Consumption.” Journal of Consumer Determinants of Satisfaction and the Likelihood of Using Research, 44 (March), 582-97. a Sharing Economy Option Again”, Journal of Consumer Bauman, Zygmunt (2000), Liquid modernity, Cambridge: Polity. Behaviour, 14 (May), 193-207. ——— (2007), Liquid times, Cambridge: Polity. Moeller, Sabine, and Kristina Wittkowski (2010), “The Burdens Belk, Russell W. (2007), “Why not Share Rather than Own?” of Ownership: Reasons for Preferring Renting,” Marketing Annals of the American Academy of Political and Social Service Quality, 20 (March), 176–91. Science, 611 (May), 126–40. PricewaterhouseCoopers (2015), “The Sharing Economy — Sizing ——— (2014), “You are What you can Access: Sharing and the Revenue Opportunity”, http://www.pwc. co.uk/issues/ Collaborative Consumption Online”, Journal of Business megatrends/collisions/sharingeconomy/ the-sharing-economy- Research, 67 (August), 1595-600. sizing-the-revenue-opportunity.html Botsman, Rachel, and Roo Rogers (2010), What’s Mine Is Riessman, Catherine K. (1993), Narrative analysis, Newbury Park, Yours: The Rise of Collaborative Consumption, New York: CA: Sage. HarperCollins. 772 / Consumers’ Journey into Access-Based Consumption: A Narrative Analysis Approach

Rubin, Herbert J. and Irene S. Rubin (1995), Qualitative Stern, Barbara B., Craig J. Thompson, and Eric J. Arnould (1998), Interviewing: The Art of Hearing Data, Thousand Oaks, CA: “Narrative Analysis of a Marketing Relationship: The Sage. Consumer’s Perspective”, Psychology & Marketing, 15 (May), Rudmin, Floyd (2016), “The Consumer Science of Sharing: A 195–215. Discussant’s Observations”, Journal of the Association for Van Boven, Leaf and Thomas Gilovich (2003), “To Do or to Have? Consumer Research, 1(April), 198-209. That Is the Question,” Journal of Personality and Social Schensul, Stephen L., Jean J. Schensul, and Margaret Diane Psychology, 85 (December), 1193–1202. LeCompte (1999), Essential Ethnographic Methods 2: Wallendorf, Melanie and Russell Belk (1989), “Assessing Observations, Interviews, and Questionnaires, Walnut Creek, Trustworthiness in Naturalistic Consumer Research,” in CA: Altamira Press. Interpretative Consumer Research, ed. Elizabeth Hirschman, Schor, Juliet B., Connor Fitzmaurice, Lindsey B. Carfagna, Provo, UT: Association for Consumer Research, 69–84. Will Attwood-Charles, and Emilie Dubois Poteat (2016), Zolfagharian, Mohammadali, and Atefeh Yazdanparast (2017), “Paradoxes of Openness and Distinction in the Sharing “The Dark Side of Consumer Life in the age of Virtual and Economy,” Poetics, 54 (February), 66-81. Mobile Technology”, Journal of Marketing Management, 33 (October), 1304-35. Product Search on Crowded Retail Shelves: Impact of Vertical Product Location on Search Performance Ana Scekic, HEC Paris, France A. Selin Atalay, Frankfurt School of Finance and Management, Germany Cathy Liu Yang, HEC Paris, France Peter Ebbes, HEC Paris, France

EXTENDED ABSTRACT In study 2, 127 participants completed the study on a lab com- In the context of FMCGs, consumers often arrive at the store puter, and were asked to find the toothpaste for smokers on a shelf with a choice of a product in mind. Their task then is to find their of 15 toothpastes arranged on a 3(columns) x 5(shelves) planogram. already chosen product among the products present at the point of The VL of the target was manipulated in a between-participant de- purchase. In the current paper, we focus on this task of visual search sign. Results showed a significant effect of VL on task duration for a target product on a shelf with a large assortment (i.e., a crowded (F(2,118) = 3.30, p = 0.04), perceived task difficulty (F(2,118) = shelf). Specifically, we investigate the influence of vertical location 3.75, p = 0.026), and satisfaction with task duration (F(2,118) = 5.20, (VL) of a product on a crowded shelf on consumers’ search perfor- p = 0.007). Planned comparisons suggested again that MS is a posi- mance. We provide initial evidence that, when the task is to find a tion of disadvantage for product search. target product on a shelf, the middle shelf (MS) is a position of dis- Eye-movement data showed that, when the target is on MS, advantage, when compared to the top shelf (TS) and the bottom shelf the total number of fixations needed to explore the assortment is (BS). the highest (F(2, 118) = 3.93, p = 0.022), the total duration of all Finding a target product on a crowded shelf is not easy. As the fixations is the longest (F(2, 118) = 3.18, p = 0.045), and the total number of objects in any scene increases, it becomes difficult to dis- duration of fixations that land on the target (i.e., total time needed tinguish the objects in that scene from one another. This is known as to look at the target before “finding” it) is the longest F(2, 118) = the “crowding effect” (Levi 2008; Whitney and Levi 2010). Among 4.03, p = 0.02). The time to first fixation on the target did not differ the factors influencing the strength of the crowding effect, is the loca- significantly when the target was on MS or BS. However, the time tion in the scene and the visual field. We investigate the influence of elapsed since the first fixation on the target until the end of the task product location on search performance. was shorter when the target was on BS than when it was on MS, sug- In a product choice context, products on MS and TS have been gesting that even though it takes the same time to look at the target shown to get more attention and positive evaluation (Chandon et al. product for the first time, the time necessary to recognize the target 2009; Valenzuela and Raghubir 2009; Atalay et al. 2012). The impact and complete the task is longer when the target is on MS than on BS. of product location on search, however, is underexplored. A follow-up study, with a more pictorially depicted target that According to the crowding literature, the ability of identify- did not necessarily require reading to be found, provided additional ing an object in a crowded scene depends on that object’s location evidence. Forty-nine participants searched for a box of almond- by way of two mechanisms: anisotropy and asymmetry (Levi 2008; flavored cookies on a 5x5 planogram of cookies. Results confirmed Whitney and Levi 2010). Anisotropy suggests that the crowding ef- that, when the target was on MS, rather than on TS or BS, the search fect differs along the vertical axis of the visual field, and, specifically, task took the longest to complete (F(2, 46) = 3.39, p = 0.042). is less strong in the lower visual field (He et al. 1996). Hence, the Taken together, our studies provide evidence for location effects ability to identify an object’s features is improved when the object is on product search and a different take on location quality: while in a located in the lower visual field. Asymmetry refers to the role of sur- choice context the lower shelves are regarded as a position of disad- rounding (flanker) objects: a target object is more difficult to identify vantage and the middle as a position of advantage, our results suggest when it is presented with two flankers, positioned one on each side that when the task is to search for a target product, MS may not be of the target, than when it is presented with only one flanker on one the best location, while BS gains value. Further research on factors side (Bouma 1970; Bouma 1973). Thus, in a display of objects, the influencing product search is needed, as it is a question relevant to objects on the outer sides of the display will be easier to identify than consumers, retailers, and brand managers. the inner objects. Drawing on these theories, we predict that there will be a main REFERENCES effect of vertical product location on search performance and that Atalay, A. Selin, H. Onur Bodur, and Dina Rasolofoarison (2012), products on both top and bottom shelves (i.e., products in the outer “Shining in the Center: Central Gaze Cascade Effect on locations) will be easier to find than products located on the middle Product Choice”, Journal of Consumer Research, 39(4), 848- shelf. 866. In study 1, 668 MTurk participants were asked to find a target Bouma, H. (1970), “Interaction Effects in Parafoveal Letter (i.e., “sugar free” jam) among 25 different jams. The location of the Recognition”, Nature, 226, 177–178. target was manipulated in a between-participant design. On a 5x5 Bouma, H. (1973), “Visual Interference in the Parafoveal planogram, we rotated the target across the jams on TS, on MS, and Recognition of Initial and Final Letters of Words”, Vision on BS. Search performance was measured by search task duration, Research, 13, 767–782. perceived duration of the task, perceived task difficulty, and satisfac- Chandon, Pierre, J. Wesley Hutchinson, Eric T. Bradlow, and tion with task duration. Results showed a main effect of VL on task Scott H. Young (2009), “Does In-Store Marketing Work? duration (F(2,653) = 5.12, p = 0.006), perceived duration of the task Effects of the Number and Position of Shelf Facings on Brand (F(2,653) = 3.19, p = 0.042), perceived task difficulty (F(2,653) = Attention and Evaluation at the Point of Purchase”, Journal of 2.69, p = 0.069), and satisfaction with task duration (F(2,653) = 2.33, Marketing, 73 (6), 1-17. p = 0.099). Planned comparisons suggested that MS is a position of disadvantage for product search when compared to BS and TS.

He, Sheng, Patrick Cavanagh, and James Intriligator (1996), Valenzuela, Ana, and Priya Raghubir (2009), “Position-Based “Attentional Resolution and the Locus of Visual Awareness”, Beliefs: The Center-Stage Effect”,Journal of Consumer Nature, 383, 334-337. Psychology, 19(2), 185-196. Levi, Dennis M (2008), “Crowding – an Essential Bottleneck for Whitney, David, and Dennis M. Levi (2011), “Visual Crowding: Object Recognition: A Mini-Review”, Vision Research, 48, A Fundamental Limit on Conscious Perception and Object 635-654. Recognition”, Trends in Cognitive Sciences, 15, 160-168. The Psychological Impact of Annuities: Can Pension Payout Choice Influence Health Behavior? Anja Schanbacher, London Business School, UK David Faro, London Business School, UK Simona Botti, London Business School, UK Shlomo Benartzi, UCLA, USA

EXTENDED ABSTRACT disadvantages of the alternative) would increase the subjective likeli- Across the world, a growing number of people face the im- hood of choosing this option. As intended, participants in the annuity portant financial decision whether to take out their pension in the (vs. lump-sum) condition reported being significantly more likely to form of a one-time lump-sum payment or as monthly income, also choose an annuity. referred to as an annuity. In the US and Australia, for instance, people To measure health behavior, we then presented five scenarios have traditionally had the freedom to choose between these payout involving a choice between a more and a less healthy behavioral op- options, whereas some countries where annuitized pensions are tra- tion, and participants indicated which behavior they would be more ditionally more common such as the UK have recently introduced likely to choose. The behaviors included selecting a costly compre- the option of cashing out one’s pension as a lump-sum. Besides this hensive vs. a free basic health check, and joining a more strenuous increase in choice, the demographic development with aging popula- exercise program that particularly benefits long-term health vs. con- tions implies that more and more people face the decision between tinuing one’s regular exercises. an annuity and a lump-sum. The results confirmed our central hypothesis: Across scenarios, Policy makers have been debating whether people should be participants in the annuity condition were significantly more likely given a choice between annuitization and lump-sum payout, and to select the healthy behavioral options than participants in the lump- whether choice of annuities should be encouraged or mandated. One sum condition. consideration in this debate is self-selection. Specifically, it has been Study 2 argued that individuals are to some extent aware of their own life The purpose of Study 2 was to replicate the effect that choice of expectancy, and that since the financial payout of annuities increases an annuity vs. a lump-sum boosts health-related behavior in the con- with growing length of life, those with a higher (compared to lower) text of real behavior. Participants were randomly assigned to a lump- life expectancy are more likely to choose an annuity. Indeed, litera- sum- or annuity-condition. Upon arriving at the lab, participants ex- ture has documented a positive relationship between receiving an an- ercised on a stationary bike for two minutes. A combined measure of nuity and longevity (Finkelstein and Poterba 2002). the amount of calories burnt and distance travelled (z-scored) within In the present research, we raise the possibility that in addi- these two minutes was included as a covariate in the main analysis to tion to self-selection, a causal effect of choosing an annuity over a control for individual differences in exercising performance. lump-sum on longevity may underlie this relationship. This proposal Next, participants were told they would participate in an un- is based on the argument that for annuity (but not lump-sum) re- related study while taking a break between exercising tasks. In fact cipients, extending their lives implies a financial gain. Further, los- they completed the task manipulating perceived likelihood of choosing a substantial part of one’s savings in the event of early death is a ing an annuity vs. a lump-sum from Study 1 – our manipulation of central concern regarding annuities (Hu and Scott 2007). As a result, the financial decision. As intended, participants in the annuity (vs. positive consequences of reaching a high age may be particularly lump-sum) condition indicated being more likely to choose an annu- salient among annuity recipients. Hence, we hypothesize that indi- ity for their own retirement. viduals who perceive a higher likelihood of receiving an annuity will After this manipulation was administered, participants exer- be more likely to engage in more (versus less) healthy behaviors as a cised again on the bike for up to ten minutes. They were told that it means to increase life expectancy. was up to them how vigorously they wanted to exercise. Our main Study 1 dependent variable was the combined measure of the amount of calo- We recruited UK-based, currently employed participants ries burnt and distance travelled (z-scored) within the ten minutes. through the website ProlificAcademic. Participants were randomly In addition, participants indicated how likely they were to exercise assigned to either a lump-sum- or an annuity- condition. The instruc- later that day. tions stated that a recent policy change in the UK had given pen- Addressing our main hypothesis, participants in the annuity- sioners the freedom to choose between a lump-sum and an annuity. condition had significantly higher values on the exercising measure We then presented a short explanation of the two payout options. To than participants in the lump-sum-condition, indicating that they had facilitate comprehension, each participant also saw a table indicating exercised more vigorously. In addition, participants in the annuity- the cumulative amounts of money that they would have received by (vs. lumps-sum-) condition reported being significantly more likely certain ages with each option. The annuity payout was determined to exercise later that day. using current annuity rates. Both options were presented in counter- Prior literature has focused on antecedents (Payne et al., 2013; balanced order. Shu, et al., 2016) rather than consequences of choosing a lump-sum Next, participants in the annuity condition listed the one or two vs. an annuity pension. We manipulate expectations of choosing an most important reasons for choosing an annuity and the one or two annuity (vs. a lump-sum) and find evidence of a causal effect on most important reasons against choosing a lump sum, whereas par- health-related behaviors. Such potential consequences should be ticipants in the lump-sum condition listed the one or two most im- considered in the public policy debate on pension payout choice. portant reasons for choosing a lump sum and the one or two most important reasons against choosing an annuity. Our reasoning was that enhancing the accessibility of advantages of an option (and of

Advances in Consumer Research 775 Volume 46, ©2018 776 / The Psychological Impact of Annuities: Can Pension Payout Choice Influence Health Behavior? REFERENCES Payne, John W., Namika Sagara, Suzanne B. Shu, Kirstin C. Appelt, Finkelstein, Amy and James Poterba (2002), “Selection Effects and Eric J. Johnson (2013), “Life Expectancy as a Constructed in the United Kingdom Individual Annuities Market,” The Belief: Evidence of a Live-To or Die-By Framing Effect,” Economic Journal, 112(476), 28-50. Journal of Risk and Uncertainty, 46(1), 27-50. Hu, Wei-Yin and Jason S. Scott (2007), “Behavioral Obstacles in Shu, Suzanne B., Robert Zeithammer, and John W. Payne (2016), the Annuity Market,” Financial Analysts Journal, 63(6), 71- “Consumer Preferences for Annuity Attributes: Beyond Net 82. Present Value,” Journal of Marketing Research, 53 (2), 240- 62. Born to Shop? A Genetic Component of Deal Proneness Robert M. Schindler, Rutgers University - Camden, USA Vishal Lala, Pace University, USA Jeanette Taylor, Florida State University, USA

EXTENDED ABSTRACT of six deal types. The items measured the enjoyment and pleasure Price discounts, cents-off coupons, and other retail sales promo- that was received from the deal types. For example, one of the six tions have been found capable of eliciting strong consumer enthusi- deal types represented was free gift (e.g., “I enjoy buying products asm. Research on this sales promotion enthusiasm has shown that that come with a free gift”) and another was coupon (e.g., “Redeem- people have enduring differences in their likelihood of feeling it. The ing coupons makes me feel good”). tendency to feel strong promotion enthusiasm has been termed deal The twins in the 73 MZ twin pairs in our sample showed a high proneness (Webster 1965). Given that deal proneness is both an in- degree of similarity with each other in their scores on the 32-item teresting behavioral phenomenon as well as descriptive of an impor- deal proneness scale (r = 0.55, p < 0.001). The twins in the 35 DZ tant market segment, it is worthwhile to gain a better understanding twin pairs in our sample showed a much lower degree of deal prone- of the origin of this consumer disposition. ness similarity (r = 0.13, p = 0.46). This greater similarity among An early view concerning the origin of deal proneness is that the MZ twins than among the DZ twins is statistically significant consumers learn it from the shopping environment. However, virtu- (Fisher’s z = 2.30, p = 0.024) and is evidence for a high degree of ally all consumers are exposed to sales promotions, but only some of heritability in deal proneness. Using Falconer’s (1989) h2 measure these consumers become deal prone. The possibility that a financial of heritability, our results indicate that 84 percent of the variability in need for price discounts leads people to become deal prone conflicts deal proneness is associated with genetic factors. with data indicating that deal-prone consumers tend to be higher The MZ-DZ correlation differences did differ among the six income, not lower income, individuals (e.g., Bawa and Shoemaker deal types. There was considerably more similarity among the MZ 1987). Rather, the tendency for some consumers to become deal twins than among the DZ twins in their interest in coupons, rebates, prone is likely to be based on differences among consumers as to how and items being on sale. For free gift deals, there was consider- they experience retail sales promotions. For example, some consum- ably more correlation among MZ twins than among DZ twins, but ers may enjoy deals more than others because of the ego satisfaction this difference was not statistically significant. For buy-one-get- they receive from them (Schindler 1998). one-free promotions and contests and sweepstakes, there were only In addition to the possibility of acquiring deal proneness from small, non-statistically significant differences between the similarity the shopping environment, consumers could acquire it from their among MZ twins and that among DZ twins. This suggests an effect parents. Schindler, Lala, and Corcoran (2014) found that deal-prone of shared environment on those types of deals rather than heritability. consumers are likely to have deal-prone parents. Although they The greater deal-proneness similarity among MZ twins than found evidence that factors related to the quality of family interac- among DZ twins provides strong evidence for the existence of an tions affected the degree of deal-proneness similarity between con- inherited component to deal proneness. This finding provides a basis sumers and their parents, the consumer-parent similarity raises the for approaching the question of what factors make deals attractive question of whether there is an inherited component to deal prone- to consumers. For example, Simonson and Sela’s (2011) finding ness. Supporting this are recent findings of inherited components to regarding the compromise effect suggests some heritability of the various types of consumer and financial decision-making (Cesarini degree to which a consumer uses external stimulus information (e.g., et al. 2012; Cronqvist and Siegel 2014; Simonson and Sela 2011). whether an alternative is the middle option) over internal preference In this study, we test for an inherited component to deal prone- information (e.g., his or her liking of the alternative). Such use of ness. We use a classical twin design to compare the similarity of external stimulus information over internal preference information monozygotic (MZ) and dizygotic (DZ) twins reared together. Be- may also be important in how promotional deals are experienced. cause MZ twins have all of their genes in common, but DZ twins For marketing practice, the finding of the heritability of deal prone- have, on average, only half of their segregating genes in common, a ness is further evidence that this trait tends to run in families and higher correlation in the amount of a trait between the members of suggests that there might be a particular effectiveness of promotional MZ twin pairs than between the members of DZ twin pairs indicates deals that encourage interactions among consumers and their family some heritability of the trait. If MZ and DZ twin pairs are relatively members. similar in their correlations, then shared environmental influences are inferred to be at work in shaping the trait. REFERENCES The twins included in this study were recruited with the assis- Bawa, Kapil and Robert W. Shoemaker (1987), “The Coupon-Prone tance of the registrar at Rutgers University. All pairs of students who Consumer: Some Findings Based on Purchase Behavior Across attended the university between the years 2016-2018 and who had Product Classes,” Journal of Marketing, 51 (October), 99-110. the same last name, same birth date, and same gender were contacted Cesarini, David, Magnus Johannesson, Patrik K.E. Magnusson, and by email and were invited to participate. A total of 108 twin pairs Bjӧrn Wallace (2012), “The Behavioral Genetics of Behavioral responded. To determine the zygosity of each set of twins, we asked Abnormalities,” Management Science, 58 (January), 21-34. each twin to complete a five-item self-report questionnaire that has Cronqvist, Henrik and Stephan Siegel (2014), “The Genetics of been validated against DNA-based zygosity determination (Lykken Investment Biases,” Journal of Financial Economics, 113, et al. 1990). By these questions, 73 twin pairs were classified as MZ 215-234. and 35 were classified as DZ. Falconer, Douglas S. (1989), Introduction to Quantitative Genetics, Each of the twins in the sample completed the deal-proneness New York: Longman. measure used by Schindler et al. (2014). The scale included 32 seven-point agree-disagree items with five or six items addressing each

Lykken, D. T., T. J. Bouchard, Jr., M. McGue, and A. Tellegen Schindler, Robert M, Vishal Lala, and Colleen Corcoran (2014), (1990), “The Minnesota Twin Registry: Some Initial “Intergenerational Influence in Consumer Deal Proneness,” Findings,” Acta Geneticae Medicae et Gemellologiae: Twin Psychology & Marketing, 31 (5), 307-320. Research, 39 (January), 35–70. Simonson, Itamar and Aner Sela (2011), “On the Heritability of Schindler, Robert M. (1998), “Consequences of Perceiving Oneself Consumer Decision Making: An Exploratory Approach for as Responsible for Obtaining a Discount: Evidence for Smart- Studying Genetic Effects on Judgment and Choice,”Journal of Shopper Feelings,” Journal of Consumer Psychology, 7 (4), Consumer Research, 37 (April), 951–966. 371–392. Webster, Frederick E., Jr. (1965), “The ‘Deal-Prone’ Consumer,” Journal of Marketing Research, 2 (2), 186-189. “Eww, It Has a Face!” Anthropomorphizing Food Products Deteriorates Consumption Experience Roland Schroll, University of Innsbruck, Austria

EXTENDED ABSTRACT post-purchase responses (Lee and Tsai 2014; Wu et al. 2017). Sec- Anthropomorphism – the attribution of human characteristics ond, while extant anthropomorphism research has mainly focused to nonhuman objects – is a popular marketing strategy used for vari- on the consequences of moral agency (i.e., the capacity to perform ous products (Aggarwal and McGill 2007; Epley, Waytz, and Ca- intentional actions), the current research is among the few excep- cioppo 2008). This research focuses on the anthropomorphization of tions that examine the consequences of experience – the second di- food products – a fairly neglected area of research. To promote food mension of mind perception (Gray et al. 2007; Waytz et al. 2010b). products, companies frequently rely on anthropomorphized brand This is relevant because many anthropomorphized entities do not act characters such as Mr. Peanut, Tony the Tiger, or the M&M mascots themselves but are acted upon (especially in a consumption context). and endow products with humanlike characteristics. For example, Third, the current research has implications for prior research on Nestlé promotes its cinnamon toast crunch cereals by making the ce- consumption-related consequences of anthropomorphism. Research reals appear alive and Dunkin’ Donuts recently introduced a smiley has demonstrated that anthropomorphizing a tempting consumption face donut. In addition, governmental institutions such as the United object increased consumption amount because anthropomorphizing States Department of Agriculture (USDA) anthropomorphize fruits reduced experienced conflict among consumers with a strong diet- and vegetables to promote healthy eating to children. ing goal (Hur et al. 2015). Our work extends this line of research by What are the consequences of anthropomorphizing food prod- uncovering an additional mechanism that impacts consumers’ con- ucts? While extant anthropomorphism research has focused pre- sumption-related responses. We find that anthropomorphizing a food dominantly on consumer responses before purchase (e.g., liking, product inhibits consumption and leads to more negative emotional purchase; e.g., Wen Wan et al. 2017), the current research takes a outcomes because consumers feel as if they would harm the anthro- new perspective and investigates how anthropomorphizing food pomorphized product by eating it. We believe that findings of Hur products impacts consumption-related outcomes. Drawing from et al. (2015) do not contradict our results, but rather suggest an ad- mind perception theory, the central proposition of this research is ditional boundary condition. That is, for consumer with a strong diet- that anthropomorphizing leads to the perception of Experience – that ing goal, the negative effect of anthropomorphism on consumption is, capacity to feel pain, pleasure and emotions (Gray, Gray, and We- enjoyment should be attenuated because feelings of reduced conflict gner 2007). Experience transforms an entity into a moral patient. A counteract perceptions of inflicted harm. suffering moral patient triggers feelings of care and concern (Gray Lastly, the current research offers implications for marketers et al. 2007; Waytz et al. 2010b), which should negatively influence and public policy makers. For example, a negative consumption ex- consumption-related responses. perience might lead to lower re-purchase likelihood and frequency, Two studies provide evidence for the deteriorating effect of which is why marketers need to be aware of post-purchase conse- anthropomorphism on consumption experience. Study 1 (n = 46), a quences of an anthropomorphism strategy. field study, was presented as a new product evaluation task. Partici- pants were instructed to choose one of two cookies (control vs. an- REFERENCES thropomorphized), eat it, and, then, indicate how much they enjoyed Aggarwal, Pankaj and Ann L McGill (2007), „Is That Car it and how they felt after eating it. The results show that consumers Smiling at Me? Schema Congruity as a Basis for Evaluating enjoyed eating the anthropomorphized cookie significantly less and Anthropomorphized Products,“ Journal of Consumer felt significantly worse after eating the anthropomorphized cookie Research, 34 (4), 468-79. compared to the nonanthropomorphized version (all p’s < .05). Epley, Nicholas, Adam Waytz, Scott Akalis, and John T Cacioppo Next, study 2 shed light on the underlying mechanisms. Study 2 (2008), „When We Need a Human: Motivational Determinants (n = 116; MTurk sample) used a single-factor (control vs. anthropo- of Anthropomorphism,“ Social Cognition, 26 (2), 143-55. morphized) between-subjects design. As stimulus, we used a recent Gray, Heather M, Kurt Gray, and Daniel M Wegner (2007), Chips Ahoy! cookie advertisements. For the anthropomorphized con- “Dimensions of Mind Perception,” Science, 315 (5812), 619- dition, we used the original ad which featured a cookie with human 19. face features. For the control condition, we removed the human face Gray, Kurt, Liane Young, and Adam Waytz (2012), “Mind features from the cookie. The results showed that while anthropomor- Perception Is the Essence Morality,” Psychological Inquiry, 23 phizing enhances consumption experience due to greater whimsical (2), 101-24. cuteness, it simultaneously, and even more so, deteriorates consump- Hayes, Andrew F (2013), Introduction to Mediation, Moderation, tion experience because of feelings of inflicted harm. Consumers feel and Conditional Process Analysis: A Regression-Based as if they would hurt the anthropomorphized food product by eating Approach, New York: Guilford. it, which negatively influences consumption-related responses. Hur, Julia D, Minjung Koo, and Wilhelm Hofmann (2015), In doing so, the current research contributes to anthropomor- „When Temptations Come Alive: How Anthropomorphism phism literature in several ways. First, while existing anthropo- Undermines Self-Control,“ Journal of Consumer Research, 42 morphism research focused predominantly on consumer responses (2), 340-58. before consumption (e.g., liking, evaluation), the current research Kim, Sara and Ann L McGill (2011), „Gaming with Mr. Slot or outlines important downstream consequences of anthropomorphism. Gaming the Slot Machine? Power, Anthropomorphism, and To the best of our knowledge, this is the first research to systemati- Risk Perception,“ Journal of Consumer Research, 38 (1), 94- cally document deteriorating effects of anthropomorphism on post- 107. purchase consumer responses. This adds a novel chapter to an evolving stream of research investigating consumers differential pre- and

Lee, Leonard and Claire I Tsai (2014), „How Price Promotions Wen Wan, Echo, Rocky Peng Chen, and Liyin Jin (2017), „Judging Influence Postpurchase Consumption Experience over Time,“ a Book by Its Cover? The Effect of Anthropomorphism on Journal of Consumer Research, 40 (5), 943-59. Product Attribute Processing and Consumer Preference,“ Nenkov, Gergana Y and Maura L Scott (2014), „“So Cute I Could Journal of Consumer Research, 43 (6), 1008-30. Eat It Up”: Priming Effects of Cute Products on Indulgent Winkielman, Piotr and John T Cacioppo (2001), „Mind at Ease Consumption,“ Journal of Consumer Research, 41 (2), 326-41. Puts a Smile on the Face: Psychophysiological Evidence That Waytz, Adam, John Cacioppo, and Nicholas Epley (2010a), „Who Processing Facilitation Elicits Positive Affect,“Journal of Sees Human? The Stability and Importance of Individual Personality and Social Psychology, 81 (6), 989-1000. Differences in Anthropomorphism,“Perspectives on Wu, Freeman, Adriana Samper, Andrea C Morales, and Gavan J Psychological Science, 5 (3), 219-32. Fitzsimons (2017), „It’s Too Pretty to Use! When and How Waytz, Adam, Kurt Gray, Nicholas Epley, and Daniel M Wegner Enhanced Product Aesthetics Discourage Usage and Lower (2010b), „Causes and Consequences of Mind Perception,“ Consumption Enjoyment,“ Journal of Consumer Research, 44 Trends in Cognitive Sciences, 14 (8), 383-88. (3), 651–72. When Lack of Belonging Means Bad News for the Planet: The Consequences of Low Belonging on Ethical Product Purchases Ainslie Schultz, Providence College, USA Kevin Newman, Providence College, USA Scott Wright, Providence College, USA

EXTENDED ABSTRACT represent higher amounts. As predicted, we found a significant two- Consumers increasingly report feeling disconnected from oth- way interaction (t(1, 254) = 2.42, p = .02). Individuals who viewed ers. They live farther from family, belong to fewer social groups, the ethical product and were primed with low belongingness were and are more likely to live alone than ever before (Pew Research willing to pay less for the ethical product than controls (Mlow belongingness,

Center 2015). Low belongingness is an aversive state for consumers, ethical = .36, Mcontrol, ethical = .29; t(1, 254) = -2.18, p = .03). since belonging is a human need (like food, water, and shelter), and To explore the underlying mechanism, we conducted dual mod- relationships with others provide access to survival benefits such as erated mediation using Hayes Process Model 86. The results con- resources and protection (Buss 1990). Consumers often use ethical firmed that both focus on others and concern with social acceptance products to help them experience belonging since these products can mediated the moderated effect ([.001, .018]). Among participants signal altruism and convey status to others (Griskevicius et al. 2010; primed with low belongingness, the effect was significant (Effect = Van Vugt et al. 2007). But for individuals low in belongingness, in- -.01, 95% CI = [-.015, -.002]), suggesting that participants low in vesting in ethical products may not be worthwhile since they cannot belongingness are less willing to pay for ethical products because harness social benefits of these products such as elevated group sta- they are less focused on others, which leads to less concern with tus and increased access to resources within groups. social acceptance. Among controls, the effect was the opposite. Dual Some literature suggests that low belongingness decreases pro- mediation was not significant when the order of the mediators was social actions (Baumeister et al. 2007). In this research, we predict reversed. that low belongingness may reduce interest in ethical products, lead- Experiment 3 provides further evidence for the underlying ing individuals to view ethical products less favorably and pay less mechanism, concern with social acceptance, by testing whether for them. We expect that these effects occur due to a lower focus on individuals who feel low belongingness gravitate toward minority others, which results in less concern for social acceptance. (vs. control) endorsed products. 133 participants were recruited on Experiment 1 predicts that consumers who experience lower (vs. Mturk. Participants completed the writing task from experiment 2 higher) feelings of social belonging will evaluate ethically branded to prime low belongingness (vs. control). Next, participants saw one products less favorably. 80 students participated in a lab study. Par- of two advertisements about a green brand, EcoTowel, encouraging ticipants read a passage about an invented brand called Ony. The them to purchase it and decrease their urban footprint. The advertise- passage allowed us to manipulate perception of the brand as either an ments were identical except for the manipulation: 1) Only 19% of ethical, socially responsible brand, or as a typically branded control. American households look for green products when they shop (mi- Next, participants indicated their overall evaluation of the Ony nority endorsed), or 2) no statement (control). Participants indicated brand: bad/good, unfavorable/favorable, and negative/positive (α = their willingness to pay for the EcoTowel. .94) as our dependent variable. Following, participants completed We transformed willingness to pay; lower values represent Twenge et al.’s belonging scale (2007, α = .87). higher amounts. We found a significant two-way interaction (F(1, Using Hayes Process Model 1 (2013), we conducted a multiple 129) = 4.47, p = .04). Among individuals primed to feel low in social regression. As expected, the analysis revealed a marginally signifi- belonging, those who viewed the minority endorsed advertisement cant two-way interaction (ethical or control × belongingness, b = .25, (vs. control) reported a higher willingness to pay for the ethical prod- t(1,76) = 1.94, p = .06). Participants lower in belongingness (-1 SD) uct (Mminority endorsed, low belongingness = .15, Methical control, control = .21; F(1, 129) = 6.41, p = .01). showed less favorable evaluations of the ethical brand (Mlow belongingness Experiment 3 shows that individuals low in belongingness grav- = 4.70) than participants higher in belongingness (Mhigh belongingness = 5.57, +1 SD, t = 2.22, p = .03). itate toward ethical products that endorse the minority opinion, sug- Experiment 2 tests the underlying mechanism “focus on oth- gesting that these individuals may be less concerned with others and ers” and “concern for social acceptance” and suggests that individu- social acceptance. This finding provides an important contribution to als low in belongingness may display lower willingness to pay for past research on social influence, which has found that individuals ethical products due to a lower focus on others that leads them to are more likely to engage in ethical behavior, such as recycling and lack concern for social acceptance. 258 participants were recruited donating (Goldsmith, 2015; Goldstein, Cialdini, and Griskevicius, on Mturk. Participants first completed a writing task to prime low 2008), when majority norms are used. We suggest that this effect may belongingness (vs. control). In the low belongingness condition, par- vary for individuals low in belongingness. ticipants wrote about a time they felt they did not belong; in the con- Overall, our results reveal that individuals who feel low in be- trol condition, participants wrote about a TV show. Next, participants longingness (vs. controls) evaluate ethical products less favorably saw the Ony ethical product manipulations from experiment 1. As and are less willing to pay for them. Applied, our results suggest the dependent variable, participants indicated how much they would that feelings of isolation and low belongingness may engender unin- be willing to pay for a cup of Ony coffee. Afterward, participants tended consequences for society and the environment. answered the focus on others (from Winterich and Barone, 2011, α = .89) and concern with social acceptance scales (from Chen, 2017, α REFERENCES = .96), the mediators. Baumeister, Roy F. et al. (2007), “Thwarting the need to belong: Using Hayes Process Model 1 (2013), we conducted regression Understanding the interpersonal and inner effects of social to predict willingness to pay. Willingness to pay was transformed to exclusion,” Social and Personality Psychology Compass, 1(1), account for extreme values. Due to the transformation, lower values 506-20.

Buss, David M. (1990), “The evolution of anxiety and social Preacher, Kristopher J., Derek D. Rucker, and Andrew F. Hayes exclusion,” Journal of Social and Clinical Psychology, 9(2), (2007), “Addressing moderated mediation hypotheses: 196-201. Theory, methods, and prescriptions,” Multivariate Behavioral Goldsmith, Eizabeth B. (2015), Social Influence and Sustainable Research, 42(1), 185-227. Consumption, New York: Springer International Publishing. Twenge, Jean M., Roy F. Baumeister, C. Nathan DeWall, Natalie Goldstein, Noah J., Robert B. Cialdini, and Vladas Griskevicius J. Ciarocco, and J. Michael Bartels (2007), “Social exclusion (2008), “A room with a viewpoint: Using social norms to decreases prosocial behavior,” Journal of Personality and motivate environmental conservation in hotels,” Journal of Social Psychology, 92(1), 56-66. Consumer Research, 35(3), 472-82. Van Vugt, Mark, Gilbert Roberts, and Charlie Hardy (2007), Griskevicius, Vladas, Joshua M. Tybur, and Bram Van den Bergh “Competitive altruism: Development of reputation-based (2010). “Going green to be seen: status, reputation, and cooperation in groups,” in Handbook of Evolutionary conspicuous conservation,” Journal of Personality and Social Psychology, eds. R. Dunbar & L., Oxford University Press: Psychology, 9(8), 392-404. Oxford, 531-40. Hayes, Andrew F (2017), Introduction to mediation, moderation, Winterich, Karen Page and Michael J. Barone (2011), “Warm and conditional process analysis: A regression-based glow or cold, hard cash? Social identity effects on consumer approach, New York, NY: Guilford Publications. choice for donation versus discount promotions,” Journal of Marketing Research, 48(5), 855-68. When Less is More - How Making Products More Personal Can Decrease Purchase Intention Michael Schulz, University of Cologne, Germany

EXTENDED ABSTRACT The aim of Study 2 was to replicate the findings of Study 1 Personalization (i.e., tailoring products and services to individual and to identify potential downsides of personization. Research in the consumer needs) is central to many successful businesses (Kavadias, domain of personalization suggests that over-personalization can in- Ladas, and Loch 2016). Research on personalization has predominant- duce negative consumer reactions (Doorn and Hoekstra 2013; White ly investigated the effects of consumers sharing personal information et al. 2008). To explain negative consequences of personization, with companies (e.g., Zhang and Wedel 2009). Yet in various indus- the author builds on social presence theory and the role of intimacy tries, companies are also sharing personal information with consumers (Short, Williams, and Christie 1976). Personization might enable (i.e., the flow of information is reversed): The furniture retailer IKEA an object to become a social participant in the environment of the provides names of product designers aside its products in its retail and consumer. Following social presence theory, the author hypothesizes online stores. The food manufacturer Ruegenwalder Muehle displays that personization can violate consumers’ privacy in categories as- pictures and job descriptions of product developers on its products’ sociated with a high level of intimacy (e.g., toiletries, hygiene prod- packages. The FMCG company Beiersdorf shares videographic sto- ucts). Study 2 (N = 209) followed the same between-subjects design ries of product developers online. These examples illustrate a novel as Study 1. Shower gel was chosen as a product associated with a approach to personalization (hereafter referred to as “personization” high level of intimacy in usage. In both conditions, participants saw (Fuchs et al. 2016)), which is different from traditional personalization a bottle of shower gel containing a new formula. In the treatment in several aspects: Foremost, companies share personal information condition, participants were informed by whom the new formula was with consumers. Thus, the direction of information flow is reversed. developed (i.e., two product developers of the company). Addition- Furthermore, personization might make products appear to be more ally, a picture of the two developers was shown on the bottle. In the personal, yet they do not become more individual. Finally, consum- control condition, participants received no information on the prod- ers cannot opt out of personization as there is no proactive consent uct developers. Reconfirming the findings of Study 1, participants necessary. Thus, the author believes that the existent research on per- reported higher perceptions of anthropomorphism (t(207) = 1.90, p = sonalization is only partially applicable to this matter. To the best of .059) and uniqueness (t(207) = 5.48, p < .001) in the treatment condi- the author’s knowledge, there is only one study specifically dedicated tion. Also, participants perceived the treatment condition to be more to this topic (Fuchs et al. 2016), which demonstrates that consumers intrusive to privacy (t(207) = 7.87, p < .001). Subsequently, a media- choose personized over non-personized products in a series of choice- tion model was estimated. In line with Study 1, personization has based experiments. Considering that personization used in practice yet positive indirect effects on purchase intention via anthropomorphism under-researched, the author examines positive and negative effects of (ba1×b1 = 0.21, CI90% [0.03, 0.40]) and uniqueness (ba2×b2 = 0.15, CI90% personization on consumer behavior in two studies. [0.05, 0.26]). Additionally, a negative indirect effect of personization

The aim of Study 1 was to establish the underlying psychologi- on purchase intention via intrusion to privacy (ba3×b3 = -0.19, CI90% cal processes of personization. On one side, the author hypothesizes [-0.34, -0.06]) was found. The total effect of personization on pur- that personization influences perceptions of anthropomorphism. An- chase intention was negative (bc = -0.44, 90% CI90% [-0.75, -0.12]). thropomorphism “describes the tendency to imbue the real or imag- These results suggest that for products associated with a high level ined behavior of nonhuman agents with humanlike characteristics, of intimacy, personization is perceived to be intrusive to consumers’ motivations, intentions, or emotions” (Epley, Waytz, and Cacioppo privacy and thus decreases purchase intention. 2007, 864). In marketing literature, different forms of anthropomor- These findings indicate that personization is a double-edged phism and their effects on consumer behavior have been discussed sword. While personization increases perceptions of anthropomor- (e.g., Kim and McGill 2011). Also, research suggests that even non- phism and uniqueness, it can be used in practice to differentiate prod- diagnostic information, for example knowing a person’s name, can ucts by making them appear to be more human and unique. However, influence consumers (e.g., Jiang et al. 2010). On the other side, the this approach seems not to be appropriate for all categories. The find- author hypothesizes that personization influences perceptions of ings highlight that for categories associated with a high level of in- uniqueness (Snyder and Fromkin 1980). Compared to the majority timacy, personization can decrease purchase intention as consumers of products, personization still might be a distinctive feature making feel violated in their privacy. products unique. To test these hypotheses, data for Study 1 (N = 211) was collected in a large Western European country. In a between- REFERENCES subjects design, participants were randomly assigned to one of two Doorn, Jenny and Janny Hoekstra (2013), “Customization of Online conditions. In both conditions, participants saw a snack foods prod- Advertising: The Role of Intrusiveness,” Marketing Letters, 24 uct. In the treatment condition, participants were informed how and (4), 339-51. by whom (i.e., the product developer’s name was explicitly men- Epley, Nicholas, Adam Waytz, and John Cacioppo (2007), tioned) the product was developed. In the control condition, partici- “On Seeing Human: A Three-Factor Theory of pants were only informed how the product was developed. Results Anthropomorphism,” Psychological Review, 114 (4), 864-86. suggest that participants had higher perceptions of anthropomor- Fuchs, Christoph, Martin Schreier, Ulrike Kaiser, and Stijn van phism (t(209) = 4.90, p < .001) and uniqueness (t(209) = 4.74, p < Osselaer (2016), “Reducing Consumer Alienation: The Effect .001) in the treatment condition. While the findings of a mediation of Making Product Producers Personal,” in Advances in model suggest that personization has indirect effects on purchase in- Consumer Research, Vol. 44, ed. Page Moreau and Stefano Puntoni, Duluth, MN: Association for Consumer Research, tention via anthropomorphism (ba1×b1 = 0.50, CI90% [0.32, 0.69]) and 118-22. uniqueness (ba2×b2 = 0.16, CI90% [0.06, 0.29]), there is no overall effect

Jiang, Lan, Joandrea Hoegg, Darren Dahl, and Amitava Short, John, Ederyn Williams, and Bruce Christie (1976), The Chattopadhyay (2010), “The Persuasive Role of Incidental Social Psychology of Telecommunications, London: John Similarity on Attitudes and Purchase Intentions in a Sales Wiley & Sons Ltd. Context,” Journal of Consumer Research, 36 (5), 778-91. Snyder, Charles and Howard Fromkin (1980), Uniqueness: The Kavadias, Stelios, Kostas Ladas, and Christoph Loch (2016), “The Human Pursuit of Difference, New York: Plemum. Transformative Business Model,” Harvard Business Review, White, Tiffany, Debra Zahay, Helge Thorbjørnsen, and Sharon 94 (10), 90-98. Shavitt (2008), “Getting Too Personal: Reactance to Highly Kim, Sara and Ann McGill (2011), “Gaming with Mr. Slot or Personalized Email Solicitations,” Marketing Letters, 19 (1), Gaming the Slot Machine? Power, Anthropomorphism, and 39-50. Risk Perception,” Journal of Consumer Research, 38 (1), 94- Zhang, Jie and Michel Wedel (2009), “The Effectiveness of 107. Customized Promotions in Online and Offline Stores,”Journal of Marketing Research, 46 (2), 190-206. A Conceptual Framework of Violation of Trust and Negative Emotional Responses during Brand Transgressions Karthik Selvanayagam, Indian Institute of Technology, India Varisha Rehman, Indian Institute of Technology, India

EXTENDED ABSTRACT drives the consumer into adopting the brands identity comes from extending the knowledge about the brand into a more personal identi- Summary: fication (Tanis and Postmes 2005). At this resonance stage bonding, Though parallels to interpersonal relationships are drawn to there exists a trust equilibrium that maintains the relationship at its explain various facets of consumer-brand relationships, seldom fo- strongest level of development. An instability in this state of trust cus has been shown on explaining the nature of trust and its role results from the perception of the behaviour of one party by the other in emotional responses, particularly its violations and negative emo- as violating the mutual trust. This instability causes the perceiver to tions. By drawing from exchange and penetrations principles of so- assess the situation at a cognitive and emotional level, leading to the cial psychology theories, we build a conceptual model describing relationship’s decline or eventual disruption. Brand transgressions the development and decline of trust, and its possible outcomes in create instability in this state of trust in the relationship. the consumer-brand dyad. We outline the growth of this relationship Transgression and Decline of Trust – Dissatisfaction versus through Knapp’s (1978) ten stage model and the evolution of trust over these stages by employing Lewicki’s (1995) development and Violation of Trust decline model of trust in relationships. We further posit the emer- Conflicts and misunderstandings arise out of situations where gence of negative emotional responses from the consumers at times the behaviour of a partner does not fall into the boundary of the ex- of breach of this trust. pected behaviour, leading to a sense of distress (Lewicki & Bunker, 1995). Dissatisfaction resulting from unmet performance expecta- Conceptualization: tions as against a sense of loss arising from violation of trust would lead to distinct outcomes on the relationship (Reimann et al. 2018) Development of Trust two studies shed new light into the experience of brand betrayal. A Extant consumer behaviour literature predominantly exam- large-scale psychometric study shows that brand betrayal (vs. dis- ines the phenomenon of trust as purchase situation characteristics satisfaction. Though dissatisfaction causes negative emotional re- between exchange partners (Dwyer, Schurr, and Oh 1987; Das and sponses from the consumer, however, it would not be as strong or Teng 1998; Sirdeshmukh, Singh, and Sabol 2002), or alongside con- deeply rooted as responses resulting from a sense of moral viola- sumer commitment to organizations as a result of marketing practic- tion (Romani et al., 2012; Reimann et al., 2018)discontent, dislike, es (see: Morgan and Hunt 1994; Shankar and Ron 1997). However, embarrassment, sadness, and worry. While reactions to moral vio- there is a need for a perspective of understanding the nature of trust lations could lead to immediate termination of a relationship, dis- in a consumer-brand relationship through stages of its development, satisfaction would more likely follow a path of separation much in to better establish its similarities and distinctions from interpersonal similarity with an interpersonal relationship that would fall out of relationships. To this end, we correlate Lewicki’s (1995) model of equilibrium, taking the five coming-apart stages of Knapp’s (1978) development of trust with Knapp’s (1978) stages of formation and ten stage relationship model (Anderson, Avtgis, and West 1998). disintegration of relationships. Knapp’s (1978) model lists five stages ofcoming-apart as a result of Knapp’s (1978) ten stage relationship model describes the misunderstandings and conflicts in a relationship. It is also seen that coming-together and coming-apart stages of an interpersonal re- stronger derived emotions emerge in the emotional-activity stage of lationship. It is through this model we describe the developmental the dynamics of trust violation in case of violation of trust, while stages of a consumer brand relationship, progressing from initiating the assignment of importance to the transgressing event happens in and experimenting, to integrating and intensifying, and finally reach- the cognitive-activity stage (Lewicki 1995). The role of transgressor ing the bonding stage, where the relationship partners are able to response as a moderator is also pertinent, wherein the nature of the form a shared identity. As the relationship continues to develop, the response from the brand could drive the outcome towards rupture, nature of trust evolves alongside (Lewicki 1995). Lewicki’s (1995) recalibration or restoration of the relationship. model describes the evolution of trust in relationships over three stages. The first stage of building of trust is the Calculus-Based Trust, Major findings and Conclusions: where the validity of trust is confirmed through consistent meetings Drawing upon principles of social exchange and penetrations of expectations of both involved parties. The expectations in this from social psychology theories, this work contributes in differen- phase are limited to calculated value in exchanges between the part- tiating the distinct outcomes of a performance related transgression ners (Das and Teng 1998). The second stage of evolution of trust is as against a perceived violation of trust by the brand, resulting in the Knowledge-Based Trust, where trust develops by gaining more negative emotional responses from the consumer, and their subse- knowledge of the partners’ needs, preferences, and priorities (Bhat- quent impact on the consumer-brand relationship. Further concep- tacharya, Devinney, and Pillutla 1998). By reassuring the promised tual and empirical contributions on the nature of this trust and its value propositions and creating a consistent brand personality in the role in the context of different consumer-brand relationship types consumers’ minds over the period of the relationship, the consumer and stages, across cultures and product categories, will strengthen internalizes knowledge about the brand creating a stronger trust in the field of humanizing brands that is increasingly gaining attention it (Keller 2008). Consistency in this stage leads the consumer into from marketing scholars. Furthermore, the role of trust as antecedent developing a much stronger relationship with the brand, where the of consumer negative emotions towards brands, and its implications consumer beings to substitute a portion of the self-concept using the on consumption as well as consumer well-being also can be explored brand, creating a shared identity. The Identification-Based trust that

Advances in Consumer Research 785 Volume 46, ©2018 786 / A Conceptual Framework of Violation of Trust and Negative Emotional Responses during Brand Transgressions to build on the similarities, and distinctions, between interpersonal Lewicki, Roy J and Barbara Benedict Bunker (1995), “Trust in and consumer-brand relationships. Relationships: A Model of Development and Decline.,” in Understanding the role of trust, and the negative emotional re- Conflict, Cooperation, and Justice: Essays Inspired by the sponses associated with a violation of trust in a consumer brand re- Work of Morton Deutsch, ed. JZ & ASSOCIATES (Eds.) IN lationship would provide a predictive strength to practioners aiming BUNKER, BB, RUBIN, San Francisco, CA,: Jossey-Bass, at creating strong bonds with consumers in a crowded marketplace. 133–73. Morgan, Robert M and Shelby D Hunt (1994), “The Commitment- REFERENCES Trust Theory of Relationship Marketing,” The journal of Anderson, Traci L., Theodore a. Avtgis, and Daniel V. West (1998), marketing, 20–38. “Relationship Stages: An Inductive Analysis Identifying Reimann, Martin, Deborah J. MacInnis, Valerie S. Folkes, Arianna Cognitive, Affective, and Behavioral Dimensions of Knapp’s Uhalde, and Gratiana Pol (2018), “Insights into the Experience Relational Stages Model,” Communication Research Reports, of Brand Betrayal: From What People Say and What the Brain 15(3), 280–87. Reveals,” Journal of the Association for Consumer Research, Bhattacharya, Rajeev, Timothy M Devinney, and Madan M Pillutla 3(2), 000–000. (1998), “A Formal Model of Trust Based on Outcomes,” Romani, Simona, Silvia Grappi, and Daniele Dalli (2012), Academy of management review, 23(3), 459–72. “Emotions That Drive Consumers Away from Brands: Das, Tarun K and Bing-Sheng Teng (1998), “Between Trust and Measuring Negative Emotions toward Brands and Their Control: Developing Confidence in Partner Cooperation in Behavioral Effects,”International Journal of Research in Alliances,” Academy of management review, 23(3), 491–512. Marketing, 29(1), 55–67. Doney, Patricia M. and Joseph P. Cannon (1997), “An Examination Shankar, Ganesan and Hess Ron (1997), “Dimensions and Levels of the Nature of Trust in Buyer-Seller Relationships,” Journal of Trust : Implications for Commitment to a Relationship,” of Marketing, 61(2), 35. Marketing Letters, 8(4), 439–48. Dwyer, F Rober, Paul H. Schurr, and Sejo Oh (1987), “Developing Sichtmann, Christina (2007), “An Analysis of Antecedents and Buyer-Seller Relationships,” American Journal of Marketing, Consequences of Trust in a Corporate Brand,” European 51(2), 11–27. Journal of Marketing, 41(9/10), 999–1015. Keller, Kevin Lane (2008), Strategic Brand Management, 58 Sirdeshmukh, Deepak, Jagdip Singh, and Barry Sabol (2002), Brand. “Consumer Trust, Value, and Loyalty in Relational Knapp, Mark L (1978), Social Intercourse: From Greeting to Exchanges,” Journal of marketing, 66(1), 15–37. Goodbye, Allyn & Bacon, Incorporated. Tanis, Martin and Tom Postmes (2005), “A Social Identity Lewicki, Roy J (1995), “Trust in Relationships: A Model of Approach to Trust: Interpersonal Perception, Group Development and Decline.,” in Conflict, Cooperation, and Membership and Trusting Behaviour,” European Journal of Justice: Essays Inspired by the Work of Morton Deutsch, Social Psychology, 35(3), 413–24. 133–73. Round it up: Preference Exists for Rounded Totals (PERT) Varun Sharma, Bocconi University, Italy Aradhna Krishna, University of Michigan, USA Zachary Estes, Bocconi University, Italy

EXTENDED ABSTRACT Study 3 In recent years, charitable giving has stagnated. In the US, on (N=397) tests if PERT effect also occurs when participants must average individuals only donate about 2% of their annual dispos- mentally sum the purchase and donation amounts? We again used the able income, a level of contribution that has remained fairly constant art and craft scenario from the prior studies, using a 2 (total amount: across the past four decades (Giving USA 2017). Such charitable present, absent) × 2 (rounded number: present, absent) between-par- contributions are positive and admirable, of course, but there re- ticipants design with only two options of donation or no donation in mains a considerable and untapped upside of potential contributions all conditions. in donors’ mental budgets. We therefore present a simple method for Moreover, in all conditions the only the ‘donation’ option varied nudging more people to donate more money to charities. across conditions and was presented either as a total (i.e., total cost) Our main hypothesis builds from prior research showing that or as its component costs (i.e., purchase + donation). Thus, the ‘dona- rounded numbers (e.g., $3.00) subjectively “feel right” (Wadhwa and tion’ option was $30.00 in the rounded-present condition, and $29.84 Zhang 2015), and that people prefer offers with rounded prices (Yan in the nonrounded-present condition, while in these total-absent con- and Pena-Marin 2017). We investigate charitable donations that, ditions, it was $27.60 + $2.40 in the rounded-absent condition, and when added to the price of a product, sum to a rounded total i.e. $27.60 + $2.24 in the nonrounded-absent condition. for example, when purchasing a product for $7.60, might people be In the total present conditions, there was a significant prefer- more willing to donate an additional $1.40 than $1.20? ence for donating in the rounded-present condition (n=70, 71%), To the best of our knowledge, no research has investigated χ2(1)=16.98, p<.001but not in nonrounded-present condition (p=.55). whether the roundness of the total amount affects donations, and if More importantly, in the total-absent conditions there was a pref- so, by what psychological process it operates. Moreover, although erence (marginally) for donating in the rounded-absent condition consumer researchers have revealed many important factors affect- (n=59, 60%), χ2(1)=3.65, p=.056 but not in nonrounded-absent con- ing charitable donations (e.g. Charness and Holder 2018; Koschate- dition ( p=.19). The difference in donation choices was significant Fischer, Stefan and Hoyer 2012; Krishna 2011; Goswami and Ur- among the two total-present conditions (χ2(1)=11.54, p<.001) as well minsky 2016; Hagtvedt and Patrick 2016), no prior research has as for total-absent conditions (χ2(1)=5.18, p<.05). There was no sta- examined donation behavior in the context of add on to a purchase. tistical evidence of moderation (p=.33). We show in this study that Here we focus on the roundness of a total amount, hypothesizing that PERT effect is strongly associated with rounded totals. a preference exists for rounded totals (henceforth PERT effect). While rounded numbers could have positive effects, on the oth- Study 4 er hand, however, rounded numbers are also perceived to be bigger (N=197) tested whether processing fluency mediates the PERT than nonrounded numbers (Thomas, Simon and Kadiyali 2010). This effect. With the same scenario, participants were assigned to one of could lead consumers to feel that a rounded total is psychologically two conditions (total amount: rounded, nonrounded) between-partici- more costly than a comparable but nonrounded number. Thus, the pants. The ‘donation’ option was $30.00 ($2.36 donation) and $29.82 presumed PERT effect is not a foregone conclusion; theoretically, the ($2.18 donation) in the rounded and nonrounded conditions respec- opposite effect could occur. tively. Participants also rated the total amount of the ‘donation’ option (i.e., $30.00 in the rounded condition, $29.82 in the nonrounded condi- Study 1 tion) in terms of simplicity (2 items) and ease of processing (2 items). (N=99) tested whether a rounded total amount (product price + We again found support for PERT effect. There was a significant charity donation) is preferred over a nonrounded total, even though preference to donate in the rounded condition (n=66, 67%; χ2(1)=11.00, the rounded total was a higher overall cost. p=.001), but not in the nonrounded condition (n=52, 53%; p=.54), and Participants were asked to imagine that they visited an arts and this preference for donating differed marginally between the rounded crafts fair and purchased products costing $27.30 (including taxes). and nonrounded conditions, χ2(1)=3.80, p=.051. Bootstrap mediation They were asked whether they would like to donate to a charity asso- analysis (Hayes 2015, model 4, 10K samples) revealed a significant ciated with the fair from options: $28.70 (donation of $1.40), $30.00 indirect effect,B =.32, SE=.12, 95% CI [.12, .62] with fluency (α=.92). (donation of $2.70), or no donation. The rounded total increased fluency (B=.65, p<.001), which in turn in- Despite the rounded total being a higher overall cost, about half creased the likelihood of donating (B=.49, p=.001). Thus, these results of the participants (n=49; 50%) chose to donate with this rounded suggest that rounded totals increase donations because they are more total ($30.00) while only 26% chose the other donation option fluently processed than nonrounded totals. ($28.70). This preference for the higher, rounded total was signifi- In the current research, we show that people prefer rounded to- cant, χ2(1)=7.05, p<.01.These results provide a first demonstration tal amounts over nonrounded ones. This makes them choose a char- of the PERT effect. ity donation option in addition to a purchase which makes the total amount rounded. In addition, we show that if the option to donate is Study 2 presented as a total amount which is rounded, more people choose to (N=303) replicated PERT effect, but instead examined thelikeli - donate. We also show that this effect exists as people find the rounded hood of donating. We show that more people prefer to donate if the total amount to be simpler and easier to process. Our research has donation option has a rounded total rather than a nonrounded total. important implications for consumer research related to charity, as we show an easy way to make more people to donate and to increase donation amount if a higher donation option is rounded.

Advances in Consumer Research 787 Volume 46, ©2018 788 / Round it up: Preference Exists for Rounded Totals (PERT) REFERENCES Krishna, Aradhna (2011), “Can supporting a cause decrease Charness, Gary, and Patrick Holder (2018), “Charity in the donations and happiness? The cause marketing paradox,” laboratory: Matching, competition, and group identity,” Journal of Consumer Psychology, 21(3), 338-345. Management Science, Forthcoming. Thomas, Manoj, Daniel H. Simon, and Vrinda Kadiyali (2010), Giving USA (2017), “Giving USA 2017,” Retrieved from https:// “The price precision effect: Evidence from laboratory and givingusa.org. market data,” Marketing Science, 29(1), 175-190. Goswami, Indranil, and Oleg Urminsky (2016), “When should the Wadhwa, Monica, and Kuangjie Zhang (2015), “This number just ask be a nudge? The effect of default amounts on charitable feels right: The impact of roundedness of price numbers on donations,” Journal of Marketing Research, 53(5), 829-846. product evaluations,” Journal of Consumer Research, 41(5), Hagtvedt, Henrik, and Vanessa M. Patrick (2016), “Gilt and guilt: 1172-1185. Should luxury and charity partner at the point of sale?” Yan, Dengfeng, and Jorge Pena-Marin (2017), “Round Off the Journal of Retailing, 92(1), 56-64. Bargaining: The Effects of Offer Roundness on Willingness to Koschate-Fischer, Nicole, Isabel V. Stefan, and Wayne D. Hoyer Accept,” Journal of Consumer Research, 44(2), 381-395. (2012), “Willingness to pay for cause-related marketing: The impact of donation amount and moderating effects,”Journal of Marketing Research, 49(6), 910-927. Neural Pattern Similarity Reveals Brand Equity Feng Sheng, University of Pennsylvania, USA Michael Platt, University of Pennsylvania, USA

EXTENDED ABSTRACT sis across the brain to find brain regions where the neural pattern Memory about a brand is associated with memory about the similarity between a car brand and car was predictive of brand recall. category of the brand’s product (Hutchinson, 1983). Consequently, Neural representational similarity analysis revealed a whole brain when thinking about products in that category, that brand is likely map for each car brand that illustrated their neural similarity with the to be recalled by the activation of the associative network and to be product category car. These whole-brain maps showed the variance considered for choice (Nedungadi, 1990). Therefore, the strength of of neural similarity across different brain regions and the gradient of mental association between a brand and its product category deter- neural similarity across different car brands. Survival analysis revealed mines the top-of-mind accessibility of the brand in the product cat- that brand recall was predicted by neural similarity between car brand egory and constitutes a fundamental aspect of consumer-based brand and car across a distributed brain network including ventral medial equity (Keller, 1993). Despite its importance, the structure of mental prefrontal cortex (VMPFC), posterior cingulate cortex (PCC), and lat- association between a brand and a product category is largely invis- eral temporal cortex (LTC). Neural similarity at these brain regions ible in traditional behavioral measurements. were still predictive after neural similarity between car brand and com- We posit that mental representation for a brand and for its product puter were included as a control variable, suggesting that car brand category can both be directly visualized as neural representations, and recall was selectively predicted by neural similarity between car brand that mental association between the brand and the product category and the product category car rather than other product categories. can be operationally defined as the similarity of the neural representa- By performing pattern analysis with neural data acquired by tions between the brand and the product category. Furthermore, we fMRI, we illustrated the way to operationally define the structure hypothesize that neural representational similarity between brand and of mental association between brand and product category by cal- product category will be predictive of mental accessibility of brand culating neural similarity between brand and product category. Fur- assessed by behavioral measurement such as brand recall. To demon- thermore, we showed a neural network where the neural similarity strate our neural definition of brand-category association and to test between brand and product category was predictive of the accessibil- our hypothesis about the relationship between neural similarity and ity of brands to consumers. The brain regions found to be predictive mental accessibility, in an fMRI study, we recorded consumers’ brain of brand recall in our study, including VMPFC, PCC and LTC, are activity in response to car brands and the concept of car in an MRI largely overlapped with the neural network underpinning semantic scanner as well as recorded the process of brand recall when they were knowledge (see Binder, et al., 2009 for a quantitative meta-analysis). cued by the product category “car” outside the scanner. This implies that the degree to which the knowledge about a brand is In the experiment, participants first completed a brand recall incorporated with the concept of a product category is critical for the task without brain imaging. The brand recall task was designed to mental accessibility of the brand in that product category. be done first because we intended to protect participants’ memory Echoing with recent effort to imaging brain equity with neural of brands from being contaminated by any experimental exposure pattern analysis (Chen, Nelson & Hsu, 2015), our measurement of to brands. In the recall task, participants first saw a word indicating neural similarity between brand and product category has potential the product category (i.e., “car”) on the screen of a laptop, and then to act as a new indicator for brand equity in the future. Importantly, recalled and typed down brand names of the product category in 10 estimating consumer-based brand equity with neural activity will be minutes. The time at which they typed down the first letter of a brand beneficial for grounding the concept of brand equity. In a more broad was taken as the recall time for the brand. sense, redefining key concepts in marketing such as brand equity Thereafter, in the MRI scanner, we recorded blood-oxygen-level with neural evidence will guide marketing science to advance in a dependent (BOLD) signals in participants’ brains when they were pre- more quantitative and scientific direction. sented with each of 14 car brands, including Ford, Toyota, Chevrolet, Honda, Nissan, Hyundai, Subaru, Volkswagen, Mercedes-Benz, Maz- REFERENCES da, BMW, Chrysler, Volvo and Fiat. In addition to car brand names, Binder, Jeffrey R., Rutvik H. Desai, William W. Graves, and Lisa we also recorded their brain activity to the word of the target product L. Conant (2009), “Where is the Semantic System? A Critical category, “car”, as well as their brain activity to the word of a control Review and Meta-Analysis of 120 Functional Neuroimaging product category, “computer”. Each of the 14 car brand names and the Studies,” Cerebral Cortex, 19(12), 2767-2796. 2 product category words was presented 24 times in two runs in a pseu- Chen, Yu-Ping, Leif D. Nelson, and Ming Hsu (2015), “From dorandom order. The presentation of each word lasted 2 s, followed by “Where” to “What”: Distributed Representations of Brand a fixation of varying durations (mean=2 s, ranging from 1 s to 10 s). Associations in the Human Brain,” Journal of Marketing Participants were instructed to think about the meaning of each word Research, 52 (August), 453-466. when they saw it. To monitor participants’ attention, twenty four non- Hutchinson, J. Wesley (1983), “Expertise and the Structure of Free sense words were randomly inserted and participants were instructed Recall,” in Advances in Consumer Research, Vol. 10, eds., to press a key when they saw a nonsense word. Richard P. Bagozzi and Alice M. Tybout, Ann Arbor, MI: fMRI data analysis primarily included three steps. In the first Association for Consumer Research, 585- 589. step, we obtained a whole-brain pattern of activation to each word Keller, Kevin Lane (1993), “Conceptualizing, Measuring, and and to each product category by estimating a general linear model on Managing Customer-Based Brand Equity,” Journal of BOLD signals. In the second step, we performed a searchlight repre- Marketing, 57 (January), 1-22. sentational similarity analysis across the brain to compute the simi- Nedungadi, Prakash (1990), “Recall and Consumer Consideration larity of regional neural pattern between each brand and each product Sets: Influencing Choice without Altering Brand Evaluations,” category. In the third step, we conducted a voxel-wise survival analy- Journal of Consumer Research, 17(December), 263-276.

Advances in Consumer Research 789 Volume 46, ©2018 Seeing Brands Through Rose-Colored Lenses: When Fear of Being Taken Advantage Of Leads to Increased Trust Steven Shepherd, Oklahoma State University, USA Gavan J. Fitzsimons, Duke University, USA

EXTENDED ABSTRACT Study 1 (N=150) had participants complete the VS scale and People often face issues and domains that they do not under- then presented them with either a simple or complex description of stand. As such, people are often dependent on experts to manage how the sun’s rays affect one’s skin. Participants then rated their per- complex issues (Deci and Ryan 1985; Iyengar and Lepper 2000). In ceptions of how easy it would be for a sunscreen manufacturer to the context of products and services, a lack of knowledge and having take advantage of consumers (e.g., the product that does not actu- to rely on others brings with it risk and the potential to be taken ad- ally work). Results showed that when the domain was complex (vs. vantage of (Balafoutas, Beck, Kerschbamer, and Sutter 2013; Darby simple), those high in VS saw it as more possible for a company to and Karni 1973; Ford, Smith, and Swasy 1990; Mitra, Reiss, and take advantage of consumers. Capella 1999; Ostrom and Iacobucci 1995). Given that domain complexity leads high VS consumers to Uncertainty such as this is uncomfortable and people are mo- see increased opportunity to be taken advantage of, and given their tivated to avoid it (Hogg 2007; Kruglanski and Webster 1996). Al- concerns about being victimized, one would expect that high VS though increased skepticism and minimizing interdependence is an consumers would respond with decreased trust. However, based on intuitive strategy for managing this uncertainty, many domains exist past work on defensive trust, we posited that high VS consumers that consumers cannot easily opt out of. Therefore, counterintuitive- will address this concern by bolstering trust. Study 2 (N=201) sup- ly, some consumers may instead respond by putting on rose-colored ported this; high VS consumers showed increased trust in a brand of lenses and bolster trust in those they are reliant on. The idea that in- sunscreen when the domain (how the sun’s rays affect one’s skin) terdependence-related concerns are regulated by bolstering the value was complex as opposed to simple. Study 3 (N=299) replicated this of one’s partner is found in a variety of literatures (de Jong, Van der effect by first describing harmful bacteria in either simple or complex Vegt, and Molleman, 2007; Murray et al. 2008; Murray and Holmes terms, and then measuring trust in a brand of kitchen cleaner. 2009; Shepherd and Kay, 2012; van der Toorn et al. 2011). Finally, if high VS consumers manage concerns about exploita- Who is most likely to engage in this process? Those high in tion (i.e., when domain complexity is high) by bolstering trust, then victim sensitivity (VS) have a basic need to trust, but are also particu- we should observe that reducing the opportunity to be taken advan- larly sensitive to being victimized (Gollwitzer and Rothmund 2011). tage of will eliminate the positive effect of VS on trust. We also pre- However, our reliance on products to manage complex domains is dicted that high VS consumers would show increased anxiety when a relationship that is not always easy to opt out of. Because those thinking about their knowledge of the domain at hand, but that this high in VS are concerned about being taken advantage of, facing a anxiety only translates into increased trust when complexity is high complex domain that one does not understand (and can therefore be and concerns about being taken advantage of are present. taken advantage of) may increase anxiety and actually lead them to Study 4 (N=536) supported these hypotheses. When the oppor- bolster trust. In other words, bolstering trust in a brand’s intentions tunity to be taken advantage of was present, we replicated the effects and competency may be a defensive strategy that helps alleviate con- observed in Studies 2 and 3; high VS consumers showed higher lev- cerns about being taken advantage of. els of trust in their brand of sunglasses when the sun’s effect of one’s eyes was described in complex (vs. simple) terms. However, this ef- Hypothesis 1 When a domain is complex, those high (vs. low) fect disappeared when participants were told that industry standards in VS will see greater opportunity to be taken ad- and regulations ensure that most sunglasses protect against UV rays vantage of. (i.e., when the opportunity to be taken advantage of was reduced). Additionally, when the domain was complex and the opportunity to be taken advantage of was present, we found an indirect effect from Hypothesis 2 When a domain is complex, those high (vs. low) VS, to anxiety, to trust, such that those higher in VS felt more anx- in VS will bolster trust in a product/product ious about their lack of knowledge regarding how the sun affects manufacturer that manages the domain at hand. one’s skin, and this anxiety predicted increased trust. This indirect If this increase in trust is defensive in nature, such that it serves effect was not present in any of the other three conditions. These ef- to manage anxiety and concerns about being taken advantage of, then fects further support the prediction that because those high in VS are it follows that: especially concerned about being taken advantage of, they ironically bolster trust under conditions that one might expect to decrease trust; Hypothesis 3 When the opportunity to be taken advantage of that is, when a domain is complex and poorly understood by the con- is less present, the effect posited in H2 will be sumer, and when the opportunity to be taken advantage of is present. eliminated. Although existing consumer behavior research would not easily predict these effects, the current findings make sense in the context of a broader literature on domain complexity, interdependence, and de- Hypothesis 4 When domain complexity is high (vs. low) and fensive trust. This addresses a gap in the marketing literature regard- the opportunity to be taken advantage of is ing our understanding of consumer trust. Moreover, we introduce VS present, VS will predict anxiety regarding one’s as a relevant construct to the marketing literature. knowledge of a domain, which will in turn predict increased trust.

Advances in Consumer Research 790 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 791 REFERENCES Mitra, Kaushik, Michelle C. Reiss, and Louis M. Capella (1999), Balafoutas, Loukas, Adrian Beck, Rudolf Kerschbamer, and “An Examination of Perceived Risk, Informational Search and Matthias Sutter (2013), “What Drives Taxi Drivers? A Field Behavioral Intentions in Search, Experience, and Credence Experiment on Fraud in a Market for Credence Good,” The Services,” Journal of Services Marketing, 13 (3), 208—228. Review of Economic Studies, 80 (3), 876—891. Murray, Sandra L., Jaye L. Derrick, Sadie Leder, and John G. Darby, Michael R., and Edi Karni (1973), “Free Competition Holmes (2008), “Balancing Connectedness and Self-Protection and the Optimal Amount of Fraud,” The Journal of Law & Goals in Close Relationships: A Levels-of-Processing Economics, 16 (1), 67—88. Perspective on Risk Regulation,” Journal of Personality and de Jong, Simon B., Gerben S. Van der Vegt, and Eric Molleman Social Psychology, 94 (3), 429—459. (2007), “The Relationship Among Asymmetry in Task Murrary, Sandra L., John G. Holmes, Maya Aloni, Rebecca Dependence, Perceived Helping Behavior, and Trust,” Journal T. Pinkus, Jaye L. Derrick, and Sadie Leder (2009), of Applied Psychology, 92 (6), 1625—1637. “Commitment Insurance: Compensating for the Autonomy Ford, Gary T., Darlene B. Smith, and John L. Swasy (1990), Costs of Interdependence in Close Relationships,” Journal of “Consumer Skepticism of Advertising Claims: Testing Personality and Social Psychology, 97 (2), 256—278. Hypotheses from Economics of Information,” Journal of Ostrom, Amy, and Dawn Iacobucci (1995), “Consumer Trade-Offs Consumer Research, 16 (4), 433-441. and the Evaluation of Services,” Journal of Marketing, 59, Gollwitzer, Mario, and Tobias Rothmund (2011), “What Exactly are 17—28. Victim-Sensitive Persons Sensitive To?” Journal of Research Ryan, Richard M., and Edward L. Deci (2000), “Self-Determination in Personality, 45, 448—455. theory and the Facilitation of Intrinsic Motivation, Social Hogg, Michael A. (2007), “Uncertainty-Identity Theory,” in Development, and Well-Being,” American Psychologist, 55 Advances in Experimental Social Psychology, ed. M. P. Zanna, (1), 68—78. San Diego, CA: Academic Press. Shepherd, Steven, and Aaron C. Kay (2012), “On the Perpetuation Iyengar, Sheena S., and Mark R. Lepper (2000), “When Choice of Ignorance: System Dependence, System Justification, and is Demotivating: Can One Desire Too Much of a Good the Motivated Avoidance of Sociopolitical Information,” Thing?” Journal of Personality and Social Psychology, 79 (6), Journal of Personality and Social Psychology, 102 (2), 995—1006. 264—280. Kruglanski, Arie W., and Donna M. Webster (1996), “Motivated van der Toorn, Jojanneke, Tom R. Tyler, and John T. Jost (2011), Closing of the Mind: ‘Seizing’ and ‘Freezing’,” Psychological “More than Fair: Outcome Dependence, System Justification, Review, 103 (2), 263—283. and the Perceived Legitimacy of Authority Figures,” Journal of Experimental Social Psychology, 47, 127—138. The Effect of Future Focus on Self-Control is Moderated by Self-Efficacy Rafay A. Siddiqui, Hong Kong Polytechnic University, China Jane Park, University of California-Riverside, USA Frank May, Virginia Tech, USA

EXTENDED ABSTRACT more when focusing on the future, but the reverse emerged for those For consumers, sometimes the future may be very salient, high in self-efficacy. while other times it may not be (Shipp, Edwards and Lambert 2009). Study 2 replicated the effects we found in study 1 using the Future focus is critical to self-control, as one cannot self-regulate same future focus manipulation in the domain of spending. After the without at least some consideration of future goals or consequences future focus manipulation, participants imagined that they were try- (Baumeister and Heatherton 1996). Indeed, companies that support ing to save money but had come across a pair of shoes that they really self-control pursuits highlight this notion. For example, a Fidelity liked. After indicating whether they would purchase the shoes, they Investments advertisement featured the tagline “The Future is the completed a financial self-efficacy scale. Those low in self-efficacy Present,” while Weight Watchers highlights “My Bright Future” on were less likely to purchase the shoes when focusing on the future, its website. Not surprisingly, research has found that focusing on the but the reverse emerged for those high in self-efficacy. future can be beneficial. For instance, future focus can lead to the Study 3 shows that our effects hold for real choice when both choice of virtue foods (Winterich and Haws 2011), a greater likeli- factors are manipulated. Participants completed a self-efficacy ma- hood of investing in retirement (Nenkov, Inman and Hulland 2008) nipulation in which the scale to which they responded created the il- and increased patience (Bartels and Urminsky 2011). lusion of low or high success in goal achievement. Then, participants However, other research hints at special circumstances under were asked to write about what comes to mind when they think about which focusing on the future may actually lead to indulgence. For either the future or the present, and made a choice to receive either a instance, when consumers are asked to consider future progress to- Dairy Queen or Subway gift-card in a lottery. In the low self-efficacy wards goals, it may lead to indulgence in the present (Fishbach and condition, participants chose the unhealthy option, Dairy Queen, Dhar 2005; Laran 2010; Zhang, Fishbach and Dhar 2007). Similarly, less when focusing on the future (vs. the present), but the reverse simply anticipating a future opportunity to choose a healthy option emerged for those in the high self-efficacy condition. can lead to indulgence (Khan and Dhar 2007). This research suggests In study 4, we demonstrate process via mediation. Future focus that future focused individuals may lose self-control in the present if was manipulated, then participants imagined that they were trying they foresee themselves balancing out the negative effects of current to save money, but had come across a pair of headphones that they indulgence. could purchase. Next, we measured the degree to which participants Thus, it is not entirely clear as to when future focus will lead perceived that the purchase would negatively impact their savings to greater self-control, or what types of people will benefit from goals. Finally, participants completed a self-efficacy scale. The self- future focus. In addressing these issues, we posit that consumers’ control results mirrored those of past studies, and these effects were self-efficacy (Bandura 1994; Wood and Bandura 1989) will moderate mediated by the extent to which participants felt the purchase would the effect of future focus on self-control. Focusing on the future will negatively impact their savings goals. lead to higher self-control when one foresees the negative outcomes In study 5, we demonstrate process via moderation. That is, of current indulgence, but will lead to lower self-control when one we predict that our effects will diminish for those who believe that foresees successful counteractions in response to negative outcomes the future does not bring about change (i.e., entity theorists). In the of current indulgence. We posit that self-efficacy can determine the low (high) self-efficacy condition, participants were asked to solve extent to which people will focus on negative outcomes versus suc- nine puzzles, and thereafter were told that they performed better than cessful counteractions when thinking of the future. 30% (90%) of all other participants. Then, participants in the entity Self-efficacy is defined as the belief in one’s capabilities to (incremental) condition were asked to write a statement in agree- undertake actions to meet situational demands and combat goal ob- ment with a passage that stated people’s personalities could not be stacles (Wood and Bandura 1989). A higher self-efficacy means a changed (could be changed) even late in their lives. Participants then greater belief in one’s capability to successfully cope with goal set- imagined a scenario where they were determined to perform well in backs through action. However, individuals low in self-efficacy tend a course, and had an exam coming up, and chose between studying to visualize scenarios of goal failure (Bandura 1994). When one has versus going to a party. Participants subsequently completed a future an opportunity to engage in a goal inconsistent action, a focus on focus scale (Shipp et al. 2006). Our effects persisted in the incremen- the future should bring to mind the consequences of indulgence. We tal condition, but were attenuated in the entity condition. expect that a future focused individual with low (high) self-efficacy would visualize goal failure and its consequences (successful coun- Conclusion teractions to indulgence). Hence, we predict that when one is low Our findings have significant implications. Theoretically, we (high) in self-efficacy, focusing on the future will increase (decrease) delineate when and for whom a future focus may be detrimental to self-control. self-control. This has implications for consumer welfare, as asking consumers to focus on the future might not always be beneficial. Studies In study 1, we manipulated participants to either write a state- REFERENCES ment about focusing on the future or office products. Next, they Bandura, A. (1994). “Self-efficacy”. In V. S. Ramachaudran (Ed.), imagined a scenario in which they had to choose between chocolate “Encyclopedia of Human Behavior” (Vol. 4, pp. 71-81). New cake and fruit, and completed a self-efficacy scale (Chen, Gully and York: Academic Press. (Reprinted in H. Friedman [Ed.], Eden 2001). Those low in self-efficacy chose the healthier option “Encyclopedia of Mental Health” San Diego: Academic Press, 1998).

Bartels, Daniel M. and Oleg Urminsky (2011), “On Intertemporal Nenkov, Gergana Y., J. Jeffrey Inman, and John Hulland (2008), Selfishness : How the Perceived Instability of Identity “Considering the Future: The Conceptualization and Underlies Impatient Consumption,” Journal of Consumer Measurement of Elaboration on Potential Outcomes,” Journal Research, 38(1), 182–98. of Consumer Research, 35, 126–41. Baumeister, Roy F. and Todd F. Heatherton (1996), “Self- Shipp, Abbie J., Jeffrey R. Edwards, and Lisa Schurer Lambert Regulation Failure: An Overview,” Psychological Inquiry, (2009), “Conceptualization and Measurement of Temporal 7(1), 1–15. Focus: The Subjective Experience of the Past, Present, and Chen, Gilad, Stanley M. Gully, and Dov Eden (2001), “Validation Future,” Organizational Behavior and Human Decision of a New General Self-Efficacy Scale,”Organizational Processes, 110(1), 1–22. Research Methods, 4(1), 62–83. Winterich, Karen Page and Kelly L. Haws (2011), “Helpful Fishbach, Ayelet and Ravi Dhar (2005), “Goals as Excuses or Hopefulness: The Effect of Future Positive Emotions on Guides: The Liberating Effect of Perceived Goal Progress on Consumption,” Journal of Consumer Research, 38(3), 505–24. Choice,” Journal of Consumer Research, 32(3), 370–77. Wood, R and A Bandura (1989), “Impact of Conceptions of Ability Khan, Uzma, and Ravi Dhar (2007), “Where There is a Way, is on Self-Regulatory Mechanisms and Complex Decision There a Will? The Effect of Future Choices on Self-Control,” Making.,” Journal of Personality and Social Psychology, Journal of Experimental Psychology: General, 136(2), 277- 56(3), 407–15. 288. Zhang, Ying, Ayelet Fishbach, and Ravi Dhar (2007), “When Laran, Juliano (2010), “Choosing Your Future: Temporal Distance Thinking Beats Doing: The Role of Optimistic Expectations in and the Balance between Self-Control and Indulgence,” Goal-Based Choice,” Journal of Consumer Research, 34(4), Journal of Consumer Research, 36, 1002–15. 567–78. The Asymmetry between Time and Money Compensation effect when feeling Scarcity: Time helps the Money Poor, but Money doesn’t help the Time Poor Jane So, University of Washington, USA Nidhi Agrawal, University of Washington, USA

EXTENDED ABSTRACT we asked participants to indicate their willingness to accept advice We examine two important aspects of scarcity. First, we identify from the reviewer. As predicted, we found that compared to the con- a new scarcity effect on consumers’ advice-taking tendency. Second, trol condition, the time scarcity condition and the money scarcity we show whether the abundance of one resource alters the scarcity condition showed lower advice-taking tendency. effect of another resource in the domain of advice-taking and wheth- In study 2, we investigated the compensation effect. We used er such interplays vary depending on the resource type (e.g. time 2(scarcity: time vs. money) × 2(abundance of alternate resource: abundance compensating for money scarcity vs. money abundance abundance vs. nonabundance) + control condition between-subjects compensating for time scarcity). design. We used the same wine bar scenario from study 1. Results Researchers found that feelings of resource scarcity systemati- showed that when people felt only time or money scarcity but no cally affect how consumers think, perceive, feel, and behave through feelings of abundance, we replicated our previous effect that scar- scarcity mindsets (Shah et al. 2012; Shah et al. 2015; Roux et al. 2015). city leading to lower advice-taking tendency. However, when money Adding to this line of research, we suggest that when feeling time or scarce participants also felt time abundance, time abundance com- money scarcity, which are the two common scarcities that people expe- pensated for money scarcity and increased participant’s advice- rience, the scarcity mindset affects consumers’ advice-taking tendency. taking tendency. Moreover, when time scarce participants also felt We argue that when people are feeling time or money scarcity, people money abundance, money abundance did not compensate for time are less likely to change their pre-existing opinions based on someone scarcity, showing asymmetric compensation effect. else’s advice. Resource scarcity signals that the environment is harsh In study 3, we investigated whether such asymmetric compensa- (Laran and Salerno 2013; S. E. Hill et al., 2012; Griskevicius et al. tion effect is due to how people perceive value of time versus money. 2013), such that people need to compete with others for the scarce re- We used the 2(scarcity: time vs. money) × 2(abundance of alternate source (Kristofferson et al. 2017; Roux et al. 2015). Thus when people resource: abundance vs. nonabundance) between-subjects design. are feeling scarcity, the self becomes more important and people act In this study, we asked participants which option between diets and in accordance to one’s own benefit (Roux et al. 2015). Thus, time or exercising is better for health and then showed an article about an money scarcity can make people focus on the self more, thereby lead- opposing opinion. We then asked how willing they are to take advice ing to lower advice-taking tendency. from the article writer. After a filler task, we asked measured par- Another important gap in the literature is that research on re- ticipants’ belief about value of time compared to money. The results source scarcity examines the effect of scarcity in comparison to that showed a significant three-way interaction. For those who believe of resource abundance (Shah et al. 2012; Shah et al. 2015; Mehta and time is more valuable than money, participants who felt only money Zhu 2016; Sharma and Alter 2010; Zhu and Ratner 2015). However, scarcity showed marginally lower advice-taking tendency than those consumers often experience both resource abundance and scarcity at who felt only time scarcity. However, among those who believe time the same time. For example, busy people may feel time scarcity but is more valuable than money, people who felt money scarcity and they may also feel money abundance. Poor people may feel money time abundance showed greater advice-taking tendency than those scarcity but they may also feel time abundance. Thus in this article, who felt time scarcity and money abundance. As expected, this effect we examine how feelings of resource abundance interplay with feel- was not significant for those who believe money is more valuable ings of resource scarcity. than time. We suggest that time abundance compensates for money scarci- Finally in study 4, we examined the underlying process of ty effect while money abundance does not compensate for time scar- asymmetric compensation effect by manipulating whether time and city effect. We suggest that this asymmetric compensation of time money are substitutable. In this study, we primed people the concept (vs. money) abundance when feeling money (vs. time) scarcity is due of time and money substitutability or imperfect substitutability. We to the difference in the nature of time (vs. money). First, people per- used the same wine bar scenario from the previous studies. As ex- ceive money is a substitutable resource whereas time is not (Leclerc pected, when primed with money can be substituted with time, the et al. 1995). Due to the greater fungibility of money, money scarce results replicated our previous findings that time abundance com- people may have a mindset that money can be substituted even with pensated for money scarcity. However, when the imperfect substi- a different resource such as time. However, time is a less fungible re- tutability is primed, the time abundance no longer compensated for source than money, thus time scarce people may have a mindset that money scarcity. time cannot be substituted with any resources. Second, people may perceive time to greater value than money. Time is more laden with REFERENCES personal meaning than money (Aaker, Rudd and Mogilner 2011; Aaker, Jennifer L., Melanie Rudd, and Cassie Mogilner. “If money Mogilner and Aaker 2009; Reed et al. 2007; Reed et al. 2015) and does not make you happy, consider time.” Journal of consumer thus people may assign greater value to time than money. If this is the psychology 21, no. 2 (2011): 126-130. case, having such valuable resource, which is time, may compensate Griskevicius, Vladas, Joshua M. Ackerman, Stephanie M. for lack of less valuable resource, which is money. Cantú, Andrew W. Delton, Theresa E. Robertson, Jeffry Study 1 first examined whether feeling time or money scarcity A. Simpson, Melissa Emery Thompson, and Joshua M. leads to lower advice-taking tendency. We used a one-way ANOVA Tybur. “When the economy falters, do people spend or (money scarcity vs time scarcity vs. control). After manipulation, we save? Responses to resource scarcity depend on childhood asked participants to imagine a wine bar that they did not like. Next, environments.” Psychological science 24, no. 2 (2013): 197- we showed a review that was very positive. After reading the review, 205.

Hill, Sarah E., Christopher D. Rodeheffer, Vladas Griskevicius, Reed II, Americus, Adam Kay, Stephanie Finnel, Karl Aquino, Kristina Durante, and Andrew Edward White. “Boosting and Eric Levy. “I don’t want the money, I just want your beauty in an economic decline: Mating, spending, and time: How moral identity overcomes the aversion to giving the lipstick effect.” Journal of personality and social time to prosocial causes.” Journal of personality and social psychology 103, no. 2 (2012): 275. psychology 110, no. 3 (2016): 435. Kristofferson, Kirk, Brent McFerran, Andrea C. Morales, and Roux, Caroline, Kelly Goldsmith, and Andrea Bonezzi. “On the Darren W. Dahl. “The dark side of scarcity promotions: psychology of scarcity: When reminders of resource scarcity How exposure to limited-quantity promotions can induce promote selfish (and generous) behavior.” Journal of consumer aggression.” Journal of Consumer Research 43, no. 5 (2016): research 42, no. 4 (2015): 615-631. 683-706. Shah, Anuj K., Sendhil Mullainathan, and Eldar Shafir. “Some Laran, Juliano, and Anthony Salerno. “Life-history strategy, food consequences of having too little.” Science 338, no. 6107 choice, and caloric consumption.” Psychological Science24, (2012): 682-685. no. 2 (2013): 167-173. Shah, Anuj K., Eldar Shafir, and Sendhil Mullainathan. “Scarcity Leclerc, France, Bernd H. Schmitt, and Laurette Dube. “Waiting frames value.” Psychological Science 26, no. 4 (2015): 402- time and decision making: Is time like money?.” Journal of 412. Consumer Research 22, no. 1 (1995): 110-119. Sharma, Eesha, and Adam L. Alter. “Financial deprivation prompts Macdonnell, Rhiannon, and Katherine White. “How construals consumers to seek scarce goods.” Journal of Consumer of money versus time impact consumer charitable Research 39, no. 3 (2012): 545-560. giving.” Journal of Consumer Research 42, no. 4 (2015): 551- Tully, Stephanie M., Hal E. Hershfield, and Tom Meyvis. 563. “Seeking lasting enjoyment with limited money: Financial Mehta, Ravi, and Meng Zhu. “Creating when you have less: The constraints increase preference for material goods over impact of resource scarcity on product use creativity.” Journal experiences.” Journal of Consumer Research 42, no. 1 (2015): of Consumer Research 42, no. 5 (2015): 767-782. 59-75. Mogilner, Cassie, and Jennifer Aaker. ““The time vs. money Zhu, Meng, and Rebecca K. Ratner. “Scarcity polarizes effect”: Shifting product attitudes and decisions through preferences: the impact on choice among multiple items in personal connection.” Journal of Consumer Research 36, no. 2 a product class.” Journal of Marketing Research 52, no. 1 (2009): 277-291. (2015): 13-26. Reed, Americus, Karl Aquino, and Eric Levy. “Moral identity and judgments of charitable behaviors.” Journal of Marketing71, no. 1 (2007): 178-193. The Subjective Experience of Goal Failure: How Choosing the Lesser Evil Eradicates the Negative Consequences of Goal Failure Kamila Sobol, Concordia University, Canada

EXTENDED ABSTRACT ilarly as those who forwent a pastry, subsequently selected a more Imagine that you’re attending a morning meeting and are of- healthy dessert than those who initially selected the unhealthy snack. fered an assortment of pastries: muffins, donuts, and brownies. While Study 2 deciding whether to indulge in one of these tempting options you The main objective of Study 2 was to provide mediational evi- remember your fitness goal, and therefore opt for the muffin since dence for the proposed model. After the goal prime, participants read you perceive it as the healthiest of the three options – i.e. lesser evil. a scenario in which they were asked to choose a snack at the movie How will this choice affect your subsequent food-related decisions theater: popcorn (lesser evil), M&Ms, or no snack. After their snack throughout the day? Will your initial indulgence encourage or dis- selection, participants were asked to report their perceptions of goal courage subsequent healthy eating? (fitness) digression. Finally, participants were asked to select a bev- The goal literature suggests that once we violate an activated erage to go with their snack: water or soda (using sliding scale: -50 goal, the resulting feelings of guilt and despair generally impel fur- to 50). A mediational analysis indicated a significant partial media- ther transgressions (McGonigal 2011; Soman and Cheema 2004). tion supporting our premise that choosing the lesser evil attenuates Based on these findings, we would predict that the individual de- perceptions of goal digression, preventing goal disengagement, as picted in the introductory scenario who violated her fitness goal by manifested by a higher likelihood of choosing water over soda. eating a muffin is likely to temporary disengage from the focal fitness goal, and continue to indulge throughout the day. Study 3 In the present paper, I suggest that violating a goal does not In study 2, it is plausible that participants perceived popcorn as always demotivate behavior. Focusing on the decision context that an authentically healthy snack, and therefore the results are a simple offers exclusively vice alternatives (i.e. options that violate a goal), replication of past research demonstrating that choosing the virtuous I suggest that the subjective experience of goal failure is dampened (versus vice) option improves self-regulation (Knight and Boland when choosing the lesser evil option – i.e. option that is least detri- 1989). In study 3, I show that choosing popcorn evokes the typi- mental to the foal goal, by attenuating perceptions of goal digression. cal demotivating effects of goal violation when it stands alone. This In this sense, even though a goal violation transpired, selecting the demotivating effect disappears when the popcorn is presented along- lesser evil can actually motivate subsequent goal-congruent choices, side a greater evil. A 2 (choice: lesser evil; greater evil) x 2 (number similarly as if no transgression was committed. of vices presented: one; two) between subject design was used. In The present project contributes to the literature in at least three the first two conditions, participants were exposed to, and selected ways. First, while past research generally primed goal failure (i.e. one snack option: popcorn or M&Ms. In the remaining conditions, goal failure versus no failure conditions), in the present paper, I participants were told that they have a choice between the two snacks manipulate the subjective perception of goal failure, and show that and either selected popcorn (condition 3) or M&Ms (condition 4). goal failure leads to further goal violation only when subjectively Participants were subsequently asked to make a beverage selection: experienced. Second, I introduce an alternative mechanism that ex- water or soda. Findings confirmed that when only popcorn was of- plains reform after goal violation (Taylor and Wilson 2015), namely fered, subsequent self-regulation was impaired, but when both pop- perceptions of goal digression. Third, most goal research examines corn and M&Ms were offered, selecting popcorn did not induce goal consumer motivation within a single decision task and focuses on the disengagement, but rather continued to motivate healthy choices. conflicting choice between a “vice” and “virtue” (Wang and Huang Study 4 2018). In the present study, I show how choosing the lesser evil The objective of study 4 was to replicate the main effect with within a two vice context motivates subsequent sequential consumer real consumption. Participants read a scenario in which they either decisions. selected popcorn or M&Ms at the movie theater. In a seemingly un- Findings from four studies show that choosing the lesser evil related task, participants were asked to evaluate the packaging for attenuates perceptions of goal digression and as such prevents goal a fictitious chip brand. While completing the questionnaire, partici- disengagement that generally occurs after a goal transgression. This pants were given a bowl of chips and instructed to eat as many chips effect is only prevalent when the lesser evil is considered alongside a as they wanted. The consumption dependent variable was assessed greater vice, versus in isolation. by subtracting the weight of the bowl before and after the study was completed, representing the amount of chips the participants ate. STUDIES Findings showed a significant difference of consumption between Study 1 the two conditions, supporting that choosing the lesser evil enhances The objective of Study 1 was to verify whether choosing the the likelihood to resist temptation, demonstrating goal perseverance, lesser evil improves motivation compared to choosing an alternative instead of goal disengagement. that is more damaging to the end goal. After a fitness goal prime, The present research shows that consumers can reform after participants read a scenario in which they either selected the healthi- a minor goal violation since choosing the lesser evil prevents con- est pastry (lesser evil), the unhealthiest pastry or did not buy a pastry sumers from recognizing a goal digression, and therefore evades the for breakfast (control). In the scenario, they proceeded to go about negative motivational effects of typical goal failure. their day, have lunch and now had a choice of dessert. The selected dessert’s calorie count served as the dependent variable. Findings revealed that participants who initially chose the healthiest pastry, sim-

Advances in Consumer Research 796 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 797 REFERENCES Taylor, Jill and Janet Clare Wilson (2016). “Failing Time after Knight, Linda and Fred Boland (1989). “Restrained Eating: An Time: Time Perspective, Procrastination, and Cognitive Experimental Disentanglement of the Disinhibiting Variables Reappraisal in Goal Failure,” Journal of Applied Social of Perceived Calories and Food Type,” Journal of Abnormal Psychology, 46 (10), 557-564. Psychology, 98 (4), 412-420. Wang, Chen, and Yanliu Huang (2018). “I Want to Know the McGonigal, Kelly (2011). The Willpower Instinct: How Self- Answer! Give me Fish ‘n’ Chips!”: The Impact of Curiosity Control Works, Why it Matters, and What you can do to get on Indulgent Choice,” Journal of Consumer Research, 44 (5), More of it. Penguin Group: New York. 1052-1067. Soman, Dilip, and Amar Cheema (2004). “When Goals are Counterproductive: The Effects of Violation of a Behavioral Goal on Subsequent Performance,” Journal of Consumer Research, 31 (June), 52-62. Is All That Glitters Gold? The Effect of Product Surface Glossiness on Consumer Judgments Jiaqi Song, Hong Kong Polytechnic University, China Yuwei Jiang, Hong Kong Polytechnic University, China Gerald J. Gorn, Hong Kong Polytechnic University, China

EXTENDED ABSTRACT new version). The new version of product in each pair either always Glossiness describes how much the surface of an object reflects looks glossier (the consistent-association condition) or looks more light in a specular (i.e. mirror-like) way (Nicodemus et al. 1977). matted (the inconsistent-association condition) than the old version. Notwithstanding the importance of glossiness in product design, Participants then completed a product evaluation task in which a compared with other visual marketing elements such as shape, color, glossy or a matted power bank (i.e. portable battery for mobile de- symmetry and boundary, there is little research on glossiness in the vices) was evaluated. As expected, preferences for the glossy over marketing realm, except the work by Meert, Pandelaere, and Patrick the matted power bank were significant and mediated by perceived (2014), in which they explored the impact of glossiness originating newness only when the primed association was consistent with the from its biological association with water resources. In an attempt default. to fill this void, this research proposes that consumers have a gen- Study 4 investigated another moderator, i.e. feelings of nostal- eral preference for products with a glossy, compared to a matted gia. Nostalgia refers to a situation in which people reminisce about appearance, an effect which is mediated by psychological newness, a positive life episode in the past (e.g., Sedikides, Wildschut, and and moderated by priming of the inconsistent association, feelings of Baden 2004). Consumers are likely to prefer older products to new nostalgia, and durability concerns. ones when nostalgic feelings are induced. Thus, we predict that the Glossiness can be mentally associated with newness due to effect of glossiness on product attitude will be weakened among nos- embodied cognition. Embodied cognition arises when a conceptual talgic consumers. Employing a 2 (glossy vs. matted) × 3 (nostalgia metaphor is built up due to the repeated association between a con- vs. control vs. baseline) between-subjects design, 397 undergraduate crete sensory feeling (e.g., visual glossiness in this research) and a students listened to and rated three pretested old songs (the nostalgia more abstract cognition (e.g., psychological newness; e.g., Lakoff condition), recent songs (the control condition), or skipped this task and Johnson 1999). In particular, we argue that human beings may (the baseline condition). Afterwards, participants evaluate a glossy have learned that glossy things are usually newer. For example, au- or matted chair. Consistent with our expectation, participants in the tomobiles, jewelry, and tableware can be very glossy when they are control and baseline conditions judged the glossy chair more favor- newly manufactured, but will become dull after usage and abrasion. ably, whereas the matted chair was rated more favorably in the nos- Consequently, the mental association between glossiness and new- talgia condition. ness is established (associative network model; e.g., Wickelgren The last study adopted a 2 (glossy vs. matted) × 2 (durability 1981). Meanwhile, as a pervasive logical fallacy in philosophy, argu- concern vs. baseline) between-subjects design on mTurk (N = 281) mentum ad novitatem (appeal to novelty; Bennett 2012) argues that to explore the moderating effect of durability concerns. Durability individuals may pursue new things because of a belief that newness is one of the major quality dimensions (Garvin 1987). Glossy prod- implies improvements, even if it is unnecessarily always the case (Jie ucts could be seen as less durable than their matted counterparts, as and Li 2017). Hence, we suggest that the perceived newness induced glossiness induces the cross-modal perception of lighter weight and by glossiness should result in more favorable product evaluations. thinness (Cloonan and Decré 2015), which we reasoned as conceptu- In five experiments, we found support for this proposition and ally related to brittleness. In this study, participants in the durability evidence for three moderating factors. In Study 1, university students concern condition first imagined that their previous mug was broken (N = 176) were randomly given either a glossy version or a matted after just one week, whereas participants in the baseline condition version of the same computer mouse, and were instructed to bid for were not given this information. Then all participants were given a it, where the winner could purchase the mouse at his/her indicated set of glossy or matted mugs to evaluate. The results confirmed that price. Following the bid task, chronic newness-seeking tendency was participants’ attitudes were more favorable toward the glossy (vs. measured. The results revealed that the participants bid higher prices matte) mugs in the baseline condition, but this effect was dismissed for the glossy, rather than the matted mouse, and this effect was more when durability concerns were elicited. salient among those who has a higher chronic newness-seeking in- Taken together, the five experiments in the current research clination. supported that consumers generally judge glossy products more Study 2 tested the mediating role played by psychological new- favorably than matted ones, an effect promoted by a perception of ness. One hundred mTurk participants were randomly given the im- newness and moderated by priming of the inconsistent association, age of a glossy or a matted set of mugs, and reported their evaluations nostalgic feelings, and durability concerns. The research contributes and perceived newness of the mugs. Consistent with predictions, new knowledge to visual marketing, and to an attribute that thus far glossiness (vs. matte) induced increased perceptions of newness and has hardly attracted any research, i.e. visual glossiness. It provides subsequently resulted in more favorable product judgments. insight into not only why consumers generally prefer glossy prod- In spite of the prevailing glossiness-newness mental associa- ucts, but also the circumstances under which they are likely not to tion, priming of the inconsistent association (i.e. matte-newness) have this preference. In doing so, it provides managerial guidance should at least temporarily weaken the positive link between product for designing products. glossiness and evaluations. To verify this, in Study 3, mTurk workers (N = 309) were randomly assigned to one of the 2 (glossy vs. matted) REFERENCES × 2 (consistent vs. inconsistent association) between-subjects condi- Bennett, Bo (2012), Logically Fallacious: The Ultimate Collection tions. Participants first performed a memory task with eight pairs of of Over 300 Logical Fallacies (Academic Edition), Boston: products (one labeled as the old version and the other labeled as the eBookIt.com Press.

Cloonan, Caroline and Gwenaelle B. Decré (2015), “The Cross- Meert, Katrien, Mario Pandelaere, and Vanessa M. Patrick (2014), Modal Effects of Packaging Glossiness on Haptic Perception,” “Taking a Shine to It: How the Preference for Glossy Stems in Asia Pacific Advances in Consumer Research Volume from an Innate Need for Water,” Journal of Consumer 11, eds. Echo Wen Wan and Meng Zhang, Duluth, MN: Psychology, 24 (April), 195–206. Association for Con sumer Research, Pages: 124–25. Nicodemus, Fred E., Joseph C. Richmond, Jack J. Hsis, Warren Garvin, David A. (1984), “Product Quality: An Important Strategic I. Ginsberg, and Peter T. Limperis (1977), Geometrical Weapon,” Business Horizons, 27 (June), 40–43. Considerations and Nomenclature for Reflectance, Jie, Joseph, Y., and Ye Li (2017), “Newness for Newness’s Sake: Washington: National Bureau of Standards Press. Consumers’ Bias toward Newer Options,” University of Sedikides, Constantine, Tim Wildschut, and Denise Baden (2004), California, Riverside, working paper. “Nostalgia: Conceptual Issues and Existential Functions,” in Lakoff, George, and Mark Johnson (1999), Philosophy in the Flesh: Handbook of Experimental Existential Psychology, ed. Jeff The Embodied Mind and Its Challenge to Western Thought, Greenberg, Sander L. Koole, and Tom Pyszczynski, New New York: Basic Books Press. York: Guilford, 200–14. Wickelgren, Wayne A. (1981), “Human Learning and Memory,” Annual Review of Psychology, 32 (February), 21–52. Decisional Conflict Predicts Myopia Paul Edgar Stillman, Ohio State University, USA Melissa Ferguson, Cornell University, USA

EXTENDED ABSTRACT vestigate this, for each participant and for each trial, we used their Despite the integral role conflict plays in the subjective experi- discount rate to calculate their subjective value for both the larger- ence of self-control decisions, most work investigating conflict does later and smaller sooner options. We then calculated the distance be- not measure it directly. In the present paper, we use mouse-tracking tween the two subjective values and predicted trial-by-trial conflict (Freeman & Ambady, 2009, 2010; Wojnowicz, Ferguson, Dale, & from this difference, predicting that trials for which the subjective Spivey, 2009) – an emerging technique in which researchers mea- values are more similar should produce more conflict than trials for sure participants’ computer-mouse movements while making a deci- which the subjective values are further apart. Consistent with this, sion – to demonstrate the utility of a real-time measure of decisional across all four studies, the further apart the two subjective values conflict for studying self-control. Specifically, we investigated the were, the less conflicted participants’ mouse movements were (ps directness of participants’ mouse movements while they made in- < .001). tertemporal choices (e.g., $5 today vs. $10 tomorrow), with greater conflict inferred for decisions in which participants’ trajectories were The Predictive Power of Conflict less direct (i.e., greater veering towards the unchosen option). Past We next wanted to demonstrate that conflict within decisions was work suggests that these deflections are a sensitive measure of re- predictive of myopia more generally. Specifically, we hypothesized sponse conflict (for a recent review, see Freeman, in press). that curvature while resolving self-control conflicts should predict Across four studies, we first test how the relative subjective value the participants’ discount rates. To test this, for each participant, we of both options influences conflict as measured by mouse-tracking, divided their trials based on whether they selected the larger-later or predicting that the more similar the subjective values, the greater the the smaller-sooner option. We then calculated the average curvature conflict and therefore the less direct mouse trajectories will be. We then within these two groups, yielding two metrics per participant -- aver- test whether (and how strongly) directness of these decisions predicts age conflict when choosing the larger-later option, and average conflict participants’ myopia, as quantified by their hyperbolic discount rate. when choosing the smaller-sooner option -- and used this to predict individuals’ discount rate, k (log transformed, Kirby, Petry, & Bickel, Methods 1999). Consistent with our hypotheses, across all studies we find a strong relationship between both measures of conflict -- conflict when Procedure choosing larger-later and conflict when choosing smaller-sooner -- and Participants completed 180-210 (depending on the study) inter- log discount rates. The greater conflict participants displayed when temporal choice trials that pitted a smaller amount of money avail- ultimately choosing the larger-later (i.e., delaying gratification), the able immediately (“smaller-sooner”) against a larger amount avail- higher their discount rates, and the greater conflict participants dis- able at some delay (“larger-later”). Participants completed trials in played when ultimately choosing the smaller-sooner (i.e., indulging), blocks of 15, in which the delay of the larger-later was kept constant. the lower their discount rates (ps < .001). Further, across studies we At the start of each block, we told participants how long the delay explain a large amount of the variance in discount rates with our two would be for that block. metrics of conflict, R2 between .49 and .70. Each trial began with a screen that was blank except for a black box in the bottom center labeled “Start”. Once participants clicked Discussion on the start box, their mouse cursor disappeared, and the delay in- Although conflict plays an integral role in both the subjective formation appeared in the top-left and top-right areas of the screen, experience and theoretical understanding of self-control, conflict it- followed by the magnitude information (e.g., $25 and $50). Partici- self is rarely studied using real-time approaches. In the present pa- pants’ mouse-cursors then reappeared in the bottom-center of the per, we used mouse-tracking to more directly tap into conflict while screen, and they made their selection using the mouse. participants decided between immediate gratification and delayed rewards in an intertemporal choice paradigm. Across four studies, Results we show that the difference in the subjective values of the smaller- Metrics of conflict and myopia sooner and larger-later options strongly predicts conflict as indexed To gauge conflict on a given decision, we analyzed the area by the degree to which cursor movements deviated from a direct path under the curve (AUC). This metric quantifies the area between an towards the chosen option. These findings suggest that mouse-track- ideal trajectory (i.e., straight toward the chosen option) and the par- ing is a sensitive measure of the conflict present in a given decision. ticipants’ actual trajectory. To gauge myopia, we estimated partici- We further find that average conflict -- both when electing the pants’ hyperbolic discount rate (k) and inverse temperature param- larger-later reward over the smaller-sooner, and vice versa -- strongly eter (beta). predicts participants’ myopia as indexed by their discount rate. Nota- bly, our metrics of conflict are relatively uninformed -- they quantify The relationship between conflict and trial difficulty only the average amount of conflict on decisions in which partici- We first wanted to demonstrate that mouse trajectory- direct pants ultimately chose the larger-later or the smaller-sooner, and do ness was influenced by the response conflict present during a given not account for information regarding the magnitudes or delay for a decision (see also Dshemuchadse, Scherbaum, & Goschke, 2013; given trial. Even with this relatively blunt measure of conflict, we are O’Hora, Carey, Kervick, Crowley, & Dabrowski, 2016). Specifically, able to account for a large proportion of the variance in discount rate, we predicted that the differences in subject values on a given trial with R2 ranging from .49 to .70. Together, these results demonstrate would predict greater conflict as measured by the degree to which the predictive strength of direct measures of conflict for studying mouse movements were attracted to the unchosen alternative. To in- self-control.

Advances in Consumer Research 800 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 801 REFERENCES Kirby, K. N., Petry, N. M., & Bickel, W. K. (1999). Heroin addicts Ahn, W.-Y., Haines, N., & Zhang, L. (2017). Revealing have higher discount rates for delayed rewards than non-drug- neurocomputational mechanisms of reinforcement using controls. Journal of Experimental Psychology. General, learning and decision-making with the hbayesdm package. 128(1), 78–87. Computational Psychiatry, 1, 24–57. Loewenstein, G., & Prelec, D. (1992). Anomalies in intertemporal Ainslie, G. (1975). Specious reward: A behavioral theory of choice: Evidence and an interpretation. The Quarterly Journal impulsiveness and impulse control. Psychological Bulletin, of Economics, 107(2), 573–597. 82(4), 463–496. McKinstry, C., Dale, R., & Spivey, M. J. (2008). Action dynamics Dshemuchadse, M., Scherbaum, S., & Goschke, T. (2013). How reveal parallel competition in decision making. Psychological decisions emerge: action dynamics in intertemporal decision Science, 19(1), 22–24. making. Journal of Experimental Psychology: General, Mischel, W. (2014). The Marshmallow Test: Mastering Self- 142(1), 93–100. Control. New York: Little, Brown and Company. Emmons, R. A. (1986). Personal strivings: An approach to Muraven, M., & Baumeister, R. F. (2000). Self-regulation and personality and subjective well-being. Journal of Personality depletion of limited resources: Does self-control resemble a and Social Psychology, 51(5), 1058–1068. muscle? Psychological Bulletin, 126(2), 247–259. Freeman, J. B. (in press). Doing psychological science by hand. O’Hora, D., Carey, R., Kervick, A., Crowley, D., & Dabrowski, Current Directions in Psychological Science. M. (2016). Decisions in Motion: Decision Dynamics during Freeman, J. B., & Ambady, N. (2009). Motions of the hand expose Intertemporal Choice reflect Subjective Evaluation of Delayed the partial and parallel activation of stereotypes. Psychological Rewards. Scientific Reports, 6, 20740. Science, 20(10), 1183–1188. Stroebe, W., Mensink, W., Aarts, H., Schut, H., & Kruglanski, A. Freeman, J. B., & Ambady, N. (2010). MouseTracker: Software for W. (2008). Why dieters fail: Testing the goal conflict model studying real-time mental processing using a computer mouse- of eating. Journal of Experimental Social Psychology, 44(1), tracking method. Behavior Research Methods, 42(1), 226–241. 26–36. Inzlicht, M., Schmeichel, B. J., & Macrae, C. N. (2014). Why self- Wojnowicz, M. T., Ferguson, M. J., Dale, R., & Spivey, M. control seems (but may not be) limited. Trends in Cognitive J. (2009). The self-organization of explicit attitudes. Sciences, 18(3), 127–133. Psychological Science, 20(11), 1428–1435. Love Is Blind: How Sensory Liking Impacts Perceptions of Unbranded Products Jennifer L. Stoner, University of North Dakota, USA Maria Rodas, University of Minnesota, USA

EXTENDED ABSTRACT participants. Participants tasted either regular strawberry yogurt or Brand elements have been shown to impact perceptions of strawberry yogurt that had been sweetened. They responded to four things even unrelated to the brand or product such as the consumer’s items regarding their mindfulness during consumption on a 7-point environment (Rahinel and Nelson 2016) and group identity (Torelli scale. Yogurt loving participants in the brand present condition that et al. 2017). Even though brands have been demonstrated to impact consumed the novel-tasting, sweetened product rated themselves as perception, companies regularly use “blind” testing in the product more mindful than those that consumed the normal strawberry yo- development process. We suggest that this may be a mistake as it gurt (M = 78.64 vs. 58.53, t(122) = 3.07, p <.01). There was no dif- may lead to a different sensory experience in the testing process than ference in mindfulness for yogurt lovers who had not been presented would be experienced after purchase, when a brand is present. with a brand (M = 6.04 vs. 5.88, t(136) = 0.60, ns). While individuals may sense the same thing, their perceptions Study 4 combines the findings of the first three studies but re- of these sensations may differ drastically. Individual perceptions of a searching both strength perceptions and mindfulness in the same sensory experience (i.e. how sweet, salty, strongly scented/flavored study. Strong or weak coffee was used and new 7-point rating scales a stimulus is) are malleable, in that they are often influenced by en- of both strength and mindfulness. Once again coffee lovers rater the vironmental factors. Judgments about the taste or smell of a prod- strong and weak coffee as equal in strength when no brand was pres- uct have been shown to be affected by things such as ambient light- ent (M = 3.61 vs. 3.33, t(156) = 0.84, ns) but different in strength ing (Oberfeld et al. 2009) and package design (Becker et al. 2010). when they had read a description of a brand prior to consumption (M Given the prevalence of brand imagery and the malleability of sen- = 4.11 vs. 3.10, t(156) = 3.87, p <.001). Furthermore, the coffee lov- sory perceptions, we hypothesize that the presence of a brand dur- ers rated themselves as more mindful of the consumption experience ing consumption can impact perceptions of the sensory experience. when a brand description had been presented versus not (M = 4.70 Research on cultural symbolism supports this idea by demonstrating vs. 3.80, t(160) = 3.67, p <.001). that taste evaluations can be impacted by the values a product em- Our findings demonstrate the importance of understanding the bodies (Allen, Gupta, and Monnier2008) and other product informa- interaction of sensory love and brands on judgments of sensory per- tion (Lahne, Trubek, and Pelchat 2013). ceptions. By showing that individual who love a certain scent or taste There are groups of consumers who love a certain sensory ex- may have difficulty in judging different sensory stimuli – in our case, perience (e.g. chocoholics). It seems likely that those who love a the strength of the scent or taste. However, the presence of a novel sensory experience will have very different perceptions of that ex- brand can make these consumers more mindful of the consumption perience than those who do not. In our case, we are limiting this experience and thus better able to judge their sensory perceptions. sensory “love” to merely a strong positive attitude valence which has been considered one dimension of the brand love construct (Rausch- REFERENCES nabel et al. 2015). We therefore hypothesize that the presence (ver- Allen, Michael W., Richa Gupta, and Arnaud Monnier (2008), “The sus absence) of a brand will differentially impact those who love the Interactive Effect of Cultural Symbols and Human Values specific sensory experience (i.e. chocolate lovers) versus those that on Taste Evaluation,” Journal of Consumer Research, 35(2), do not. We hypothesize that those who love a sensory experience 294-308. will be reveling in that experience when they consume an unbranded Becker, Liza, Thomas J.L. van Rompay, Hendrik N.J. Schifferstein, product and be less able to judge perceptions of that experience. Con- and Mirjam Galetzka (2011), “Tough Package, Strong Taste: versely, when an unfamiliar brand is present consumers who love the The Influence of Packaging Design on Taste Impressions and particular scent or flavor being presented will be more mindful to the Product Evaluations,” Food Quality and Preference, 22(1), consumption experience in order to integrate the new product infor- 17-23. mation with their existing knowledge and experiences. Lahne, Jacob, Amy B. Trubek, and Marcia L. Pelchat (2014), In Study 1, using strong or weak scented peppermint shampoo “Consumer Sensory Perception of Cheese Depends on and a fictitious brand, we demonstrated that high liking participants Context: A Study Using Comment Analysis and Linear Mixed (+1 SD above the mean on a generally liking scale) rated both sham- Models,” Food Quality and Preference, 32, 184-197. poos as equal in strength on a 100-point scale when there was no Oberfeld, Daniel, Heiko Hecht, Ulrich Allendorf, and Florian brand present (M = 81.41 vs. 80.40, t(126) = .25, ns). Conversely, Wickelmaier (2009), “Ambient Lighting Modifies the Flavor high liking participants who read a fictitious brand description prior of Wine,” Journal of Sensory Studies, 24(6), 797-832. to smelling the shampoo accurately judged the stronger scented sham- Rahinel, Ryan, and Noelle M. Nelson (2016), “When Brand poo as being stronger (M = 89.44 vs. 80.77, t(126) = 2.18, p < .05). Logos Describe the Environment: Design Instability and the Study 2 replicates this effect using the sense of taste and mild Utility of Safety-Oriented Products,” Journal of Consumer or sharp cheddar cheese as the stimuli. Once again there was not Research, 43(3), 478-496. difference between judgments of strong or weak for people who Rauschnabel, Philipp, Aaron Ahuvia, Björn Ivens, and Alexander loved cheese when no brand had been presented (M = 74.15 vs. Leischnig (2015), “The Personality of Brand Lovers” Consumer 69.17, t(122) = .81, ns); but cheese lovers who had read a fictitious Brand Relationships, Palgrave Macmillan, London, 108-122. brand descriptions accurately rated the sharp cheddar cheese as being Torelli, Carlos J., Rohini Ahluwalia, Shirley Y.Y. Cheng, Nicholas J. stronger tasting (M = 78.64 vs. 58.53, t(122) = 3.07, p <.01). Olson, and Jennifer L. Stoner (2017), “Redefining Home: How Study 3 attempted to find initial evidence for the role of mind- Cultural Distinctiveness Affects the Malleability of In-Group fulness as the mechanism behind the effect. Additionally, it used Boundaries and Brand Preferences,” Journal of Consumer a real brand, Yoplait, with a product that was described as new to Research, 44(1), 44-61.

Advances in Consumer Research 802 Volume 46, ©2018 Attentional Breadth Affects In-Store Exploration and Unplanned Purchasing Mathias Streicher, University of Innsbruck, Austria Zachary Estes, Bocconi University, Italy Oliver Büttner, University of Duisburg-Essen, Germany

EXTENDED ABSTRACT not significantly differ for planned purchases in all studies reported Maximizing shoppers’ in-store travel has become an impor- hereafter. Nevertheless, attentional breadth affected unplanned cat- tant strategy to trigger unplanned purchasing (Gilbride, Inman, and egory purchases. The broad attention group purchased more cat- Stilley 2015; Hui et al. 2013) because longer in-store travel exposes egories unplanned compared to both the narrow the narrow atten- shoppers’ to relatively more products which may remind about for- tion group and the control group. The narrow attention and control gotten needs or may even evoke new ones (Inman, Winer and Ferraro groups did not differ. The same pattern of significance emerged for 2009). No research has, however, addressed the role of attentional unplanned products and unplanned spending as dependent variable. breadth in shaping store exploration and unplanned purchasing. Study 2 used the same recording procedures as in Study 1 ex- Attentional breadth refers to the focus that is directed to a cept that we placed a pedometer on participants after the attentional wider or more limited area in processing visual scenes (Friedman manipulation to record their travel distance and shopping time (N et al. 2003). It is a malleable state and may become narrowed or = 74). Activating broad attention significantly increased unplanned broadened by prior visual processing. Relative to narrow attention, categories and the pattern of significance was identical for unplanned consumers with broad visual attention explore a larger proportion products and unplanned spendings, respectively. Participants of the visual field, which increases their susceptibility to environ- shopped longer with broad attention than with narrow attention and mental stimuli (Wadlinger and Isaacowitz 2006). In retail setting, also travelled farther with broad attention than with narrow atten- this should increase the likelihood to perceive objects of interest in tion. A mediation analyses with attentional breadth as independent the background, such as products in other areas of the store or signs variable, distance travelled as mediator, and unplanned categories as and promotions. Because objects in background are spatially more dependent variable revealed a significant indirect effect. remote than foreground objects approaching relatively more back- Study 3 used the same recording procedures in a 3000 m2 su- ground products should increase the average travel distance per trip. permarket (N = 100). Participants additionally reported their chronic And because longer travel distance further increases the likelihood buying impulsiveness (Rook and Fisher, 1995). Activating broad to see a tempting product or one that reminds a consumer about for- attention significantly increased unplanned categories, unplanned gotten needs (Gilbride, Inman, and Stilley 2015; Hui et al. 2013), products, and unplanned spending. Chronic buying impulsiveness we hypothesized that activating broad (vs. narrow) attention should moderated the effect of attentional focus on these variables. Specifi- cally, the attentional manipulation significantly affected those who increase unplanned purchasing (H1) by increasing the distance trav- scored high in chronic buying impulsiveness than those who scored eled in-store (H2). We further predicted that the effect of attentional breadth on unplanned purchasing would be accentuated among more low in chronic buying impulsiveness. In sum, activating broad (vs. narrow) attention increased un- impulsive buyers (H3) because prior research has shown that impulsive (vs. non-impulsive) shoppers are more susceptible to non-focal planned purchasing and increased in-store travel of shoppers (H1 products (Büttner, Florack et al. 2014). and H2). This effect was more pronounced for chronically impulsive

In all studies except Study 2, attentional breadth was manipu- shoppers (H3). Theoretically, our studies complement prior theoriz- lated by showing participants 20 object pairs on a computer display, ing on unplanned purchasing by showing that attentional breadth af- one pair at a time (2 sec). Critically, each pair included one object fects store exploration and unplanned purchasing. Our studies further in the center and one in a random corner (periphery) of the display. imply that retailers may stimulate unplanned purchases by influenc- Participants in the narrow group were instructed to name the object ing shoppers’ attentional breadth via digital devices such as displays appearing in the center, whereas participants in the broad group on shopping trolleys and store aisles. named all objects appearing in the periphery. All participants saw and named the same objects. In Study 2, the manipulation was pre- REFERENCES sented as in-store advertisement using 20 products of the store and Büttner, Oliver B., Arnd Florack, Helmut Leder, Matthew Paul, instead of presenting object pairs participant saw only one product at Benjamin G. Serfas, and Anna M. Schulz (2014), “Hard to a time (without naming it). ignore impulsive buyers show an attentional bias in shopping To validate our attentional manipulation, a pilot study was con- situations,” Social Psychological and Personality Science, ducted in a 400 m2 retail shop using eye-tracking glasses (N = 79). 5(3), 343-351. After the attentional manipulation, participants were guided to an Friedman, Ronald S., Ayelet Fishbach, Jens Förster, and Lioba aisle (∼15 meter) and asked to walk down the aisle as if they were Werth (2003), “Attentional priming effects on creativity,” shopping. Shoppers with broad attention spent significantly more Creativity Research Journal, 15(2-3), 277-286. time walking through the aisle than with narrow attention, they fixat- Gilbride, Timothy J., Jeffrey J. Inman, and Karen Melville Stilley ed on more products than the narrow group, and also exhibited more (2015), “The role of within-trip dynamics in unplanned versus cross-aisle saccades (e.g., visual switches between the left and right planned purchase behavior,” Journal of Marketing, 79(3), shelves) than the narrow group. Thus, our manipulation successfully 57-73. affects subsequent store exploration. Hui, Sam K., Jeffrey J. Inman, Yanliu Huang, and Jacob Suher Study 1 was conducted in a 3600 m2 supermarket. First, shop- (2013), “The effect of in-store travel distance on unplanned pers (N = 192) indicated their planned purchases on category level spending: Applications to mobile promotion strategies,” and then were released for shopping after the attentional manipula- Journal of Marketing, 77(2), 1-16. tion except those in a control group. After their shopping, the experimenters recorded their unplanned purchases. Attentional breadth did

Inman, Jeffrey J., Russell S. Winer, and Rosellina Ferraro (2009), Wadlinger, Heather A., and Derek M. Isaacowitz (2006), “Positive “The interplay among category characteristics, customer mood broadens visual attention to positive stimuli,” characteristics, and customer activities on in-store decision Motivation and Emotion 30(1), 87-99. making,” Journal of Marketing, 73 (5), 19-29. Rook, Dennis W., and Robert J. Fisher (1995), “Normative influences on impulsive buying behavior,” Journal of Consumer Research, 22(3), 305-313. How The Past Shapes The Present: The Assimilation of Enjoyment to Similar Past Experiences Anika Stuppy, Erasmus University Rotterdam, The Netherlands Bram Van den Bergh, Erasmus University Rotterdam, The Netherlands

EXTENDED ABSTRACT novice coffee drinkers lack this reference system (mean, minimum Throughout their lives, consumers sample experiential prod- and maximum), their enjoyment of coffee might be less attuned to ucts (e.g., movies, coffee). Obviously, the type and number of ex- hedonic value. Because evaluability theory and other theories on periences that consumers accumulate varies greatly. Some gather value sensitivity were mostly tested with relatively simple numeric a great variety, others only a few. Some collect the most amazing stimuli, it is unclear whether these theories’ predictions will hold up product experiences, others only experience mediocre products. A for more complex, multidimensional experiential stimuli. Indeed, coffee aficionado, for instance, has sampled elite and abysmal cof- previous work has shown that consumers use different strategies fee while an inexperienced coffee drinker might have mostly visited when evaluating the value of experiences rather than numbers (Mar- Starbucks. Past work found that consumers learn distributional in- tin et al. 2016). formation about numeric stimuli such as prices (e.g., the range, the This investigation pursued two goals. First, we aimed to dem- mean) through prior experiences, which boosts sensitivity for value onstrate that more experienced consumers are more sensitive to the (i.e., knowledge factor, general evaluability theory; Hsee and Zhang hedonic value of experiential products than less experienced con- 2010). While distributional information about past prices might af- sumers and thus, that greater knowledge might not be universally fect perceptions of how cheap/expensive a product is, it is unclear positive (e.g., the “curse of expertise” or the flipside, “blessed igno- whether these findings apply to more experiential products. Our in- rance”). Second, we strived to elucidate how accumulating experi- vestigation is the first to test whether prior experiences boost sensi- ences boosts sensitivity. If consumers draw back on past experiences tivity for the hedonic value of experiences. We also examine how to gauge present enjoyment, do they: past experiences trigger sensitivity. a) compare the new experience to the past average experience Consistent with evaluability theory, we suggest that prior expe- (Helson 1964; Hsee and Zhang 2010), riences make consumers’ enjoyment more attuned to the hedonic val- b) do they use range information (minimum, maximum past ex- ue of experiential products (figure 1; A). Their enjoyment of higher perience; Hsee and Zhang 2010), vs. lower value products differs more strongly. To illustrate, a coffee c) do they determine the rank of the new experience (Stewart aficionado should enjoy coffee of higher hedonic value more than et al. 2010), coffee of lower hedonic value while a coffee novice should be more d) or do they use a different strategy? insensitive. While evaluability theory assumes that greater knowledge would generally benefit consumers (e.g., greater knowledge Study 1: Mesuring The Distribution helps determine whether a price is favourable or not), we predict Study 1 explored how the distribution of consumers’ past ex- that having (vs. lacking) knowledge not only has hedonic benefits periences with gaming apps affects their enjoyment of two target (“blessing of expertise”), but also hedonic costs (“curse of exper- games . We measured the participants’ past experiences with gaming tise”). apps with a distribution builder. Then, participants rated how much they enjoyed two target games of different hedonic value (lower vs. How do Past Experiences Shape Present Enjoyment? higher). Study 1 served to explore how different features of - par Prior research on the accumulation of experiences focused on ticipants’ distribution of past experiences (mean, sum, range, and processes during the knowledge acquisition phase, such as effects standard deviation [SD]) would affect their enjoyment of the target of the serial position of stimuli during sequential evaluation (e.g., games. We counterbalanced whether the distribution builder was primacy/recency or hedonic contrast effects; Ghoshal et al. 2015, completed before or after evaluating the target games to alleviate Zellner et al 2002). Our research question differs substantially concerns that we artificially activated past experiences. from these studies by examining what happens after, not during, the Procedure knowledge acquisition phase. By comparing more experienced with Participants (n=156) either reported their past experiences with less experienced consumers, our inquiry extends previous work by gaming apps before or after we measured their enjoyment of two examining whether and how the distributional information of past target games. The distribution of past experiences with gaming apps experiences affects enjoyment, after knowledge is acquired. was measured with a distribution builder. Participants allocated balls The distributions of experienced (figure 1; B) and inexperienced (1 ball = 1 gaming app) across ten hedonic value categories (1 = (figure 1; C) consumers differ in many ways. Given that distributions horrible to 10 = world class) to indicate how many gaming apps entail multiple features (e.g., sum, min, max, mean, rank, etc.), the they had experienced in each hedonic value category. They could question begs: Which aspect of consumers’ distribution of prior ex- allocate between 0 and 100 balls in total. We computed the range periences will affect present enjoyment? (Mrange=5.87), the standard deviation (Mdeviation=1.78), the sum Evaluability theory suggests that consumers become sensitive (Msum=12.65) and the average valence of all past experiences (Mav- to hedonic value because their past experiences provide information erage=6.21). about the range and average of hedonic value in the marketplace. As We measured participant’s enjoyment of two new gaming apps. such, coffee aficionados may judge whether a new coffee is more or Participants watched two videos that introduced two gaming apps less enjoyable than their average coffee experience (Helson 1964; in random order. These apps were selected to be as similar as pos- Hsee and Zhang 2010). Or, they might gauge how much better or sible and involved navigating a green monster through a laboratory. worse the new coffee is compared to their all-time best or worst cof- However, a pretest confirmed that their hedonic value differed con- fee experience (Hsee and Zhang 2010). Or they might compare it to siderably (84% vs. 24% enjoyment rating). Participants’ enjoyment the experience they typically have (i.e., the mode). Because more

Advances in Consumer Research 805 Volume 46, ©2018 806 / How The Past Shapes The Present: The Assimilation of Enjoyment to Similar Past Experiences of each game was measured on 10-point scales (1=horrible game to drawings. In the irrelevant range condition, participants saw 15 ani- 10=world class game). mal photographs. Participants rated their enjoyment of each stimulus on slider scales (1=do not enjoy at all to 100=enjoy extremely). Pre- Results sentation order of stimuli was randomized. We computed several separate mixed linear models where we Next, all participants completed six dot-estimation filler tasks. regressed enjoyment on each distribution statistic (range, SD, sum, Thereafter, participants were randomly assigned to one of two novel, mean), game hedonic value (higher vs. lower), order (distribution but similar drawings: one of lower or one of higher hedonic value. builder first vs. games first), and all interactions. Game hedonic Those constituted a gift to take home. A pretest ensured that the tar- value was the within-subjects variable. No model revealed a 3-way gets were similarly enjoyable as drawings of lower and higher he- interaction (all p>.27). However, 2- way interactions between the donic value in the experience building phase (M=60 and M=79 re- range measures (standard deviation statistic and range statistic) and spectively). Enjoyment was measured with three items: How happy game hedonic value emerged which remained significant after con- are you with this drawing, How much do you like this drawing, How trolling for the mean and sum statistics (all p<.08). As an example, beautiful is this drawing? (1=not at all to 9=extremely). we dissected the interaction for the standard deviation measure. Par- ticipants’ enjoyment was more sensitive (the difference in enjoyment Results of the good and bad game was larger) when they had consumed a We regressed enjoyment on gift hedonic value (lower vs. high- broad (+1SD; β=1.61, p<.001) rather than narrow range of gaming er), range (coded to compare the irrelevant range [animal pictures] to apps in the past (-1SD; β=.430, p=.231). The two-way interactions be- the other two conditions), and all interactions. Two significant 2-way tween game hedonic valence and sum or average were insignificant interactions between gift hedonic value and relevant range – higher (all p>.243). quantity (vs. irrelevant range) and relevant range – lower quantity Study 1 explored which feature of the distribution of past ex- (vs. irrelevant range) emerged (all p<.001). Participants’ enjoyment periences would predict sensitivity to hedonic value. We find that was more sensitive to hedonic value when they had experienced all sensitivity was only predicted by distribution features that captured 15 drawings (p<.001) as compared to animal pictures (p =.18). Par- the range (range and standard deviation) but not distribution features ticipants’ enjoyment was also more valence sensitive when they had not directly related to range (sum and average). As such, the results exclusively seen extreme drawings (p<.001) as compared to animal raise doubts as to whether consumers compare new experiences with pictures. Sensitivity did not differ between the higher and lower the average experience and whether the quantity of past experiences quantity conditions (p=.99). is important. Next, we tested whether present enjoyment was assimilated to similar or contrasted away from dissimilar past experiences. We Study 2: Reducing The Number of Experiences computed each participant’s average enjoyment of the three less Study 2 aimed to conceptually replicate study 1 by manipulat- (M=42.80) and the three more positively valued (M=85.48) drawing, instead of measuring, the range of past experiences in a “knowl- ings in the KAP. Next, we examined the correlations between these edge acquisition phase” (KAP). We also strived to provide process averages and gift enjoyment. The correlation pattern suggest assimi- evidence. lation to similar past experiences. We find no evidence for hedonic Study 1 showed that the range, but not the sum or average, pre- contrast. dicted sensitivity for hedonic value. In study 2, we thus eliminated Consistent with study 1, not the quantity but the range of past moderate experiences from the participants’ distribution of past ex- experiences increased sensitivity for the hedonic value of a target periences. Thereby reducing the number, while keeping the range of drawing. Participants with 6 past experiences were as sensitive as experiences constant. those with 15 past experiences. This finding raises doubts as to To test whether, and how participants would compare a target whether the sum of experiences or the average experience are rel- experience to past experiences, we measured enjoyment in the KAP. evant for enjoyment. We then examined the correlations between earlier enjoyment in the Extending study 1, we demonstrated which past experiences are KAP and enjoyment of the target experience. Given that we provid- driving the effects on enjoyment through measurement. Participants ed a range of experiences, two comparison processed are possible: assimilated present enjoyment to similar drawings they had seen in Consumers might assimilate to similar past experiences (hedonic the past. Our findings are inconsistent with the possibility that sensi- assimilation: “This is as enjoyable as this other past experience”) tivity was driven by hedonic contrast. or contrast away from dissimilar past experiences – those at the opposite end of the range (hedonic contrast: “This is much more (less) Study 3: Manipulating The Range of Experiences enjoyable than this other past experience”). Study 3 aimed to test whether consumers rank experiences by manipulating the range of experiences in the KAP. We compared Procedure participants with a narrower range of experiences (15 moderate val- Study 2 manipulated prior experiences with drawings (relevant ue drawings) with participants with a broader range (the 15 drawings range – higher quantity, relevant range – lower quantity, irrelevant from study 2). If the rank within the distribution determines enjoy- range) and the hedonic value of a target drawing (lower vs. higher) ment (Stewart et al. 2016), then consumers in the narrower range in a 3X2 between-subjects design. First, participants (n=297) com- condition should be sensitive to hedonic value. Because the higher pleted the KAP. In the relevant range – higher quantity condition, (lower) value target ranks highest (lowest) in their distribution, they participants saw 15 mountain drawings. The drawings’ hedonic should enjoy it very much (not at all). value was normally distributed to ensure a broad range. Participants evaluated three drawings of lower (enjoyment score pretest: M=56 Procedure on a 100-point scale), nine drawings of moderate (M=68), and three Study 3 applied a 2 (range: broader vs. narrower) x 2 (hedonic of higher hedonic value (M=81). In the relevant range lower quantity value: lower vs. higher) between-subjects design (n=193). Partici- condition, participants only evaluated the three drawings of lower pants in the broader range condition saw the 15 artworks from study and the three drawings of higher hedonic value but no moderate 2 (3 high, 6 moderate and 3 low) while participants in the narrower Advances in Consumer Research (Volume 46) / 807 range condition saw 15 moderately enjoyable artworks (all similar to (study 3). A finding that might be specific for experiential stimuli. the 6 moderately enjoyable drawings from the broader range condi- When judging prices, for instance, consumers were sensitive to pric- tion). The drawings were presented in random order. To rule out an es within and outside their range of past experiences (Janiszewski explanation in terms of response scale during the KAP, participants and Lichtenstein 1999). did not rate their enjoyment in the KAP. We used study 2’s items to Although many consumers strive to accumulate more product assess enjoyment of the - randomly assigned –gift. experiences to acquire knowledge, having this knowledge involves a cost. A coffee aficionado might need to walk past four Starbucks, Results sacrificing time and energy, to reach a coffee shop that offers elite We examined the effect of range (narrower vs. more broad), coffee. gift hedonic value (lower vs. higher), and their interaction gift enjoyment. A significant 2-way interaction between hedonic value and REFERENCES range emerged (p=.016). Participants’ enjoyment was more sensitive Alba, Joseph W. and J. Wesley Hutchinson (1987), “Dimensions to hedonic value when they had experienced the broader (p<.001) as of consumer expertise,” Journal of Consumer Research, 13 compared to the narrower range (p=.09) of drawings. (March), 411-454. Study 2 demonstrated that experienced participants assimilated Ghoshal, Tanuka, Eric Yorkston, Joseph C. Nunes, and Peter to similar past experiences. The ability to draw back on similar past Boatwright (2014), “Multiple reference points in sequential experiences may therefore be what distinguishes novices from ex- hedonic evaluation: an empirical analysis,” Journal of perts and creates value sensitivity for experiential products. Study Marketing Research, 5 (October), 563-577. 3 supports this notion. When similar past experiences were absent, Helson, Harry (1964), “Current trends and issues in adaptation- sensitivity declined. These findings again raise doubts as to whether level theory,” American Psychologist, 19 (January), 26 consumers compare their experiences to an average or rank them. Hsee, Christopher K. and Jiao Zhang (2010), “General evaluability theory,” Perspectives on Psychological Science, 5 (August), GENERAL DISCUSSION 343-355. Our experimental paradigm allowed us to manipulate knowl- Janiszewski, Chris and Donald R. Lichtenstein (1999), “A range edge. Consistent with evaluability theory, we show that acquiring theory account of price perception,” Journal of Consumer product knowledge creates value sensitivity. However, we also show Research, 25 (March), 353-368. that in the domain of experiential products, knowledge has hedon- Martin, Jolie M., Martin Reimann, and Michael I. Norton (2016), ic benefits (“blessing of expertise”) but also entails hedonic costs “How desserts are like losses,” Journal of Experimental (“curse of expertise”). Knowledgeable consumers enjoy elite prod- Psychology: General, 11 (November), 1460 -1472. ucts more than novices, but are less able to enjoy mundane products. Medin, Douglas L. and Marguerite M. Schaffer (1978), “Context Extending beyond evaluability theory, we show which aspect of theory of classification learning,”Psychological Review, 85 knowledge contributes to sensitivity. Our findings suggests that the (May), 207-238. process of acquiring knowledge for experiences differs from the pro- Stewart, Neil, Nick Chater, Gordon DA. Brown (2006), “Decision cess of acquiring knowledge for numbers. Sensitivity was not driven by sampling,” Cognitive Psychology, 53 (August), 1-26. by the sum, or average experience, or the rank of an experience. Zellner, Debra A., Brett B. Kern, and Scott Parker (2002), Rather, a broad range of past experiences seems crucial because it “Protection for the good: Subcategorization reduces hedonic provides similar past experiences that serve as assimilation standards contrast,” Appetite, 38 (June), 175-180. (study 2). More importantly, our findings imply that expertise for hedonic experiences may be local. Consumers seemed sensitive to experiences within but not outside of their range of past experiences A Complete Consumer Journey: Tracking Motivation in the Marketplace Jacob Suher, Portland State University, USA Szu-chi Huang, Stanford University, USA Leonard Lee, National University of Singapore, Singapore

EXTENDED ABSTRACT In study 2, the Prolific research platform recruited 120 partici- Traditional models of shopper motivation assume that a con- pants to complete a shopping trip in a simulated online grocery store. sumer arrives at a store with a particular goal in mind and shops We randomly assigned participants to one of two pre-trip planning according to that goal. As a result, the majority of studies on in-store conditions (List vs. No-List). In the list condition, participants re- decision making have treated shopper motivation as a stable trip-lev- ceived a budget of $35 and created a list of four categories. In the el variable measured either before or after a shopping trip (Kaltcheva no-list condition, participants received the same instructions but did and Weitz 2006). In contrast, consumer behavior studies demonstrate not create a shopping list. The store included 15 categories with six that examining motivation and choice through a series of actions and items from a national US grocery retailer (Kroger) presented in a across multiple time points helps to uncover the inherent dynamics fixed order. As in study 1, the dependent variable was inter-choice of such behavior (Dhar et al. 2007; Khan and Dhar 2006). Studying time and we followed the same procedure to create a regression in-store behavior in a continuous manner from the beginning to the model. The final sample for analysis included 715 purchases made end, therefore, is imperative to understanding consumer motivation by 102 participants who remained within the specified budget. We in the marketplace. again observed an interaction between squared trip duration and the In this research, we capture one facet of consumer motivation list contrast code (β = 2.2482, Wald χ2(595) = 14.88, p = .0001). In in the marketplace—the amount of time shoppers spend between the list condition, there was a positive linear effect of trip duration product choices within a shopping trip. We refer to this measure (β = 0.7363, Wald χ2(595) = 14.99, p = .0001), while the quadratic as inter-choice time. A field study and two online shopping experi- effect was not significant (β = 0.8258, Waldχ 2(595) = 0.91, p = .34). ments contribute to the measurement and understanding of consumer In the no-list condition, there was a negative quadratic effect of trip motivation in the marketplace. First, we advance prior literature on duration (β = −3.6705, Wald χ2(595) = 14.92, p = .0001), while the shopper motivation by measuring consumers’ in-store behavior in a linear effect was not significant (β = 0.3015, Wald χ2(595) = 1.78, continuous manner. Second, we add to growing research that uses p = .1826). We replicated these patterns with another 211 Prolific shopper-tracking technology by drawing attention to an antecedent participants using the same paradigm as study 2 except the store ran- of in-store behavior—pre-trip planning—and capturing its divergent domized category order and the list condition used a pre-determined impact on consumers’ motivation over the course of their shopping shopping list (N = 1,376 purchases from 190 participants). trips. Third, our findings add to the dialogue that consumer motiva- An in-store field study and an online shopping experiment tion can follow a linear trend (Kivetz, Urminsky, and Zheng 2006) or (plus a replication study) found that shoppers’ motivational patterns a curvilinear trend (Bonezzi, Brendl, and De Angelis 2011). We show diverged depending on whether they had a shopping list. Whereas that pre-trip planning determines whether a curvilinear motivational consumers with shopping lists exhibited a deceleration of shopping pattern emerges as a consumer progresses from one product choice speed over the course of their trip, consumers without lists showed a to another in the store. nonmonotonic increase then decrease in inter-choice time (i.e., a cur- In study 1, 250 shoppers at a grocery store participated in a re- vilinear pattern). These patterns reveal the role that reference points search study that included an entrance and exit survey as well as vid- may serve for consumers in the marketplace. The moderating effect eo tracking to record the timing and categories of all product choices. of list usage on shopper behavior adds a new angle to the literature The dependent variable was the amount of time leading to each prod- demonstrating that pre-trip factors influence the dynamic patterns of uct choice (i.e., inter-choice time). We calculated inter-choice time in-store behavior. Future research can explore other consumer-based as the time a participant began shopping to making the first product antecedents (e.g., the presence of a shopping partner or group; dif- choice, and then the time from the first product choice to the second, ferent purpose of the trip) and capture their effects on consumers’ and so on, until the time between the final two product choices. Inter- motivational patterns in the store. Capturing the dynamics of in-store choice time was regressed on the cumulative trip duration at the mo- motivation is critical to providing real-time and relevant messages to ment of purchase, the square of the trip duration, a contrast code for enhance the retail experience and consumer welfare (Lee et al. 2018). usage of a shopping list (List = 1; No-List = −1), and the two-way interactions between the list code and trip duration and the list code REFERENCES and squared trip duration. The final data set consisted of 2,285 prod- Bonezzi, Andrea, C. Miguel Brendl, and Matteo De Angelis uct choices made by 237 shoppers (video files from 13 participants (2011), “Stuck in the Middle: The Psychophysics of Goal were unusable). There was a positive interaction between squared Pursuit,” Psychological Science, 22 (5), 607-12. trip duration and the list code (β = 0.0010, Wald χ2(1959) = 34.58, p Dhar, Ravi, Joel Huber, and Uzma Khan (2007), “The Shopping < .0001), suggesting that whether a shopper’s motivation followed a Momentum Effect,”Journal of Marketing Research, 44 (3), curvilinear trend depended on pre-trip planning. In the list condition, 370-78. there was a positive linear effect of trip duration (β = 0.0247, Wald Kaltcheva, Velitchka D., and Barton A. Weitz (2006), “When χ2(1959) = 37.79, p < .0001); the quadratic effect of trip duration was Should a Retailer Create an Exciting Store Environment?” not significant (β = −0.0002, Wald χ2(1959) = 0.81, p = .37). In the Journal of Marketing, 70 (1), 107-18. no-list condition, in contrast, there was a negative quadratic effect of Khan, Uzma, and Ravi Dhar (2006), “Licensing Effect in Consumer trip duration (β = −0.0023, Wald χ2(1959) = 56.02, p < .0001); the Choice,” Journal of Marketing Research, 43 (2), 259-66. linear effect of trip duration was not significant (β = 0.0053, Wald χ2(1959) = 1.24, p = .27).

Kivetz, Ran, Oleg Urminsky, and Yuhuang Zheng (2006), “The Lee, Leonard, Jeffrey Inman, Jennifer J. Argo, Tim Böttger, Utpal Goal-Gradient Hypothesis Resurrected: Purchase Acceleration, Dholakia, Timothy Gilbride, Koert van Ittersum, Barbara Illusionary Goal Progress, and Customer Retention,” Journal Kahn, Ajay Kalra, Donald R. Lehmann, Leigh M. McAlister, of Marketing Research, 43 (1), 39-58. Venkatesh Shankar, and Claire I. Tsai (2018), “From Browsing to Buying and Beyond: The Needs-Adaptive Shopper Journey Model,” Journal of the Association for Consumer Research, 3 (3), in press. Dehumanization: Coping with Embarrassment in Consumer Purchases Yixia Sun, Zhejiang University, China Xuehua Wang, East China Normal University, China Joey Hoegg, University of British Columbia, Canada Darren Dahl, University of British Columbia, Canada

EXTENDED ABSTRACT Study 1 tested whether participants would choose a doctor with Consumers often find themselves in encounters with service pro- robotic (vs. warm) characteristics when seeking an embarrassing viders that can spark feelings of embarrassment. For example, they treatment (H2). Female participants imagined doing either a pelvic may need to speak to a pharmacist or a doctor about an embarrassing exam or an ear exam in a clinic. They were told that only male doc- health condition. They may need to purchase embarrassing personal tors were available and were asked to choose one between two doc- products (Blair and Roese 2013). In situations such, consumers em- tors who were equally knowledgeable and professional but varied in ploy coping strategies to deal with the feelings of embarrassment, style. Doctor A worked in an efficient, almost robotic way, whereas such as buying additional non-embarrassing items (Blair and Roese Doctor B was chatty and asked lots of questions. As predicted, par- 2013; Lewittes and Simmons 1975), shopping in less crowded en- ticipants seeking embarrassing treatment were more likely to choose vironments, or avoiding salespeople (Brackett 2004). The current the robotic doctor than those in the control condition (59.3% vs. research considers another way that consumers cope with embarrass- 38.5%; χ2 = 6.88, p < .01). ing interactions with service professionals: dehumanization. Study 2 tested whether embarrassing service contexts lead con- Given that thoughts about what others are thinking leads to em- sumers to dehumanize service providers (H1). Participants read a barrassment (Dahl, Manchanda, and Argo 2001), a strategy to cope scenario similar to study 1 but rather than having a choice of doctor, with embarrassment is perceiving others to be lacking in thoughts or participants simply read that the doctor performed the exam and were feelings (Epley and Waytz 2010). For example, when buying con- asked to judge his style. Results confirmed that participants rated the doms, consumers may feel embarrassed if they believe that the ca- doctor as more robotic in the embarrassment condition than in the shier infers they are promiscuous (Dahl et al. 2005). We propose that neutral condition (3.84 vs. 3.10; F(1,82) = 6.36, p = .014). consumers in such situations reduce self-consciousness by thinking Study 3 was a field study conducted in a convenience store built of the service provider as less like a human being and more like a in the lab. Each participant selected, at random, an envelope that robot who is simply “doing the job”. Someone who is not fully hu- contained the name of a product and redeemed it with a coupon. In man should be more focused on the task and less likely to think about fact, each envelope identified either an embarrassing product (e.g., the consumer and the potentially embarrassing events that may have condom) or a normal product (e.g., gum) to be redeemed. Results led the consumer to need the particular product or service. Thus, by indicated that participants who bought embarrassing products per- perceiving the service provider as more robotic and less human, the ceived the cashier to be more robotic than those who bought normal embarrassment of the situation can be mitigated. In this paper, we in- products (4.40 vs. 3.39; F(1, 113) = 17.64, p < .001). vestigate this subtle form of dehumanization as a coping mechanism Study 4 tested the effect of dehumanization on self-disclosure consumers employ when buying socially embarrassing products or in a field experiment. (H3). An experimenter pretended to be a sales- seeking embarrassing services. We have four hypotheses stemming person and invited passersby to do a survey for a small payment. from our conceptualization: Participants were told that they could stop answering questions any- time they wanted, without influencing their payment. A list of thirty- Hypothesis 1 When consumers buy embarrassing products or two personal questions was developed and the number of questions consume such services, they will dehumanize the answered by each participant was recorded as the self-disclosure service provider to reduce embarrassment. measure. Participants were told that the product being surveyed was either about condoms or gum, as a manipulation of embarrassment. The salesperson was trained to act in a warm or robotic manner. In Hypothesis 2 When consumers buy embarrassing products or the warm condition, the experimenter focused on the participant’s consume such services, they will prefer a robotic eyes and spoke enthusiastically. In the robotic condition, he looked at service provider to a warm one. However, this the participant’s face without focus and spoke in a monotone voice. preference will be reversed when they buy nor- Results confirmed that when the survey was about condoms, partici- mal products or consume normal services. pants answered more questions when the salesman behaved roboti- cally than when he behaved warmly (19.47 vs. 13.54; F(1, 133) = Hypothesis 3 Consumers will be more willing to make self-dis- 18.19, p < .001). This pattern reversed when the survey was about closures about an embarrassing product when gum (22.35 vs. 28.56; F(1, 133) = 4.81, p < .05). the service provider is robotic than when he/she A final study tested the moderation effect of self-consciousness is warm. However, they will be more willing to (H4) and found that the effect of embarrassment on dehumanization make self-disclosures about a normal product of salespeople is mitigated for individuals low in chronic self-con- when the service provider is warm than when he/ sciousness. she is robotic.

Hypothesis 4 The effect of embarrassment on dehumanization (H1) will be more evident when consumers are of high (vs. low) self-consciousness.

EXTENDED ABSTRACT using tokens) were excluded since they are not relevant to testing a difference between endowment pricings. Furthermore, studies that Previous Meta Research did not involve endowment (either actual or hypothetical) were also While the discussion on the phenomenon of endowment effect excluded. Few studies, such as Peck and Shu (2009), examined the has been on-going, it is surprising that few meta-analyses have ex- effect of physical touch on perceived ownership. While the authors amined this phenomenon exclusively. Three previous meta-analyses argue that theoretically physical touch enhances one’s psychological have investigated on the disparity that exists between Willingness- ownership of the object and therefore may lead to an endowment to-Pay (WTP) and Willingness-to-Accept (WTA). Experimental and effect, the endowment here is the dependent variable and not what is survey studies often find that WTA is much greater than WTP; how- resulting in the price difference. Hence, such studies were excluded. ever, this disparity may not come exclusively from endowment of There are 42 studies included in this analysis with 89 total experi- an object but could also be attributable to seller/buyer perspective, ments. The studies included are listed in the Appendix. income effects, transaction costs (Randall and Stoll, 1980), commit- Dependent Variable: Endowment Price Ratio ment costs (Zhao and Kling, 2004), or other psychological variables Similar to Tuncel and Hammitt (2014) which used a WTA/WTP such as framing (Thaler, 1980). ratio, this analysis uses the ratio of the mean price of the Endowed to The most recent meta-analysis on the WTP/WTA gap, by Tun- mean price of Unendowed for each experiment in each study. Since cel and Hammit (2014), is based on a previous meta-analysis by there is no information on the price elicited from each participant in Horitz and McConnel (2002), which reviewed 45 studies and exam- the studies, this analysis focusses on the group means in each experi- ined the effects of type of good and experimental conditions. Horitz ment instead. Natural log is then used to transform this Endowed/ and McConnel’s main finding showed that the WTP/WTA disparity Unendowed price ratio. The resulting Ln-Price-Ratio is a semi- nor- was greater for public goods or none-market goods compared to ordi- malized distribution. No outliers were identified during this vetting nary private goods. Tuncel and Hammit expanded the previous anal- process and so no studies were excluded as a result. ysis by including new studies up until early 2012 with a total of 76 studies. Tuncel and Hammit replicate prior finding that there exists Independent Variables systematic differences in the WTP/WTA disparity by type of good. Type of Good: Ordinary Private Goods, Lottery, and Others However, they showed that this disparity is related to participants’ The type of good in each experiment was classified into three experience (real world experience and experimental trial experience) categories, ordinary private good, lottery, and others. Ordinary pri- in valuing the good. Furthermore, the magnitude of this parity has vate goods are items that are often seen and bought in everyday decreased overtime. The third meta-analysis by Sayman and Onculer transactions. Three most common used items used in endowment (2005) examined 39 studies and found that incentive-compatible de- experimental studies are pens, mugs, and chocolate bars, all of which signs (such as the goods are tangible or intangible) decreased the are ordinary private goods. Lotteries are a type of good that carries WTP/WTA disparity. a probability of winning some pre-set face value. In endowment ex- However, all of these previous three meta-analyses focused on periments, participants could either be endowed with such a lottery the WTP/WTA disparity. While Tuncel and Hammit’s (2014) analy- which they have some chance of winning a pre-set value of winnings sis does include studies that focus on the endowment effect, their (and have the option to sell such a lottery) or not have such a lottery analysis and discussion do not offer any specific insight on the actual but have the option to buy it. Others type of good in this category phenomenon of endowment. This current study is an updated and captures items that were used that were not explicitly captured by the extension of Tuncel and Hammit but with a focus on the endowment other two categories (such as time) or explained as to what was used effect. It expands the sample by including newer studies. Similar to in the experiments. Tuncel and Hammit, this meta-analysis uses the logarithm of the ratio of the elicited price from endowed and unendowed participants. This Endowment Method: Real vs. Hypothetical analysis examines several factors that may play a role in contributing In some studies, the endowment experiment takes place with ac- to the endowment effect in experimental studies, such as country of tual items given to the participants; in other studies, the experiment is origin, type of good, endowment method, price elicitation method, scenario based and participants are asked to imagine to hypothetical- usage of actual endowment, and experimental procedures. Further- ly possess (or not possess) some item. Past studies and meta-analysis more, regression analysis shows that unlike in Tuncel and Hammit’s have shown that the endowment effect is robust regardless of real case where the WTP/WTA gap has diminished by publication date, or hypothetical method of endowment (Tuncel and Hammitt, 2014). the price gap due to endowment has increased in recent publications. Physical Endowment: Physical vs. Non-physical Current Meta-Analysis In some studies, participants are given actual items to touch and use, while in other studies, participants may be told that the item they Data are endowed with will be provided to them at the end of the study. To identify related studies, PsycINFO was used with keywords While the later experimental methods still uses real endowment “WTA,” “WTP,” and “Endowment Effect” up to mid-2015. Since where the participants are given an actual item, these participants do this analysis focuses on the study of the price gap due to endowment, not have the opportunity to physically engage with the item itself. none experimental studies and experimental studies that did not in- Prior studies have suggested that while endowment effect may still volve explicit pricing of a good or object were excluded. For ex- occur regardless of physical (or lack of) engagement with an item, ample, in Sen and Block’s (2008) study, the experimenters used rat- physical contact does enhance the endowment effect (Shu and Peck, ing scales and probability to measure participants’ inclination to keep 2011; Peck and Shu, 2009). an endowed object. This and other induced-value experiments (e.g.,

Advances in Consumer Research 811 Volume 46, ©2018 812 / A Meta-Analysis on the Endowment Effect in Experiments Price Solicitation Method: Solo, Variable Range, and Other on the publication year (Md = 2011), a marginal interaction is found In solo pricing method, participants are given an open ended between publication date and price elicitation method (F = 2.44, p = question on how much they are willing to spend on or sell the item .09). This marginal effect may be attributable tothe increase in Other and are only required to provide one single price. In variable range price elicitation method since the year 2011 (Mbefore2011 = .33, method, participants are presented with several ranges of prices in a Mafter2011 = .87) since Solo and Variable Range elicitation methods sequential manner and are asked under which range they would most did not change much. likely to pay for or sell the item. The other solicitation method is a catchall for experiments that were not explicit on how pricing was Discussion elicited from the participants. This meta-analysis takes a closer examination at the effect of endowment on pricing. While it shares some similarities to prior Experimental Procedure: KKT, BDM, and Other research on WTP and WTA gap, it is theoretically different in the In their seminal paper on the endowment effect and loss aver- manipulation and the examination of the resulting pricing gap. This sion, Kahneman, Knetsch, and Thaler (1991) outlined the meth- study is unique in that it examines exclusively the experiments con- odology in which they conducted their experiments. In the KKT ducted on the endowment effect with the endowment takes place ei- procedure, participants are randomly assigned to the endowed and ther in real world situation or in hypothetical scenario. Furthermore, unendowed groups; the endowed participants are the sellers and are it adds more recent studies while focussing on the experimental stud- asked for their WTA price and the unendowed participants are the ies in the area of endowment. buyers and are asked for their WTP price. Alternatively, endowment This result of analysis replicate the findings from Tuncel and experiments could use a variation based on the Becker-DeGroot- Hammitt (2014) and Horowitz and McConnel (2002) that type of Marschak (BDM) procedure (Becker, DeGroot, Marschak, 1964). good used in experiments has an effect on the gap between prices due The BDM procedure is commonly acknowledged as more incentive- to endowment. This analysis shows that lottery endowment leads to compatible where it has higher accuracy in measuring one’s WTP. In greater price gap compared to ordinary private good endowment, a the variations of the BDM procedure, participants are asked to state caveat that was missed by prior meta-analyses. This has some inter- their best selling or buying price for the endowed or unendowed item esting implications for future research. Since lottery carries inherent with the knowledge that there is a market clearing price (either based risks, it is possible that risk aversive individuals will demand greater on a pre-set price by the experimenter or the mean of the other sellers compensation in the form of price gap between sellers and buyers to and buyers in the experiment). If the participants’ prices match the compensate for the potential risk. This is also consistent with the loss market price, they will have the opportunity to buy or sell the item. aversion account of the endowment effect where individuals demand Country of Origin: Western vs. Eastern greater value for losing (or selling) an object compared to gaining Finally, this analysis also examines the country where the ex- (or buying) the same object (Khaneman, Knetsch, and Thaler, 1991). periments were conducted in each study. Western countries are cat- The publication year also appear to play a role in influencing the egorized based on region (such as North America and Europe) and effect of endowment on pricing where the pricing gap appears to in- Language (such as English); Eastern countries are based on region crease as publication year increased. This result is contrary to Tuncel which is mainly from oriental cultured countries (such as China, Ja- and Hammitt’s (2014) where they showed a diminishing WTP/WTA pan, and Singapore). price gap as a result of increased publication year. One reason for this difference may be because Tuncel and Hammitt’s meta-analysis is on Results the WTP/WTA price gap whereas the current meta-analysis focussed The association between the independent variables and the on the endowment effect price gap. Whereas the WTP/WTA isa dependent variable (Ln of ratio between endowed and unendowed methodological price elicitation process, the endowment price gap is prices) is examined by univariate analysis. attributed to fundamental psychological variables such as loss aver- Similar to prior research (Tuncel and Hammitt, 2014; Horitz sion (Khaneman, Knetsch, and Thaler, 1991), ownership effect Ag- and McConnel, 2002), the analysis reveals a statistically significant garwal 2004; Aggarwal and Law 2005), and even self-enhancement effect of type of good on the endowment price ratio. Pair-wise com- (Dommer and Swaminathan, 2013; Chatterjee, Irmak, and Rose, parison analysis indicate that the main significance comes from the 2013). There may be two potential explanations for this endowment between ordinary private good and lottery (M = difference difference price gap as a result of publication year. First, participants in general .31, SE = .13, p = .017). One potential explanation for this greater are becoming more loss averse or attached to their endowed objects. LN price ratio for lottery compared to ordinary private good is the This may have some interesting implication in terms of sociology or inherent risk involved in a lottery. Since most people are risk-averse, personal psychology on how our society may be evolving compared it is reasonable to assume that owners and non-owners of the lottery to years past. Second, since the current endowment effect price gap would demand greater compensation for their selling and buying of seems to be marginally moderated by the change in price elicitation the lottery, leading to the observed greater price ratio. method, there may a change in experimental settings that is result- While there is an apparent statistical significance in the elicita- ing in greater endowment price gap. This may have some negative tion method used by the studies, pair-wise comparison reveals that implication in which experimenters are artificially inflating the effect and that no the Other method is the main driver of the significance of endowment through experimental settings and procedures. significance exists between Solo and Variable Range methods (M dif- This current meta-analysis has some limitations. First, while = .06, SE ference = .1, p = .57). it focusses on the endowment effect in the experimental setting, it Using publication year as a factor, a regression analysis was ignores the effect through other research methods such as quasi- run on the endowed/unendowed price ratio. The regression analysis 2 experiments and surveys. This reduces the power and the scope that show a significant effect of publication year on price ratio (R = could have brought out to examine a larger picture of the endowment .045, b = .21, t = 2.0, p = .05). While only 4.5% of the change in effect. Second, since the data used are means from experiments (or price ratio may be explained by the publication year, a one standard the ratio of the price means), it lacks some statistical tools to further deviation increase in publication year leads to a .21 unit increase in refine the relationship between the factors. the endowed/unendowed price ratio. By using a median split based Advances in Consumer Research (Volume 46) / 813

One potential way to address both of these limitations is to ex- Horowitz, J.K. & McConnell, K.E. (2002). A review of WTP/ amine the endowment effect but through a different psychological WTA studies. Journal of Environmental Economics and variable: psychological ownership, the degree to which an individual Management, 44, 426–447. feels or believes that he or she owns an item with or without actual Kahneman, D., Knetsch, J., & Thaler, R. H. (1991). Anomalies: or physical ownership. Furthermore, the dependent variable could The Endowment Effect, Loss Aversion, and Status Quo Bias. be changed to include not only pricing gaps but also rating or prefer- Journal of Economic Perspectives, 5(1), 193-206. ence gaps that are a result of psychological ownership. This would Nayakankuppam, D. & Mishra, H. (2005). The Endowment Effect: expand the level of analysis beyond experimental studies and include Rose-Tinted and Dark-Tinted Glasses. Journal of Consumer correlational studies, which more refined meta-analytical tools are Research, 32, 390-395. available for detailed analysis of the data. Pierce, J. L., Kostova, T., & Dirks, K. T. (2003). The State of Psychological Ownership: Integrating and Extending a REFERENCES Century of Research. Review of General Psychology, 7 (1), Aggarwal, P. (2004), The Effects of Brand Relationship Norms 84-107. on Consumer Attitudes and Behavior. Journal of Consumer Randall, A., Stoll, & J.R., (1980). Consumer’s surplus in Research, 31 (June), 87-101. commodity space. American Economic Review, 70 (3), Aggarwal, P. & Law, S. (2005), Role of Relationship Norms in 449–455. Processing Brand Information. Journal of Consumer Research, Reb, J. & Connolly, T. (2007). Possession, feelings of ownership 32 (December), 453-464. and the endowment effect.Judgement and Decision Making, Ariely, D., Huber, J., & Wertenbroch, K. (2005). When Do Losses 2(2), 107-114. Loom Larger Than Gains? Journal of Marketing, 41, 134-138. Saqib, N., Frohlich, N., & Bruning, E. (2010). The influence of Becker, G. M., DeGroot, M. H., & Marschak, J. (1964). Measuring involvement on the endowment effect: The moveable value Utility by a Single-Response Sequential Method. Behavioral function. Journal of Consumer Psychology, 20, 335-368. Science, 9 (3), 226-232. Sayman, S. & Onculer, A. (2005). Effects of study design Belk, R. W. (1988). Possessions and the Extended Self. Journal of characteristics on the WTA–WTP disparity: A meta analytical Consumer Research, 15(2), 139-168. framework. Journal of Economic Psychology, 26(2), 289-312. Brenner, L., Rottenstreich, Y., Sood, S. & Bilgin, B. (2007). On Shu S. B. & Peck, J. (2011). Psychological ownership and affective the Psychology of Loss Aversion: Possession, Valence, and reaction: Emotional attachment process variables and the Reversals of the Endowment Effect.Journal of Consumer endowment effect.Journal of Consumer Psychology, 21, Research, 34, 369-376. 439–452 Carmon, Z. & Ariely, D. (2000). Focusing on the Forgone: How Strahilevitz, M. A. & Loewenstein, G. (1998). The Effect of Value Can Appear So Different to Buyers and Sellers.Journal Ownership History on the Valuation of Objects. Journal of of Consumer Research, 27 (3) 360-370. Consumer Research, 25, 276-289. Chatterjee, P., Irmak, C., & Rose, R. L. (2013). The Endowment Thaler, R.H. (1980). Toward a positive theory of consumer choice. Effect as Self-Enhancement in Response to Threat. Journal of Journal of Economic Behavior and Organization, 1, 39–60. Consumer Research, 40(3), 460-476. doi:10.1086/671344 Tuncel, T. & Hammitt, J. K. (2014). A New Meta-Analysis on the Dittmar, H. (1992). The social psychology of material possessions: WTP/WTA Disparity. Journal of Environmental Economics To have is to be. New York: St. Martin’s Press. and Management, 68, 175-187. Dittmar, H. (1991), “Meanings of material possessions as Zhao, J., & Kling, C.L. (2004). Willingness to pay, compensating reflections of identity: gender and social-material position in variation, and the cost of commitment. Economic Inquiry, 42, society”, Journal of Social Behavior and Personality, 6 (6), 503–517. 165- 86. Dommer, S. L. & Swaminathan, V. (2013). Explaining the Endowment Effect through Ownership: The Role of Identity, Gender, and Self-Threat. Journal of Consumer Research, 39, 1034–1050. Uncertain Reward Campaigns Impact Consumers’ Size Choices Nükhet Taylor, York University, Canada Theodore J. Noseworthy, York University, Canada Ethan Pancer, Saint Mary's University, Canada

EXTENDED ABSTRACT large) revealed that participants in the uncertain promotion condition Many companies annually launch uncertain reward campaigns, chose larger cup sizes (M = 2.08, SD = .83) compared to participants where consumers are given a chance to win a prize with each pur- in the no uncertain promotion condition (M = 1.87, SD = .62), t(204) chase. Uncertain reward campaigns offer consumers a probability = 2.10, p < .05, d = .29. Consistent with the supersizing anecdote, (e.g., one in four chances) to win a gradient of prizes, ranging from this mean difference was driven predominantly by variations in the nominal to valuable. For example, Tim Hortons’ Roll-Up-The-Rim largest offering. Specifically, the extra-large cup was chosen by 8% campaign gives explicit odds of a 1 in 6 chance to win, with the of participants in the uncertain promotion condition compared to prizes ranging from a hot beverage to a new car (Roll-Up-the-Rim only 1.9% in the no uncertain promotion condition, χ2 (1) = 4.16, p < Website, 2017). Similar uncertain reward campaigns include Coca .05, φ = .14. Reliable choice shifts for other product sizes were not Cola’s Sip & Scan, Pepsico’s Win Every Hour, M&M’s When We observed (ps > .12). Hence, the results suggested that consumers are Win, You Win, and Wendy’s Dip & Squeeze and Win. The gradients indeed more likely to supersize their product purchases during an in the prizes exist to elicit interest, but there is anecdotal evidence to uncertain reward campaign. suggest that these campaigns may be changing consumer behavior. Larger products are usually more expensive than their smaller Specifically, a recent article in the Huffington Post suggests that alternatives. Thus, one plausible explanation for this behavior is consumers are purchasing larger products (or ‘supersizing’) during the somewhat rational belief that firms are housing larger, and thus Tim Horton’s Roll-Up-The-Rim campaign (Yum, 2013). Although more costly, prizes in the most profitable options. One way of rul- the company regularly advertises otherwise, it thus appears that cusing out this possibility is by holding prices constant across product tomers behave as if larger beverage sizes bring better odds (Yum, sizes. As such, in Study 2, participants (N = 148) were randomly 2013). This speculation has incited customers to catalogue the prize assigned to one of two conditions (Uncertain Promotion vs. No Un- frequency and distribution across cup sizes (Roll Up the Stats Web- certain Promotion). All participants were told that, for a limited time, site, 2015). The results of over 14,000 crowdsourced reports reveal Starbucks was conducting a promotion where all product sizes are no statistical variation of wins by cup size (Aspler, 2016). Yet this equally priced. An independent samples t-test with average cup size supersizing tendency persists. selected as the dependent variable (1 = Short to 4 = Venti) revealed In the current article, we propose the intriguing possibility that that participants in the uncertain promotion condition chose larger supersizing during uncertain reward campaigns is occurring because product sizes (M = 3.39, SD = .86) compared to participants in the consumers infer that their odds of winning the most elusive prize are no uncertain promotion condition (M = 3.11, SD = .95), t(146) = better in the larger sized offerings. Thus, we propose that it is not 1.91, p = .058, d = .31. Follow-up contrasts confirmed that this effect that consumers are rejecting or ignoring explicit information about occurred because participants were more likely to choose Venti in their overall odds of winning, but rather they are engaging in tactics the uncertain promotion condition (59.5%) than in the no uncertain to elevate a sense of control over a desirable, but elusive outcome. promotion condition (43.2%), χ2 (1) = 3.90, p < .05, φ = .16. Reliable Of course, from a rational standpoint (e.g., Expected-Utility Theory; choice shifts for other product sizes were not observed (ps > .25). Bernoulli, 1738; von Neumann and Morgenstern, 1945), if the goal The results of study 2 suggest that the supersizing tendency during is to maximize one’s chances of winning, consumers should purchase uncertain reward campaigns goes beyond just-world beliefs, percep- the smallest product offering and enter the lottery twice for an equiv- tions of equity, or other more rational inferences based on price (e.g., alent price. Yet if the goal is to gain a sense of control over seem- ulterior motives). ingly insurmountable odds, then such a tactic may serve an adaptive If price inferences did not underscore the supersizing phenom- function. enon, then it could be that consumers inherently associate larger Across four studies, we sought to test the veracity of the super- products with greater chances of winning. Further, it was possible sizing phenomenon and reveal the underlying process that in fact that this association was specific to the elusive grand prize. Indeed, nudges consumers to purchase larger products during uncertain re- prior literature suggests that consumers often focus on their odds ward campaigns. As a whole, our findings support the existence of a of winning the grand prize, because nominal prizes are far too fre- supersizing tendency during uncertain reward campaigns and further quently won (Yan & Muthukrishnan, 2014). To test these possibili- support that it may be a byproduct of consumers’ need to increase a ties, Study 3 employed a 2 (Prize: small vs. large) × 3 (Product Size: sense of control over the elusive grand prize. Small vs. Medium vs. Large) between-subjects factorial design. All In study 1, we begin by confirming the Huffington Post article, participants (N = 240) read that Lays was currently conducting an validating that consumers do indeed tend to purchase larger prod- uncertain reward campaign. Participants were then shown all sizes of ucts during uncertain reward campaigns. Undergraduate students (N Lays chips included in the campaign (small, medium, and large), but = 206) were randomly assigned to one of two conditions (Uncertain were randomly asked to rate the likelihood of winning one prize with Promotion vs. No Uncertain Promotion). While participants in the no only one of the bags. A two-way ANOVA on likelihood revealed no uncertain promotion condition chose the coffee cup size they would difference across product sizes on perceived likelihood of winning purchase from Tim Hortons on a normal day, participants in the un- the nominal prize (p = .99). However, participants’ perceived likeli- certain promotion condition chose the coffee cup size they would hood of winning the grand prize did vary by product size, F(2, 234) purchase during a time when Roll Up the Rim campaign was active = 10.62, p < .001, η2 = .08. As expected, participants believed the (which offered a chance to win one of the prizes that ranged from a larger bag was more likely to win the grand prize (M = 6.29, SD = cup of coffee to a car). An independent samples t-test with average 3.54) relative to the small (M = 3.97, SD = 2.68), F(1, 234) = 13.09, cup size selected as the dependent variable (1 = small to 4 = extra- p < .001, η2 = .05, or medium bags (M = 3.56, SD = 2.29), F(1, 234)

= 18.13, p < .001, η2 = .07. These results converged with studies 1 to range from the subtle to the outlandish (Damisch et al., 2010), and 2 to show that supersizing corresponds with biased inferences of researchers have yet to explore how a variety of novel ways of en- chance, and these inferences are restricted to the chance of winning tering a lottery might affect consumer beliefs about their chances. the grand prize. This represents a unique contribution of the current work. Rationally In our final study, we then turned to confirm that the phenom- speaking, the decision of how to enter a lottery should not affect enon relates to consumers establishing a greater sense of control in one’s odds. And yet, if the goal is to gain a sense of control, super- the presence of the elusive and desirable grand prize. Participants (N sizing may serve as a way of coping with the insurmountable odds = 300) were randomly assigned to one of four conditions in a 2 (No associated with the desirable grand prize. Uncertain Promotion vs. Uncertain Promotion) × 2 (Power: High vs. Baseline) between subjects factorial design. The procedures, stimuli, REFERENCES and choice task were the same as study 1 with one exception. In Aspler, S. (2016). 15 Things You Didn’t Know About Roll Up The an effort to extend beyond measuring our mediator, we introduced Rim. Buzzfeed Canada. Retrieved August 23, 2017, from a pretested power manipulation before the main scenario (adapted https://www.buzzfeed.com/sarahaspler/things-you- didnt- from Galinsky et al., 2003). Because high feelings of power reduce know-about-roll-up-the-rim?utm_term=.ojJXXvlrX2#. consumers’ motivation to acquire control (Inesi et al. 2012), we ex- jnyYYxD1Yw pected high power condition to serve as a boundary condition to su- Bernoulli, D. (1738). “Specimen Theoriae Novae de Mensura persizing phenomenon. Sortis,” Comentarii Academiae Scientiarum Imperialis The results converged with our expectations. That is, partici- Petropolitanae, 5, 175–92 (translated by Louise Sommer pants in the baseline power condition selected larger product sizes on [1954] in Econometrica, 22 [1], 23–36). average in the uncertain promotion condition (M = 2.63, SD = .90) Fast, N. J., Gruenfeld, D. H., Sivanathan, N., & Galinsky, A. D. than in the no uncertain promotion condition (M = 2.33, SD = .83), (2009). Illusory control: A generative force behind power’s F(1, 296) = 4.34, p < .05, η2 = .014. Critically, this pattern was elimi- far-reaching effects.Psychological Science, 20(4), 502- 508. nated in the high power condition, p = .36. Similarly, participants in Inesi, M. E., Botti, S., Dubois, D., Rucker, D. D., & Galinsky, the baseline condition indicated feeling more in control when choos- A. D. (2011). Power and choice: Their dynamic interplay ing a size during the uncertain promotion (M = 3.17, SD = 1.39) than in quenching the thirst for personal control. Psychological when there was no uncertain promotion (M = 2.69, SD = 1.47), F(1, Science, 22(8), 1042-1048. 296) = 4.65, p < .05, η2 = .015. This pattern was eliminated in the high Roll up the Stats Website. (n.d.). Retrieved August 21, 2017, from power condition, p = .16. Finally, we conducted a mediated mod- http://rollupthestats.com/ Rucker, D. D., & Galinsky, A. D. eration analysis (Hayes 2012; Model 8). Perceived control mediated (2008). Desire to acquire: Powerlessness and compensatory size choices in the baseline condition (95% CI: .002, .129), but not consumption. Journal of Consumer Research, 35(2), 257-267. in the high power condition (95% CI: −.096, .003). These results Tencer, D. (2017). Tim Hortons Brings Back ‘Roll Up The Rim’ To confirmed that consumers supersize as a way of enhancing control Prove We’ve Reached Peak Canada. Huffington Post Canada. over a desirable but extremely unlikely outcome, which is eliminated Retrieved August 21, 2017, from http://www.huffingtonpost. when consumers’ feeling of power is reinforced. ca/2017/06/28/tim-hortons-brings-back-roll-up-the-rim-to- Promotional lotteries account for over $1.8 billion in annual prove-we-ve-reached_a_23006161/ marketing expenditures in the U.S. alone (Smith, 2009). McDonald’s Tim Hortons Roll Up the Rim to Win Official Website. Monopoly and Tim Hortons’ Roll-up-the-Rim are but two examples (n.d.). Retrieved August 21, 2017, from https://www. of a prevalent industry practice. Our findings converge with anec- rolluptherimtowin.com/ dotal evidence to suggest that consumers do indeed tend to super- Von Neumann, J., & Morgenstern, O. (1945). Theory of games and size their product choices during lottery promotions. Consistent with economic behavior. Princeton, NJ: Princeton University Press. prior evidence suggesting a link between larger products and control Yan, D., & Muthukrishnan, A. V. (2014). Killing hope with good (Dubois et al., 2011), we found that the act of supersizing elevates a intentions: the effects of consolation prizes on preference sense of control in a scenario where people would otherwise have no for uncertain promotion promotions. Journal of Marketing power to control the odds. Research, 51(2), 198-204. Possessing a feeling of control has long been considered a fun- Yum, K. (2013). Roll Up The Rim Odds 2013: Size Does Matter, damental human motivation. Given that the odds of winning the 1,100 Readers Tell Us (PHOTOS). Huffington Post Canada. grand prize in any lottery is extraordinarily small, it is not surprising Retrieved August 16, 2017, from http://www.huffingtonpost. that people look to cues that they believe influences the outcomes ca/2013/03/27/roll-up-the-rim-odds-survey_n_2962737.html beyond their reach. While the cues and superstitions that people look Green Experiences: Using Green Products Improves the Accompanying Consumption Experience Ali Tezer, HEC Montréal, Canada H. Onur Bodur, Concordia University, Canada

EXTENDED ABSTRACT Participants who cleaned the tableware with the green sanitiz- The current research examines how using green products (e.g., ing product (M = 5.78, SD = 2.02) enjoyed the experience more than environmentally friendly headphones) influences the enjoyment of those who used the conventional sanitizing product (M = 4.70, SD η 2 the accompanying consumption experience (e.g., listening to mu- = 2.37; F(1, 78) = 4.78, p = .032, p = .058). Mediation analysis sic). Results reveal that consumers enjoy accompanying experiences indicated an indirect-only mediation through feelings of warm glow β more with green (vs. conventional) products and feelings of warm ( indirect = .70, SE = .33, 95% CI = [.07; 1.38]). glow drive this effect. Experiment 4 tests whether negative beliefs about green prod- Green products are available to consumers in various product ucts serves as a boundary condition. One hundred and eighty-one categories but research on how green products influence consumer undergraduate students were given a grip ball to do a set of hand behavior at the consumption stage is scarce. In order to predict how exercises (54.7% females; Mage = 21.75; SD = 3.53). Prior to the hand using green products would influence the consumption experience, exercises, half of the participants learned about a fictitious research we draw from research on warm glow. report which found that green products have negligible environmen- Theory of warm glow giving posits that people feel good upon a tal impact. Participants in the baseline condition did not receive this prosocial behavior, including proenvironmental behavior (Andreoni information. In the green product condition, participants were told 1990; Giebelhausen et al. 2016). One thing common in research on that the grip ball was environmentally friendly. Participants in the warm glow is that people make a conscious choice to behave in a conventional product condition did not receive such information. prosocial manner (e.g., donation). Extending this stream of research, Next, participants did four hand exercises and reported enjoyment (1 we suggest that simply using a green product, even if the product is = not at all, 9 = very much) and warm glow. given to consumer without choice, will elicit warm glow feelings. In In the baseline condition, participants who used the green prod- case the green product (e.g., a pair of headphones) has an accompa- uct (M = 6.15; SD = 1.41) enjoyed the hand exercise more than those nying experience (e.g., listening to music), the warm glow feelings who used the conventional one (M = 5.28; SD = 1.62; F (1, 177) = η 2 that arise due to using green products will enhance the enjoyment of 6.40; p = .012, p = .035). However, among participants who were the consumption experience. told that the environmental impact of green products is negligible, In the first experiment, sixty undergraduate students participat- the enjoyment of the exercise did not vary as a function of product type (M = 5.10 vs. M = 5.40, F (1, 177) < .1, p > .30, η 2 ed in a music consumption study (43.3% females; Mage = 21.53; SD green conventional p = 2.80) and listened to 45-second excerpts from three songs. Results < .01. Furthermore, in the baseline condition, a significant indirect β revealed that participants who were told the headphones were pro- effect of warm glow was observed ( indirect = .25; SE = .11; 95% CI duced from recycled materials enjoyed listening to songs more than [.04, .49]). In the negligible impact condition, the indirect effect of β those who did not receive such an information (1 = not at all enjoy- warm glow was absent ( indirect = -.03; SE = .12; 95% CI [-.27, .20]). The current research examines the effect of green products at able, 7 = enjoyable; Mgreen = 4.64; SD = .84 vs. Mconventional = 4.09; SD η 2 consumption stage. More importantly, vast amount of research docu- = 1.02; F (1, 58) = 5.33; p = .025; p = .084). The second experiment was a short-essay task. Sixty five un- ments potential negative effects of green product attributes (Luchs et dergraduate students were given a pen to write a short essay on how al. 2011; Newman et al. 2014). The current research presents a novel and positive effect of green products on the consumption experience they spend their spare time (49.2% females; Mage = 20.95; SD = 2.18). In the green product condition, participants were told that the pen and extends research on green products. The current research further was produced from recycled materials. In the comparison group, the contributes to research on warm glow by showing that even without pen was presented as the recipient of “The Most Practical Product making a conscious choice to engage in a prosocial behavior, using of the Last Decade” award. After writing the short essay, enjoyment green products leads to warm glow. From a managerial standpoint, of writing (1 = not at all, 9 = very much) and feelings of warm glow the current research shows that when it comes to products used to (α = .85). consume experiences, brands can benefit from going green. Results revealed that participants in the green pen condition (M η 2 REFERENCES = 5.55; SD = 2.14) enjoyed writing the short-essay, p = .068). Fur- thermore, the indirect effect of product type on enjoyment of experi- Andreoni, James (1990), “Impure Altruism and Donations to Public ence through warm glow ratings revealed an indirect-only mediation Goods: A Theory of Warm-Glow Giving,” The Economic β Journal, 100 (401), 464-77. ( indirect = .75; SE = .31; 95% CI = [.26; 1.51]. Experiment 3 tests whether the positive effect of using green Giebelhausen, Michael, HaeEun Helen Chun, J. Joseph Cronin Jr, (vs. conventional) on the enjoyment of the accompanying consump- and G. Tomas M. Hult (2016), “Adjusting the Warm-Glow tion experiences can be generalized to strength-related products. Thermostat: How Incentivizing Participation in Voluntary Eighty undergraduate students participated in a dishware sanitization Green Programs Moderates Their Impact on Service Satisfaction.” Journal of Marketing, In Press, doi: http://dx.doi. task (40% females; Mage = 20.69; SD = 2.37) were given a dishware sanitizer. In the green product condition, participants learned that the org/10.1509/jm.14.0497 dishware sanitizer is made with plant-based and biodegradable ingre- Newman, George E., Margarita Gorlin, and Ravi Dhar (2014), dients. Participants in the conventional product condition did not re- “When Going Green Backfires: How Firm Intentions Shape ceive this piece of information. Next, participants sanitized dishware the Evaluation of Socially Beneficial Product Enhancements,” and reported how much they enjoyed cleaning the dishes (1 = not at Journal of Consumer Research, 41 (October) 823-39. all, 9 = very much) and warm glow (α = .96).

Advances in Consumer Research 816 Volume 46, ©2018 Indigenous Trust and Readiness towards Development Ding Hooi Ting, Universiti Teknologi PETRONAS, Malaysia Chin Chuan Gan, Sunway University, Malaysia Amir Zaib Abbasi, Capital University of Science and Technology, Pakistan Sohel Ahmed, Universiti Teknologi PETRONAS, Malaysia

EXTENDED ABSTRACT nature bonding (Kals, Schumacher, & Montada, 1999; Schultz, The primary consumption of indigenous people across the 2001), social bonding (Kyle, Graefe, & Manning, 2005), connections world is related mostly to the jungle and its produce. The consump- to family and friends (Raymond, 2009) and the life satisfaction scale tion practices that stem from this primary consumption and its relat- (Diener, Emmons, Larsen, & Griffin, 1985), on a seven-point Likert edness to the jungle are vivid in most indigenous culture and practic- scale (1 = “strongly disagree,” 7 = “strongly agree”). es. In other words, consumption is demonstrated by the active social From the data collected, Cronbach’s alpha measurement for in- clustering (Coleman & Williams, 2013)—the dynamics of the inter- ternal consistency was computed for the different dimensions. Reli- actions between individuals and dyadic ties, and other social interac- ability values for all dimensions were above .80. A simple ANOVA tions between actors. The interaction of these will jointly develop the analysis to test the surveyed questionnaire (whether there are any social identity and attachment. differences among the indigenous groups). A test on the equality of According to Lichrou, O’Malley, and Patterson (2014) sense of means suggests that they are insignificant (p > .05), hence there are place integrates place attachment and place meanings. The jungle be- no differences among the indigenous groups on their perceptions. comes their sanctuary and living sources. The reliance on the jungle In study 2, the researchers spent about six days in each village becomes inseparable from their lives. Scannell and Gifford (2017) or settlements. In each village or settlements, between 12 – 15 indig- asserted that most people have at least one place to which they are enous people were recruited to understand the study context. Alto- emotionally attached. Place attachment refers to the experience of gether, 117 informants spoke about their life-story. The informants long-term affective bond to a specific geographic area and the mean- were asked about their life in the jungle, the symbiosis between them ing assigned to that bond (Ramkissoon, Weiler, & Smith, 2012) and and the jungle, the disruption caused by external parties, and how that place becomes one anchor of identity (Anton & Lawrence, 2014; they manage their life after a disruption. Since the researchers stayed Hay, 1998) and to regulate affect (Schmalz & Orth, 2012). But with with the indigenous, the interviews and dialogues take place through- the new developments, the urge to modernize the internal and remote out the stays and visits around the places in the villages. areas becomes one of the development agendas. Interview findings from 117 informants revealed that the indig- The objective of this study is to explore the influence of place enous community associates the jungle with their life. The consump- attachment (the emotional and functional ties that connect the indig- tion of the jungle produce has provided them with the strong attach- enous communities to their home), trust towards development and ment as their life lingers around the jungle. The following themes the readiness of the community to embrace change. The study areas and categories were extracted from the analyses: Attachment; Life- are non-tourist areas. style; Sustenance; Integration; Disruption; Coping; Readiness. In Malaysia, the development of the lands of the indigenous With the vast development and the philosophies of the govern- communities has spelt out scepticism and distrust towards the main ment underpinning the economic and welfare booster of the indig- streams. The following excerpts painted the initial picture (Mah enous communities, many plans to develop the remote areas and the Meri), “My child, I hope that you are here with good intention…” communities have taken place. Though the planned development The impact of change (disruption) to place attachment has al- could be beneficial, for many tribal communities, the conceptualiza- tered the bonds between person and location that are typically lation and implementation of development are themselves disparate tent (Chen, Dwyer, & Firth, 2014; Devine‐Wright, 2009). The So- from the indigenous epistemologies and indigenous values and sys- cial Representations theory (Moscovici, 1988) is used to understand tems (Chino & DeBruyn, 2006). However, the indigenous commu- the study context—the emphasis placed by the disruption within the nities have a different set of definition as opposed to those ofthe “common senses of the scientific inquiries” (a contradiction between mainstream. The value system and the consumption practices of the scientific knowledge), and the consensual universe of social repre- indigenous communities are far underexplored. In most cases, the sentation through power relations between different actors. development mindsets are different from what the indigenous needs. The respondents from this study are indigenous who were born The limitation of this study lies in the original understanding and raised up in the remote areas. In order to ensure the reliability of the indigenous practices and value system. Though the research- of the data collected, these indigenous must be either still staying in ers have good exposure and sound knowledge on the indigenous the remote areas and must have at least stayed in the remote areas communities, the underpinning indigenous values and narratives are for 15 years. Five of the ethnic group in Sarawak, Malaysia (Kayan, never understood by external parties, which compromises the rich- Penan, Kenyah, Iban and Kelabit indigenous ethnic groups) that ness of the study. the researchers visited is located far into the remote areas—five to This study will shed light on the mainstream and development eight hours of four-wheel rides into the bumpy forest tracks that are planners to re-consider their ambitious development thrust unto the opened by timber loggers. The other four ethnic group in Peninsula indigenous homes and jungle. The sensitivity of the indigenous com- Malaysia (Mah Meri, Kensiu, Temuan and Kintaq ethnic) are easily munities and the environment should be fully understood before accessible through tar road. further adversities towards the wealth of the indigenous values and There were two study phases in this study; study 1— basic system vanish. questionnaire survey and study 2— phenomenology. A basic questionnaire on the place attachment was distributed to 525 respondents in these villages—place identity and place dependence (Williams, Patterson, Roggenbuck, & Watson, 1992; Williams & Vaske, 2003),

Advances in Consumer Research 817 Volume 46, ©2018 818 / Indigenous Trust and Readiness towards Development REFERENCES Lichrou, M., O’Malley, L., & Patterson, M. (2014). On the Anton, C. E., & Lawrence, C. (2014). Home is where the heart marketing implications of place narratives. Journal of is: The effect of place of residence on place attachment Marketing Management, 30(9-10), 832-856. and community participation. Journal of Environmental Moscovici, S. (1988). Notes towards a description of social Psychology, 40, 451-461. representations. European journal of social psychology, 18(3), Chen, N. C., Dwyer, L., & Firth, T. (2014). Conceptualization and 211-250. measurement of dimensionality of place attachment. Tourism Ramkissoon, H., Weiler, B., & Smith, L. D. G. (2012). Place Analysis, 19(3), 323-338. attachment and pro-environmental behaviour in national parks: Chino, M., & DeBruyn, L. (2006). Building true capacity: The development of a conceptual framework. Journal of Indigenous models for indigenous communities. American Sustainable Tourism, 20(2), 257-276. journal of public health, 96(4), 596-599. Raymond, C. (2009). Assessment of Rural Landholder and NRM Coleman, N. V., & Williams, P. (2013). Feeling like my self: Staff Attitudes, Goals, and Place Values in the Eyre Peninsula Emotion profiles and social identity. Journal of Consumer Region, South Australia. Adelaide: Enviroconnect Pty Ltd, Research, 40(2), 203-222. Adelaide. Devine‐Wright, P. (2009). Rethinking NIMBYism: The role of Scannell, L., & Gifford, R. (2017). Place attachment enhances place attachment and place identity in explaining place‐ psychological need satisfaction. Environment and Behavior, protective action. Journal of community & applied social 49(4), 359-389. psychology, 19(6), 426-441. Schmalz, S., & Orth, U. R. (2012). Brand attachment and consumer Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The emotional response to unethical firm behavior. Psychology & satisfaction with life scale. Journal of personality assessment, Marketing, 29(11), 869-884. 49(1), 71-75. Schultz, P. W. (2001). The structure of environmental concern: Hay, R. (1998). Sense of place in developmental context. Journal of Concern for self, other people, and the biosphere. Journal of Environmental Psychology, 18(1), 5-29. Environmental Psychology, 21(4), 327-339. Kals, E., Schumacher, D., & Montada, L. (1999). Emotional Williams, D. R., Patterson, M. E., Roggenbuck, J. W., & Watson, affinity toward nature as a motivational basis to protect nature. A. E. (1992). Beyond the commodity metaphor: Examining Environment and Behavior, 31(2), 178-202. emotional and symbolic attachment to place. Leisure sciences, Kyle, G., Graefe, A., & Manning, R. (2005). Testing the 14(1), 29-46. dimensionality of place attachment in recreational settings. Williams, D. R., & Vaske, J. J. (2003). The measurement of place Environment and Behavior, 37(2), 153-177. attachment: Validity and generalizability of a psychometric approach. Forest science, 49(6), 830-840. The Effect of Fertility on Women’s Word-of-Mouth Behavior Sevincgul Ulu, Rutgers University, USA Kristina Durante, Rutgers University, USA Jonah Berger, University of Pennsylvania, USA Aekyoung Kim, Rutgers University, USA

EXTENDED ABSTRACT Study 2 was designed to further investigate whether the fertility- The current research explores whether fertility influences wom- regulated-increase in women’s desire to share information is targeted en’s desire to share word-of-mouth (WOM). We propose that women toward other women, men, or both. This study used a forced-choice share more WOM near the fertile phase of the ovulatory cycle and question to determine the gender of target person and a different do so with other women, not men. Sharing more near ovulation may method of assessing fertility status. Five hundred and twenty-one function to form alliances with other women and mitigate mate com- women (Mean age = 29.9) participated for payment via MTurk. Con- petition. sistent with our prediction, a logit analysis with target gender and The ovulatory cycle spans about 28 days and women can be- conception probability revealed a significant main effect of concep- come pregnant only during the days of peak fertility near ovulation. tion probability (β = -5.93, z = -2.09, p < 0.04). In addition to a main Research finds that ovulation influences women’s mating motivations effect of target gender, there was no effect of fertility on general so- and consequently women’s behavior (Durante et al. 2011; Thornhill ciability, p = .39, and no interactions emerged when sociability was and Gangestad 2008). For instance, near ovulation, women are more entered into the full model. This provides additional evidence that sexually attracted to men who are socially dominant and have sym- the effect of fertility on WOM is more strongly related to intra-sexual metrical, attractive faces (Durante et al. 2012; Gildersleeve, Hasel- competitive motivations at high fertility, as opposed to fertility acti- ton, and Fales 2014). And, fertility also enhances women’s desire vating a general desire to socialize with others. to compete with other women (Durante et al. 2014). For example, Study 3 aimed to replicate the findings of Study 1 and Study fertility enhances women’s motivation to appear more attractive and 2 in an experimental design. We tested whether fertility increases dress sexier especially when there are many other desirable women women’s desire to share information in a lab environment and used in the local area (Durante et al. 2011). an online chat platform option for the communication tool (high vs. Some research hints at the possibility that women may become low fertility-confirmed via hormone tests). Participants indicated more social at ovulation. Research finds that ovulating women are whether they wanted to share information about a product or ser- more likely to volunteer and have an increased desire to attend social vice they recently bought in a live online chat with another woman. events (Fessler 2003; Haselton and Gangestad 2006). But, it is not Within subject analysis showed that women were more interested in clear whether women share more WOM near ovulation and if so with sharing at high versus low fertility (p = .006). whom. Because the first hurdle to attracting a high quality mate is to Study 4 tested whether fertility-regulated WOM functions to mitigate competition with rivals (Durante et al. 2014), we propose form alliances with other women. To determine whether the effect that an increase in WOM at ovulation may be related to competi- is specific to mating contexts, Study 4 included both mating and tion, not courtship. If so, then women’s WOM at ovulation should be non-mating experimental conditions. Participants imagined being at directed toward women, not men. Finally, because WOM is a form a party and having a conversation with two women—Sarah and Tra- of social grooming/bonding (Berger 2014; Dunbar 1998), and fe- cy—and one man. Half of the participants were told that Sarah and male non-human primates form alliances with select females to gain Tracy were fellow partygoers and that the man was their romantic support in competitive interactions with rivals (Seyfarth 1980), we partner (mating condition). The other half was told Sarah and Tracy predicted that women’s WOM at high fertility is likely related to al- were work colleagues and the man (Jon) was their supervisor (non- liance formation. mating condition). In both situations, Sarah was the rival and Tracy Study 1 was a field study and tracked women’s Facebook ac- the potential ally. In the non-mating condition, Sarah was up for the tivity across the cycle to test the following predictions. First, we same promotion as the participant. In the mating condition, Sarah predicted a positive association between fertility within the cycle was flirting with the participant’s partner. Participants then made two and women’s activity on social media. Second, we tested whether forced-choice decisions about who to share WOM with: Sarah or women’s activity on social media was directed toward men or other Tracy. There was a significant fertility by condition interaction (p = women. Participants provided a screenshot of their Facebook activ- .007). In the mating condition, women were significantly more likely ity log for a 30-day period and we tracked each woman’s fertility to choose to share WOM with Tracy (non-rival choice) compared to by using hormone tests to detect ovulation. There was a positive as- Sarah (rival choice) (p = .0006), suggesting that women are using sociation between fertility and social media activity (p = .03). As WOM to form an alliance when faced with a mating-related threat fertility within the cycle increased, so too did women’s activity on at high fertility. There was no effect of fertility in the non-mating Facebook. Further analysis using the activity directed toward other condition (p = .82). users revealed a significant fertility by target gender interaction (p Four studies found that fertility positively influenced women’s = .035). There was a positive effect of fertility on Facebook activ- social media activity and desire to share WOM with other women, ity directed toward other women (p = .024, Figure 1), not men (p not men. The effect of fertility on WOM may function to build alli- = .43). One limitation of the Facebook study is that there might be ances with other women to ensure coalitionary support in mating- more females on Facebook (Lin and Lu 2011) and this may be the related competitive interactions. Our findings can offer guidelines for reason fertility was significantly related to communicating with other practitioners to optimize the benefits of WOM by personalizing the women, but not men. Thus, study 2 was designed to further test the timing of the deals/campaigns in accordance with women’s cycles. boundary condition of target gender in an experiment using a large Ovulating women might be especially responsive to advertising, pro- sample of women motions, and messages that emphasize friendship and alliance messages relative to other women.

Advances in Consumer Research 819 Volume 46, ©2018 820 / The Effect of Fertility on Women’s Word-of-Mouth Behavior Implications intriguing question to test whether ovulation differently affects desire Broadly, documenting a shift in women’s tendency for WOM in information sharing behavior when women are single or in com- communication at the peak fertility has implications for understand- mitted relationships. ing how mating goals in general affect women’s information sharing Finally, to enhance ecological validity and to demonstrate in behavior. The results offer important managerial insights regarding the real world the effect of women’s fertility on WOM behavior, re- the role of hormonal changes on women’s WOM behavior. searchers might consider directly investigating real purchase second- Our findings can offer guidelines for practitioners to optimize ary data. For example, women’s responses to email promotions using the benefits of WOM by personalizing the timing of the deals/cam- various types of messages across the cycle can show the real world paigns in accordance with women’s cycles. More than 200 fertility impact of the fertility. It is also possible to examine the relevant pat- tracking apps are available in the market and they are downloaded terns of WOM behavior by observing a wide range of behaviors dur- by more than 200 million people. Companies can partner with these ing their cycle. Women might post on social network sites about the apps and use their data to track women’s fertility. Additional to apps, brands and recommend brand pages to her friends more frequently big companies can use their own database to pattern women custom- near ovulation. ers’ shopping behavior and calculate their cycles by just tracking behavior change of the women. Ovulating women might be especially REFERENCES responsive to advertising, promotions, and messages that emphasize Berger, Jonah (2014), “Word-of-Mouth and Interpersonal friendship and alliance messages relative to other women. For ex- Communication: A Review and Directions for Future ample, messages like “show how good friend you are” or “gain other Research,” Journal of Consumer Psychology, 24(4), 586-607. women’s trust” might be more impactful to catch their attention. Dunbar, Robin (1998), Grooming, Gossip, and the Evolution of In other words, companies can benefit from hormone cycles with Language. Cambridge, MA: Harvard University Press. manipulation of message content. Similarly, marketers may receive Durante, Kristina M., Vladas Griskevicius, Stephanie M. Cantú, different responses from female consumers depending on where the and Jeffry A. Simpson (2014), “Money, Status, and the latter are in their menstrual cycle. Although marketers are unable to Ovulatory Cycle,” Journal of Marketing Research, 51(1), recognize women’s cycle information directly, they can serve pro- 27–39. motional messages directly through various fertility tracking apps or Durante, Kristina M., Vladas Griskevicius, Jeffry A. Simpson, use individual purchase histories to forecast cycle phase and adjust Stephanie M. Cantu, and Norman P. Li (2012), “Ovulation the timing of promotional mails/emails. To persuade female consum- Leads Women to Perceive Sexy Cads as Good Dads,” Journal ers more efficiently, marketers might send different types of promo- of Personality and Social Psychology, 103 (2), 292-305. tion messages for different phases of the cycle. Durante, Kristina M., Vladas Griskevicius, Sarah E. Hill, Carin Additional to promotions, since high fertile women have a Perilloux, and Norman P. Li (2011), “Ovulation, Female tendency to share the news with other women, companies can ask Competition, and Product Choice: Hormonal Influences on female consumers to share their feedback or leave a review during Consumer Behavior,” Journal of Consumer Research, 37 (6), their fertile period. Companies can create more buzz and better feed- 921-934. back system by managing their strategies according to the women Fessler, Daniel M. T. (2003), “No Time To Eat: An Adaptationist cycles. Account of Periovulatory Behavioral Changes.” The Quarterly Review of Biology 78 (1), 3-21. Limitation and Future Studies Gildersleeve, Kelly, Martie G. Haselton, and Melissa M. Fales Not unlike other empirical research efforts, the results presented (2014), “Do Women’s Mate Preferences Change across the in this study are limited by a number of factors—many of which can Ovulatory Cycle? A Meta-Analytic Review. Psychological be addressed in future studies. First, we do not know which type of Bulletin, 5, 1205-1259. WOM women would share during ovulation (e.g., negative or posi- Haselton, Martie G. and Steven W. Gangestad (2006), “Conditional tive information). Second, why do women share more during ovula- Expression of Women’s Desires and Men’s Mate Guarding tion? There are two potential reasons. First, it can be related to status across the Ovulatory Cycle” Hormones and Behavior, 49 (4), orientation. High fertile women would like to show other women 509-18. that “they are better” by sharing the information. The second possi- Seyfarth, Robert M. (1980), “The Distribution of Grooming bility is that women can want to be friends a with others and expand and Related Behaviours Among Adult Female Vervet their network for future communication. They can be examined as a Monkeys,” Animal Behaviour, 28, (3), 798-813. future study. Thornhill, Randy and Steven W. Gangestad (2008), The We did not examine how ovulation alters women’s WOM be- Evolutionary Biology of Human Female Sexuality, New York: havior depending on their current relationship status. It will be an Oxford University Press. Cohesion or Coercion? Why Coordinated Behavior Backfires in Marketing Contexts Noah VanBergen, University of Cincinnati, USA

EXTENDED ABSTRACT brands in determining the details of the campaign (Mvaried=5.48 vs.

Consumers often see brands and other organizations in the mar- Mcoordinated=6.15; p=.017), that each sub-brand had more freedom ketplace engaging in various actions. Moreover, subordinate repre- (Mvaried=3.04 vs. Mcoordinated=2.39; p=.028), and that each sub-brand sentatives of an overarching organization (e.g., sub-brands, product contributed more to overall success (Mvaried=4.37 vs. Mcoordinated=3.78; lines, branches, departments, etc.) can each engage in actions that p=.029). Further, agency perceptions mediated the impact of varied may or may not be the same as other representatives from the same (vs. coordinated) actions on perceived campaign success. organization. For example, NFL teams have raised awareness for Study 3 extended the investigation of varied vs. coordinated ac- breast cancer for many years, and each team performed the same tions beyond the donation context. I also aimed to rule out a poten- actions to promote this unified cause. However, in 2017, the cam- tial alternative explanation for the above results, whereby consum- paign changed such that teams could elect to support different types ers respond most favorably to scenarios that depict both a source of cancers—teams performed different actions. The present research of variability and a source of coordination (e.g., donating different asks which strategy consumers perceive to be more effective: Should amounts to the same cause, or vice versa). To rule this out, study 3 organizational actors coordinate their actions, or should they engage used a 3-cell design with N=179 undergraduate business students in in various actions? which I described university departments implementing a campaign Previous research in entitativity provides one potential answer. to make campus greener: In the same-same condition, the same goal Entitativity refers to the extent to which individuals are perceived to (increase recycling) was served by the same actions. In the same-dif- form a single coherent group (Hamilton and Sherman 1996). This lit- ferent condition, this same goal was served by four different actions. erature has demonstrated that when group members coordinate their In the different-different condition, different goals (recycling, energy actions, the group is perceived as more competent (Callahan and use, water conservation, and emission reduction) were served by Ledgerwood 2016; Ip, Chiu, and Wan 2006). Thus, entitativity sug- four different actions. The dependent measure was perceived impact gests that organizational actions should be coordinated, not varied. of the campaign, and I again measured agency perceptions. Results However, I propose that consumers apply a different set of as- showed that both the same-different (M=5.13) and different-different sumptions about the motivating force behind actions performed by (M=5.23) conditions led to greater perceived impact than the same- organizational entities (vs. individual people): Individual people are same condition (M=4.41, ps<.01). This pattern was also found for assumed to have agency over their own actions; therefore, coordina- agency perceptions, which mediated perceived impact. tion is a signal of group cohesiveness and increases perceived com- My final study aimed to demonstrate when varied (vs. coordi- petence (Ip et al. 2006). On the other hand, organizational actions nated) actions will no longer improve perceived success of organiza- are determined by superordinate levels of the hierarchy. Therefore, tional actions by moderating the process. According to my conceptu- coordination may signal coercion rather than cohesiveness, whereas alization, if agency is signaled by cues other than varied/coordinated non-coordination may signal non-coercion—i.e., that subordinate actions, variation should no longer improve perceptions. To test this, entities have agency over their actions. If this is the case, observing study 4 used a 2 (action: coordinated vs. varied) x 2 (agent: corporate varied (vs. coordinated) actions should increase the perceived impact vs. individual) between-subject design involving N=203 Mturk par- of the actions. ticipants. I manipulated coordinated vs. varied actions through a non- I tested these predictions across four studies. Study 1 tests my profit campaign, similar to studies 1 and 2. I manipulated the agent main hypothesis using a three-cell design with N=106 Mturk partici- by describing who made the announcement about the campaign: In pants. I described a fictitious corporation comprised of four brands the corporate (individual) condition, the company’s CEO (manager engaging in a charitable campaign. I manipulated whether each of each sub-brand) made the announcement. I again measured per- brand performed the same actions in this campaign (i.e., donating ceived impact and agency. Results showed a significant two-way in- the same percentage and running the campaign during the same time teraction on both items (ps<.05), such that varied (vs. coordinated) period) or performed different actions (i.e., donating different per- actions led to greater perceived impact and greater perceived agency centages during a different time period). I also included a control in the corporate condition (ps<.05) but not the individual condi- condition that did not specify donation percentages or time frames. tion (ps>.08). Again, perceived agency mediated perceived impact. The dependent measure was the perceived impact of the campaign. My studies demonstrate that coordination is not always a signal of Results showed significantly higher perceived impact of varied ac- cohesiveness, and lack thereof is not necessarily negative. These re- tions (M=4.85) than coordinated actions (M=4.08; p=.025). The con- sults contribute to research on branding and entitativity by showing trol condition was equivalent to the varied condition (M=4.74). that, in hierarchical organizational contexts, varied (vs. coordinated) Study 2 built on study 1 by using a different (non-numerical) actions signal agency and increase perceived impact. Further, these manipulation of actions in a two-cell design with N=106 undergrad- findings relate to a wide range of consumer research interests span- uates. I described a fictitious brand comprised of four sub-brands ning such topics as branding, charitable behavior, entitativity, and engaging in a philanthropic campaign. Each sub-brand donated the organizational behavior. same percentage of revenue in each condition; however, in the coordinated action condition, all the money raised went to the same REFERENCES cause, whereas four different targets received donations in the varied Callahan, Shannon P., and Alison Ledgerwood (2016), “On the action condition. I measured perceptions of the campaign’s success Psychological Function of Flags and Logos: Group Identity as well as perceptions of each sub-brand’s agency in the campaign. Symbols Increase Perceived Entitativity,” Journal of Results showed that varied actions (M=4.92) significantly increased Personality and Social Psychology, 110 (4), 528-50. Hamilton, David L. and Steven J. Sherman (1996), “Perceiving perceived success (Mcoordinated=4.35, p=.04). I also found that varied (vs. coordinated) actions increased the perceived role of the sub- Persons and Groups,” Psychological Review, 103 (2), 336-55.

Ip, Grace Wai-man, Chi-yue Chiu, and Ching Wan (2006), “Birds of a Feather and Birds Flocking Together: Physical Versus Behavioral Cues May Lead to Trait- Versus Goal-Based Group Perception,” Journal of Personality and Social Psychology, 90 (3), 368-81. Nostalgiacising: A Performative Theory of Nostalgic Consumption Ela Veresiu, York University, Canada Thomas Derek Robinson, City University London, UK Ana Babić Rosario, University of Denver, USA

EXTENDED ABSTRACT talgia and prefer objects that provide short-term distraction from the Nostalgia—commonly conceptualized as a longing for the present. Nostalgia can thus serve as a means of consumer escapism past—plays an important role in shaping contemporary consumer (Goulding 2000; Stewart 1988). behavior and market offerings (e.g., Belk, Ger, and Askegaard 2003; From a producer-oriented perspective, nostalgia presents nu- Brown, Kozinets, and Sherry 2003; Thompson and Tian 2008). How- merous possibilities for the creation of unique market offerings. ever, purely retrospective nostalgia is difficult to frame in relation For instance, Callcott and Alvey (1991) illustrate how advertisers’ to marketing as a discipline, which is overall future-oriented by fo- prime consumers’ personal memories and elicit nostalgia with the cusing on “unrealized potentialities” among consumers and ways to use of characters. Brown, Kozinets, and Sherry (2003) demonstrate translate them into “personal needs and desires” (Zwick and Cayla that consumers resort to retro brands to connect to communities that 2011, 7) through branding, advertising, and other marketing mecha- once shared those brands, which has led Brown (2007) to develop a nisms. retro-dominant logic of marketing. Furthermore, nostalgia has been This conceptual paper aims to establish a performative theory used in heritage attractions (Goulding 1999), in high street shopping of nostalgic consumption, which we term nostalgiacising. We define (Maclaran and Brown 2001), and in cinematic experiences like re- nostalgiacising as reflective (future-oriented) consumer nostalgia makes, sequels, and prequels (Brown 2001). From the perspective of that is not innate or an emotion, but rather performatively enacted producers or memory makers, consumers “interact with the cultural through five specific dimensions. Extant marketing and consumer meanings of nostalgic brands and expand them into social universes research typically theorize that nostalgic consumption and consum- composed of stories fuelled by the yearning for what is gone but ers’ memory work, while occurring in the present, are retrospective not forgotten” (Giesler et al. 2014). Hence, more than one type of in nature, and therefore inherently about the past. In contrast to this nostalgia exists. dominant view, we bridge Butler’s (1990) conceptualization of per- Yet, the distinctions of consumer-driven or producer-driven formativity with the age of crisis concept and Boym’s (2008) notion nostalgic consumption summarized above all focus on the past-ori- of reflective nostalgia to theorize how contemporary individuals are entation of nostalgia rather than also including nostalgia’s future-fac- increasingly engaging in a performative consumption of nostalgia ing potential. Using extant theory outside the field of marketing and not to relive or remember a bygone past, but rather to playfully re- consumer research, we offer an extension to the study of nostalgic flect on or reenact possible futures during a present time. consumption: reflective consumer nostalgia. As such, we are able Boym (2001; 2008, 61) was the first to theorize reflective nos- to shift the focus to future-oriented nostalgic consumption. Follow- talgia as a playful, funny, and creative narrative that articulates “the ing Butler’s (1990, 25) conceptualization of gender performativity, relationship between past, present, and future.” Bradbury (2012, where “gender is always a doing, though not a doing by a subject 341) subsequently noted that, “nostalgia is not only a longing for who might be said to pre-exist the deed,” we argue that nostalgia is the way things were, but also a longing for futures that never came, not a pre-existing (innate) emotion, but rather a narrative that emerg- or for horizons of possibilities that seem to have been foreclosed by es as individuals engage in doing the act of reflecting across five dif- the unfolding of events.” However, we know very little about what ferent dimensions which shape the construction of self in an age of forms of reflectivity consumer nostalgia can take in the current age crisis: language, space, social systems, community, and style. Lastly, characterized by more intense and frequent global crises, ranging we argue that reflective consumer nostalgia, rather than stressing from coffee crises (Vega, Rosenquist, and Collins 2003) to trust cri- continuity or flow of consumer time, plays an important structuring ses (Humphreys and Thompson 2014). Hence, in extension, we ask: role in producing the social imaginary resources necessary for articu- What dimensions of reflectivity mediate nostalgic consumption in the lating a temporal rupture. That is, nostalgiacising allows consumers age of crisis? to playfully engage with the future, while living in the present, using To begin answering our research question, we first engaged in a past-oriented market offerings. qualitative meta-analysis of extant consumer culture theory research We see reflective consumer nostalgia as making “a project of (e.g., Castilhos, Dolbec and Veresiu 2017; MacInnis 2011) on nos- oneself” (Zwick and Cayla 2011, 7) by introducing new possibilities talgic consumption. Specifically, we coded 36 relevant articles for for the future. Thus, rather than merely being mobilized in regard to emerging themes and theoretical gaps. This allowed us to uncover “specific preferences and limitations of the capitalist present” (Brunk, what we believe is a sizeable gap in the literature, which has so far Giesler, and Hartman 2017, 2), reflective consumer nostalgia engag- focused on either consumer-driven or producer-driven but past-ori- es with the social imaginary of the future (Castoriadis 1987). This ented nostalgic consumption. process plays a pivotal role by replenishing the symbolic resources From a consumer-oriented perspective, nostalgic consumption available for “imaginative anticipation of, or speculation about, the has implied a “preference toward objects (people, places, or things) future” (Campbell 1987, 83). As such, nostalgiacising contributes to that were more common (popular, fashionable, or widely circulated) the work investigating the dimension of time in consumer behavior when one was younger […] or even before birth” (Holbrook and (e.g., Woermann and Rokka 2015). However, unlike Woermann and Schindler 1991, 330). According to Belk (1988; 1990), consumers Rokka (2015) which emphasizes the present and continuity of time- transcend their immediate confines by incorporating objects from flow, our study shows how the past can inform imagined futures and their past and physical environment into their present identities. Re- horizons of expectation. searchers have demonstrated that there is something about the past that brings comfort (e.g., Rutherford and Shaw 2011). When individuals feel sad and powerless, they experience higher levels of nos-

Advances in Consumer Research 823 Volume 46, ©2018 824 / Nostalgiacising: A Performative Theory of Nostalgic Consumption REFERENCES Giesler, Markus, Tatiana Astray, Ana Babić, Anton Siebert, and Ela Belk, Russell W. (1988), “Possessions and the Extended Self,” Veresiu (2014), “Zombie Brands: The Science Behind Undead Journal of Consumer Research, 15(2), 139–68. Market Icons,” The Huffington Post, URL: goo.gl/aTauUo ——— (1990), “The Role of Possessions in Constructing and Goulding, Christina (1999), “Heritage, Nostalgia, and the Maintaining a Sense of Past,” ACR North American Advances ‘Grey’ Consumer,” Journal of Marketing Practice: Applied in Consumer Research, 17(1), 669–76. Marketing Science, 5(6/7/8), 177–199. ———, Güliz Ger, and Søren Askegaard (2003), “The Fire of ——— (2000), “The Museum Environment and the Visitor Desire: A Multisited Inquiry into Consumer Passion,” Journal Experience,” European Journal of Marketing, 34(3/4), 261-78. of Consumer Research, 30(3), 326–51. Holbrook, Morris B. and Robert M. Schindler (1991), “Echoes of Boym, Svetlana (2001), “Nostalgia and its Discontents,” The the Dear Departed Past: Some Work in Progress on Nostalgia,” Hedgehog Review, 9(2), 7-18. ACR North American Advances in Consumer Research. ——— (2008), The Future of Nostalgia. Basic Books. Bradbury, Humphreys, Ashlee and Craig J. Thompson (2014), “Branding Jill (2012), “Narrative Possibilities of the Past for the Future: Disaster: Reestablishing Trust through the Ideological Nostalgia and Hope,” Peace and Conflict: Journal of Peace Containment of Systemic Risk Anxieties,” Journal of Psychology, of Narratives and Nostalgia, 18(3), 341–50. Consumer Research, 41(4), 877–910. Brown, Stephen (2001), “Marketing for Muggles: Harry Potter and Maclaran, Pauline and Stephen Brown (2005), “The Center Cannot the Retro Revolution,” Journal of Marketing Management, Hold: Consuming the Utopian Marketplace,” Journal of 17(5–6), 463–79. Consumer Research, 32(2), 311–23. ——— (2007), “Are We Nearly There Yet? On the Retro-Dominant MacInnis, Deborah J. (2011), “A Framework for Conceptual Logic of Marketing,” Marketing Theory, 7(3), 291–300. Contributions in Marketing,” Journal of Marketing, July, 2011, ———, Robert V. Kozinets, and John F. Sherry (2003), “Teaching 136-54. Old Brands New Tricks: Retro Branding and the Revival of Rutherford, Jana and Eric H. Shaw (2011), “What Was Old Is Brand Meaning,” Journal of Marketing, 67(3), 19–33. New Again: The History of Nostalgia as a Buying Motive in Brunk, Katja H., Markus Giesler, and Benjamin J. Hartman (2018), Consumption Behavior,” Marketing History in the New World, “Creating a Consumable Past: How Memory Making Shapes 157–66. Marketization,” Journal of Consumer Research, 44 (6), 1325- Stewart, Kathleen (1988), “Nostalgia—A Polemic,” Cultural 42. Anthropology, 3(3), 227–41. Butler, Judith (1990), Gender Trouble: Feminism and the Thompson, Craig J. and Kelly Tian (2008), “Reconstructing the Subversion of Identity, New York: Routledge. South: How Commercial Myths Compete for Identity Value Callcott, Margaret F. and Patricia A. Alvey (1991), “Toons Sell... through the Ideological Shaping of Popular Memories and and Sometimes They Don’t: An Advertising Spokes-Character Counter Memories,” Journal of Consumer Research, 34 Typology and Exploratory Study,” in Proceedings of the 1991 (February), 595–612. Conference of the American Academy of Advertising, New Vega, Fernando E., Eric Rosenquist, and Wanda Collins (2003), York: D’Arcy Masius Benton and Bowles, 43–52. “Global Project Needed to Tackle Coffee Crisis: Comments Campbell, Colin (1987), The Romantic Ethic and the Spirit of and Opinion,” Nature. Modern Consumerism. Oxford: Basil Blackwell. Woermann, Niklas and Joonas Rokka (2015), “Timeflow: How Castilhos, Rodrigo B., Pierre-Yann Dolbec and Ela Veresiu (2017), Consumption Practices Shape Consumers’ Temporal “Introducing a Spatial Perspective to Analyze Marketing Experiences,” Journal of Consumer Research, 41(6), 1486- Dynamics,” Marketing Theory, 9-29 1508. Castoriadis, Cornelius (1987), The Imaginary Institution of Society. Zwick, Detlev and Julien Cayla (2011), Inside Marketing: Cambridge: Cambridge Polity Press. Practices, Ideologies, Devices, Oxford University Press. Social Class and Prosocial Behaviors Yan Vieites, FGV/EBAPE, Brazil Rafael Goldszmidt, FGV/EBAPE, Brazil Eduardo B. Andrade, FGV/EBAPE, Brazil

EXTENDED ABSTRACT positive relationship with social class, monetary donations were Consumer researchers have long been interested in the asso- more nuanced. Although money donations increased with social ciation between social class and consumer preferences or “tastes” class when no opportunity and resource controls were included, the (Henry and Caldwell 2018; Martineau 1958). They have also shown relationship vanished when these two variables were accounted for. interest in prosocial behavior and what influences prosociality (Go- The subsequent studies sought to conceptually replicate the swami and Urminsky 2016; Simpson, White, and Laran 2018). With findings from study 1 in a consequential scenario. Since it is difficult inequality growing in many regions of the world (OECD 2015)”title” to assess all 13 behaviors, studies 2 and 3 allowed participants to : “In it Together: Why Less Inequality Benefits All”, “type” : “book” help or to donate to a food donation cause and/or a blood donation }, “uris” : [ “http://www.mendeley.com/documents/?uuid=abd51e4f- cause. If the results obtained in study 1 held true, we would expect a22a-49e1-b3a9-76b7169a303d” ] } ], “mendeley” : { “format- a negative relationship between social class and prosocial behavior tedCitation” : “(OECD 2015 and charity institutions increasingly when the cause was food-related, but a positive relationship when the competing for funds (Weerawardena, McDonald, and Mort 2010), cause was blood-related. understanding the role played by social class on prosocial acts be- In study 2, higher- and lower-class individuals were offered the comes of major importance. In the present research, we examine how opportunity to expend their time/effort to help an NGO. Four hundred social class relates to prosociality. twenty-five high- and low-social class individuals completed a filler Although psychologists and economists have recently delved questionnaire and were then told that the purported study had come into this topic, there are a few distinguishable gaps. First, the evi- to an end. However, before they had the chance to leave, respondents dence is mixed and still incipient. While some have documented that learned that interviewers were also collecting information about those who have less give more (Chen, Zhu, and Chen 2013; Côté, blood and food donation on behalf of a partner NGO. Participants House, and Willer 2015; Guinote et al. 2015; Piff et al. 2010)includ- then indicated how many of 24 possible questions they would like ing many children left behind by parents who had migrated to ur- to answer about each topic to help the institution. Unbeknownst to ban areas for work, played the DG. Stickers comprised the resource, participants, this choice served as our dependent variable. The inter- while recipients in the game were assumed to be either their friends action between social class and type of donation was consistent with or strangers, with the social distance (i.e., strangers compared to our prediction (β=-.77, t(416)=-4.00, p<.001). Higher social class friends, others have proposed and shown that affluent people donate, individuals responded less questions about food when compared to help, and volunteer more than those with less resources (Benenson, their lower-class counterparts (β=-.56, t(416)=-2.03, p=.043), but the Pascoe, and Radmore 2007; Gittell and Tebaldi 2006; Korndörfer, tendency reversed and failed to reach significance for blood dona- Egloff, and Schmukle 2015). Second, the vast majority of the stud- tions (β=.21, t(416)=.74, p=.46). ies have been conducted in nations where the levels of poverty and In study 2, higher- and lower-class individuals were offered inequality are relatively low, which leaves open the question of how the opportunity to donate money to food- and/or blood-related cam- social class and prosociality would relate to one another in societ- paigns. Three hundred and four participants received five R$2.00 ies where extreme inequality prevails and resource scarcity can be bills in return for their participation in an unrelated study. Upon com- shockingly high. Finally, the literature has paid little attention to the pletion of the filler questionnaire, respondents learned that research potential interactions between social class and type of prosocial be- assistants were collecting resources for two campaigns, one related havior. to blood donation and the other to food donation. Participants were We address these gaps in a series of three field studies by exam- then given two envelopes and were instructed to allocate the money ining the extent to which coming from the opposite ends of the so- they were willing to donate (including any) using the adequate en- cioeconomic spectrum in a highly unequal region influences different velopes and then deposit them in an urn placed a few meters away. prosocial behaviors as well as the same prosocial behavior serving Again, the interaction between social class and type of donation con- different purposes. Given the incipiency of the phenomenon under firmed our prediction (β=-2.17, t(296)=-3.50, p=.001). Higher class investigation, the plausibly of the conflicting mechanisms, and the individuals donated less money than their lower-class counterparts to array of prosocial acts examined, we deliberately chose to follow an a cause related to food donation (β=-1.12, t(296)=-2.89, p=.004), but inductive path in study 1. More precise hypotheses are established donated more money to a cause related to blood donation (β=1.04, prior to studies 2 and 3. t(296)=2.20, p=.028). Study 1 examined the relationship between social class and pro- This research demonstrates that despite the general negative sociality on 13 behaviors originated from a qualitative pilot study trend, the relationship between social class and prosocial behavior (n=56). Four hundred forty-six residents from wealthy (south zone) may vary as a function of the behavior under analysis, the social and poor (Favela da Maré) areas of Rio de Janeiro idicated how often cause one aims to help, and even the empirical approach (i.e., wheth- they performed each of the 13 prosocial behaviors in the previous er resources and opportunities for prosociality are considered). Taken year, reported the frequency of opportunities and the amount of re- together, these findings join a burgeoning, but still very limited, body sources they had to perform these behaviors, and filled a socio-de- of research that progressively reconciles previous conflicting results mographic questionnaire. A series of ordered logistic regressions in- by showing that social class is imbued with both selfishness and dicate that social class is overwhelmingly negatively associated with generosity and the degree to which these contrasting inclinations are prosociality. In fact, those occupying higher social classes reported manifested across the social spectrum is highly context-dependent behaving less prosocially in all behaviors, except for donating money (Côté et al. 2015; Kraus and Callaghan 2016; Whillans, Caruso, and and donating blood. While donating blood presented a consistently Dunn 2017)contexts that can foster a sense of entitlement among

Advances in Consumer Research 825 Volume 46, ©2018 826 / Social Class and Prosocial Behaviors higher-income individuals that, in turn, reduces their generosity. Kraus, Michael W. and Bennett Callaghan (2016), “Social Class Analyzing results of a unique nationally representative survey that and Prosocial Behavior: The Moderating Role of Public Versus included a real-stakes giving opportunity (n = 1,498. Private Contexts,” Social Psychological and Personality Science, 7(8), 769–77. REFERENCES Martineau, Pierre (1958), “Social Classes and Spending Benenson, Joyce F., Joanna Pascoe, and Nicola Radmore (2007), Behaviour,” Journal of Marketing, 23(2), 121–30. “Children’s Altruistic Behavior in the Dictator Game,” OECD (2015), In It Together: Why Less Inequality Benefits All, Evolution and Human Behavior, 28(3), 168–75. Paris: OECD Publishing, http://www.oecd.org/mexico/ Chen, Yongxiang, Liqi Zhu, and Zhe Chen (2013), “Family Income OECD2015-In-It-Together-Highlights-Mexico.pdf. Affects Children’s Altruistic Behavior in the Dictator Game,” Piff, Paul K., Michael W. Kraus, Stéphane Côté, Bonnie Hayden PLoS ONE, 8(11), 1–8. Cheng, and Dacher Keltner (2010), “Having Less, Giving Côté, Stéphane, Julian House, and Robb Willer (2015), “High More: The Influence of Social Class on Prosocial Behavior,” Economic Inequality Leads Higher-Income Individuals to Journal of Personality and Social Psychology, 99(5), 771–84. Be Less Generous,” Proceedings of the National Academy of Simpson, Bonnie, Katherine White, and Juliano Laran (2018), Sciences, 112(52), 15838–43. “When Public Recognition for Charitable Giving Backfires: Gittell, Ross and Edinaldo Tebaldi (2006), “Charitable Giving: The Role of Independent Self-Construal,” ed. Vicki Morwitz Factors Influencing Giving in U.S. States,”Nonprofit and and Sharon Shavitt, Journal of Consumer Research, 44(6), Voluntary Sector Quarterly, 35(4), 721–36. 1257–73. Goswami, Indranil and Oleg Urminsky (2016), “When Should Weerawardena, Jay, Robert E. McDonald, and Gillian Sullivan the Ask Be a Nudge? The Effect of Default Amounts on Mort (2010), “Sustainability of Nonprofit Organizations: An Charitable Donations,” Journal of Marketing Research, 53(5), Empirical Investigation,” Journal of World Business, 45(4), 829–46. 346–56. Guinote, Ana, Ioanna Cotzia, Sanpreet Sandhu, and Pramila Siwa Whillans, Ashley V., Eugene M. Caruso, and Elizabeth W. Dunn (2015), “Social Status Modulates Prosocial Behavior and (2017), “Both Selfishness and Selflessness Start with the Egalitarianism in Preschool Children and Adults,” Proceedings Self: How Wealth Shapes Responses to Charitable Appeals,” of the National Academy of Sciences, 112(3), 731–36. Journal of Experimental Social Psychology, 70, 242–50. Henry, P and M Caldwell (2018), “The Hidden Hand of Social Class,” in The Routledge Companion to Consumer Behavior, ed. M.R. Solomon and T.M. Lowey, United Kingdom: Routledge. Korndörfer, Martin, Boris Egloff, and Stefan C. Schmukle (2015), “A Large Scale Test of the Effect of Social Class on Prosocial Behavior,” PLoS ONE, 10(7), 1–48. Small but Sincere: The Impact of Firm Size and Gratitude on the Effectiveness of Cause-Marketing Campaigns Eline L.E. De Vries, University Carlos III Madrid, Spain Lola C. Duque, University Carlos III Madrid, Spain

EXTENDED ABSTRACT Expressions of gratitude also reinforce benefactors for their generos- Companies are increasingly forced to respond to various morality. These social features distinguish gratitude from related emotions ity concerns of consumers and stakeholders to “do good” and be- such as happiness and feelings of indebtedness. Evolutionary theo- have ethically. Our study focuses on a specific strategy to do good, ries propose that gratitude is an adaptation for reciprocal altruism known as Cause-Marketing (CM). CM consists of “marketing activi- (the sequential exchange of costly benefits between nonrelatives. ties that are characterized by an offer from the firm to contribute a Gratitude motivates to behave pro-socially towards the benefactor specified amount to a designated philanthropic cause” (Varadarajan (McCullough et al., 2001). and Menon 1988, p. 60). CM enables consumers to contribute to a Feelings of gratitude arise even if consumers do not directly or cause while the firm acts morally and gains profit (Pracejus and Ol- tangibly benefit from the action (Fredrickson, 2004). Within the con- sen 2004). text of CM campaigns, the tangible benefit goes to the philanthropic CM seems a promising marketing strategy (Nielsen 2015). But cause. But on top of the firm’s contribution to the cause, we argue is it effective irrespective of firm size? And what roles do perceived that consumers may be grateful towards companies engaging in CM sincerity and feelings of gratitude play? The great majority of CM campaigns as buying their products provides a way to contribute to research has focused on big firms (File and Prince 1998; Lepoutre the cause themselves and to satisfy their own social goals (Romani, and Heene 2006), producing little insight into the effectiveness of Grappi, and Bagozzi 2013). CM campaigns for smaller firms. This gap in knowledge is particu- Our reasoning implies that, if we are right that smaller firms larly relevant as the proportion of firms that is small lies around 99% engaging in CM are perceived as doing a bigger effort than larger across Europe and the United States (European Commission, 2015; firms, and hence are perceived as more sincere, consumers will ex- Grover and Suominen 2014). And even more important, research perience stronger feelings of gratitude towards small firms engaging into the affective response of consumers to CM campaigns is lack- in CM and hence are more likely to reciprocate by purchasing the ing. The current research addresses both issues. CM-promoted product, translating into greater CM effectiveness for small compared to larger firms. Theoretical background An obvious shortcoming of small compared to big businesses Methods, Results and Discussion is the more limited cash budgets and access to external financial In Experiment 1, a field study, we examined the impact of firm resources. This implies that, even if small businesses may want to size (small/big/control) on CM effectiveness (actual purchase behav- engage in CM campaigns to support a philanthropic cause, they may ior of CM-promoted product) and the mediating role of feelings of need to dedicate their capital to more immediate needs. gratitude. A mediation analysis (Hayes, 2013) confirmed that the in- Another shortcoming for small firms is their lack of visibility. fluence of firm size on actual purchase behavior is mediated by feel- When spending time and effort on social causes, they should rela- ings of gratitude. These findings directly support our hypotheses that tively heavily invest in marketing expenditures to make consumers the effectiveness of CM may be higher for small than for big firms, aware of their campaign. This means a further increase in costs for as consumers have greater feelings of gratitude towards small than small firms, required to capitalize on their responsible actions (Spen- towards big firms engaging in CM. ce, Jeurissen, and Rutherfoord 2000). In Experiment 2 we delved deeper into the underlying reasons. Taking these arguments together, we hypothesize that small Participants got a scenario describing a CM initiative. We expected firms, disadvantaged compared to big firms in terms of financial and and found that the effect of firm size (big/small) on CM effectiveness human resources necessary to engage in charitable donations, are (purchase intention) is first mediated by perceived effort, then by per- perceived as doing a bigger effort and hence are perceived as more ceived sincerity of motives and then by feelings of gratitude (Model sincere once they do contribute to a worthy cause. More specifically, 6, Hayes 2013; see Figure 1). The results confirmed our hypotheses. we posit that consumers will infer from the company’s perceived ef- Compared to big firms, small firms engaging in CM are perceived fort put into contributing to a cause, the perceived sincerity of the as doing a bigger effort and hence score higher on perceived sincer- company’s motives (Morales 2005). With higher levels of effort per- ity. As a result of these sincerity perceptions, consumers experience ceived as more sincere and trustworthy, signaling a real societal con- stronger feelings of gratitude towards small firms engaging in CM, cern and genuine intention to support the cause. translating into greater purchase intentions of CM-promoted prod- We further argue that the perceived sincerity of a firm’s motives ucts of small compared to big firms. to engage in CM is particularly relevant for CM effectiveness as con- In conclusion, the current research is of academic and manage- sumers who are skeptical about the company’s underlying motives rial relevance by being the first that examines the impact of firm size to engage in the charitable behavior, will not experience feelings of on CM effectiveness, identifying the important role of gratitude that gratitude, and so will not feel motivated to reciprocate the company consumers feel towards small firms for providing a charitable dona- for its efforts (McCullough et al., 2001; Morales, 2005). tion opportunity. Feelings of gratitude have been defined as a “positive emotion that typically flows from the perception that one has benefited from REFERENCES the costly, intentional, voluntary action of another person” (Mc- European Comission (2015). European SMEs Starting Hiring cullough, Kimeldorf, and Cohen 2008, p. 4)intentional, voluntary ef- Again. Research Report. fort on their behalf. Experiencing gratitude motivates beneficiaries to repay their benefactors and to extend generosity to third parties.

Fredrickson, Barbara L. (2004), “Gratitude, like Other Positive Nielsen (2015). The Sustainability Imperative: new Insights Emotions, Broadens and Builds,” in The Psychology on Consumer Expectations. Global Sustainability Report. of Gratitude, ed. Robert A. Emmons and Michael E. Retrieved from: http://www.nielsen.com/content/dam/ McCullough, Oxford: Oxford University Press, 145–66. nielsenglobal/dk/docs/global-sustainability-report-oct-2015. Grover, A., & Suominen, K. (2014). 2014 Summary – State of pdf SME Finance in the United States. TradeUp, (January), Pracejus, John W. and G. Douglas Olsen (2004), “The Role of 1–38. Retrieved from: http://www.growadvisors.com/ Brand-Cause Fit in the Effectiveness of Cause-Related uploads/2/7/9/9/27998715/state_of_sme_finance_in_the_ Marketing Campaigns,” Journal of Business Research, 57(6), united_states_-tradeup_2014.pdf 635–40. Hayes, Andrew F. (2013), Introduction to Mediation, Moderation, Romani, Simona, Silvia Grappi, and Richard P. Bagozzi (2013), and Conditional Process Aanalysis: A Regression-Based “Explaining Consumer Reactions to Corporate Social Approach, The Guilford Press. Responsibility: The Role of Gratitude and Altruistic Values,” McCullough, Michael E, Marcia B Kimeldorf, and Adam D Cohen Journal of Business Ethics, 114(2), 193–206. (2008), “An Adaptation for Altruism?,” Current Directions in Spence, Laura J., Ronald Jeurissen, and Robert Rutherfoord (2000), Psychological Science, 17(4), 281–85. “Small Business and the Environment in the UK and The The McCullough, Michel E., Shelley D. Kilpatrick, Robert A. Emmons, Netherlands: Toward Stakeholder Cooperation,” Business and David B. Larson (2001), “Is Gratitude a Moral Affect?,” Ethics Quarterly, 10(4), 945–65. Psychological Bulletin, 127(2), 249–66. Varadarajan, P. Rajan, and Anil Menon (1988), “Cause-Related Morales, Andrea C. (2005), “Giving Firms an ‘E’ for Effort: Marketing: A Coalignment of Marketing Strategy and Consumer Responses to High‐Effort Firms,”Journal of Corporate Philanthropy,” Journal of Marketing, 52(3), 58–74. Consumer Research, 31(4), 806–12. Unexpected Benefit Framing: Impact of Framing a Benefit as Unexpected on Product Desire Monica Wadhwa, Temple University, USA JeeHye (Christine) Kim, HKUST, China Amitava Chattopadhyay, INSEAD, France Wenbo Wang, HKUST, China

EXTENDED ABSTRACT and White 1994). We argue that the unexpected framing effect should Companies commonly perform pre-launch product testing. Dur- be demonstrated among those high on BAS, but not for those low ing such product testing, brands could learn that the product can de- on BAS. In Study-3, participants viewed an antioxidant ad, which liver originally unintended benefits. We explore how brands should presented two product benefits— supports healthy skin and healthy frame such unexpectedly discovered benefits in their marketing com- eye-function. Healthy eye-function benefit was either framed as un- munication. Should they frame such a benefit as unexpected or not? expected or as expected. Research on intentionality suggests that when consumers be- Participants indicated their WTP for the supplement, and finally lieve a product benefit was intentionally designed, as opposed to responded to the BAS-scale. A regression showed a significant two- unexpectedly discovered, they should infer that the company has a way interaction between BAS and benefit-frame (B =3.93,p =.011. A higher ability and motivation to develop it (Malle and Knobe 1997). spotlight analysis indicated that those high on BAS (+1SD) indicated Thus, this research would suggest that unexpected framing should higher WTP when unexpected framing was used than when it was hurt consumer preferences. An alternate possibility, arising from re- not used (B=4.79, p=.045). Those low on BAS (-1SD) revealed a search on reward motivation, is that unexpected framing could posi- non-significant effect of the benefit-framep ( > .1). tively impact consumer preferences. Specifically, this research raises In Studies1-3, the benefit framed as unexpected was congru- the possibility that when one is in a heightened motivation to seek ent with the product category; that is, it was not unexpected, given rewards, a benefit framed as unexpected might act as a cue leading the product-category. It is likely that the unexpected framing effect prospective consumers to anticipate other potential benefits from the should be attenuated when the unexpected benefit is incongruent product and, thereby, increase their desire to experience the product and, therefore, is already unexpected, given the product-category. (Spanagel and Weiss 1999, McClure, Berns and Montague 2003). We tested this prediction in Study-4. In sum, while research on intentionality suggests that unexpect- Participants imagined that they were buying a body lotion for ed benefit framing should hurt consumer preferences, research on a vacation and saw description for a lotion brand; this description reward motivation suggests the opposite. The current research exam- presented two benefits. While skin-firmness was presented across ines the two opposing predictions. conditions, the second benefit was either even toning (congruent- Study-1 was run in collaboration with a retailer on WeChat. Pro- benefit) or aids in sleeping better (incongruent-benefit). Additionally, spective customers on the retailer’s mobile store were sent an adver- either both the benefits were described as expected or one of the two tisement for a sunscreen, positioned against a hedonic backdrop. Ad benefits, even-toning in the congruent condition and sleep-aid in the presented two benefits of the sunscreen—UVA-rays protection and incongruent condition, was framed as unexpected. Participants indi- even-toning. Prospective customers either received the version of the cated their WTP for the lotion. ad that framed even-toning as unexpected or the version that did not In the congruent-benefit conditions, participants indicated a frame it as unexpected. higher WTP when unexpected framing was used (M=11.08), com- A logistic regression revealed that the purchase rate for the pared with when it was not used (M=8.95; p = .052). In the incon- sunscreen spray was higher when unexpected framing was used gruent-benefit conditions, unexpected framing effect was eliminated (5.70%), compared with when it was not used (4.73%; Wald χ2 = (p >.12). These findings show that unexpected framing is a more 9.96, p=.002). These findings, thus, support predictions arising from effective communication approach, but only when the unexpected research on reward motivation—unexpected framing increased sales benefit is congruent with the product-category. when the reward sensitivity was heightened (when the consumption- Study-5 examined the impact of framing an undesirable charac- context was hedonic). teristic (side effect) as unexpected on product desirability. We exam- Study-2 examined the unexpected framing effect in both hedon- ined whether unexpected framing could negatively impact product ic and utilitarian consumption contexts. Participants imagined that desire for those who have a heightened sensitivity to losses (mea- they were buying a sunscreen for either an upcoming vacation (he- sured by the BIS scale; Carvery and White 1994). donic-consumption context) or a business trip (utilitarian-consump- Participants read a description of a fictitious collagen-supple- tion context; Wadhwa and Zhang 2015). Participants read description ment brand, which informed participants of both the benefits and the of a fictitious sunscreen brand, which communicated two benefits, potential side effects of the supplement. One of the side effects was UVA-rays protection and even-toning. Even-toning was framed ei- either framed as unexpected or neither side-effect was framed as un- ther as unexpected or as expected. Participants indicated their WTP expected. (in USD) for the sunscreen. Participants indicated their likelihood to purchase the supple- In the consumption-context hedonic condition, participants indi- ment and then responded to the BIS scale. A regression showed a cated a higher WTP when unexpected framing was used (M=11.54), significant interaction between BIS and side-effect frame (B=- compared with when it was not used (M=8.11; p =.013). No such .43, p=.071). A spotlight analysis indicated that those high on BIS difference was found for the utilitarian consumption-context (p=.34). (+1SD) indicated lower likelihood to purchase the supplement when Thus, unexpected framing enhanced WTP when the consumption unexpected framing was used, compared with when it was not used context was hedonic, but not when it was utilitarian. (B=-.80, p=.042). A similar spotlight analysis revealed a non-sig- Research suggests that those high, versus low, on behavioral ac- nificant effect of the frame factor for those low on BIS (p>.5). This tivation system (BAS) exhibit heightened reward sensitivity (Carver study shows that framed a negative feature as unexpected can reduce

Advances in Consumer Research 829 Volume 46, ©2018 830 / Unexpected Benefit Framing: Impact of Framing a Benefit as Unexpected on Product Desire product desirability, especially amongst those who have a heightened sensitivity to negative outcomes. The current research delineates when framing a product benefit as unexpected versus not, a task that can be managed by marketers during the process of marketing communication, can enhance the potential consumers’ desire. The Upside of Incompetence: How Discounting Luxury Affects Retailer Price Image Karen Wallach, Emory University, USA Ryan Hamilton, Emory University, USA Morgan Ward, Emory University, USA

EXTENDED ABSTRACT The six services and their prices were presented in a table, along “Luxury” is typically defined partly by its premium pricing with reference prices for the same services at the average hair salon. (Bain and Co. 2014; Grossman & Shapiro 1998; Kim & Ko 2010). In the discounted luxury condition, the price of each of the 3 luxury Theories on luxury management emphasize that consumers draw in- services was less than the average salon prices for each luxury ser- ferences on the basis of the price of luxury offerings, and there may vice, whereas for the discounted non-luxury condition, the price of be a perceived skepticism when luxury products are sold at a lower each of the 3 non-luxury services was less than the average. In each price (Kapferer 2015, Uggla 2017). group, the discounted prices were the same (i.e. $17, $19, $24) and While historically luxury items were rarely discounted, in re- the non-discounted prices were the same (i.e. $30, $32, $35). Thus, cent times, retailers are breaking with tradition. For instance, both the total potential savings in each condition was the same. The order Macy’s and Lord & Taylor have recently announced plans to dis- in which services were presented in the table was randomized across count luxury cosmetics brands like Bobbi Brown and Estée Lauder participants. across all stores (Kapner and Terlep, 2017). Likewise, in a differ- After reviewing the prices of the hair salon, all participants ent category, Carnival Cruise Lines has pursued a similar strategy in rated the overall price image on two items: “How would you rate pricing offerings on its ships. Goods and services from “luxury” cat- the prices at this salon?” and “How would you evaluate the prices at egories (e.g., steak & lobster) tend to be priced low relative to other this salon compared to other salons?” measured on a 10-point scale cruise lines, where more pedestrian offerings (e.g., beach towels and anchored on 1=inexpensive and 10=expensive. These questions were sweatshirts) are comparable to competitors’ prices. collapsed into a single measure of overall price image (α = .826). The purpose of this research is to investigate the effects of dis- Throughout this research, this two-question index serves as the pri- counting luxury products, on the retailer’s overall reputation for pric- mary measure of price image. ing. Specifically, we examine whether discounting luxury offerings As predicted, consumers in the discounted luxury condition will reduce the retailer’s price image more than taking similar price formed a lower price image of the retailer than those in the discount- reductions on non-luxury offerings. ed non-luxury condition. Consistent with this prediction, an analysis We propose that discounting luxury items will indeed have a of variance including discounted luxury vs. discounted non-luxury larger influence on price image, than discounts of the same size on as the predictor variable and overall price image as the dependent non-luxury items. This prediction is based on the notion that, given variable revealed that participants in the discounted luxury condi- the importance of premium prices in defining luxury, retailers that tion had a lower price image of the hair salon than participants in the reduce the price of luxury items will be seen by consumers as less discounted non-luxury condition (Mlux = 3.79 vs. Mnonlux = 4.58; F competent, because they don’t understand the value of their own of- (1, 94) = 4.7, p = .03). ferings. Specifically, we propose that discounting on luxury (versus This finding was replicated in Experiment 1B, in the scenario not luxury goods) will result in lower price image perceptions, which of a cruise line that sold non-branded products from both luxury and is mediated by a reduction in the perceived competence of the retail- non-luxury categories. Three of the items were luxury offerings (i.e. er. We tested these predictions in a series of five experiments across lobster dinner, 3-course steakhouse dinner, bottle of champagne). four distinct retail settings: a liquor store, a hair salon, a cruise line, Three of the items were non-luxury offerings (i.e. cruising tote bag, and a mass retailer. beach towel, souvenir sweatshirt). In each group, the discounted pric- Our first experiment tested our prediction that discounted luxu- es were the same (i.e. $20, $26, $35) and the non-discounted prices ry goods and services are stronger signals of a lower price image than were the same (i.e. $45, $50, $60). Consistent with our results from are similarly discounted non-luxury goods and services. The scenar- Study 1A, data revealed that participants in the discounted luxury io for Experiment 1A was a hair salon, a service provider offering condition had a lower price image of the cruise line than participants non-branded services, which were described in either luxury or non- in the discounted non-luxury condition (Mlux = 4.08 vs. Mnonlux = luxury descriptive terms. We recruited participants for a paid online 4.95; F (1, 99) = 9.211, p = .003). survey through Amazon Mechanical Turk. We randomly assigned The objective of Study 2 was to replicate our main effect in a participants to one of two conditions: discounted luxury goods or different domain and to test the psychological phenomenon behind it. discounted non-luxury goods. Participants were given a hypothetical We further examined the impact of discounting luxury by using the scenario in which they were told to imagine that they were visiting a hedonic category of liquor. An inherent part of luxury is the hedonic new city and stopped into a local hair salon. They were shown a list nature of the products. Indeed, alcohol purchase or consumption is of six services that the hair salon offered. Three of the services were predominantly related to pleasure, rather than utility. For the stimuli luxury services, described using language meant to convey extrava- in Experiment 2, we test to see if the luxury vs. non-luxury brands gance (hair rejuvenating indulgence regimen, relaxation & hydration within this category would still signal a lower price image. In this scalp massage, balayage platinum smoothing treatment). Three of experiment, participants were shown six liquor options, described the services were non-luxury services, described using functional by brand name, size, and price. Three of the brands were luxury of- language (classic bang & neck cut, basic shampoo & style, standard ferings (i.e. Veuve Clicquot champagne, Grey Goose vodka, and St. woman’s cut & trim). Before selecting our stimuli, we performed Germain Elderflower liqueur). Three of the brands were non-luxury a pretest of these hair salon services to measure perceived luxury. offerings (Popov Vodka, Everclear Grain alcohol, and Pabst Blue Results revealed that participants considered the luxury services to Ribbon). Pretests and a manipulation check confirmed that these indeed be more of a luxury than the non-luxury services. brands were appropriately categorized as luxury and non-luxury.

In each group, the discounted prices were the same (i.e. $16, $18, image than did discounts of the same size on non-luxury items (Mlux $21) and the non-discounted prices were the same (i.e. $26, $30, = 4.11 vs. Mnonlux = 4.95; F (1, 266) = 15.122, p = .00). This experi- $36). Bottle sizes were manipulated to ensure appropriate pricing to ment also replicated and extended the mediation results from Experi- marketplace standards across discounted luxury and discounted non- ment 2. The results showed a significant indirect effect for a media- luxury conditions. For example, a 375 ml bottle of Veuve Clicquot tion path through perceived competence and perceived discount size: champagne and a 1.5 L bottle of Popov vodka were both $30. discounted luxury items (vs. non-luxury items) resulted in a decrease Consistent with our predictions, an analysis of variance includ- in perceived competence of the retailer, which in turn increased the ing discounted luxury vs. discounted non-luxury as the predictor size of the perceived discount at the retailer, ultimately leading to a variable and overall price image as the dependent variable revealed lower price image. that participants in the discounted luxury condition had a lower price This research contributes to the literature on behavioral pricing image of the liquor store than participants in the discounted non- by identifying discounted luxury as a category with a disproportion- luxury condition (Mlux = 4.07 vs. Mnonlux = 4.59; F (1, 144) = 4.76, ate influence on a retailer’s price image. For retailers looking for p = .031). efficient ways to lower their price image, this research suggests that Subsequently, we measured perceived competence using an in- discounting luxury items may be an effective strategy. Using a se- dex of 3 questions based on specific personality traits from Aaker’s rial mediation model, we found that perceived competence causes (1997) brand personality scale. The 3 questions measured on 1-10 the price discounts to seem larger, resulting in a lower retailer price Likert scales were: (1) “How intelligently do you think this retailer image. However, this strategy comes with a cost to the retailer: re- is managing prices?” (2) “What do you think of this retailer’s strat- ductions in the retailer’s perceived competence in response to these egy?” and (3) “What do you think of this store’s owner?” After re- discounts. verse coding the answers to #1, the 3 questions were collapsed into a simple measure of overall perceived competence (α = .89). The data REFERENCES revealed that perceptions of competence significantly mediated the Aaker, J. L. (1997). Dimensions of Brand Personality. Journal of relationship between our conditions (discounted luxury/discounted Marketing Research, 34(3), 347. non-luxury) and overall price image. In a bootstrap analysis, we Bain & Co (2014). World luxury market report. Paris: Bain & Co. found that the 95% bias-corrected confidence interval for the overall Hamilton, R., & Chernev, A. (2010). The impact of product line size of the indirect effect excluded zero (95% CI = .0755 to .4555), extensions and consumer goals on the formation of price suggesting a significant indirect effect. Our mediation analysis dem- image. Journal of Marketing Research, 47(1), 51–62. onstrated that the overall price image of the retailer was indeed medi- Desai, K. K., & Talukdar, D. (2003). Relationship between product ated by the perceived competence of the retailer. groups’ price perceptions, shopper’s basket size, and grocery Experiment 3 featured a 2 (time duration: short term vs. long store’s overall store price image. Psychology & Marketing, term) x 2 (discounted group: luxury vs. non-luxury) between-sub- 20(10), 903–933. jects design. Participants were shown 6 prices from a hypothetical Grewal, Dhruv and Michael Levy (2007), “Retailing Research: department store that carried both luxury (e.g. Versace jacket) and Past, Present, and Future,” Journal of Retailing, 83 (4), non-luxury (e.g. Rug Doctor vacuum) goods. In each group, the dis- 447–64. counted prices were the same (i.e. $335, $400, $550) and the non- Grossman, G. M., & Shapiro, C. (1998). Counterfeit-product trade. discounted prices were the same (i.e. $670, $800, $1100). One objec- American Economic Review, 78, 59–75. tive of Experiment 3 was to see if the duration of the discount (short Kapferer J N (2015), Kapferer on Luxury: How Luxury Brands Can vs. long term) would impact the inferences of price image. In the Grow Yet Remain Rare, Kogan Page, England. short-term condition, the discount was available for only a limited Kapner, Suzanne, and Sharon Terlep (2017), “The Cosmetics time, while in the long-term condition, the discount was discussed Counter Was Long Immune to Discounting. Not Anymore.” as being part of a permanent pricing strategy for the retailer. As it The Wall Street Journal (July 2017), (accessed July 18, 2017). turned out, the length of the discount did not affect participants price Kim, A. J., & Ko, E., (2010). Impacts of luxury fashion brand’s image perceptions: short vs. long term pricing strategies were not social media marketing on customer relationship and purchase significantly different (Mshort = 4.59 vs. Mlong = 4.46; F (1, 266) = intention. Journal of Global Fashion Marketing, 1, 164–171. .367, p = .545). Uggla, H. (2017). The Price of Luxury. Journal of Brand Experiment 3 replicated the central finding of the previous ex- Management, 14(2). periments: discounting luxury items resulted in a lower overall price Perceptions of Epistemic vs. Aleatory Uncertainty Affect Stock Investment Daniel Walters, INSEAD, France Gülden Ülkümen, University of Southern California, USA Carsten Erner, FS Card, Inc., USA David Tannenbaum, University of Utah, USA Craig R. Fox, UCLA, USA

EXTENDED ABSTRACT In Study 3, we manipulated perceptions of stock uncertainty The decision to invest in a stock entails uncertainty. Prior work by asking participants to read a short article about the importance shows that people intuitively distinguish two independent dimen- of knowledge in investing (which enhanced perceived epistemic sions of uncertainty: (i) “epistemic” uncertainty that is seen as in- uncertainty) or how experts are often incorrect (which diminished herently knowable, given sufficient information; and (ii) “aleatory” perceived epistemic uncertainty). Next, as part of an ostensibly dif- uncertainty that is seen as inherently random. We hypothesized that: ferent study, participants invested $50 over the next 6 months across (1) as investors perceive more epistemic uncertainty in stocks, they three unfamiliar stocks. Participants then were offered research re- will be more sensitive their own level of ignorance, and will man- ports on the three companies for $2, which would be deducted from age this uncertainty by seeking expert advice; and (2) as investors the value of their investments at the end of 6 months. Finally, par- perceive more aleatory uncertainty in stocks, they will be more sen- ticipants made an investment allocation across the three companies. sitive to their exposure to risk, and will manage this uncertainty by As predicted, priming participants with higher (vs. lower) epistemic diversifying their portfolios. We tested these predictions using four uncertainty led them to be more (less) likely to pay for financial in- incentive-compatible lab studies, one survey of self-reported invest- formation. Moreover, perceived epistemic uncertainty partially me- ment behaviors from lay investors, and one survey of practicing fi- diated this relationship, whereas aleatory uncertainty did not. nancial advisors. In Study 4, we predicted that investors would be more sensi- In Study 1A, 354 investors with at least $1,000 in stock invest- tive to financial information to the extent that they perceived stock ments assessed their perceptions of stock market uncertainty using uncertainty to be epistemic (but not aleatory). Participants indicated the Epistemic-Aleatory Rating Scale (EARS; Fox et al., 2018) and whether they thought Apple stock in 6 months would be: (1) above reported their own investment behaviors. Confirming our predic- its current price of $140 or (2) the same or below as its current price tions, investors who perceive stock market uncertainty to be more of $140, and then chose between: (a) $50 for sure, or (b) $150 if epistemic were more likely to pay for financial advice and traded their prediction about Apple is correct. Next, participants were pre- stocks more frequently. In contrast, investors who perceive stock un- sented with a real research report predicting that Apple stock would certainty as more aleatory held less concentrated portfolios (i.e. a appreciate in the coming months, and then repeated their Apple fore- greater number of distinct stocks). cast and investment choice. As predicted, company information had Study 1B further examines the perceptions of uncertainty across greater impact on willingness to invest in Apple among participants experts and lay investors. To do this, we compared perceptions of un- who rated stock uncertainty as more epistemic; meanwhile, aleatory certainty of lay investors in study 1A to those of 37 practicing finan- uncertainty had no such effect. cial advisors (recruited through an executive education program at In Study 5, we explored the impact of perceived aleatory uncer- UCLA) who also evaluated stock market investing using the EARS. tainty on behavior. We first assessed risk preference by asking partic- In a separate sample t-test we find that financial advisors perceived ipants to choose among chance gambles. We next presented partici- lower epistemic uncertainty in stock investing, compared to non-pro- pants one-paragraph descriptions of eight companies (from Reuters), fessional investors. However, perceptions of aleatory uncertainty did then asked them to assess uncertainty concerning each stock’s return not differ between financial advisors and non-professional investors. over the coming week on the EARS. Finally, participants predicted This result suggests that less experienced investors may be overesti- whether each stock would appreciate more or less than the S&P 500 mating the predictability of the stock market. over the coming week, and chose between: (a) $30 for sure, or (b) In Study 2, we recruited 201 participants from a Qualtrics panel, $90 if their stock forecast is correct. As predicted, willingness to ac- who read the quarterly earnings release for two companies. We except investment prospects significantly increased with the interaction perimentally manipulated epistemic uncertainty by having partici- between rated aleatoriness of stocks and the participant’s risk toler- pants consider future company performance by elaborating on either ance; meanwhile there was no significant interaction between rated a single outcome (singular reasoning) or a set of possible alternative epistemicness and risk tolerance. This finding suggests that to the outcomes (distributional reasoning), (Tannenbaum, Fox, and Ülkü- extent that an investor sees stock movements as random, her willing- men, 2017). Participants then estimated the revenues of these com- ness to bet on stock movements more closely resembles her prefer- panies in the same quarter following year, and place 90% confidence ence to bet on chance outcomes. intervals around their estimates. We had the participants rate their In sum, we find that investment behaviors are influenced by in- uncertainty about the following year revenue using the EARS. Next, vestors’ subjective perception of the two independent dimensions of we asked participants to imagine investing $100 dollars between the uncertainty. Our results across six studies show that investors who two companies. Some participants were randomly selected to receive believe that stock returns are relatively more epistemic seek informa- the actual payout of their investment decisions after six months. tion and expert advice to manage their uncertainty, whereas investors Confirming our prediction, participants primed with aleatory who believe that stock returns are relatively more aleatory diversify uncertainty provided wider confidence intervals, than those primed their assets to manage their uncertainty. These findings hint at the with epistemic uncertainty participants. Also as predicted, partici- possibility that shifting beliefs about epistemic and aleatory market pants primed with aleatory uncertainty concentrated their invest- uncertainty may prove to be a useful intervention for improving in- ments less than participants primed with epistemic uncertainty. vestment decisions.

Advances in Consumer Research 833 Volume 46, ©2018 834 / Perceptions of Epistemic vs. Aleatory Uncertainty Affect Stock Investment REFERENCES Tannenbaum, David, Craig R. Fox, and Gülden Ülkümen (2017), Fox, Craig R., David Tannenbaum, Gülden Ülkümen, Dan Walters, “Judgment Extremity and Accuracy under Epistemic and Carsten Erner (2018), “The Epistemic-Aleatory Rating versus Aleatory Uncertainty,” Management Science, 63(2) Scale (EARS) in Judgment and Choice,” UCLA Anderson (February), 497-518. School working paper series. To Trace is to Trust: From Product Traceability to Brand Trust Jing Wan, University of Groningen, The Netherlands Pankaj Aggarwal, University of Toronto, Canada Min Zhao, Boston College, USA

EXTENDED ABSTRACT actually “traced” production/origin information about the milk; the As consumers become more conscious of and concerned with information consisted of positive or negative facts about the produc- the ingredients that constitute their products and how their purchases tion process. The brand was rated on its trustworthiness. The pres- can impact the environment, they are paying more attention to the ence of the traceable label improved trust ratings compared to con- origin of their products and the process by which these products are trol, with or without positive production information, suggesting that made. The present paper examines how merely labelling products as participants infer positive production processes in the presence of a “traceable” to their origin can help brands establish their transpar- traceable label, even without factual evidence. Negative information ency and build trust between consumers and brands. decreased trust compared with the trace label and positive-trace con- The concept of tracing products to their origin has gained trac- ditions; however, the negative-trace condition did not significantly tion over the last two decades (e.g., knowing where/how the product differ from the control condition, suggesting that providing negative was made; van Rijswijk and Frewer 2008). Although traceability can information about production practices under the label of traceability evoke thoughts of safety and quality to consumers (van Rijswijk et does not decrease trust compared with giving no information at all. al. 2008), the ability to trace products does not necessarily provide Experiment 2b used a 2 traceability label (absent vs. present) consumers with actual safety information or the exact production x 2 valence of information (positive vs. negative) between subjects process (Hobbs et al., 2005). Currently, there is little research in the design. All participants received information about the milk brand’s marketing domain of how consumers react to traceability labels and production process that was either positive (above standard) or nega- how brands should make use of these labels to build relationships tive (below standard). Half of the participants saw a traceability label with consumers. before the production information. The other half received the pro- Individuals infer that their relationship with a partner is more duction information without the traceable label. Participants rated the intimate and close when personal information is divulged (human-to- brand on its trustworthiness. Adding a traceable label to the positive human interactions: Aron et al. 1997; Laurenceau et al. 1998; human- information did not change trust ratings; however, when the produc- to-computer: Moon 2013). Similarly, the presence of a traceability tion information was negative, preceding the negative information label signals the willingness of a brand to disclose information that with a traceable label increased trust ratings compared with showing is typically kept private, allowing consumers to infer that they may the negative information alone have a closer and more trusting relationship with the brand. Further, Experiment 3 investigates if traceability increases consumers’ research on signaling has indicated that consumers infer low store willingness to reciprocate by sharing their own information with the price from price-matching policies (Srivastava and Lurie 2001), sug- brand, through increased brand trust. Participants read about a choc- gesting that a mere signal in the absence of factual information can olate brand, and the chocolate was either traceable or not. Partici- still induce positive evaluations. In addition, consumers feel more pants perceived the brand offering traceability as more trustworthy confident about products when they know they can retrieve infor- and they were more willing to share their personal information with mation about them from the Internet (Bhargave et al. 2016). Hence, the traceable brand. Level of trust mediated the relationship between we propose that in the presence of a traceable label, which gives presence of traceability and reciprocation. consumers the option to retrieve information about production origin Experiment 4 examines whether the mere presence of traceabil- online, consumers would infer greater transparency of the produc- ity can improve overall product evaluation, through increased brand tion process, more positive qualities about the brand and its products, trust. Participants read about a baking soda brand, and the baking even in the absence of actual information. Five experiments were soda was either traceable or not. Participants perceived the brand as conducted to confirm this predicted effect. more trustworthy and the product as more desirable when the baking Experiment 1 tests how the traceability label, in the absence of soda was traceable. Level of trust mediated the relationship between actual production information, can affect quality perceptions of the presence of traceability and product evaluation. brand. Further, it tests how a traceability label which does not convey Overall, although the mere presence of traceability is not associ- any ethical information measures up against an ethical label like fair ated with any specific ethical practice, consumers infer that the will- trade. Participants read about a chocolate brand and were assigned to ingness of a brand to potentially disclose private information must one of three conditions: control (no production information), trace- signal transparent and socially responsible practices, similar to actual able label (label-only, signaling that tracing is a possibility), and ethical labels such as fair trade. Brands that employ traceability la- fair trade label. The brand was rated on its trustworthiness, honesty, bels indicate to consumers that they can retrieve production informa- transparency, and social responsibility. Participants in both treatment tion on-demand, and hence, the mere presence of the label is enough conditions reported significantly higher brand trust scores compared to evoke trust within consumers. Our research sheds insight into the with the control condition; in addition, participants viewed the trace- consumer inference process and offers practical recommendations to able brand just as trustworthy as the fair trade brand. brands on strengthening their relationships with their customers. Experiment 2a and 2b examine how providing traceable negative information affects trust and whether or not having a traceable REFERENCES label can mitigate the negative effects of poor production processes. Aron, A., Melinat, E., Aron, E., Vallone, R., & Bator, R. (1997). In Experiment 2a, participants read about a milk brand and were as- The Experimental Generation of Interpersonal Closeness: A signed to one of four conditions: control, traceable label (label-only), Procedure and Some Preliminary Findings. Personality and positive traceable information, and negative traceable information. Social Psychology Bulletin, 23(4), 363-377. In the latter two conditions, participants saw the traceable label and

Bhargave, R., Mantonakis, A., & White, K. (2016). The Cue-of- Moon, Y. (2003). Don’t Blame The Computer: When Self- the-Cloud Effect: When Reminders of Online Information Disclosure Moderates The Self-Serving Bias. Journal of Availability Increase Purchase Intentions and Choice. Journal Consumer Psychology, 13(1), 125-137. of Marketing Research, 53(5), 699-711. Srivastava, J. & Nicholas L. (2001). “A consumer perspective on Hobbs, J., Bailey, D., Dickinson, D., & Haghiri, M. (2005). price-matching refund policies: Effect on price perceptions Traceability in the Canadian Red Meat Sector: Do Consumers and search behavior.” Journal of Consumer Research, 28(2), Care?. Canadian Journal of Agricultural Economics/Revue 296-307. Canadienne D’agroeconomie, 53(1), 47-65. van Rijswijk, W., & Frewer, L. (2008). Consumer perceptions of Laurenceau, J., Barrett, L., & Pietromonaco, P. (1998). Intimacy as food quality and safety and their relation to traceability. British an interpersonal process: The importance of self-disclosure, Food Journal, 110(10), 1034-1046. partner disclosure, and perceived partner responsiveness in van Rijswijk, W., Frewer, L., Menozzi, D., & Faioli, G. (2008). interpersonal exchanges. Journal of Personality and Social Consumer perceptions of traceability: A cross-national Psychology, 74(5), 1238-1251. comparison of the associated benefits.Food Quality and Preference, 19(5), 452-464. Show Me More! Powerlessness Drives Variety Seeking Wangshuai Wang, Shanghai University of International Business and Economics, China Raj Raghunathan, University of Texas at Austin, USA Dinesh Gauri, University of Arkansas, USA

EXTENDED ABSTRACT you like blue and thus often wear blue ones (Warren and Campbell, People play different power roles every day. A stern boss who 2014). We followed Levav and Zhu (2009) to measure variety seek- runs a company may be a docile son in the presence of his parents; a ing by asking participants to indicate their likelihood of donating to tender wife at home can turn into a tough professor when supervising unfamiliar charities. her students. The rapid switch of power state leads to an interesting A two-way ANOVA yielded a significant interaction effect (F and important research question, i.e., the downstream consequences (2,133) = 3.12, p < .05). A contrast analysis revealed significant of power on human behaviors. Yet, little research speaks to the po- differences in the need for autonomy unmet condition (F (2,133) = tential effect of power on variety seeking behavior, leaving an in- 5.68, p < .001), such that low power participants had a higher variety triguing question unexplored. seeking intention compared with those in the high and control power We propose that feeling powerless will enhance variety-seeking conditions. In contrast, variety seeking behavior did not vary across behavior and that need for autonomy accounts for this effect. Prior different power levels when need for autonomy was satiated. research provides initial theoretical support for these hypotheses. Study 5 provides evidence using scanner data. The data comes For example, researchers consistently find that feeling powerless from Nielsen Scanner Panel data for one year on two product catego- is aversive (e.g., Keltner, Gruenfeld, and Anderson 2003; Rucker ries: Yogurt and Laundry Detergent. We used occupation information and GalinskY 2008). In particular, Lammers et al. (2016) show that as the indicator of power (Rucker, Galinsky, and Dubois 2012). To powerless experiences threaten individuals’ perceived autonomy and classify variety seekers, we followed Trivedi and Morgan (2003) by freedom. Furthermore, Levav and Zhu (2009) document that free- computing the number of distinct brands that the panelists purchased. dom deprivation leads to variety seeking. We conducted five studies The results showed that in high power occupations, there were to test these hypotheses. more low variety seekers compared to high variety ones (55.71 % vs. In Study 1, 77 consumers were randomly assigned to high ver- 43.06 % in Yogurt; 54 % vs. 38.3 % in Laundry Detergent) while in sus low power conditions. First, they recalled an experience in which low power occupations, there were more high variety seekers com- they had or lacked power (Galinsky, Gruenfeld, and Magee 2003). pared to low variety ones (31.94 % vs. 20 % in Yogurt; 51.06 % Then we measured need for autonomy. After that, participants were vs. 32 % in Laundry Detergent). We then estimated an ordered logit told to imagine they would be served one box of chocolate in the model and found a significant impact of high versus low power occu- following 5 weeks and decide which one they would like to have for pations (.55, p < .10 for Yogurt; .62, p < .05 for Laundry Detergent). each week (Huang and Wyer 2015). Results revealed that low-power The findings from the current research carry important theoreti- participants sought more variety than high-power ones (F (1, 75) = cal contributions. First, while much contemporary literature exam- 4.78, p < .05). Furthermore, a 5,000 bootstrap revealed a significant ines the interpersonal consequences of power, the current research is indirect effect via need for autonomy, 95% CI = [–.26, –.01]. one of a burgeoning stream of studies that investigate how power ex- In Study 2, 48 consumers were randomly assigned to high ver- erts an influence on intrapersonal consumption behavior. Second, we sus low power conditions. In this study, we adopted a more subtle add to the literature on compensatory consumption by demonstrating manipulation of power (Chen, Lee-Chai, and Bargh 2001), whereby how variety seeking serves as a subtle way to deflect powerless state. participants either sat in a tall chair (high power condition) or on an Third, our research contributes to the variety seeking literature by ottoman (low power condition). Then they completed a survey that documenting consumers’ power state as a new psychological deter- included questions measuring a general interest in variety seeking minant of variety seeking behavior. (Fishbach, Ratner, and Zhang 2011). Specifically, participants indicated whether they preferred the same or different products in various REFERENCES consumption situations. Results revealed that low-power participants Chen, Rui, Yuhuang Zheng, and Yan Zhang (2016), “Fickle Men, sought more variety than high-power ones (F (1, 46) = 3.87, p = .05). Faithful Women: Effects of Mating Cues on Men’s And In Study 3, 62 undergraduate students were randomly assigned Women’s Variety-Seeking Behavior in Consumption,” Journal to high versus low power conditions. Another manipulation of power of Consumer Psychology, 26(2), 275–82. was adopted (Jin, He, and Zhang 2014). Specifically, each participant Chen, Serena, Annette Y. Lee-Chai, and John A. Bargh (2001), was told that he or she was best suited for the role of manager (high “Relationship Orientation as a Moderator of the Effects of power) or subordinate (low power). Then they were asked to choose Social Power,” Journal of Personality and Social Psychology, their gifts for participating in the experiment. We provided four fla- 80 (2), 173–87. vors of lollipops as gifts and they could choose three lollipops of any Fishbach, Ayelet, Rebecca K. Ratner, and Ying Zhang (2011), flavors (Chen, Zheng, and Zhang 2015). Again, results revealed that “Inherently Loyal or Easily Bored? Nonconscious Activation low-power participants sought more variety than high-power ones (F of Consistency versus Variety-Seeking Behavior,” Journal of (1, 60) = 3.32, p = .07). Consumer Psychology, 21 (1), 38–48. 139 undergraduate students participated in Study 4. They were Galinsky, Adam D., Deborah H. Gruenfeld, and Joe C. Magee randomly assigned to the cells of a 3 (power: high vs. low vs. control) (2003), “From Power to Action,” Journal of Personality and × 2 (need for autonomy: satiated vs. unmet) between-subjects de- Social Psychology, 85 (3), 453–66. sign. In the high (low) power condition, participants were told to imagine being a boss (employee) at a company (Rucker, Dubois, and Galinsky 2011). Then participants in the satiated condition were told to imagine: It is a social norm to wear black clothes. However,

Huang, Zhongqiang Tak, and Robert S. Wyer Jr (2015), “Diverging Rucker, Derek D., and Adam D. Galinsky (2008), “Desire to Effects of Mortality Salience on Variety Seeking: The Acquire: Powerlessness and Compensatory Consumption,” Different Roles of Death Anxiety and Semantic Concept Journal of Consumer Research, 35 (2), 257–67. Activation,” Journal of Experimental Social Psychology, 58, Rucker, Derek D., David Dubois, and Adam D. Galinsky (2011), 112–23. “Generous Paupers and Stingy Princes: Power Drives Jin, Liyin, Yanqun He, and Ying Zhang (2014), “How Power States Consumer Spending on Self versus Others,” Journal of Influence Consumers’ Perceptions of Price Unfairness,” Consumer Research, 37 (6), 1015–29. Journal of Consumer Research, 40 (5), 818–33. Rucker, Derek D., Adam D. Galinsky, and David Dubois (2012), Keltner, Dacher, Deborah H. Gruenfeld, and Cameron Anderson “Power and Consumer Behavior: How Power Shapes Who and (2003), “Power, Approach, and Inhibition,” Psychological What Consumers Value,” Journal of Consumer Psychology, 22 Review, 110 (2), 265–84. (3), 352–68. Lammers, Joris, Janka I. Stoker, Floor Rink, and Adam D. Galinsky Trivedi, Minakshi, and Michael S. Morgan (2003), “Promotional (2016), “To have control over or to be free from others? The Evaluation and Response among Variety Seeking desire for power reflects a need for autonomy,” Personality and Segments,” Journal of Product & Brand Management, 12(6): Social Psychology Bulletin, 42(4), 498-512. 408–25. Levav, Jonathan, and Rui (Juliet) Zhu (2009), “Seeking Freedom Warren, Caleb, and Margaret C. Campbell (2014), “What through Variety,” Journal of Consumer Research, 36 (4), Makes Things Cool? How Autonomy Influences Perceived 600–10. Coolness,” Journal of Consumer Research, 41(2): 543-63. Dimming the Light Offers A Creative Lens: The Impact of Ambient Illuminance on Creativity Assessment Chen Wang, Drexel University, USA Ravi Mehta, University of Illinois at Urbana-Champaign, USA Rui (Juliet) Zhu, Cheung Kong Graduate School of Business, China Jennifer Argo, University of Alberta, Canada

EXTENDED ABSTRACT pants were presented with a survey packet containing five print ad- Creative ideas are prevalent in everyday consumption. The cre- vertisements, one per page in random order. They were asked to rate ative professionals always hope that consumers could understand the creativity of the advertisements. As hypothesized, we found that and acknowledge the creativity displayed in their work. But is it al- those in the dim (vs. bright) condition assessed the presented adver- ways the case? In other words, can consumers usually get the “Aha!” tisements more creative. moment when viewing a creative work? Unfortunately, creativity is Study 2 examined the underlying process by testing the mediat- only in the eye of the beholder. As consumers are not directly ining role of inhibition. Participants completed the study in either a volved in the creation process, oftentimes they cannot share the per- dimly (150 lux) or a brightly (1500 lux) lit room. To measure inhibi- spectives of the creators and appreciate their creativity (West, Kover, tion, they were asked to complete Hayling’s sentence completion test & Caruana, 2008). (Burgess & Shallice, 1996; Kemps & Wilsdon, 2010). Specifically, This has raised a fundamental yet important question: How to participants were presented with two sets of sentences in which the make consumers assess creativity in a more effective way? Creative last word was omitted. For each sentence, there was a high prob- assessment is the attempt to identify or recognize creative character- ability of one particular response (e.g., “George must keep his dog istics in the assessment target (Hennessey & Amabile, 1999). While on a _____”). In the first set, participants were asked to provide a existing research has suggested some solutions such as inviting con- response that would complete the sentence in a reasonable way (e.g., sumers into the creation process (Kover, et al., 1997), it usually re- “leash”). By contrast, in the second set, they were asked to provide quires vast investment in labor and finance. Given that contextual a word that made no sense (e.g., “book”). That is, they must inhibit variables are usually inexpensive and easy to control, if researchers the rational response that easily comes to mind to have a correct an- can start to identify ambient factors that impact consumers’ creativity swer. Participants were instructed to respond as quickly as possible assessment, it would offer valuable insights into how to help con- within 60 seconds for each set. Hence, participants’ inhibition level sumers appreciate innovation, and ultimately enhance their product was scored as the ratio of the number of correctly completed sen- desirability and purchase intention (Horn & Salvendy, 2009). The tences in the second set (i.e., nonsensical responses) on the number current research aims to address this void by being the first work to of correctly completed sentences in the first set (i.e., reasonable re- identify ambient illuminance as a contextual variable to influence sponses), controlling for the individual difference in response rate. A consumers’ creative assessment. higher score indicated greater inhibition (Kemps & Wilsdon, 2010). The central thesis is that dim (vs. bright) illuminance reduces Next, all participants assessed 10 novel products on how creative individual’s inhibition, which subsequently leads to higher creativ- they thought they were. As expected, those in the dim (vs. bright) ity assessments. When the illuminance is dim (vs. bright), the mind condition rated the products more creative. Further, their inhibition is less alert to distinguish between differential concepts, and thus score mediated the effect of illuminance on creativity assessment. less likely to filter and restrain (i.e., inhibit) irrelevant thoughts or Finally, Study 3 offered additional evidence of the underlying concepts. Prior research has provided support for this argument. For process by examining the joint effect of illuminance and inhibition example, it has been shown that darkness induces a sense of illu- (measured as trait Openness to Experience) on creativity assessment. sory anonymity thereby reducing inhibition (Zhong et al., 2010). Re- Participants first completed the Openness to Experience scale (John search on sleep behaviors also suggests that dim illuminance helps & Srivastava 1999) in a room with typical illuminance (450 Lux). sleepiness (Lowden, Åkerstedt, & Wibom, 2004) by slowing down Next, they were led to a room where illuminance was either dim or physiological reactions in terms of heart rate, body temperature, bright. Here they completed a product evaluation task, in which they and cortisol production, which are highly correlated with alertness indicated willingness to buy a new and innovative product over a and inhibition (Lockley et al., 2006). We next argue that such re- traditional one for five pairs of products, a behavioral measure of cre- duced inhibition subsequently enhances creativity assessment. This ativity assessment. We found that individuals who were high in open- is because reduced inhibition enables people to loosen the boundar- ness (i.e., chronically low on inhibition) showed higher willingness ies between concepts and consequently embrace less filtering and to buy innovative products irrespective of the illuminance condition. more open-minded processing (Enright & Beech, 1993; Green & However, replicating results form previous studies, dim (vs. bright) Williams, 1999). As a result, when consumers are presented with illuminance induced higher likelihood of buying innovative products creative ideas, they likely do not consider these concepts isolated for low openness (i.e., chronically high on inhibition) individuals. and distinct. Rather, they are able to recognize and understand the Taken together, this research advances prior work by demon- underlying remote connections between those distal concepts. Such strating the impact of illuminance on consumer’s creativity assess- comprehension of the latent inter-conceptual links enables them to ment and providing evidence of the underlying mechanism. Our give higher creativity ratings of the assessment target. A set of three findings also have important practical implications for creative pro- studies provides support for our hypotheses. fessionals and marketers selling novel and innovative products. Study 1 demonstrates the main effect of ambient illuminance on creativity assessment. The ambient illuminance was manipulated REFERENCES by the number of lights in the room. Specifically, in the dim (bright) Burgess, P. W., & Shallice, T. (1996). Response suppression, condition, the room was lit by 2 (20) fluorescent light tubes mounted initiation and strategy use following frontal lesions. to the ceiling, with the total luminance of 150 (1500) lux. Partici- Neuropsychologia, 34(4), 263–273.

Enright, S. J., & Beech, A. R. (1993). Further evidence of reduced Kover, A. J., James, W. L., & Sonner, B. S. (1997). To whom do cognitive inhibition in obsessive-compulsive disorder. advertising creatives write? An inferential answer. Journal of Personality and Individual Differences, 14(3), 387-395. Advertising Research. 37(1), 41-53. Green, M. J., & Williams, L. M. (1999). Schizotypy and creativity Lockley, S. W., Evans, E. E., Scheer, F. A., Brainard, G. C., as effects of reduced cognitive inhibition.Personality and Czeisler, C. A., Aeschbach, D. (2006). Short-Wavelength Individual Differences, 27, 263-276. Sensitivity for the Direct Effects of Light on Alertness, Hennessey, B. A., & Amabile, T. M (1999). Consensual assessment. Vigilance, and the Waking Electroencephalogram in Humans. In M.A. Runco & S.R. Pritzker (Eds.), Encyclopedia of Sleep, 29(2), 161–168. Creativity (pp. 347-361). San Diego, CA: Academic Press. Lowden, A., Åkerstedt, T., & Wibom, R. (2004). Suppression of Horn, D, & Salvendy, G. (2009). Measuring Consumer Sleepiness and Melatonin by Bright Light Exposure During Perception of Product Creativity: Impact on Satisfaction Breaks in Night Work. Journal of Sleep Research, 13, 37–43. and Purchasability. Human Factors and Ergonomics in West, D. C., Kover, A. J., & Caruana, A. (2008). Practitioner and Manufacturing, 19 (3), 223-240. Customer Views of Advertising Creativity: Same Concept, John, O. P., & Srivastava, S. (1999). The Big-Five trait taxonomy: Different Meaning?Journal of Advertising, 37(4), 35-45. History, measurement, and theoretical perspectives. In L. A. Zhong, C., Bohns, V. K., & Gino, F. (2010). A good lamp is the Pervin & O. P. John (Eds.), Handbook of personality: Theory best police: Darkness increases self-interested behavior and and research (Vol. 2, pp. 102-138). New York: Guilford Press. dishonesty. Psychological Science, 21(3), 311-314. Kemps, E., & Wilsdon, A. (2010). Preliminary evidence for a role for impulsivity in cognitive disinhibition in bulimia nervosa. Journal of Clinical and Experimental Neuropsychology, 32(5), 515-521. Restrained by Resources: The Joint Effect of Scarcity Cues and Childhood Socioeconomic Status (SES) on Consumer Preference for Feasibility Lili Wang, Zhejang University, China Yanfen You, New Mexico State University, USA Chun-Ming Yang, Ming Chuan University, Taiwan

EXTENDED ABSTRACT those with a relatively high childhood SES (+1 SD above mean), Consumers’ current socioeconomic status (SES) has been construal level did not mediate the effect of scarcity cues on product widely acknowledged as an important factor that affects various as- preference. pects of their behavior, such as aggression (Greitemeyer and Sagio- Study 4 further tested the mediation role of construal level by glou 2016), prosocial behavior (Piff et al., 2010) and variety-seeking directly manipulating construal level. Taking low construal level (Yoon and Kim 2017). However, scant marketing literature examines condition as the base group, analysis revealed a significant three- whether and how consumers’ childhood SES affects their behavior. way interaction among resource availability, childhood SES and Compared to consumers’ current SES, childhood SES represents construal level. Specifically, in the control condition, we replicated consumers’ early life environments and functions more like a latent the previous interaction effect. In the low construal level condition, factor. Thus, once activated by situational factors, the behavior pat- only the main effect of scarcity cues was significant. However, in terns that people have formed in childhood are likely to affect their the high construal level condition, scarcity cues led people with low decision-making regardless of their current SES. Therefore, it is im- childhood SES to prefer low feasibility. These results speak to the portant to put childhood SES in the spotlight and to examine its inter- important role of construal level in driving low childhood-SES con- active effect with important situational factors in shaping consumer sumers’ preference after exposure to scarcity cues. behavior. Across four experiments, we examined how childhood SES To address this void, we study the interplay of childhood SES and resource scarcity cues jointly affect consumers’ construal level and resource scarcity cues. Specifically, drawing from research on and their subsequent product choices. This research contributes to scarcity cues (Mehta and Zhu 2016; Roux, Goldsmith, and Bonezzi research on childhood SES, construal level theory, and resource scar- 2015), life history theory (Griskevicius et al. 2013; Laran and Saler- city cues. no 2013; Mittal and Griskevicius 2014), and construal level theory (Liberman and Trope 1998), we propose and demonstrate that the REFERENCES combination of low childhood SES and resource scarcity cues leads Greitemeyer, Tobias and Christina Sagioglou (2016), “Subjective to a low construal level and thus propels consumer preference for Socioeconomic Status Causes Aggression: A Test of the feasibility in product choices. Theory of Social Deprivation.,” Journal of Personality and Study 1 consisted of a three-level between-subjects design (re- Social Psychology, 111(2), 178–94. source availability: scarcity vs. abundance vs. control) with child- Liberman, Nira and Yaacov Trope (1998), “The Role of Feasibility hood SES measured continuously. Results suggested that, for partici- and Desirability Considerations in near and Distant Future pants with lower childhood SES (-1 SD blow mean), scarcity cues Decisions: A Test of Temporal Construal Theory.,” Journal of (vs. the control condition) increased their preference for feasibility. Personality and Social Psychology, 75(I), 5–18 No significant difference was found between the abundance and con- Mehta, Ravi and Meng Zhu (2016), “Creating When You Have trol conditions. In contrast, for those with higher childhood SES (+1 Less: The Impact of Resource Scarcity on Product Use SD above mean), there was no significant difference in preference Creativity,” Journal of Consumer Research, 42(5), 767–82. for feasibility among the three conditions. As indicated by Study 1, Mittal, Chiraag and Vladas Griskevicius (2014), “Sense of there was no significant difference between the control and abun- Control under Uncertainty Depends on People’s Childhood dance conditions and that our effect was driven by scarcity. Thus, we Environment: A Life History Theory Approach.,” Journal of did not include the control condition in other studies. Personality and Social Psychology, 107(4), 621–37. Study 2 utilized counterfeit products to represent products with Piff, Paul K., Michael W. Kraus, Stéphane Côté, Bonnie Hayden feasibility (Jiang and Cova 2012). Results revealed a significant in- Cheng, and Dacher Keltner (2010), “Having Less, Giving teraction between childhood SES and resource availability. Flood- More: The Influence of Social Class on Prosocial Behavior.,” light analysis indicated that participants who scored lower than 4.64 Journal of Personality and Social Psychology, 99(5), 771–84,. Roux, Caroline, Kelly Goldsmith, and Andrea Bonezzi (2015), “On on childhood SES (BJN = .49, SE = .25, p = .05) were more likely to purchase counterfeit products after exposure to scarcity cues than the Psychology of Scarcity: When Reminders of Resource after exposure to abundance cues. Scarcity Promote Selfish (and Generous) Behavior,” Journal of Study 3 examined the underlying mechanism for the interac- Consumer Research, 42(4), 615–31. tive effect. The results revealed a significant moderated mediation. Yoon, Sunyee and Hyeongmin Christian Kim (2017), “Feeling Among those with a relatively low childhood SES (-1 SD below Economically Stuck: The Effect of Perceived Economic mean), scarcity cues primed them with a relatively low construal Mobility and Socioeconomic Status on Variety Seeking,” level, which in turn increased their preference for feasibility. Among Journal of Consumer Research, 44(5), 1141–1156.

Advances in Consumer Research 841 Volume 46, ©2018 Foods for Sharing: The Social Value of Handmade Foods Xin Wang, Nanjing University, China Chunqu Xiao, Nanjing University, China Xingyu Duan, Nanjing University, China Hong Zhu, Nanjing University, China

EXTENDED ABSTRACT or machine-made foods condition. In each condition, three of the Mass-manufactured foods are ubiquitous in busy daily life. nine foods and beverages were randomly presented. They firstly re- However, consumers still search for food products with handmade ported the hunger level, then estimated the price of a unit of foods feature (Pieniak, Perezcueto, and Verbeke 2009). Market surveys and or beverages in dollar and answered whether they would like to eat/ academic research indicate that handmade products are more attrac- drink with others with a binary decision: yes or no. Results showed tive than machine-made ones (Fuchs, Schreier, and Osselaer 2015; that participants were more willing to eat with others if handmade Ga et al. 2013). Why are handmade foods so popular? foods were available. In addition, people gave higher price estima- Numerous research have documented the necessity of involving tion for handmade foods than machine-made foods. We also found interpersonal relationship in food choices research. Food is widely the dominate process of sentimental value that led to willingness to regarded as a means of expression and a hallmark of social relation- eat with others, rather than quality and price. ships (Cramer, Greene, and Walters 2011). The type of food we se- Study 2 aimed to test the handmade effects according to main- lect and the amount of food we consume reflects social relationships tenance motives. We manipulated relationship familiarity as a proxy (White and Dahl 2006) and social motives (Meiselman 2008) when for maintenance motives (Hasford et al. 2018). Participants were 35 we eat with others. For example, when women have relationship for- male and 28 female students. The experiment was a mixed-design mation motives, their food choices are susceptical to men’ chocie with eating companions (familiar person vs. unfamiliar person) as a (Hasford, Kidwell, and Lopez-Kidwell 2018). While men are influ- between-subject factor, types of coffee (pour over vs. machine-made) enced by females’ choices when they have relaitonship maintainence as a within-subject factor. They were randomly assigned to one con- motives (Hasford et al. 2018). Choosing handmade foods may serve dition (with familiar vs. unfamiliar person) to imagine a chat with an as a cue to signal social motives, since handmade products are im- acquaintance in a coffee house, and they could choose to buy pour bued with love, which assists people in conveying affection (Fuchs over coffee, machine-made coffee or neither. The number of coffee et al. 2015). types chosen by participants was significantly different across two In the present research, we propose that the social motives to conditions. According to adjusted standardized residuals from the initiate or maintain a close relationship may direct people to choose Post Hoc testing, when with a familiar person, participants are more handmade foods over machine-made foods. The current article con- likely to choose pour over coffee, on the contrary, with an unfamiliar tributes to our understanding of how social motives influence con- person, they were more likely to choose machine-made coffee (H3). sumers’ predilection for handmade foods and beverages. We find Study 3 sought to detect the social value of handmade foods and that, consumers prefer handmade foods when they have motives to beverages according to formation motives. We induced explicit social form a relationship with a new acquaintance or maintain a relation- motive to provide the process evidence. The experiment was a 2 (social ship with a familiar person. In addition, the current article provides motives: with vs. without, between) × 2 (process type: handmade vs. insights for advertisers and managers in promoting handmade foods manufactured, within) × 2 (snack type: candy, cookie, within) mixed and beverages, and implications that could improve consumers’ well- design. The dependent variable was purchase intention. Participants being in social life. were 46 male and 63 female students. They imagined that they had high tea in a café together with a new acquaintance. In social motive Hypothesis 1 Consumers prefer handmade foods and beverag- condition, they were informed to initiate a relationship, where in no so- es to machine-made foods and beverages when cial motive condition, such information was not given. Participants in they eat with others. both conditions read the menu and indicated their intention to purchase cookies and candy with a 7-point scale separately. The interaction between process type and social motive was significant. In social motive Hypothesis 2 (formation motives) The social motive to form condition, participants had significant higher intentions to purchase a relationship with a new acquaintance will in- handmade snacks compared with machine-made snacks (H2). While crease the preference towards handmade foods in no social motive condition, participants showed significant higher and beverages compared with machine-made. intentions to purchase machine-made snacks. Foods and beverages labelled as handmade are more popular Hypothesis 3 (maintenance motives) The social motive to than machine-made in social interaction. We are probably the first to maintain a close relationship will increase their address how individuals’ social motives influence food choices pro- preferences towards handmade foods and bever- cess mode (handmade vs. machine-made) preferences. When having ages compared with machine-made foods and meals with acquaintances with whom people prefer to form or main- beverages. tain a relationship, they would prefer handmade foods. For consumer well-beings, handmade foods signal social motivations to co-eaters, Study 1 experimentally manipulated the foods production which facilitates an enjoyable part of our daily experiences in the modes (handmade vs. machine-made) to test consumers’ commensal company of other people, reduces loneliness and improves mental intention. We chose nine common foods and beverages with hand- and emotional well-being. In practical, we suggest that companies made and machine-made feature: beer, burgers, chips, chocolate, and restaurants can use our findings to advertise handmade foods in coffee, cookie, ice cream, nougat and pizza. Participants (N = 782) social interaction contexts that encourages consumers to share and from MTurk were randomly assigned to handmade foods condition interact with others, which could boost sales.

Advances in Consumer Research 842 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 843 REFERENCES Meiselman, Herbert L. (2008),“24 - Experiencing Food Products Cramer, Janet M, Carlnita P Greene, and Lynn M Walters within a Physical and Social Context,” in Product Experience, (2011),“Food as Communication Communication as Food.” ed. Hendrik N. J. Schifferstein and Paul Hekkert, San Diego: Fuchs, Christoph, Martin Schreier, and Stijn M. J. Van Osselaer Elsevier, 559-80. (2015),“The Handmade Effect: What’s Love Got to Do with Pieniak, Zuzanna, Federico J A Perezcueto, and Wim Verbeke It?,” Journal of Marketing, 79 (2), 98-110. (2009),“Association of Overweight and Obesity with Interest Ga, De Freitas Folly, da Silva E Neiva, Verner F Vieira, Dos in Healthy Eating, Subjective Health and Perceived Risk of Santos Silva Fc, and A. C. Pinheiro Volp (2013),“Acceptance Chronic Diseases in Three European Countries,” Appetite, 53 of Handmade Products Containing Nuts and (3), 399-406. Fructooligosaccharides,” Nutricion Hospitalaria, 28 (1), 86- White, Katherine and Darren W. Dahl (2006),“To Be or Not Be? 92. The Influence of Dissociative Reference Groups on Consumer Hasford, Jonathan, Blair Kidwell, and Virginie Lopez-Kidwell Preferences,” Journal of Consumer Psychology, 16 (4), (2018),“Happy Wife, Happy Life: Food Choices in Romantic 404–14. Relationships,” Journal of Consumer Research, 44 (6), 1238- 56. Cultural Values and Consumers’ Brand Preference Jessie J. Wang, Miami University, USA Ashok K. Lalwani, Indiana University, USA Devon Delvicchio, Miami University, USA

EXTENDED ABSTRACT that high (vs. low) PDB consumers preferred mass-market to niche

Consumers routinely need to choose between mass-market (e.g., brands (Mlow-PDB = 2.39, Mhigh-PDB = 2.96; t(144) = -2.02; p < .05). Häagen-Dazs) and niche brands (e.g., Halo Top), yet little research Study 2 examined the mechanism underlying the relationship has examined the factors that influence such choices (Dalgic and between PDB and brand preference. First, the effect of PDB on brand Leeuw 1994). Can consumers’ cultural background and orientation preference was significant and positive (β = .17, t(320) = 3.97, p < affect their choices? How and why? Are there any strategies firms .001), which suggested that high (vs. low) power distance consumers may use to influence consumers’ choices? In the current research, we preferred mass-market to niche brands. Next, a bootstrapping proce- propose that power distance belief (henceforth, PDB) is associated dure with 10,000 iterations (Zhao, Lynch, and Chen 2010) showed with greater risk aversion and, hence, a greater preference for mass- that the mean indirect effect of PDB on brand preference through market over niche brands. risk aversion was negative and significant (β = -.03, SE = .01, CI95 PDB is defined as the degree to which individuals accept and = -.06, -.01) as predicted. endorse hierarchy and inequalities in power (Hofstede 1984, 2001; Study 3 assessed a boundary condition based on product-usage Zhang, Winterich, and Mittal 2010). Research suggests that that context – high or moderate. A linear regression model revealed sig- people in high (vs. low) PDB societies are more likely to stay within nificant main effects of PDB and product usage contexts (ps < .05). their comfort zone and cling to the familiar (Hofstede 1984, 2001; More importantly, the interaction effect of PDB x product usage con- Carl 2004). Therefore, we expected high (vs. low) PDB consumers texts was also significant (β = -.27,t (240) = -2.29, p < .03). A follow- to be more risk averse. up analysis (Johnson and Neyman 1936; Spiller et al. 2013) revealed Risk aversion refers to one’s tendency to avoid uncertainty that there was a significant positive effect of product usage context rather than avoid loss, and a risk-averse person avoids a gamble in on brand preference for participants whose PDB score was less than favor of a sure thing (Isen, Nygren, and Ashby 1988; Kahneman and 2.91 (BJN = .33, SE = .17, p = .05), but not for those whose PDB Tversky 2013). Mass-market brands are seen as tried and tested by score was more than 2.91. Hence, low PDB consumers’ preference the market, and are accepted by a majority of consumers (Desarbo for mass-market (vs. niche brands) significantly increased in the high et al. 2002; Raj 1985; Kim and Chung 1997). We suggest that mass- product-usage risk contexts, compared to the moderate risk contexts. market brands are perceived to be less risky because following the However, high PDB consumers’ brand preferences did not change by crowd appears less risky, compared to following a new path (Deve- varying the risk involved in product usage context, supporting H3. now and Welch 1996; Olsen 1996). Following the same logic for The current research offers several theoretical and managerial mass-market brands, niche brands that lack the endorsement of the contributions. We provide novel insights on how respondents’ cul- majority may be perceived as risky (Munshi 2004; Huang and Chen tural orientation, specifically their PDB, influences preference for 2006; Desarbo et al. 2002). Because risk-averse (vs. risk-taking) con- mass-market versus niche brands. We also shed light on the underly- sumers have been found to take fewer risks associated with prod- ing mechanism by showing that consumers high (vs. low) in PDB ucts (Grewal, Gotlieb, and Marmorstein 1994; Kahn and Sarin 1988; prefer mass-market to niche brands due to a greater risk aversion. In Mandel 2003), we propose that high (vs. low) PDB consumers would addition, we highlight a boundary condition that identifies when low prefer mass-market brands to niche brands because they are more PDB consumers switch their preference to mass-market, instead of risk averse. niche, brands. Our findings have implications for brand managers’ Furthermore, because the risk associated with a brand or a prod- promotion efforts by providing insights for firms targeting consum- uct can change depending on its use (Dowling and Staelin 1994; ers who are more receptive to either mass-market or niche brands Campbell and Goodstein 2001), we suggest that product usage mod- based on their PDB. Second, managers may prime consumers to be erates the effect of PDB on brand preference. Specifically, we sug- risk-averse or risk seeking which in turn is expected to influence gest that high (vs. low) PDB consumers prefer mass-market to niche brand preference. Third, firms can vary the product usage contexts to brands in low or moderate product-usage risk contexts. However, in make some types of consumers to switch from niche to mass-market high product-usage risk contexts, low PDB consumers also become brands. risk-averse and their preference for mass-market brands increases, compared to baseline conditions. In contrast, because high PDB con- REFERENCES sumers are risk averse to begin with, they should not become more Campbell, Margaret C. and Ronald C. Goodstein (2001), “The risk averse when the product usage indicates risk (i.e., ceiling effect). Moderating Effect of Perceived Risk on Consumers’ A multi-method approach was used to establish reliability and Evaluations of Product Incongruity: Preference for the Norm,” generalizability across four studies. Our pilot study used a second- Journal of Consumer Research, 28 (3), 439-49. ary dataset from the footwear industry in 42 countries and revealed Carl, Dale, Mansour Javidan, and Vipin Gupta (2004), “Power that countries high (vs. low) in PDB have a greater market share of Distance,” in Culture, Leadership, and Organizations: The mass-market, compared to niche, brands. Study 1 further ascertained Globe Study of 62 Societies, Robert J. House, Paul J. Hanges, the link between consumers’ PDB and their preference for mass-mar- Mansour Javidan, Peter W. Dorfman, and Vipin Gupta, eds. ket versus niche brands in the lab. We also gained confidence in the Thousand Oaks, California: Sage, 513-59. causal role of PDB by manipulated it in this study. A GLM revealed Dalgic, Tevfik and Maarten Leeuw (1994), “Niche Marketing that the PDB significantly affected consumers’ brand preference Revisited: Concept, Applications and Some European Cases,” (F(1,144) = 4.37, p < .04). A follow-up independent t-test suggested European Journal of Marketing, 28 (4), 39-55.

Devenow, Andrea, and Ivo Welch (1996), “Rational herding in Kim, Chung Koo, and Jay Young Chung (1997), “Brand financial economics,”European Economic Review 40(3), 603- Popularity, Country Image and Market Share: An Empirical 615. Study.” Journal of International Business Studies, 28(2), 361- Dowling, Grahame R. and Richard Staelin (1994), “A Model of 386. Perceived Risk and Intended Risk-Handling Activity,” Journal Mandel, Naomi (2003), “Shifting Selves and Decision Making: of Consumer Research, 21 (June), 119-34. The Effects of Self-construal Priming on Consumer Risk- Grewal, Dhruv, Jerry Gotlieb, and Howard Marmorstein (1994), taking,” Journal of Consumer Research, 30 (1), 30-40. “The Moderating Effects of Message Framing and Source Munshi, Kaivan (2004), “Social Learning in a Heterogeneous Credibility on the Price-Perceived Risk Relationship,” Journal Population: Technology Diffusion in the Indian Green of Consumer Research, 21 (June), 145-53. Revolution,” Journal of Development Economics, 73 (1), Hofstede, Geert H. (1984), Culture’s Consequences: International 185–213. Differences in Work-Related Values. Newbury Park, CA: Sage. Olsen, Robert A. (1996), “Implications of Herding Behavior for _____ (2001), Culture’s Consequences: Comparing Values, Earnings Estimation, Risk Assessment, and Stock Returns,” Behaviors, Institutions and Organizations across Nations, 2nd Financial Analysts Journal, 52 (4), 37-41. ed. Thousand Oaks, CA: Sage. Raj, S. P. “Striking a Balance Between Brand” Popularity” and Huang, Jen‐Hung, and Yi‐Fen Chen (2006), “Herding in Online Brand Loyalty (1985),” The Journal of Marketing, 49 (winter), Product Choice,” Psychology & Marketing, 23 (5), 413-428. 53-59. Isen, Alice M., Thomas E. Nygren, and F. Gregory Ashby (1988), Winterich, Karen Page, and Yinlong Zhang (2014), “Accepting “Influence of Positive Affect on the Subjective Utility of Inequality Deters Responsibility: How Power Distance Gains and Losses: It Is Just Not Worth the Risk,” Journal of Decreases Charitable Behavior.” Journal of Consumer Personality and Social Psychology, 55 (5), 710-17. Research, 41 (2), 274-293. Johnson, Palmer O. and J. Neyman (1936), “Tests of Certain Linear Zhang, Yinlong, Karen P. Winterich, and Vikas Mittal (2010), “PDB Hypotheses and Their Application to Some Educational and Impulsive Buying,” Journal of Marketing Research, 47 Problems,” Statistical Research Memoirs, 1, 57–93. (5), 945-54. Kahn, Barbara E., and Rakesh K. Sarin (1998), “Modeling Zhao, Xinshu, John G. Lynch Jr, and Qimei Chen (2010), Ambiguity in Decisions Under Uncertainty,” Journal of “Reconsidering Baron and Kenny: Myths and Truths about Consumer Research, 15 (2), 265-272. Mediation Analysis.” Journal of Consumer Research, 37 (2), Kahneman, Daniel, and Amos Tversky (2013), “Prospect Theory: 197-206. An Analysis of Decision Under Risk.” In Handbook of the Fundamentals of Financial Decision Making: Part I, 99-127. Contagion and Product Physicality: A Study of Consumer Response to Recycled-content Products . Qizhou Wang, University of Connecticut, USA David A. Norton, University of Connecticut, USA Robin A. Coulter, University of Connecticut, USA William T. Ross, Jr., University of Connecticut, USA

EXTENDED ABSTRACT transfer of undesirable essence). In study 1, we showed that partici- The purchase of recycled-content products poses dilemmas for pants were more likely to purchase recycled-content products made consumers. Consumers may want to preserve the environment, but with materials from a more (vs. less) desirable source (F(1, 128) = simultaneously can be concerned about potential contamination re- 8.03, p < .1). Also, participants were more likely to purchase recy- sulting from prior use. We focus on understanding how prior history cled-linen products made with recycled materials that had less con- of recycled-content products might affect consumer perceptions and tact with past users than those that had more contact with past users consumption of recycled goods. Specifically, we attempt to answer (F(1, 128) = 2.80, p < .1). In study 2, we replicated the effect of phys- the following questions about consumer perceptions of recycled- ical contact with past users (F(1,317) = 5.62, p < .05). More impor- content products using contagion theory. To what extent do consum- tantly, we demonstrated that consumers were more likely to purchase ers devalue certain recycled-content products, but appreciate others? recycled-linen products that would incur less (vs. more) physical To what extent do consumers perceive an essence transfer from old contact with their bodies (F(1,317) = 4.85, p < .05). Moreover, our materials to new products—that is, would a recycled-content product analysis suggests that disgust is a key mechanism underlying these inherit some internal characteristics or properties from its original effects. In study 3, we sought to test the strength of these effects in a materials? What are the product factors that amplify or dampen a context where consumers are interacting with actual products (e.g., perceived essence transfer in a current consumption context? dusting cloths vs. dusting gloves). We replicated prior findings and Prior research has suggested the importance of physical inter- found a significant interaction effect between the degree of physical actions in activating contagion (e.g., Newman, Diesendruck, and contact with past users and the degree of physical contact with cur- Bloom 2011). However, most scholars have studied physical con- rent users (F(1,138) = 3.95, p < .05). That is, when a product is made tact as a dichotomous variable and assumed that contagion is dose- with used, recycled materials, consumers are more likely to purchase insensitive to contact (Rozin and Nemeroff 2002). By focusing on a product that would incur less (vs. more) physical contact with their how consumers infer the amount of physical contact from product bodies during product consumption. physicality and functions, we explore the relationship between con- Prior research has suggested that consumers are more critical of sumers anticipated physical contact and the prevailing contagion on recycled-content products than conventional products. We attempt recycled-content products. We argue that consumers should perceive to understand these attitudes through the theory of contagion. We greater contagion from recycled materials that have more physical argue that the contagion effect, derived from consumers’ magical contact with past users than from materials that have minimal physi- beliefs that certain essences can transfer from one to another, may cal contact with past users. Also, we extend this discussion to the explain consumer negativity toward certain recycled-content prod- way in which the implied physicality of recycled-content products ucts. More importantly, we argue that when purchasing recycled- informs consumers of the magnitude of contagion. We suggest that content products, consumers tend to engage in a unique evaluation consumers infer different degrees of physical contact from products process that involves both contagion-related constructs (i.e., disgust) with different forms. For example, consumers may perceive greater and heuristics for conventional purchases (i.e., quality perceptions). physical contact with a pair of gloves that cover the entire hand than Our research makes two contributions to the existing contagion lit- a pair of potholders which touches only the palm of the hand. We erature. First, whereas most of the past literature has assumed that suggest that consumers should be more attentive to contagion cues the contagion effect is insensitive to the degree of physical contact, when the use of a product suggests having more (vs. less) physical we have demonstrated that consumer perception of contagion varies contact. with the degree of physical contact. For example, we showed that the small difference in physical contact between wearing a pair of To test our hypotheses, we conducted three studies (N1 = 132; gloves versus using a pair of potholders facilitated different degrees N2 =326, N3=144). Study 1 and study 2 were conducted online using MTurk. Study 3 was conducted in a controlled laboratory setting. of disgust among consumers. Second, we have explored how aspects Textile recycling (i.e., recycling of linen products) was chosen as of a product’s physicality such as forms may influence the magnitude the context because it enabled us to identify: 1) a more (vs. less) of the contagion effect. We expect that business managers and policy desirable supplier of materials (e.g., 5-star hotel vs. a 2-star hotel); 2) makers may find our research helpful because we have identified recycled source material with more (vs. less) previous body contact ways to potentially manage consumers’ response to recycled-content (e.g., bed sheets vs. curtains); 3) recycled products with more (vs. products through the amount of perceived essence transfer. less) physical body contact with the consumer (e.g., winter gloves vs. pot holder). From a managerial perspective, it is also notable that REFERENCES recycling of linen products becomes important because of growing Newman, George E., Gil Diesendruck, and Paul Bloom (2011). expenditures on textile wastes (e.g., U.S. spent $523 million land- “Celebrity Contagion and the Value of Objects,” Journal of filling textile waste in 2013). After reading a shopping scenario on Consumer Research, 38(2), 215–228. textile recycling, participants completed several measures including Rozin, Paul and Carol Nemeroff (2002), “Sympathetic Magical purchase intention toward recycled-linen products. Thinking: The Contagion and Similarity ‘Heuristics,’” in The findings indicate that consumers engage in a contagion-ori- Heuristics and Biases, ed. Thomas Gilovich, Dale Griffin, and ented thinking process when purchasing recycled-content products Daniel Kahneman, Cambridge University Press, 201–216. that are associated with a negative context (i.e., contamination or

EXTENDED ABSTRACT game that contained a total of 12 features, six person-related (e.g., People regularly make choices about products they already “intuitive,” “smart”) and six product-related (“portable,” “smooth”). own, such as which of two sneakers bought through Zappos.com to Participants did not receive the list of words to be found rather had to retain. However, decision research has focused on situations where find as many words as they can within a 60 seconds. To test whether choice precedes ownership, or acquisition, whereas situations where person-related features are easier to process for items consumers choice follows ownership, or retention, remain understudied. think of as already (vs. not-yet) owned, participants were asked to Previous research finds that people use ownership as a source search words using a provided search cue, specifically the head- of information to categorize objects relative to the category of “self” phones that participants initially thought of either as already or as (Weiss and Johar 2013; Weiss and Johar 2016); simply thinking of an not-yet owned. Finally, all participants rated the self-importance of item as owned (unowned) fosters item classification in (outside of) the identity “audiophile.” Consistent with predictions, data analysis the self (Turk et al. 2011). Relatedly, classifying an item in a category showed a marginally significant 2-way interaction between search- makes it is easier to process category-applicable (vs. -inapplicable) cue frame and identity: framing the search cue (i.e., the headphones) features of the item (Barsalou 1982; Medin 1989). Building on these as already (vs. not-yet) owned increased the number of person-relat- findings, we suggest that simply thinking about items as already (vs. ed words found, but only when the primed identity was audiophile. not yet) owned (and thus classifying these items as “self”) would This 2-way interaction was qualified by the importance of the iden- allow one’s self-concept to facilitate the processing of information tity audiophile, yielding a significant 3-way interaction. Specifically, about certain item features. Specifically features like “resilience” or the two-way interaction pattern described above appeared only when “daringness” that commonly apply to describe people, or person- the identity “audiophile” was self-important. related features, would become easier to process and understand. Next, to test the predicted effects on decision weight, in Study Consequently, because the information that people process more (vs. 2a [2 (primed-identity) x 2 (choice-frame), between-subjects), par- less) easily has greater impact on choice (Schwarz 2004), people ticipants in the product-independent (product-dependent) identity may afford greater decision-weight to person-related features when prime wrote two things about themselves that reflect their hiker people think about items as already (vs. not yet) owned, specifically (walker) side. Participants in the retention (acquisition) choice frame when making a retention (vs. acquisition) choice. Notably, such flu- imagined buying (considering) two pairs of high-quality shoes, with ency advantage in retention (vs. acquisition) choices is not predicted the intent to keep (acquire) and use one of them. Participants then for features like “portability” or “smoothness,” which primarily are allocated 30 points among attributes according to their importance in applied to describing products, or product-related features. choosing shoes. Two of the attributes were person-related (e.g., “out- In our conceptualization, the predicted effect of thinking of doorsiness,” “daringness”) and two product-related (e.g., “smooth- items as already (vs. not-yet) owned are driven by the use of the ness,” “compactness”). Consistent with predictions, data analysis self as an organizing category for material objects (Weiss and Jo- showed a significant 2-way interaction between choice frame and har 2018). The effect therefore should appear (disappear) when con- primed identity. Specifically, framing choice as retention (vs. acqui- sumers (do not) mentally construe the self in terms of the objects it sition) increased the importance of person-related features, but only includes. Recent research shows that priming product-independent when the primed identity was hiker. Study 2b replicated these results identities attenuates usage of the self as a category for objects (Weiss in a lab study that manipulated real (rather than imagined) ownership 2018). An identity is product-independent if carrying out that identity with a different product category (winter hats). does not rely on any product. For example, carrying out the identities Next, to test the predicted effects on choice, in Study 3, the iden- of “musicophile” or “walker” only requires using the physical senses tity “audiophile” was primed for all participants. Choice frame was (e.g., one’s hearing to enjoy music) or the physical body (e.g., one’s manipulated using study 2a’s manipulation. Participants were then legs to engage in walking). This independence on the use of products presented with the rating of two headphones along 4 attributes, two render these identities product-independent. Conversely, carrying out person-related and two product-related. Ratings created a tradeoff the identities of “audiophile” or “hiker” necessitates specialized gear between the product-related and the person-related attributes (i.e., (e.g., hi-fi headphones to appreciate sound or hiking boots to hike), one set of headphones was rated higher on the person-related features rendering these identities product-dependent. Thus, we predicted and the other on the product-related features). Participants in the re- the effect of choice-frame to disappear following the activation of tention (acquisition) choice frame chose which headphones to keep product-independent (vs. product-dependent) identities. Relatedly, (buy). Finally, all participants rated the self-importance of the iden- self-important (vs. unimportant) identities are more integral to the tity “audiophile.” Consistent with predictions, data analysis showed self-concept (Reed II 2004) and thus can better support the utiliza- that framing choice as retention (vs. acquisition) increased choice tion of the self-concept for processing information. Therefore, we in the headphones with higher ratings on person-related features. expected the predicted effects to manifest mainly among individuals This effect mainly appeared when the identity “audiophile” was self- who consider the focal product-dependent identity as self-important. important, yielding a significant 2-way interaction. Implications for Four studies supported the above predictions. In study 1 ownership research and marketing practice are discussed. [2(primed-identity) x 2(search-cue-frame), between-subjects], participants in the product-independent (product-dependent) identity REFERENCES prime wrote two things about themselves that reflect their musi- Barsalou, Lawrence W. (1982), “Context-independent and context- cophile (audiophile) side. Participants in the already (vs. not-yet) dependent information in concepts,” Memory & Cognition, 10 owned frame condition imagined that they received (considered) (1), 82-93. two sets of high-quality headphones but could only keep (acquire) Medin, Douglas L. (1989), “Concepts and Conceptual Structure,” one. Participants then played an incentive-compatible word-puzzle American Psychologist, 44 (12), 1469-81.

Reed II, Americus (2004), “Activating the self-importance of Weiss, Liad and Gita V. Johar (2013), “Egocentric Categorization consumer selves: Exploring identity salience effects on and Product Judgment: Seeing Your Traits in What You judgments,” Journal of Consumer Research, 31 (2), 286-95. Own (and Their Opposite in What You Don’t),” Journal of Schwarz, Norbert (2004), “Metacognitive experiences in consumer Consumer Research, 40 (1), 185-201. judgment and decision making,” Journal of Consumer ---- (2016), “Products as Self-Evaluation Standards: When Owned Psychology, 14 (4), 332-48. and Unowned Products Have Opposite Effects on Self- Turk, David J., Kim van Bussel, Gordon D. Waiter, and C. Neil Judgment,” Journal of Consumer Research, 42 (6), 915-30. Macrae (2011), “Mine and Me: Exploring the Neural Basis of ---- (2018), “Psychological Ownership in Egocentric Categorization Object Ownership,” Journal of Cognitive Neuroscience, 23 Theory,” Psychological Ownership and Consumer Behavior. (11), 3657-68. New York: Springer. Weiss, Liad (2018), “Product-Independent vs. Product-Dependent Identity and the Use of Rentals to Substitute for the Use of Possessions,” Journal of Consumer Research, Under Review. Growing Up Rich and Insecure Makes Objects Seem Human: Childhood Material and Social Environments Predict Anthropomorphism Jodie Whelan, York University, Canada Sean T. Hingston, Western University, Canada Matthew Thomson, Western University, Canada Allison R. Johnson, Western University, Canada

EXTENDED ABSTRACT American adults participated in a two-part online survey. Part 1 The socioeconomic environments in which people grow up of the survey consisted of the anxiety and avoidance scales used in have a lasting influence on how they think and behave. Of unique study 1 and established three-item measures of childhood and current interest to this research, social class can influence people’s preferred SES (Griskevicius et al. 2011). Approximately one week later, partic- strategies of coping with environmental threats and stressors. In par- ipants completed the gadget task (Epley et al. 2008), which measures ticular, people who grow up wealthy are taught to prioritize the self the anthropomorphism of four gadgets, and the Brand Engagement in over community, independence over interdependence, and autonomy Self-Concept scale (BESC; Sprott, Czellar, and Spangenberg 2009). over relatedness (see Markus and Kitayama 2003 for a review), and Analyses revealed the predicted three-way interaction between child- their preferred coping strategies reflect these lessons. But, what if hood SES, anxiety, and avoidance: when participants had insecure these same people also grow up with unmet relationship needs? How attachment styles, childhood SES positively predicted their tendency do they simultaneously protect their prized autonomy and fulfil their to anthropomorphize, which subsequently increased the likelihood need for relatedness—a motivational quandary that, theoretically, that they include brands in their self-concept. does not exist for people growing up with fewer material resources Study 3 sought evidence for our proposed mechanism by examin- and more community-oriented models of agency? ing if the effect of childhood SES is mediated by the extent to which Here, we propose that anthropomorphism may be an especially people’s parents adopted a family communication style that empha- attractive coping strategy for people who grew up wealthy and in- sized the importance of personal autonomy (i.e., a concept-oriented secure in their personal relationships. Anthropomorphism refers to communication style; Grougiou and Moschis 2015; Moschis 1985) . the attribution of human-like characteristics, motivations, or mental American adults participated in a two-part online survey. Part 1 consist- states to nonhuman entities (for a review, see Epley, Waytz, and Ca- ed of our predictor variables: the childhood SES, current SES, anxiety, cioppo 2007). From gadgets to brands, people frequently treat non- and avoidance scales used in study 2, as well as a four-item measure human agents as human-like. Although anthropomorphized objects of concept-oriented communication style (Baker et al. 2016). Approxi- may seem like inferior relationship partners because they lack hu- mately one week later, participants completed the gadget task (Epley et man interaction, research has shown that engaging with anthropo- al. 2008). Analyses revealed that concept-oriented communication style morphized objects can not only help people cope with unmet social significantly mediated the effect of childhood SES: when participants needs (Chen, Wan, and Levy 2017; Mourey, Olson, and Yoon 2017), had insecure attachment styles, the effect of childhood SES on anthro- but also, some people may actually prefer relationships with objects pomorphism via concept-oriented communication style was significant. precisely because they do not require human intimacy (Price and Ar- These findings make a number of contributions. First, we con- nould 1999; Thomson and Johnson 2006; Whelan et al. 2016). tribute to a growing body of work examining how childhood SES in- To test this proposition, we investigate the interactive effects of fluences adult consumer behavior (Ahuvia and Wong 2002; Connell, childhood SES and attachment style, which reflects how (in)secure Brucks, and Nielsen 2014; Mittal and Griskevicius 2016; Richins people felt in their relationships with their primary caregivers, on an- and Chaplin 2015; Whelan and Hingston 2018). To the best of our thropomorphism. Specifically, we hypothesize that people who grew knowledge, not only are we the first to show that anthropomorphic up wealthy and insecurely attached will be the most likely to an- tendencies vary depending on people’s childhood socioeconomic thropomorphize. Study 1 provides initial support for this hypothesis. environments, we are also the first to empirically investigate the in- American adults participated in an online survey. The study began teractive effect of childhood SES and attachment style on adult psy- with the Individual Differences in Anthropomorphism Questionnaire chology. In addition, we answer calls to go beyond childhood SES (Waytz, Cacioppo, and Epley 2010). Next, participants completed as “merely a proxy measure” (Mittal and Griskevicius 2014, 633; the Experiences in Close Relationships Scale, which examines at- Whelan and Hingston 2018) and demonstrate that the effect of child- tachment styles as a function of two dimensions: anxiety and avoid- hood SES on people’s tendency to anthropomorphize is mediated by ance (ECR-S; Wei et al. 2007). When individuals are low on both the extent to which their parents adopted a family communication dimensions, they have a secure attachment style; when individuals style that emphasized the importance of personal autonomy. are high in one or both dimensions, they have an insecure attachment Last, our findings contribute to growing interest in the root style. causes of individual differences in anthropomorphic tendencies. To The study concluded with childhood and current family income. date, most anthropomorphism research has focused on either the ex- Analyses revealed the predicted three-way interaction between child- tent to which specific nonhuman agents are anthropomorphized, the hood family income, anxiety, and avoidance: childhood family in- accuracy of anthropomorphic inferences in describing these agents, come positively predicted anthropomorphism for those with insecure or the consequences of anthropomorphism (Waytz, Morewedge, et attachment styles (i.e., when anxiety and avoidance were both high al. 2010). Consequently, “the psychological determinants that ex- and/or at the mean). plain and predict variability in the tendency to anthropomorphize” Study 2 used a longitudinal design to replicate and extend have remained largely unaddressed (Waytz, Morewedge, et al. 2010, these findings. In particular, we examined whether the increase in 411). Accordingly, our findings suggest that individual differences anthropomorphism among participants with wealthy childhoods and in the tendency to anthropomorphize arise from differences in child- insecure attachment styles subsequently predicted the likelihood that hood material and social environments. Indeed, it’s difficult to imag- consumers include important brands as part of their self-concept. ine two more proximal causes.

Advances in Consumer Research 849 Volume 46, ©2018 850 / Growing Up Rich and Insecure Makes Objects Seem Human: Childhood Material and Social Environments Predict Anthropomorphism REFERENCES Moschis, George P. (1985), “The Role of Family Communication in Ahuvia, Aaron C. and Nancy Y. Wong (2002), “Personality and Consumer Socialization of Children and Adolescents,” Journal Values Based Materialism: Their Relationship and Origins,” of Consumer Research, 11(4), 898–913. Journal of Consumer Psychology, 12(4), 389–402. Mourey, James A., Jenny G. Olson, and Carolyn Yoon (2017), Baker, Andrew M., George P. Moschis, Edward E. Rigdon, and “Products as Pals: Engaging with Anthropomorphic Products Choong Kwai Fatt (2016), “Linking Family Structure to Mitigates the Effects of Social Exclusion,”Journal of Impulse-Control and Obsessive–Compulsive Buying,” Journal Consumer Research, 414–31. of Consumer Behaviour, 15(4), 291–302. Price, Linda L. and Eric J. Arnould (1999), “Commercial Chen, Rocky Peng, Echo Wen Wan, and Eric Levy (2017), Friendships: Service Provider-Client Relationships in “The Effect of Social Exclusion on Consumer Preference Context,” Journal of Marketing, 63(4), 38–56. for Anthropomorphized Brands,” Journal of Consumer Richins, Marsha L. and Lan Nguyen Chaplin (2015), “Material Psychology, 27(1), 23–34. Parenting: How the Use of Goods in Parenting Fosters Connell, Paul M., Merrie Brucks, and Jesper H. Nielsen (2014), Materialism in the next Generation,” Journal of Consumer “How Childhood Advertising Exposure Can Create Biased Research, 41(6), 1333–57. Product Evaluations That Persist into Adulthood,” Journal of Sprott, David, Sandor Czellar, and Eric Spangenberg (2009), “The Consumer Research, 41(1), 119–34. Importance of a General Measure of Brand Engagement on Epley, Nicholas, Scott Akalis, Adam Waytz, and John T. Cacioppo Market Behavior: Development and Validation of a Scale,” (2008), “Creating Social Connection through Inferential Journal of Marketing Research, 46(1), 92–104. Reproduction Loneliness and Perceived Agency in Gadgets, Thomson, Matthew and Allison R. Johnson (2006), “Marketplace Gods, and Greyhounds,” Psychological Science, 19(2), and Personal Space: Investigating the Differential Effects of 114–20. Attachment Style across Relationship Contexts,” Psychology Epley, Nicholas, Adam Waytz, and John T. Cacioppo and Marketing, 23(8), 711–26. (2007), “On Seeing Human: A Three-Factor Theory of Waytz, Adam, John Cacioppo, and Nicholas Epley (2010), “Who Anthropomorphism,” Psychological Review, 114(4), 864–86. Sees Human? The Stability and Importance of Individual Griskevicius, Vladas, Andrew W. Delton, Theresa E. Robertson, Differences in Anthropomorphism,”Perspectives on and Joshua M. Tybur (2011), “Environmental Contingency Psychological Science, 5(3), 219–32. in Life History Strategies: The Influence of Mortality and Waytz, Adam, Carey K. Morewedge, Nicholas Epley, George Socioeconomic Status on Reproductive Timing,” Journal of Monteleone, Jia-Hong Gao, and John T. Cacioppo (2010), Personality and Social Psychology, 100(2), 241–54. “Making Sense by Making Sentient: Effectance Motivation Grougiou, Vassiliki and George P. Moschis (2015), “Antecedents Increases Anthropomorphism,” Journal of Personality and of Young Adults’ Materialistic Values: Antecedents of Young Social Psychology, 99(3), 410–35. Adults’ Materialistic Values,” Journal of Consumer Behaviour, Wei, Meifen, Daniel W. Russell, Brent Mallinckrodt, and David L. 14(2), 115–26. Vogel (2007), “The Experiences in Close Relationship Scale Markus, Hazel Rose and Shinobu Kitayama (2003), “Models (ECR)-Short Form: Reliability, Validity, and Factor Structure,” of Agency: Sociocultural Diversity in the Construction of Journal of personality assessment, 88(2), 187–204. Action,” Nebraska Symposium on Motivation. Nebraska Whelan, Jodie and Sean T. Hingston (2018), “Can Everyday Symposium on Motivation, 49, 1–57. Brands Be Threatening? Responses to Brand Primes Depend Mittal, Chiraag and Vladas Griskevicius (2014), “Sense of on Childhood Socioeconomic Status,” Journal of Consumer Control under Uncertainty Depends on People’s Childhood Psychology, 28(3), 477–86. Environment: A Life History Theory Approach.,” Journal of Whelan, Jodie, Allison R. Johnson, Tara C. Marshall, and Personality and Social Psychology, 107(4), 621–37. Matthew Thomson (2016), “Relational Domain Switching: ——— (2016), “Silver Spoons and Platinum Plans: How Interpersonal Insecurity Predicts the Strength and Number of Childhood Environment Affects Adult Healthcare Decisions,” Marketplace Relationships,” Psychology & Marketing, 33(6), Journal of Consumer Research, 43(4), 636–56. 465–79. Paying to Purchase a Conversation Topic Hillary J.D. Wiener, University at Albany, USA Josh Wiener, Oklahoma State University, USA

EXTENDED ABSTRACT study 1, but the expensive cake had either an interesting story or no People want to express themselves (Berger and Heath 2008), story. All participants indicated on an 11-point scale how likely they and signaling research has found that they use products to do so (Es- would be to buy the expensive cake. They then answered our key calas and Bettman 2003). However, sometimes people do not want to mediator indicating how likely they would be to “tell people some silently signal to others, they want to talk to them (Kahneman et al. stories about the cake I bought.” 2004). Can products also play a role in facilitating conversation? Ini- There was a significant interaction between the situation and tial evidence suggests they can; displaying some self-relevant prod- whether the expensive product had a story (p = .036). When the cake ucts (conversation pieces) increases the likelihood that their owner had a story, participants were more likely to purchase it in the social will have an enjoyable conversation (Wiener, Bettman, and Luce condition than alone condition (p = .001); there was no effect for 2015). In this investigation, we look at whether consumers predict situation when it had no story (p = .80). The proposed mediator, like- that non-self-relevant products can also facilitate conversation, and lihood to tell stories about the product, showed the same pattern (p when they will pay a premium to purchase products that might do so. = .019), and the index of moderated mediation was significant (b = We investigate this issue by examining consumers’ likelihood to -.35; 95% CI [-.70--.08]). Likelihood to tell stories mediated the ef- purchase expensive products that are associated with product related fect of situation on likelihood to purchase the expensive cake when it stories, such as those about the product’s history or making. Impor- had a story (b = -.21; 95% CI [-.45--.02]), but not when it did not (b = tantly, the stories are about the products, rather than the purchaser, .14; 95% CI [-.06-.36]). Study 2 shows that people’s predictions that and are not self-relevant. We predict that people will pay a premium they would tell others the product’s stories are important in leading to for a story-associated product if they think they can use it to facilitate their purchase of expensive, story-associated products. conversation. Study 3 replicates the interaction with a cleaner manipulation of Conversation facilitation is not the only reason people might the situation. Study 3 (n = 453) had a 2 (product-mention allowed or pay more for a product with a story. People perceive products with forbidden) x 2 (interesting story v. no story) between-subjects design. stories more positively (van Laer, Ludwig, and Escalas 2017) and as Study 3 was similar to study 2, but for the situation manipulation, all more authentic (Newman and Dhar 2014). Therefore, we include a participants attended the potluck, and the potluck rules did or did situation moderator in all studies: whether the situation allows partic- not allow talking about what you brought. Again participants chose ipants to use the product to facilitate conversation. In studies 1 and 2, between purchasing an expensive cake with an interesting (no) story we manipulate whether they will use it alone or socially, and in study or an inexpensive cake with no story on the scale used in study 2. 3 we manipulate whether they can mention it. We predict that people There was a significant interaction between the situation and are more likely to purchase expensive products with stories when whether the expensive cake had a story (p < .0001): being allowed to they can use them to facilitate conversation than when they cannot. talk about their potluck contribution increased likelihood to purchase Study 1 examines whether people are more likely to purchase an the expensive cake when the expensive cake had a story (p < .0001), expensive product with an interesting story associated with it when but not when it did not (p = .24). they will have the opportunity to use it in a social situation than when In three studies, we show that people are willing to pay a pre- they will not. Study 1 (n = 245) had a 2 (social v. alone) x 2 (interest- mium to purchase products that are associated with interesting, ing story v. boring story) between-subjects design. Participants read conversation-facilitating stories when they can use the product in a that they had moved to a new city and were either going to a potluck social situation, providing preliminary evidence that the ability to to meet new people or spending the night alone. Then all participants foster, even non-self-relevant, conversations is a source of value to chose between purchasing a $21.00 cake with an interesting (boring) consumers. story or a $15.00 cake with no story. In all studies, the interesting story is about the history of their new city—an appealing, but not REFERENCES self-relevant, topic. Berger, Jonah and Chip Heath (2008), “Who Drives Divergence? There was a marginal interaction between the situation and Identity Signaling, Outgroup Dissimilarity, and the whether the expensive product had an interesting story (p = .074). Abandonment of Cultural Tastes,” Journal of Personality and When the expensive cake had an interesting story, 55% of partici- Social Psychology, 95 (3), 593-607. pants in the social condition, but only 24% percent of participants in Escalas, Jennifer Edson and James R. Bettman (2003), “You the alone condition, purchased it (p < .0001). When it had a boring Are What They Eat: The Influence of Reference Groups on story, there was no effect of situation on purchase (social = 37%, Consumers’ Connections to Brands,” Journal of Consumer alone = 28%; p = .27), suggesting that interesting stories are more Psychology, 13 (3), 339-348. valuable in social situations. Kahneman, Daniel, Alan B. Krueger, David A. Schkade, Norbert In study 2 we replicate the results of study 1 with a different Schwarz, and Arthur A. Stone (2004), “A Survey Method for story manipulation and look for mediation. We predict that the addi- Characterizing Daily Life Experience: The Day Reconstruction tional value of a product with a story in a social situation comes from Method,” Science, 306 (5702), 1776-1780. its ability to facilitate conversation. Therefore participants should Newman, George E. and Ravi Dhar (2014), “Authenticity is predict that they would tell the stories associated with the product Contagious: Brand Essence and the Original Source of in the social situation, and this prediction should mediate the effect Production,” Journal of Marketing Research, 51 (3), 371-386. of the interaction on likelihood to purchase the expensive product. Study 2 (n = 664) had a 2 (social v. alone) x 2 (interesting story v. no story) between-subjects design. Study 2 was nearly identical to

Advances in Consumer Research 851 Volume 46, ©2018 852 / Paying to Purchase a Conversation Topic van Laer, Tom, Stephan Ludwig, and Jennifer Edson Escalas (2017), “Understanding the Value of Stories in Experiential Reviews,” Cambridge, Massachusetts: Marketing Science Institute. Wiener, Hillary J.D., James R. Bettman, and Mary Frances Luce (2015), “Products As Ice Breakers: The Value of Conversation Pieces,” in NA - Advances in Consumer Research Volume 43, eds. Kristin Diehl and Carolyn Yoon, Duluth, MN: Association for Consumer Research, Pages: 297-301. A Model of Consumer Self-Regulation Failure Keith Wilcox, Columbia University, USA

EXTENDED ABSTRACT which has been shown to induce a sense of goal achievement that Self-regulation is fundamental to consumers’ ability to achieve reduces self-regulation (Wilcox et al. 2011). Participants wrote about their goals. Most existing self-regulation models focus on the psy- a “pride” experience or their typical day (control group). They then chological processes that promote goal-consistent behavior. These participated in the same auction as study 1. As predicted, participants models often assume that when individuals are motivated and have who experienced pride submitted higher bids than the control group the ability to regulate their behavior, they are more likely to act con- when they were highly focused on performance monitoring. Pride sistently with their goals (Shiv and Fedorikhin 1999). However, there did not influence spending for those who were less focused on per- is abundant evidence that external factors can lead to self-regulation formance monitoring. failures, even when consumers are sufficiently motivated and have Study 3 sought to demonstrate that the inhibition of behavior the ability to self-regulate. Thus, a more complete model of consum- makes consumers more susceptible to external cues that reduce the er self-regulation is needed to account for how consumers respond to perception of goal threat. Based on previous research which finds external factors. that using a currency with a small nominal value is less threatening to The current research introduces a model of consumer self-reg- spending goals (Raghubir and Srivastava 2002), it was predicted that ulation failure that explains how external factors shape consumers’ when consumers used a currency with small (large) nominal value ability to act consistent with their goals and when these factors are they would be less (more) likely to regulate their spending, an ef- more likely to undermine self-regulation. The model proposes that fect that would be strongest for participants focused on inhibiting many self-regulation failures occur because two processes involved their behavior. Participants submitted bids in an action using a new in self-regulation (performance monitoring and the inhibition of be- currency where the nominal value was set at either a multiple (large havior) make consumers more like to rely on external cues. nominal value) or a fraction (small nominal value) of the US dollar. A key process involved in self-regulation is performance moni- As expected, participants submitted higher bids when the nominal toring, which involves a comparison of behavior (past or current) value was small (vs. large) when they were highly focused on inhib- with a desired end state (Carver and Scheier 1990). During this iting their behavior. Spending did not differ between conditions for process, however, consumers may disengage from the goal, if they those who were less focused on inhibition. perceive that the goal has been achieved or violated in some way. Study 4 sought to demonstrate that the more focused consumers While consumers often determine goal achievement or failure by are on inhibiting their behavior, the more likely their self-regulation explicitly tracking their goal performance, consumers also rely on will be undermined by reward cues. Based on previous research that external cues associated with goal achievement and failure. Thus, demonstrates that the decision to reject (vs. choose) options makes when these cues are present, people who are more engaged in perfor- people more likely to elaborate on hedonic features during a deci- mance monitoring during efforts at self-regulation should act more sion (Dhar and Wertenbroch 2000), it was predicted that consumers inconsistently with their goals. would be more likely to select a higher priced option with more he- Even when consumers are sufficiently motivated to pursue their donic benefits when they rejected (vs. chose), an effect strongest for goals, effective self-regulation requires the inhibition of responses participants highly focused on inhibiting their behavior. Participants that undermine goal pursuit. When consumers inhibit alternatives made a decision between two apartments to rent: one that was afford- that are inconsistent with a goal, they are more likely to behave con- able, but not close to entertainment options and one that was more sistently with the goal (Förster and Liberman 2007). However, the expensive, but close to entertainment options. Participants in the re- process of inhibiting behavior may also undermine self-regulation jection (choice) condition were asked to decide which apartment to because inhibition not only makes consumers more sensitive to reject (choose). As predicted, participants were more likely to select threats, but it also makes them more sensitive to rewards. Thus, self- the more expensive, but hedonically rewarding apartment when they regulation efforts should be more likely to fail when external factors rejected (vs. chose), but this effect primarily emerged for consumers either reduce the perception of threat or increase consumers’ atten- who were highly focused on inhibiting their behavior. tion to rewards. This research contributes to existing models of self-regulation The purpose of study 1 was to demonstrate that performance by integrating situational factors and the decision context into self- monitoring makes consumers more susceptible to cues associated regulatory efforts. While other models suggest that external factors with goal failure. This was investigated by examining how the pres- can reduce consumers’ ability to regulate their behavior, many of ence of credit card debt, which is associated goal failure (Wilcox et these model focus on external constraints on mental resources. The al. 2011), influences spending. Whether or not consumer had credit current model suggests that in addition to the amount of resources card debt was made salient to them prior to participating in an auc- consumers have available, their interpretation of the environment tion where they submitted bids on products. The extent to which they also play a critical role in self-regulation efforts. engaged in performance monitoring during the auction was then measured. As expected, participants who were highly focused on INTRODUCTION performance monitoring who had debt submitted higher bids than Self-regulation is fundamental to consumers’ ability to achieve those who did not have debt. Spending did not differ between those their goals. It allows them to save for retirement, adhere to social who did (vs. did not) for those who were less focused on perfor- standards and maintain their health (Fishbach and Shah 2006). Self- mance monitoring. regulation failures are associated with a number of harmful behaviors The objective of study 2 was to demonstrate that performance including overspending, unhealthy eating and procrastination. Thus, monitoring makes consumers more susceptible to cues of goal understanding the factors that lead consumers to fail to regulate their achievement. Specifically, it examined whether performance moni- behavior is important because self-regulation is a significant mecha- toring would moderate consumers response to incidental pride, nism for maintaining well-being.

Most existing self-regulation models focus on the psychologi- 2009). Similarly, environmental factors associated with goal failure, cal processes that promote goal-consistent behavior. These models such as missing deadlines (Soman and Cheema 2004), can undermine often assume that when individuals are motivated and have the abil- self-regulation. Moreover, simply anticipating failure due to environ- ity to regulate their behavior, they are more likely to act consistently mental factors can undermine self-regulation (Bandura 1986). with their goals (e.g., Baumeister, Vohs and Tice 2007; Shiv and Fe- The implication of these findings is that performance monitoring dorikhin 1999). However, there is abundant evidence that external should make consumers more likely to rely on external cues associ- factors can lead to self-regulation failures, even when consumers are ated with goal achievement and goal failure. Thus, when these cues sufficiently motivated and have the ability to self-regulate (e.g., Fish- are present, people who are more engaged in performance monitoring bach and Dhar 2005). Thus, a more complete model of consumer during efforts at self-regulation should act more inconsistently with self-regulation is needed to account for how consumers respond to their goals. While this proposition has not be empirically tested, indi- external factors. rect support comes from research demonstrating that consumers high The current research introduces a model of consumer self-reg- in self-control, who should be more focused on monitoring their be- ulation failure that explains how external factors shape consumers’ havior, are more susceptible to external factors associated with goal ability to act consistent with their goals and when these factors are achievement and goal failure (Wilcox et al. 2009; Wilcox et al. 2011). more likely to undermine self-regulation. Hypothesis 1 The more focused consumers are on perfor- A MODEL OF CONSUMER SELF-REGULATION mance monitoring during efforts at self-regu- FAILURE lation, the more likely their self-regulation will be undermined by external cues associated with The Model goal achievement. As Figure 1 shows, a precondition for self-regulation is that consumers must perceive a conflict between a goal and a competing alternative (Myrseth and Fishbach 2009). Whether it is the con- Hypothesis 2 The more focused consumers are on perfor- sideration of a temptation or the presence of alternative goals, goal mance monitoring during efforts at self-regu- conflict initiates the need for self-regulation. lation, the more likely their self-regulation will Once consumers experience goal conflict, the model probe undermined by external cues associated with poses that consumers’ behavior is influence by two self-regulatory goal failure. processes: performance monitoring and the inhibition of behavior. Performance monitoring refers to the monitor of past and current Inhibition of Behavior behavior relative to a standard to ensure that behavior is consistent Even when consumers are sufficiently motivated to pursue their with a focal goal (Carver and Scheier 1990). Inhibition of behavior goals, effective self-regulation requires the inhibition of responses refers to the suppression of responses (e.g., thoughts and actions) that undermine goal pursuit. Inhibition involves directing attention that threaten goal pursuit (Förster and Liberman 2007). If individu- away from information that might interfere with the pursuit of a focal als monitor their performance and inhibit their behavior, this often goal (Förster and Liberman 2007). When consumers inhibit alterna- results in goal-consistent behavior. tives that are inconsistent with a goal, they are more likely to behave Environmental factors and decision context factors, however, consistently with the goal. can undermine self-regulation by serving as external cues. These However, the process of inhibiting behavior may also under- factors result in goal-inconsistent behavior by serving as cues of mine self-regulation because inhibition not only makes consum- goal achievement, goal failure, goal threat and reward. Performance ers more sensitive to threats, but it also more sensitive to rewards monitoring makes consumers more likely to rely on cues related to (Gendall et al. 1998; Jappe et al. 2011). Thus, self-regulation efforts goal attainment and goal failure. The inhibition of behavior makes should be more likely to fail when external factors either reduce the consumers more reliant on cues related to goal threat and reward. perception of threat or increase consumers’ attention to rewards. In sum, the model proposes that many self-regulation failures Consistent with this perspective research finds that chronic dieters, occur because two processes involved in self-regulation (perfor- who tend to be more focused on inhibiting food responses, are more mance monitoring and the inhibition of behavior) make consumers likely to overeat in the presence of external cues, such as smaller more like to rely on external cues. Testable hypotheses are developed package sizes (Scott et al. 2008), indicating that a food item is less next. threatening. Additionally, chronic dieters have been shown to be Performance Monitoring less likely to regulate their food intake because of more accessible A key process involved in performance monitoring is the detec- thoughts about food (Gendall et al. 1998). tion of goal discrepancies, which involves a comparison of behavior Hypothesis 3 The more focused consumers are on inhibiting (past or current) with a desired end state. When consumers detect their behavior during efforts at self-regulation, that their behavior is discrepant with their goals, they adjust their the more likely their self-regulation will be un- behavior to ensure that it is consistent with their goals. During this dermined by external cues that reduce the per- process, however, consumers may disengage from the goal, if they ception of goal threat. perceive that the goal has been achieved or violated in some way. Consumers often determine goal achievement or failure by explicitly tracking their goal performance. However, research demon- Hypothesis 4 The more focused consumers are on inhibiting strates that consumers also rely on external cues associated with goal their behavior during efforts at self-regulation, achievement and failure. For instance, consumers are less likely to reg- the more likely their self-regulation will be un- ulate their behavior when they derive a sense of achievement from en- dermined by external cues associated with re- vironmental and decision context factors such as acts of charity (Khan wards. and Dhar 2006) and the composition of a choice set (Wilcox et al. Advances in Consumer Research (Volume 46) / 855 STUDY 1 STUDY 3 The purpose of study 1 was to demonstrate that performance Study 3 sought to demonstrate that the inhibition of behavior monitoring makes consumers more susceptible to cues associated makes consumers more susceptible to external cues that reduce the with goal failure (H1). This was investigated by examining how the perception of goal threat (H3). Based on previous research which presence of credit card debt, which is associated goal failure (Wilcox finds that using a currency with a small nominal value is less threat- et al. 2011), influences spending. ening to spending goals (Raghubir and Srivastava 2002), it was predicted that when consumers used a currency with small (large) nomi- Participants and procedure nal value they would be less (more) likely to regulate their spending, 153 participants from an online panel were recruited for a small an effect that would be strongest for participants focused on inhibit- payment (M = 36; 46% female). After completing several demo- Age ing their behavior. graphic questions, participants were asked to indicate whether they currently had credit card debt (yes/no). Thus, goal failure was made Participants and Procedure salient before participants attempted to regulate their spending. Af- 152 participants from an online panel were recruited for a small terwards, participants were instructed that the purpose of the study payment (MAge = 36; 49% female). Participants were instructed that was to test a new auction web site and that they would be making they would be testing a new auction web site that would require them actual bids on products. They then submitted bids on two products: a to make bids using a new currency where the nominal value was set digital camera and a fitness tracker. at either a multiple (large nominal value) or a fraction (small nominal Afterwards, performance monitoring was measured on a two- value) of the US dollar. Participants then placed bids on the same item scale that asked participants the extent to which they monitored products and completed the same measures of performance monitor- their performance [progress] relative to their spending goals while ing (r = .76) and inhibition of behavior (r = .85) as previous studies. making their bids (1 = “not at all” and 7 = “very much”, r = .75). Results Inhibition of behavior was then measured on a two-item scale by The predicted nominal value by inhibition of behavior interac- having them indicate how focused they were on inhibiting [control- tion was significant (t(148) = 2.54, p = .01). Supporting hypothesis ling] their behavior while making their bids (1 = “not at all” and 7 = 3, a floodlight analysis found that participants submitted higher bids “very much”, r = .84). Inhibition of behavior was measured to rule when the nominal value was small (vs. large) for inhibition of behav- out the possibility that other processes involved in self-regulation ior scores above 5.41. Spending did not differ between nominal val- may moderate consumers’ response to goal failure. ue conditions for any score below this threshold. Performance moni- Results toring did not moderate the effect of the nominal value on spending The predicted debt by performance monitoring interaction was (interaction: t(148) = -.51, p = .61). significant (t(149) = 2.14, p = .03). Consistent with hypothesis 1, a floodlight analysis found that participants who had debt submitted STUDY 4 higher bids than those who did not have debt for performance moni- Study 4 sought to demonstrate that the more focused consum- toring scores above 4.49. Spending did not differ between those who ers are on inhibiting their behavior, the more likely their self-regu- did (vs. did not) have debt for scores below this threshold. A similar lation will be undermined by reward cues (H4). Based on previous analysis with inhibition of behavior did not find that it moderated the research that demonstrates that the decision to reject (vs. choose) effect of debt on spending (interaction:t (149) = .62, p = .54). options makes people more likely to elaborate on hedonic features during a decision (Dhar and Wertenbroch 2000), it was predicted that STUDY 2 consumers would be more likely to select a higher priced option with The objective of study 2 was to demonstrate that performance more hedonic benefits when they rejected (vs. chose). However, this monitoring makes consumers more susceptible to cues of goal effect was predicted to be strongest for participants highly focused achievement (H2). Specifically, it examined whether performance on inhibiting their behavior. monitoring would moderate consumers response to incidental pride, Participants and Procedure which has been shown to induce a sense of goal achievement that 200 participants from an online panel were recruited for a small reduces self-regulation (Wilcox et al. 2011). payment (MAge = 37; 44% female). Participants were instructed that Participants and procedure they were deciding between two apartments to rent: one that was

150 participants from an online panel were recruited (MAge = 35; affordable, but not close to entertainment options and one that was 45% female). Participants were instructed that they would be partici- more expensive, but close to entertainment options. Participants in pating in two unrelated studies. The first study manipulated pride by the rejection (choice) condition were then asked to decide which having participants write about a “pride” experience or their typical apartment to reject (choose). Participants then completed the same day (control group). The second study was the same auction from measures of performance monitoring (r = .74) and inhibition of be- study 1 that included the same measures of performance monitoring havior (r = .87) as previous studies. (r = .73) and inhibition of behavior (r = .86). Results Results The predicted decision framing by inhibition of behavior in- The predicted pride by performance monitoring interaction was teraction was significant (Wald2 χ = 22.40, p < .001). In support significant (t(146) = 2.06, p = .04). In support of hypothesis 2, a of hypothesis 4, a floodlight analysis found that participants were floodlight analysis found that participants who experienced pride more likely to select the more expensive, but hedonically reward- submitted higher bids than those in the control group for perfor- ing apartment when they rejected (vs. chose) for any inhibition of mance monitoring scores above 4.43. Spending did not differ be- behavior scores above 4.59. Selection of the expensive apartment tween those who did (vs. did not) experience pride for any score did not differ between nominal value conditions for any score below below this threshold. The inhibition of behavior did not moderate this threshold. Performance monitoring did not moderate the effect the effect of pride on spending (interaction:t (146) = -.20, p = .84). 856 / A Model of Consumer Self-Regulation Failure of the framing of the decision on spending (interaction: Wald χ2 = Fishbach, Ayelet, and James Y. Shah (2006), “Self-Control in 1.24, p = .27). Action: Implicit Dispositions Toward Goals And Away From Temptations,” Journal Of Personality and Social DISCUSSION Psychology, 90 (5) 820-32. This research introduces and tests a model of self-regulation Jappe, Leah M., Guido KW Frank, Megan E. Shott, Michael DH failure that explains how external factors influence consumers’ ef- Rollin, Tamara Pryor, Jennifer O. Hagman, Tony T. Yang, and forts to regulate their behavior. The model proposes that environ- Elizabeth Davis (2011), “Heightened Sensitivity to Reward mental factors and decision context factors can undermine self-regu- and Punishment in Anorexia Nervosa,” International Journal lation by serving as external cues for goal achievement, goal failure, of Eating Disorders, 44 (4), 317-24. goal threat and rewards. The results of four studies demonstrates that Khan, Uzma, and Ravi Dhar (2006), “Licensing Effect in Consumer self-regulation failures often occur because two key processes that Choice,” Journal of Marketing Research, 43, (2), 259-66. typically facilitate self-regulation, performance monitoring and the Förster, Jens, and Nira Liberman (2007), “Knowledge inhibition of behavior, make consumers more likely to rely on these Activation,” Social Psychology: Handbook of Basic external cues. Principles, 2, 201-31. This model contributes to existing models of self-regulation by Gendall, Kelly A., Peter R. Joyce, Patrick F. Sullivan, and Cynthia integrating situational factors and the decision context into self-reg- M. Bulik (1998), “Food Cravers: Characteristics of Those ulatory efforts. While other models suggest that external factors can Who Binge,” International Journal of Eating Disorders, 23 reduce consumers’ ability to regulate their behavior, many of these (4), 353-60. model focus on external constraints on mental resources. The current Myrseth, Kristian Ove R., and Ayelet Fishbach (2009), “Self- model suggests that in addition to the amount of resources consum- Control: A Function of Knowing When and How to Exercise ers have available, their interpretation of the environment also play a Restraint,” Current Directions in Psychological Science, 18 critical role in self-regulation efforts. (4), 247-52. This research has implications for policy makers and compa- Raghubir, Priya, and Joydeep Srivastava (2002), “Effect of Face nies who want to facilitate Self-regulation by suggesting ways to Value on Product Valuation in Foreign Currencies,” Journal of improve consumers’ ability to stick to their goals. If self-regulatory Consumer Research, 29 (3), 335-47. efforts are shaped by how consumers think about their environment, Scott, Maura L., Stephen M. Nowlis, Naomi Mandel, and Andrea self-regulation efforts should be improved by designing interven- C. Morales (2008), “The Effects of Reduced Food Size and tions that train consumers on more effective ways to respond to their Package Size on the Consumption Behavior of Restrained and environment. Unrestrained Eaters,” Journal of Consumer Research, 35 (3), One limitation of this research is that it only examines consum- 391-405. ers’ response to goal conflict in situations that involve tradeoffs be- Shiv, Baba, and Alexander Fedorikhin (1999), “Heart and Mind in tween temptations and long-term goals. However, this model should Conflict: The Interplay of Affect and Cognition in Consumer also apply to other self-regulation contexts where goal conflict does Decision Making,” Journal of Consumer Research, 26 (3), not necessarily involve temptation. Nevertheless, future research is 278-92. necessary to explore whether the same predictions hold in other self- Soman, Dilip, and Amar Cheema (2004), “When Goals are regulation contexts. Counterproductive: The Effects of Violation of a Behavioral Goal on Subsequent Performance,” Journal of Consumer REFERENCES Research, 31 (1), 52-62. Bandura, Albert (1986), Social Foundations of Thought and Action: Wilcox, Keith, Lauren G. Block, and Eric M. Eisenstein (2011), A Social Cognitive Theory, Englewood Cliffs, NJ: Prentice- “Leave Home Without It? The Effects of Credit Card Debt Hall, Inc. and Available Credit on Spending,” Journal of Marketing Baumeister, Roy F., Kathleen D. Vohs, and Dianne M. Tice (2007), Research, 48 (SPL), S78-S90. “The Strength Model of Self-Control,” Current Directions in Wilcox, Keith, Thomas Kramer, and Sankar Sen (2010), Psychological Science, 16 (6), 351-55. “Indulgence or Self-Control: A Dual Process Model of the Carver, Charles S., and Michael F. Scheier (1990), “Origins and Effect of Incidental Pride on Indulgent Choice,” Journal of Functions of Positive and Negative Affect: A Control-Process Consumer Research, 38 (1), 151-63. View,” Psychological Review, 97 (1), 19-35. Wilcox, Keith, Beth Vallen, Lauren Block, and Gavan J. Fitzsimons Dhar, Ravi, and Klaus Wertenbroch (2000), “Consumer Choice (2009), “Vicarious Goal Fulfillment: When the Mere between Hedonic and Utilitarian Goods,” Journal of Presence of a Healthy Option Leads to an Ironically Indulgent Marketing Research, 37 (1), 60-71. Decision,” Journal of Consumer Research, 36 (3), 380-93. Fishbach, Ayelet, and Ravi Dhar (2005), “Goals as Excuses or Guides: The Liberating Effect of Perceived Goal Progress on Choice,” Journal of Consumer Research, 32 (3), 370-77. Is Warm Always Trusting? The Effect of Seasonality on Trustworthiness Gretchen R. Wilroy, Penn State University, USA Margaret G. Meloy, Penn State University, USA Simon J. Blanchard, Georgetown University, USA

EXTENDED ABSTRACT Participants then completed trust measures modified from May- Sensory marketing has shown that non-conscious triggers and er and Davis (1999) on three dimensions: ability, benevolence, and sensory experiences influence consumers’ perceptions, judgments, integrity. Participants rated each statement on a 7-point scale (1 = and behaviors (Krishna 2012). The current work focuses on tempera- strongly disagree, 7 = strongly agree). Finally, we asked two manipu- ture and how it subsequently affects perceptions of trustworthiness. lation check questions, which confirmed the success of the tempera- Past research often suggests warm (vs. cold) ambient temperatures ture manipulation. increase feelings of social proximity (IJzerman and Semin 2009), An average trust score was calculated: ability (α = .980), be- and more directly, that warm temperatures increase interpersonal nevolence (α = .976), and integrity (α = .821). Neither the main effect trust (Kang et al. 2011). of temperature (F(3, 222)= 1.055, p= .369) nor season (F(3, 222)= We contend that ambient temperature is only part of the story. .370, p= .774) was significant. There was a marginally significant We incorporate research on thermoregulation, the need to maintain a interaction between temperature and season on overall trust, F(3, core internal temperature for survival (Tavassoli 2009). Bruno, Mel- 222)= 2.12, p= .098. Importantly, ability, benevolence, and integrity nyk, and Volckner (2017) found that when consumers were physi- individually all showed a significant temperature by season interac- cally cold (warm) they perceived emotionally warm (cold) stimuli tion: ability, F(1, 224)= 4.83, p= .029, benevolence F(1, 224)= 6.17, more positively. Thus, although past research has focused on ambi- p= .014, and integrity F(1, 224)= 4.77, p= .03. ent temperature in an absolute sense, it has yet to examine the ef- In winter, trust did not differ whether it was warm or cold inside. fect in a relative sense (e.g., interaction of seasonality and indoor However, in summer, trust ratings were significantly higher when temperature). In the context of evaluating trust in service providers, it was cold inside the office compared to warm: ability (MW= 4.61, we demonstrate that an unexpected ambient indoor temperature that MC= 5.10, p= .02), benevolence (MW= 4.13, MC= 4.74, p= .007), contrasts with the temperature outside affects consumers’ trust. and integrity (MW= 3.98, MC= 4.53, p= .006). This contradicts prior research findings – here, cold temperatures in the summertime led to STUDY 1: LONGITUDINAL STUDY higher trust ratings. Study 1 was conducted longitudinally to investigate whether trust and inside temperature are affected by the summer season. Us- STUDY 3: SEASONALITY, TEMPERATURE, AND ing the same MTurk participants over time allowed us to ensure that TRUST AT A RESTAURANT temperature variations were consistent within person. We predicted Similar to the design of study 2, temperature inside a restaurant that, in contrast to prior research, consumers would be more trusting was manipulated. We predicted that trust would be highest when the when it is colder inside. inside temperature met the expectations of the consumer, subcon- The study was conducted over a two-month period (June-Au- sciously aiding thermoregulation. gust). The initial survey was sent to 600 participants and 221 re- MTurk participants (n= 244, 50% female, median age=36) were mained at its conclusion (42% female, median age=34). Five con- randomly assigned to one of four conditions – season (summer/win- sumer scenarios, all focusing on service providers, were randomly ter) and inside temperature (cold/warm) in a 2x2. As in study 2, par- presented to each participant, approximately every two weeks. Par- ticipants completed trust measures modified from Mayer and Davis ticipants rated how trustworthy the representative in the scenario ap- (1999). A manipulation check question confirmed the manipulation. peared and the actual inside temperature of their current location. An average trust score was calculated for ability (α = .894), be- A fixed effects regression found a marginally significant nega- nevolence (α = .908), and integrity (α = .790). A 2(temperature: warm tive relationship between trust and inside temperature (b= -0.02, p= vs. cold) x 2(season: summer vs. winter) on the three trust measures .065). This finding provides preliminary evidence that during the indicated a marginal main effect of temperature (F(3, 238)= 2.24, p= summertime, when it is colder inside, individuals are more trusting .08). The main effect of season was not significant (F(3, 238)= 1.48, of service providers. p= .22). These main effects were qualified by a significant interaction between temperature and season, F(3, 238)= 3.65, p= .01. Each STUDY 2: SEASONALITY, TEMPERATURE, AND element of trust had a significant temperature by season interaction: TRUST AT THE DENTIST ability F(1, 240)= 7.58, p= .006, benevolence F(1, 240)= 10.52, p= Study 2 examined the interaction between seasonality and in- .001, and integrity F(1, 240)= 9.99, p= .002. In the summer, there side temperature in a medical domain. We designed an experiment was not a significant difference when the restaurant was warm or which varied the temperature inside a dentist’s office and the season. cold. However, in winter, the trust measures were significantly af- We predicted that if the inside temperature at the dental office did fected by indoor temperature. Trust was higher when it was warm not stabilize a person’s core temperature, trust in the dentist would inside the restaurant compared to cold: ability (MW= 5.13, MC= 4.37, be lower. p<.001), benevolence (MW=4.65, MC=3.79, p<.001), and integrity

Students (n = 234, 47% female, median age=19) completed the (MW= 4.43, MC= 3.79, p<.001). In this domain, the findings are con- study. Six participants failed an attention check, leaving 228 partici- sistent with prior research that warm temperatures engender feelings pants. In a 2 (inside temperature: warm, cold) x 2 (season: summer, of trust. However, we did not find this to be true in the summer sce- winter) design, participants imagined they were heading to their den- nario. These findings broaden our understanding of temperature and tal cleaning. The waiting room was described as either very warm or trust – warm does not always positively boost trust. freezing cold.

Advances in Consumer Research 857 Volume 46, ©2018 858 / Is Warm Always Trusting? The Effect of Seasonality on Trustworthiness DISCUSSION Kang, Yoona, Lawrence E. Williams, Margaret S. Clark, Jeremy While previous research has consistently demonstrated a posi- R. Gray, and John A. Bargh (2011), “Physical Temperature tive relationship between warmth and closeness/trust, we show that Effects on Trust Behavior: The Role of Insula,”Social a boundary exists. The trustworthiness of a retailer/service provider Cognitive and Affective Neuroscience,6 (4), 507-15. is reduced when season and inside temperature are not what the con- Krishna, Aradhna (2012), “An Integrative Review of Sensory sumer expects. Marketing: Engaging the Senses to Affect Perception, Judgment and Behavior,” Journal of Consumer Psychology, REFERENCES 22, 332-51. Bruno, Pascal, Valentyna Melnyk, and Franziska Volckner (2017), Mayer, Roger C. and James H. Davis (1999), “The Effect of the “Temperature and Emotions: Effects of Physical Temperature Performance Appraisal System on Trust for Management: A on Responses to Emotional Advertising,” International Field Quasi-Experiment,” Journal of Applied Psychology, 84 Journal of Research in Marketing, 34, 302-20. (1), 123-136. IJzerman, Hans and Gun R. Semin (2009), “The Thermometer of Tavassoli, N.T. (2009). Climate, Psychological Homeostasis, and Social Relations: Mapping Social Proximity on Temperature,” Individual Behaviors Across Cultures. In R.S. Wyer, Chi- Psychological Science, 20, 1214-20. yue Chiu, and Ying-yi Hong (Eds.), Understanding Culture: Theory, Research and Application (p. 211-221). New York: Psychology Press. “I Desire A Brand When I See How They are Different from Me”: Differential Effects of Blatant and Subtle Brand Rejection Jun (Wendy) Yan, University of Manitoba, Canada Fang Wan, University of Manitoba, Canada Americus Reed II, University of Pennsylvania, USA

EXTENDED ABSTRACT Banyan Tree’s target consumers, which was pretested to be different People choose brands to express their self-concepts (Carroll and from the university students. In the control condition, participants Ahuvia, 2006, pp. 82). When the brand they aspire to does not con- were told to complete a questionnaire. Next, all participants were sider them as valuable consumers, they feel being rejected. In this measured their brand desire (Gountas, et al., 2012), perceived attain- work, we conceptualize two forms of communication strategies of ability of the rejecting brand, manipulation check and demographic brand segmentation—blatant and subtle mode of brand rejection and questions. distinguish their effects on consumers’ brand desire. Compared with the control group, both BBR and SBR make

In prior work, blatant prejudice is to avoid any contact with out- participants feel they are being rejected (MBBR = 3.95, Mcontrol = 2.37, group members, whereas subtle prejudice is to emphasize cultural p < .000; MSBR = 3.19, Mcontrol = 2.37, F (1, 198) = 21.45, p < .000). differences between groups (e.g., Hamburger and Hewstone, 1997). BBR is perceived as more explicit than SBR, (MBBR = 4.67, MSBR = We conceptualize Blatant Brand Rejection (BBR) as a market seg- 3.26, p < .000). Interaction results supporting our hypothesis 1, that mentation that emphasizes consumers are “not good enough,” and the effect of brand rejection on brand desire is moderated by per- consumers get a direct and explicit signal that they are not wanted by ceived attainability (t = -1.99, p < .05). this brand as one of its important consumers. Subtle Brand Rejection Study 2. First, 284 American participants (Mage = 38, Male: (SBR), by contrast, emphasizes the brand’s target consumers, and 40%) recruited online were told that Parliament Pub is a local po- consumers get an indirect, concealed, and implicit signal that they litical-themed pub for the ____ Party (this blank was customized are different from the target consumers. based on participants’ choice about their supporting Party). Second, Both SBR and BBR reject “unwanted” brand recipients, but we participants were randomly assigned to complete one of two surveys argue that they have different effects on consumers’ psychological relevant to their volunteer (vs. work) experience with the ___ Party. and behavioral reactions. When consumers pursue a brand but en- A pretest revealed that to have volunteer (vs. work) experience for a counter BBR, the explicit signal is a threat to their positive self-con- Party was considered more attainable (Mvolunteer = 5.92, Mwork = 5.10, cept, activating defense mechanism (Baumeister, Dale and Sommer, p <.000). After that, participants were then randomly assigned to one 1998) and making consumers to avoid the rejecting brand. Whereas, of three conditions: BBR, SBR, or control, as similarly manipulated encountering a subtle mode of rejection from a brand, consumers in Study 1. Finally, brand desire, Want-to-Pay (WTP), manipulation tend to compare themselves with the target consumers. When the tar- check, and demographic questions were measured. get consumers are perceived to be a higher standard, this aspiration When the rejecting brand was perceived as attainable (vs. unat- effect motivates consumers to (1) decrease the “distance” (e.g., Van, tainable), SBR (vs. BBR) led to an increased brand desire (MSBR =

Pecher, and Zwaan, 2008), and (2) work harder to join the group of 5.17, MBBR = 3.45, t = -3.37, p < .001) and WTP (M(logWTP)SBR = target consumers (e.g., Park, Young and Eastwick, 2015). 1.57, M(logWTP)BBR = 1.09, t = -2.17, p < .05). Furthermore, the However, this positive effect of SBR is contingent on the attain- attainable (vs. unattainable) SBR even generated more brand desire ability of group standards (Dyczewski and Markman, 2011). When and WTP than control group (MSBR = 5.17, Mcontrol = 4.32, t = -2.20, the perceived attainability of becoming a group member is low, a p < .05; M(logWTP)SBR = 1.57, M(logWTP)BBR = 1.15, t = -3.13, p sense of self efficacy and self control is threatened (Coffee, Rees and < .005). Haslam, 2009). This threat will reduce their brand desire as defense Study 3. To test the mediation effect of need for belongingness, mechanism. Therefore, we hypothesize that this study replicated study 2 except that the control condition was dropped and another variable—measurement order—was added. We Hypothesis 1 SBR (vs. BBR) will increase (vs. decrease) brand predicted that participants whose brand desire was measured first desire of a rejecting brand, when the brand is would show the same results as in Study 2, while participants whose perceived to be attainable. The positive effect of belonging needs were measured first would have no increase in brand SBR is mitigated when consumers perceive the desire (White, Argo, and Sengupta, 2012), because the activated rejecting brand is unattainable. belongingness to other groups is one coping mechanism of threat (Knowles and Gardner, 2008) and the affirmation of self-concept (Cohen et al., 2000). Results supported the hypothesis that the brand Hypothesis 2 SBR (vs. BBR) will increase (vs. decrease) brand desire of consumers who experiencing attainable SBR decreased sig- desire of a rejecting brand, when the brand is nificantly when belong need was measured first than when itwas perceived to be attainable, mediated by their in- measured after brand desire (F (1, 98) = 7.08, p < .01). The opposite creased (vs. decreased) belonging needs to the pattern was observed when belong need was measured after brand target consumers. desire (MSBR = 4.77, MBBR = 2.98, p < .000). The results of Want- Study 1. It is a 3 (BBR vs. SBR vs. neutral) by 2 (perceived to-Pay (WTP) indicated a similar pattern. Therefore, belong needs attainability: high vs. low) between subject study. 297 university stu- mediate the interaction effect of mode of rejection and perceived at- dents (Mage = 20.6, Male: 50.5%) were recruited in an online study. tainability on brand desire and WTP. A hotel brand unknown to subjects is adopted. In the BBR condition: “Based on our analysis of your information, unfortunately, you REFERENCES Carroll, B., and Ahuvia, A. (2006). Some antecedents and outcomes are not a target consumer of Banyan Tree Restaurant & Hotel.” In of brand love. Marketing Letters, 17 (2), 79-89. the SBR condition, participants were shown some information about Advances in Consumer Research 859 Volume 46, ©2018 860 / “I Desire A Brand When I See How They are Different from Me”: Differential Effects of Blatant and Subtle Brand Rejection

Coffee, P., Rees, T., and Haslam, A. (2009). Bouncing back from Park, L. E., Young, A. F., and Eastwick, P. W. (2015). failure: The interactive impact of perceived controllability and (Psychological) distance makes the heart grow fonder: effects stability on self-efficacy beliefs and future task performance. of psychological distance and relative intelligence on men’s Journal of Sports Sciences, 27, 1117 –1124. attraction to women. Personality and Social Psychology Dyczewski, E.A. and Markman, K.D. (2012), General attainability Bulletin, 41(11), 1459-1473. beliefs moderate the motivational effects of counterfactual Van, D.S., Pecher, D. and Zwaan, R.A., (2008). Approach and thinking, Journal of Experimental Social Psychology, Vol. Avoidance as Action Effects,The Quarterly Journal of 48(5), pp. 1217-1220. Experimental Psychology,Vol. 61(9), pp. 1298-1306. Gountas, J., Gountas, S., Reeves, R. A., and Moran, L. (2012). White, K., Argo, J.J. and Sengupta, J., (2012), Dissociative versus Desire for fame: scale development and association with Associative Responses to Social Identity Threat: The Role of personal goals and aspirations. Psychology and Marketing, Consumer Self-Construal, Journal of Consumer Research, Vol 29(9): 680-689. 39(4), pp. 704-719. Hamberger J, and Hewstone, M. (1997). Inter-ethnic contact as a predictor of prejudice: tests of a model in four West European nations. British Journal of Social Psychology, Vol. 36(Jun), pp. 173-190. The Slippery Slope of Green Consumption: The Nonlinear Effects of Social Class Li Yan, Monash University, Australia Hean Tat Keh, Monash University, Australia Jiemiao Chen, Monash University, Australia

EXTENDED ABSTRACT (e.g., political ideology, religiosity, and ethnicity) did not change the As environmental issues grow increasingly acute, more re- results. search is needed to understand the impact of individuals’ behaviors Experiment 2 provided process evidence for the nonlinear me- on sustainability. Given that green consumption can reflect status dif- diation effect of social acceptance at the lower level of social class ferentiation (Griskevicius et al. 2010), and social class serves as a (H2a). It used a one factor (product greenness: regular vs. green‎) dimension of consumer distinctiveness (Grier and Deshpandé 2001), between-participants design. Social class was measured as a continu- we propose that social class, as a means of consumer differentiation, ous variable. A moderated hierarchical regression analysis showed a will influence their propensity for green consumption. significant and negative quadratic-by-linear effect of 2SC × product Social class is defined as a combination of individuals’ material greenness on purchase intention (β = -.72. p < .05) after controlling resources (i.e., income, education, and job status), and self-percep- for the effects of the linear terms and covariates. Decomposing the tions of their own rank in the society vis-à-vis others (Kraus et al. quadratic-by-linear interaction showed that the nonlinear effect of 2012). It has profound psychological and cultural effect on individu- social class was significant and negative for the green product (β = als’ feelings, thoughts, and behaviors (Grossmann and Huynh 2013). -.37, p < .01), but not significant for the regular product (p > .50). However, scant research has examined how social class shapes con- MEDCURVE analysis (Hayes and Preacher 2010) revealed a sig- sumer behaviors (Shavitt et al. 2016; Yoon and Kim 2018). nificant and negative effect of the quadratic term (SC2) on social ac- The present research proposes a nonlinear (i.e., inverted U- ceptance (b = -.24‎, p < .05), which was positively related to purchase shaped) effect of social class on green consumption, such that indi- intention (b = .25‎, p < .01). Analyses of the instantaneous indirect viduals at the middle position of the class continuum show greater effect (θ) at lower (-1 SD) and higher (+1 SD) levels of social class green propensity than those at the lower or upper position (H1). Al- revealed that social acceptance could explain the positive effect at though striving for a higher social rank is a fundamental motive for the lower level of social class (θ = .21, 95% CI = .057, .476), but individuals, the means for rank advancement differ (Belmi and Lau- not at the higher level of social class (95% CI = -.092, .112), which rin 2016). Specifically, lower-class individuals believe that they need supported H2a. to make prosocial contributions to advance their status, and are thus Experiment 3 examined the mediating effect of essentialism at driven by social acceptance for green consumption (H2a). However, the higher level of social class (H2b). It used a one factor (product as the intrinsic benefits associated with social hierarchy grow expo- greenness: regular vs. green) within-participant design. Social class nentially, individuals proximal to the top attach more importance to was measured as a continuous variable using subjective SES‎. Re- justifying their advantaged standing using ideological means such gression with both linear and quadratic terms (SES and SES2) in the as essentialism. Essentialism refers to the belief that social class cat- equation showed a significant and negative effect of SES2 on prefer- egory is predetermined and immutable, and such ideology helps to ence for green product (SES: β = .08‎, p > .14, SES2: β = -.14, p < justify disparities in social rank (Kraus and Keltner 2013). Thus, we .05). MEDCURVE analysis revealed a significant and positive effect propose that essentialism would demotivate individuals from engag- of SES2 on essentialism (b = .04, p < .05), which was negatively re- ing in green consumption (H2b). Finally, we propose that individu- lated to preference for the green product (‎b = -.23, p < .05). Analyses als’ belief in a just world (BJW) moderates the nonlinear effect of of the instantaneous indirect effect showed that essentialism could social class on green consumption (H3). BJW refers to the belief explain the negative effect at the higher level of social class (θ = -.05, that the world is a just place where people deserve what they get 95% CI = -.128, -.008), but not at the lower level of social class (95% and get what they deserve (Lerner 1980). We suggest that high BJW CI = -.170, .064), supporting H2b. promotes individuals’ acquiescence of perceived differences in social Experiment 4 tested the overall model and the moderating ef- class, which attenuates the nonlinear effects of social class difference fect of BJW using a similar design as experiment 3. Moderated re- on green consumption. gression results revealed a significant and negative effect of SES2 (β We tested our hypotheses in four experiments. Experiment 1 = -.20, p < .01), and a significant and positive quadratic-by-linear examined the main effect of social class (H1).Experiment 1 used a 3 effect of SES2 × BJW on preference for the green product (β = .20, (primed social class: lower vs. middle vs. upper) × 2 (product green- p < .05). Further analysis of the quadratic-by-linear interaction indi- ness: green vs. regular) between-participants design. A 3 × 2 ANOVA cated that the effect of SES2 was significant at low BJW (b = -.19, p ‎revealed only a significant interaction effect of social class and prod- < .001), but not at high BJW (p > .50). Thus, H1 and H3 were sup- uct greenness (p < .05). Decomposing the interaction showed that ported. MEDCURVE analyses showed that social acceptance could social class significantly influenced purchase ‎intentions for the green explain the positive effect at the lower level of social class (θ = .04, product (p < .05), but not for the regular ‎product (p > .20). Impor- 95% ‎CI = .000, .111) while essentialism could explain the negative tantly, polynomial contrast showed a significant quadratic effect of effect at the higher level of social class (θ = -.05, 95% CI = -.104, social class on purchase intention for the green product (p < .001), -.001), supporting H2a and H2b. ‎but no linear effect (p ‎‎> .50). Specifically, participants in middle- Taken together, four experiments using varying products and class condition showed higher purchase intention for the green prod- designs provided consistent evidence for the nonlinear effect of so- uct than those in the lower-class (p < .05) and upper-class conditions cial class on green consumption, as explained by social acceptance (p < .05). In contrast, the three class conditions were not significantly and essentialism, and moderated by BJW. This curvilinear result can different in purchase intentions for the ‎regular product (all ps > .14). help reconcile conflicting findings in the literature based on the two- Thus, H1 was supported. For all studies, controlling for the objective level approach. In particular, it addresses the middle class, an under- social class indicators (i.e., income, education) and other covariates theorized social category, which is a powerful category for self-iden-

Advances in Consumer Research 861 Volume 46, ©2018 862 / The Slippery Slope of Green Consumption: The Nonlinear Effects of Social Class tification and plays an important role in economic growth, political Hayes, Andrew F. and Kristopher J. Preacher (2010), “Quantifying transformation, and social restructuring in many societies. In sum, and Testing Indirect Effects in Simple Mediation Models our findings contribute to the literature on consumer differentiation, When the Constituent Paths are Nonlinear,” Multivariate social class and sustainable consumption, as well as have important Behavioral Research, 45, 627-60. practical implications. Kraus, Michael W. and Dacher Keltner (2013), “Social Class Rank, Essentialism, and Punitive Judgment,” Journal of Personality REFERENCES and Social Psychology, 105(2), 247-61. Belmi, Peter and Kristin Laurin (2016), “Who Wants to Get to Kraus, Michael W., Paul K. Piff, Rodolfo Mendoza-Denton, the Top? Class and Lay Theories about Power,” Journal of Michelle L. Rheinschmidt, and Dacher Keltner (2012), Personality and Social Psychology, 111 (4), 505-29. “Social Class, Solipsism, and Contextualism: How the Rich Grier, Sonya A. and Rohit Deshpandé (2001), “Social Dimensions are Different from the Poor,” Psychological Review, 119 (3), of Consumer Distinctiveness: The Influence of Social Status 546-72. on Group Identity and Advertising Persuasion,” Journal of Lerner, Melvin J. (1980), The Belief in a Just World: A Marketing Research, 38 (2), 216-24. Fundamental Delusion. New York: Plenum Press. Griskevicius, Vladas, Joshua M. Tybur, and Bram Van den Bergh Shavitt, Sharon, Duo Jiang, and Hyewon Cho (2016), “Stratification (2010), “Going Green to Be Seen: Status, Reputation, and and Segmentation: Social Class in Consumer Behavior,” Conspicuous Conservation,” Journal of Personality and Social Journal of Consumer Psychology, 26 (4), 583-93. Psychology, 98 (3), 392-404. Yoon, Sunyee and Hyeongmin Christian Kim (2018), “Feeling Grossmann, Igor and Alex C. Huynh (2013), “Where is the Culture Economically Stuck: The Effect of Perceived Economic in Social Class?” Psychological Inquiry, 24 (2), 112-19. Mobility and Socioeconomic Status on Variety Seeking,” Journal of Consumer Research, 44 (5), 1141-56. Do You Trust the System? Interaction Effect between Perceived Economic Mobility and Socioeconomic Status on Fair Market Ideology and Consumer Responses Chun-Ming Yang, Ming Chuan University, Taiwan Chia-Chi Chang, National Chiao Tung University, Taiwan

EXTENDED ABSTRACT consumers and mediated by FMI. To test our prediction, adult U.S. The increasing inequality of wealth in modern societies may consumers were randomly assigned to either a high-PEM or low- limit the ability of low socioeconomic status (SES) consumers to PEM condition. In the high-PEM and low-PEM conditions, partici- achieve a higher SES. Although there have been extensive studies pants read an article that compellingly presented arguments for high on inequality and SES in the fields of psychology and economics, and low economic mobility in the United States, respectively. The scant research has examined how consumer SES and perceived eco- articles were similar in length and style. After reading the article, nomic mobility (PEM) jointly influence decisions (Yoon and Kim participants were asked to choose between purchasing coffee from 2016, 2018). Starbucks or an independent coffee shop. Regression analyses pro- On the basis of system justification theory, the author predicts vided supporting evidence. Using FMI as the dependent variable, the that high-PEM consumers have a higher degree of fair market ideol- results replicated the findings observed in previous studies. A boot- ogy (FMI) than other consumers (Jost et al. 2003). High FMI leads to strap analysis supported the indirect path from PEM through FMI to various responses, including higher price–quality relationship beliefs the tendency to support underdogs support. and even the rejection of government health policies. FMI, defined In the literature, it has been suggested that a consumer’s attitude as the tendency to view market-based processes and outcomes as le- toward consumer-related policy is affected by system confidence gitimate and just, is a cognitive underpinning of economic system (Shepherd, Chartrand, and Fitzsimons 2015). For participants with justification. People with high FMI have a high propensity to defend a high degree of system confidence, policy arguments that appeal to and justify the existing economic system. The author argued that the dominant U.S. social ideology (i.e., power) will be more persua- consumers with high PEM would report higher FMI than those with sive than arguments appealing to universalism; simply supporting low PEM, because consumers with high PEM rely on the current these arguments may diminish consumer wellbeing. Therefore, we economic system to provide resources and opportunities to move predicted that high-PEM consumers would evaluate power appeal upward in the economic hierarchy. This argument is consistent with arguments as being more persuasive than their low-PEM counter- recent findings in social psychology (Day and Fiske 2017). parts would. Adult U.S. participants were first asked to evaluate the In the current study, it was predicted that the effect of PEM persuasiveness of an argument against a ban on trans fats by the U.S. on FMI would be stronger for consumers with a low SES. This is Food and Drug Administration. After reading the argument, partici- because, compared with high-SES consumers, low-SES consum- pants used seven items to evaluate the argument’s persuasiveness. ers have fewer resources with which they can secure future wellbe- Regression analyses suggested that PEM had a positive effect on per- ing. The current social system largely determines whether they can suasiveness. We also observed a marginally significant PEM–SES achieve a higher SES. If they have the opportunity to move upward, interaction. For low-SES consumers, PEM positively affected per- they should tend to justify the system and even accept social inequal- suasiveness; however, PEM had no effect on persuasiveness for high- ity. PEM should not be a concern for high-SES consumers, because SES consumers. The relationship between PEM and persuasiveness they have sufficient financial resources. A series of studies tested our was mediated by FMI. predictions and provided supporting evidence. Our research contributes to the literature in several ways. First, In the pilot studies, participants from Taiwan and the United we demonstrated that consumer PEM and SES are critical but under- States completed an online questionnaire. Regression analyses re- explored factors influencing consumer decisions through assessing vealed a consistent pattern in both samples. Supporting our hypoth- FMI as a novel psychological consequence. Second, the findings of esis, PEM had a positive effect on FMI. Moreover, these effects were Study 1 and 2 have crucial implications for practitioners. When using qualified by a significant PEM–SES interaction. Further analyses price to signal product quality, high-PEM consumers are the better indicated that for low-SES consumers, PEM had a positive effect on target group, and practitioners should design marketing messages to FMI; however, PEM did not affect high-SES consumers. enhance consumers’ PEM. Moreover, our findings contribute to so- In Study 1, Taiwanese consumers were randomly assigned to cial mobility literature by demonstrating that high-PEM consumers either a high-PEM or low-PEM condition. Participants first read the not only rationalize inequality but may also reject consumer-related statement “Everyone has a fair chance at moving up the economic policies if they believe that these policies are against the dominant ladder.” Participants in the high (low) PEM conditions respectively social ideology. Future research should determine which factors wrote three arguments in support of (against) this statement. Par- eliminate the effect of FMI on policy rejection. ticipants then reported their price–quality relationship belief, FMI, childhood SES, and current SES. A regression analysis indicated a REFERENCES significant PEM–SES interaction. For low-SES consumers, PEM Day, Martin V. and Susan T. Fiske (2017), “Movin’ on Up? How positively affected the perception of a price–quality relationship. Perceptions of Social Mobility Affect Our Willingness to However, the effect of PEM was not observed for high-SES consum- Defend the System,” Social Psychological and Personality ers. A significant PEM–SES interaction was also revealed in relation Science, 8(3), 267–274. to FMI. For only low-SES consumers, PEM positively affected FMI. Jost, John T., Sally Blount, Jeffrey Pfeffer, and György Hunyady The results of a bootstrap analysis confirmed the indirect path from (2003), “Fair Market Ideology: Its Cognitive-Motivational PEM through FMI to the price–quality relationship belief. Underpinnings.” Research in Organizational Behavior, 25, In Study 2, the author predicted that high-PEM consumers 53-91. would affect consumer’s willingness to support underdog brands. the author also predicted that this effect would be stronger for low-SES

Advances in Consumer Research 863 Volume 46, ©2018 864/ Do You Trust the System? Interaction Effect between Perceived Economic Mobility and Socioeconomic Status on Fair Market Ideology and Consumer Responses

Shepherd, Steven, Tanya L. Chartrand, and Gavan J. Fitzsimons ______and ______(2018), “Feeling Economically (2015), “When Brands Reflect Our Ideal World: The Values Stuck: The Effect of Perceived Economic Mobility and and Brand Preferences of Consumers Who Support versus Socioeconomic Status on Variety Seeking,” Journal of Reject Society’s Dominant Ideology,” Journal of Consumer Consumer Research, 44(5), 1141-1156. Research, 42(1), 76–92. Yoon, Sunyee, and Hyeongmin Christian Kim. (2016), “Keeping the American dream alive: The interactive effect of perceived economic mobility and materialism on impulsive spending.” Journal of Marketing Research, 53(5), 759-772. Why Do People Who Have More Enjoy Horror More? Haiyang Yang, Johns Hopkins University, USA Kuangjie Zhang, Nanyang Technological University, Singapore

EXTENDED ABSTRACT potential affective differences due to the mindset manipulation we From horror films and TV series to haunted house attractions used, might not easily explicate the pattern of results we observed. and terrifying Halloween goods, people consume a wide variety of Studies 3 and 4 tested our proposed sense-of-control based horror-related experiences and products. In fact, the market demand mechanism. In study 3, participants primed with a resource abun- for horror consumption has been rising in the past decades, making dance (vs. scarcity) mindset indicated an increased sense of control, the genre of horror one of the most widely consumed form of enter- which mediated the higher likelihood that they would consume a tainment. Yet, the extant understanding of this type of consumption haunted house experience. Study 4 directly manipulated participants’ remains relatively limited (Andrade and Cohen 2007). For example, sense of control and showed that the advantage of resource abun- when might consumers be more or less likely to engage in horror dance (vs. scarcity) mindset on horror consumption was eliminated consumption? What might underlie such difference in preference? after participants’ sense of control was reduced via a control depriva- Our research seeks to help answer these questions. tion manipulation (Whitson and Galinsky 2008). Recent research suggests that consumers frequently experience Our research findings contribute to the literature in multiple perceived scarcity (vs. abundance) in resources such as goods and ways. First, complementing the extant research on consumption of wealth (e.g., Booth 1984; Mullainathan and Shafir 2013; Shah, Mul- experiences that engender negative emotions (Andrade and Cohen lainathan, and Shafir 2012; Roux, Goldsmith, and Bonezzi 2015; 2007; Scott et al. 2017), we show that resource availability mind- Zhu and Ratner 2015; see Cannon, Goldsmith, and Roux 2018 for set can shape consumers’ preference for horror-related experiences a review). Such differences in perceived resource availability can and products. Our findings also add to this literature by demonstrat- trigger a generalized mindset, impacting subsequent consumer being that consumers’ sense of control serves as an important type of havior in unrelated contexts (e.g., Mehta and Zhu 2015; Roux et al. ‘protective frame,’ enabling them to derive more pleasure from con- 2015). Complementing and extending these prior findings, the cur- suming horror. That is, activating a resource abundance (scarcity) rent research explores how and why consumers’ activated resource mindset can increase (decrease) sense of control, thereby boosting availability mindset can shape their preference for counterhedonic (dampening) consumers’ preference for horror consumption. More- consumption. We propose that consumers with a resource abundance over, our research adds to the growing stream of research on the im- (vs. scarcity) mindset tend to have a higher preference for consuming pact of resource scarcity versus abundance on consumer behavior horror. This occurs because resource abundance (scarcity) mindset (e.g., Cannon et al. 2018; Mehta and Zhu 2016; Shah et al. 2012; increases (decreases) one’s sense of control, which serves as a type Roux et al. 2015; Zhu and Ratner 2015) by offering further evidence of ‘protective frame’—the feeling that danger and threat one is expe- of how situationally induced resource availability mindset can sub- riencing will not cause actual harm (Apter 1992; Andrade and Cohen sequently shape consumers’ preference for products and experiences 2007)—and hence allows consumers to experience more pleasure in unrelated contexts. when consuming horror. In four experimental studies, we test our propositions in a va- REFERENCES riety of consumption contexts. In study 1, using an established pro- Andrade, Eduardo B. and Joel B. Cohen (2007), “On the cedure (Roux et al. 2015), participants were primed with a resource Consumption of Negative Feelings,” Journal of Consumer abundance, scarcity, or control mindset. All participants were then Research, 34 (October), 283–300. presented with a poster ad for a haunted house experience, and were Apter, Michael J. (1992), The Dangerous Edge: The Psychology of asked to indicate how much they would enjoy visiting the haunted Excitement, New York: Free Press. house. Supporting our proposition, those primed with a resource Booth, Alan (1984), “Responses to Scarcity,” Sociological abundance/control (vs. scarcity) mindset reported higher enjoyment Quarterly, 25 (Winter), 113–24. ratings. This occurred even though a pretest showed that the particu- Cannon, Christopher, Kelly Goldsmith, and Caroline Roux (2018), lar priming procedure we used did not lead to significantly differ- “A Self-Regulatory Model of Resource Scarcity,” Journal of ent affective reactions (a pattern consistent with prior research [e.g., Consumer Psychology. Mehta and Zhu 2015; Roux et al. 2015]). Thus, potential affective Mullainathan, Sendhil and Eldar Shafir (2013),Scarcity: Why differences induced by the mindset manipulation might not easily Having Too Little Means So Much, New York: Henry Holt. account for the results. Mehta, Ravi and Meng Zhu (2015), “Creating When You Have In study 2, participants were first primed with either a resource Less: The Impact of Resource Scarcity on Product Use abundance or scarcity mindset. In a randomized order, they watched Creativity,” Journal of Consumer Research, 42 (5), 767-782. and rated two movie clips—one from the horror film The Conjur- Roux, Caroline, Kelly Goldsmith, and Andrea Bonezzi (2015), “On ing 2 and the other from the comedy-adventure film Finding Dory. the Psychology of Scarcity: When Reminders of Resource Following Andrade and Cohen (2007), participants were asked to Scarcity Promote Selfish (and Generous) Behavior,” Journal of indicate how much positive and negative emotions they experienced Consumer Research, 42, (4), 615-631. when viewing each of the clips. For the horror film clip, participants Scott, Rebecca, Julien Cayla, and Bernard Cova (2017), “Selling primed with a resource abundance (vs. scarcity) mindset experienced Pain to the Saturated Self,” Journal of Consumer Research, 44 more positive emotions when viewing the clip, but their experience (1), 22-43. of negative emotions did not differ. For the comedy clip, however, Shah, Anuj K., Sendhil Mullainathan, and Eldar Shafir (2012), no difference was found for either the positive or negative emotions. “Some Consequences of Having Too Little,” Science, 338 These results thus further supported our propositions and showed that (6107), 682–85.

Whitson, Jennifer A. and Adam D. Galinsky (2008), “Lacking Control Increases Illusory Pattern Perception,” Science, 322 (5898), 115-117. Zhu, Meng and Rebecca Ratner (2015), “Scarcity Polarizes Preferences: The Impact on Choice Among Multiple Items in a Product Class,” Journal of Marketing Research, 52 (1), 13–26. Material Gifts as Relationship Mnemonics: Why More Material Gifts Are Given Than Wanted? Adelle X. Yang, National University of Singapore, Singapore Minjung Koo, Sungkyunkwan University, Korea Jaewon Hwang, Sejong University, Korea

EXTENDED ABSTRACT intended for consumption (5-point scale), respectively. Their ratings How do people choose material and experiential purchases were highly correlated and thus averaged for each scale. Givers and when these purchases are intended as gifts for others? Following receivers’ own categorization of gifts indicated that more givers list- van Boven and Gilovich’s (2003) initial distinction between material ed a material gift than receivers did (48.6% vs. 31.2%; χ2 = 4.73, p and experiential purchases centering on “people’s intentions when = .044). Coders’ ratings of the gift descriptions further suggest that investing in their happiness”, the present research examines how gifts listed by givers were intended for possession to a greater extent people’s interpersonal motives affect their choice between material than those by receivers (Mgiver = 2.83, SD = 1.46 vs. Mreceiver = 2.22, gifts and experiential gifts. SD = 1.38; t(145) = 2.63, p = .010), while intended for consumption

Although recent research supports the superior effectiveness of to a lesser extent than those by receivers (Mgiver = 3.88, SD = 1.17 vs. experiential gifts at inducing receiver’s satisfaction (Chun and Hiang Mreceiver = 4.25, SD = 1.03), t(145) = -2.03, p = .045). These findings 2016) and fortifying relationship between the receiver and the giver suggest two parallel causes contributing to the overall giver-receiver (Chan and Mogilner 2017), givers may not have such foresight. In preference discrepancy between material and experiential gifts: giv- particular, material gifts may appear more suitable than experien- ers had a stronger preference to endow material possessions than re- tial gifts to symbolize the giver’s ingratiating action as a physical, ceivers wanted them, consistent with the mnemonic account; at the visible and durable mnemonic in the receiver’s surroundings. This same time, givers had a weaker preference to give experiences than mnemonic, however, may be viewed in different light by a giver ver- receivers wanted them, consistent with recent theorizing that givers sus a receiver. In social reciprocation, people tend to egocentrically often perceive higher social risk to choose experiential gifts to match believe that their salient input to a favor can induce feelings of ap- receivers’ uncertain preferences (Goodman and Lim 2018). All of preciation, whereas receivers tend to overlook the others’ input and these results were replicated in Study 2b, a registered replication focus on the benefits that they can derive from the favor (Flynn and (N = 276). Adams 2009; Zhang and Epley 2009). Moreover, when reminded In Studies 3 and 4, we further examined the psychological pro- of the others’ favor, the receiver tends to feel indebted, pressured, cesses underlying the mnemonic account. We found a mediation in and obliged to reciprocate, a sentiment not as desirable as the giver Study 4 (N = 490) with perspective (giver vs. receiver) as the inde- may believe (Belk and Coon 1993; Greenberg 1980; Emerson 1976). pendent variable, difference in predicted bonding instrumentality of Therefore, we theorize that givers tend to endow receivers with more gifts as the mediating variable, and gift choice as the dependent vari- material gifts (vs. experiential gifts) than receivers want them, be- able (indirect effect = .57,SE = .28, 95% CI = [.04, 1.14]). We further cause givers overestimate the extent to which material gifts facilitate manipulated motivation to bond in Study 4 (N = 445) and found that bonding as mnemonics of their kind act, compared with receivers’ a lower level of motivation to bond attenuated givers’ overestimation prediction. We present five studies and a registered replication to test of the relative bonding instrumentality of material gifts and mitigated these predictions. their relative preference for the material gift (F(1, 441) = 3.37, p = Study 1 is an incentive-compatible study with 2 (giver vs. re- .067). ceiver) between-subjects conditions. Students (N = 216) were ran- Last, we examined a practically relevant boundary condition domly paired up. After a brief interaction, those randomly assigned in Study 5 (N = 183). The mnemonic account predicts that givers to the giver condition were told that they could choose one of two desire to be thought of frequently yet would avoid being associat- options as an extra reward for their partner whereas their partner was ed with negative events in receivers’ memory. Consistent with this separately asked to indicate their preference between the same op- prediction, we found that the giver-receiver preference discrepancy tions. The choice was between a mini postcard and a mini movie fea- was mitigated when the gift was to be associated with a negative turing the same cartoon characters, rated as a typical material gift and life event instead of a positive one. A two-way ANOVA revealed the a typical experiential gift, and pretested to be similar in most aspects predicted interaction between perspective and event valence (F(1, (e.g., desirability, attractiveness, perceived cost). Givers were more 179) = 8.80, p = .003). likely to choose the postcard over the movie compared with receiv- In sum, this research demonstrates that the trade-off underlying ers (postcard choice: giver 66.4% vs. receiver 48.6%; χ2 = 6.94, p an interpersonal decision between material and experiential purchas- = .009). es systematically differs from that underlying an otherwise identical Study 2a has the same 2 (giver vs. receiver) between-subjects self-oriented decision. In particular, unlike what givers wish, their conditions. First, we asked MTurkers (N = 149) to list the name of strong desire to bond may color their judgment and yield rather coun- an acquaintance that they would like to bond with, then list a life terproductive bonding consequences. achievement important to the acquaintance and a life achievement important to themselves. After that, we randomly assigned partici- REFERENCES pants into the giver or receiver conditions and asked them to list a Belk, Russell W. and Gregory S. Coon (1993), “Gift Giving as gift that they would like to give to or receive from the listed friend, Agapic Love: An Alternative to the Exchange Paradigm Based to celebrate the desired life achievement. Participants then indicated on Dating Experiences,” Journal of Consumer Research, whether the gift was a material gift or experiential gift and briefly 20(3), 393-417. described it. Two independent coders rated the gift descriptions on Chan, Cindy and Cassie Mogilner (2017), “Experiential Gifts Foster two separate scales to determine the extent to which a gift was in- Stronger Social Relationships Than Material Gifts,” Journal of tended for possession (5-point scale), and the extent to which it was Consumer Research, 43(6), 913-931.

Chun, H. Helen and Yue Woon Hiang (2016), “Free Drink or Free Goodman, Joseph K. and Sarah Lim (2018), “When Consumers Mug? Managing Service Experience with Experiential vs. Prefer to Give Material Gifts Instead of Experiences: The Role Material Complimentary Gifts,” Service Science, 8(2), 184- of Social Distance,” Journal of Consumer Research, 45(2), 202. 365-382. Emerson, Richard M. (1976), “Social Exchange Theory,” Annual Greenberg, Martin S. (1980). A theory of indebtedness, Social Review of Sociology, 2(1), 335-362. exchange (pp. 3-26). Springer, Boston, MA. Flynn, Francis J. and Gabrielle S. Adams (2009), “Money Can’t van Boven, Leaf and Thomas Gilovich (2003), “To do or to have? buy Love: Asymmetric Beliefs about Gift Price and Feelings That is the Question,” Journal of Personality and Social of Appreciation,” Journal of Experimental Social Psychology, Psychology, 85(6), 1193. 45(2), 404-409. Zhang, Yan and Nick Epley (2009), “Self-centered Social Exchange: Differential use of Costs versus Benefits in Prosocial Reciprocity,” Journal of personality and social psychology, 97(5), 796-810. Consumer’s Local-Global Identity and Price-Quality Associations Zhiyong Yang, University of North Carolina at Greensboro, USA Sijie Sun, University of Texas at Arlington, USA Ashok K. Lalwani, Indiana University, USA Narayanan Janakiraman, University of Texas at Arlington, USA

EXTENDED ABSTRACT signed to either high or low price condition, using identical product Previous literature documents multiple factors that influence descriptions. The target brand was priced the highest (lowest) in the consumers’ price-quality associations (PQA)—their tendency to use high (low) price condition, with equal price difference toward the price as an indicator of perceived quality, including prior knowledge other two baseline brands. After reading brand descriptions, partici- (Monroe 2003), product information (Kardes et al. 2004), and cul- pants rated the target brand on the three-item quality measures as in tural background (Lalwani and Shavitt 2013). However, no research Study 1 (α’s>.85). These three items were averaged to form a quality has examined the effect of local-global identity on PQA. Extending index. A 2 (identity) × 2 (price) ANOVA with the quality index as prior research, we investigate the role of local-global identity in in- the dependent variable revealed a significant identity × price two- fluencing PQA. way interaction (F(1,118)=4.82, p<.05). As expected, locals rated the Individuals whose local identity is salient (‘locals’) are faithful target brand as having higher quality in the high (M=4.99) than in and respectful of local traditions, and identify with local commu- the low (M=3.97, t(61)=4.70, p<.01) price condition. In contrast, for nity, whereas those with a salient global identity (‘globals’) favor globals, the quality ratings did not vary across two price conditions globalization, and view the world as a “global village” (Ng and Batra (MLow=4.24, MHigh=4.54; t(57)=1.25, p=.22). 2017). We expect that locals and globals differ in perceived quality Finally, Study 3 (N=277) adopted a 2 (identity: local vs. global) × difference among brands in the marketplace which, in turn, affects 2 (price: high vs. low) × 2 (product type: services vs. goods) between- PQA. Suppose the comparison objects are two alarm clocks. When subjects design to examine the moderating role of product type. Given global identity is high (vs. low), individuals focus on standardization that services are intangible and heterogeneous, perceived quality dif- and see things as homogeneous and similar to each other, leading to ference of services is by nature greater than that of physical goods lower perceived quality differences between the alarm clocks. Fur- (Lovelock and Gummesson 2004). The greater variation in the quality ther, if the brands are perceived as similar in quality, consumers less of services (vs. goods) should increase globals’ PQA—whose base- likely use any cues (e.g., price) to distinguish between them, thereby line tendency to differentiate brands (and hence, PQA) is low and has reducing PQA. Specifically, perceived similarity between - them greater potential for increase. In contrast, since locals’ tendency to dif- selves and others helps globals iron out the differences across people ferentiate brands (and hence, PQA) is already high, there is little room and drives them to focus not on specific attributes, but rather on “a to increase it further (“ceiling effect”). As a result, they should exhibit general way of making sense of the world” (Oyserman 2009, p. 256). little change in PQA when evaluating services (vs. physical goods).

However, when local identity is high (vs. low), individuals As expected, globals exhibited PQA when evaluating services (MLow

“zoom in” on the two alarm clocks and focus on finer details, which Price=4.42, MHigh Price=5.15; t(62)=3.01, p<.01) but not physical goods subjectively increases perceived dissimilarity between these two (MLow Price=4.74, MHigh Price=4.58; t(66)=-.79, p=.43), whereas locals had clocks. Indeed, perceived difference makes locals rely more on the PQA when evaluating both services and physical goods (p’s<.05). distinction between themselves and others and leads them to focus This research contributes to the literature in two significant on specific elements and direct experiences to make sense ofthe ways. First, it represents a first attempt to explicitly examine the ef- world (Ng and Batra 2017). When evaluating products, the focus on fect of local-global identity on PQA. Second, it is also a pioneer to dissimilarities leads locals to use price—one of the most direct and uncover perceived quality difference as a new qualitative difference obvious cue for brand comparison (Lalwani and Forcum 2013)—to between local and global identity. distinguish among brands, thereby increasing their PQA. We test our hypothesis via a field study and two lab experiments. REFERENCES: Study 1 was conducted in a large shopping mall with real consumers Gao, Huachao, Yinlong Zhang, and Vikas Mittal (2017), “How (N=164). Respondents were given a brochure either describing “Think Does Local–Global Identity Affect Price Sensitivity?,”Journal Local Movement” or “Think Global Movement,” which served as lo- of Marketing, 81(3), 62-79. cal and global identity manipulation, respectively (Gao, Zhang and Kardes, Frank, Maria Cronley, James Kellaris, and Steven Posavac Mittal 2017). Afterwards, participants rated quality, reliability, and de- (2004), “The Role of Selective Information Processing in pendability (1=Very Low and 7=Very High; Lalwani and Shavitt 2013) Price-Quality Inference,” Journal of Consumer Research, of three shoes that differed in prices. These three items (α’s>.89) were 31(2), 368-374. averaged to form quality evaluation. The correlation between retail Lalwani, Ashok and Sharon Shavitt (2013), “You Get What You price and subjective quality evaluation for the shoes was calculated Pay for? Self-Construal Influences Price-Quality Judgments,” and served as the dependent variable PQA. As predicted, participants Journal of Consumer Research, 40(2), 255-267. in the local (vs. global) identity condition had significantly higher PQA Lovelock, Christopher and Evert Gummesson (2004), “Whither Services Marketing? In Search of a New Paradigm and Fresh (MLocal=.68, MGlobal=.40; t(162)=2.98, p<.01). Study 2 (N=122) featured a 2 (identity: local vs. global) × 2 Perspectives,” Journal of Service Research, 7(1), 20-41. (price: high vs. low) between-subjects design. Local-global identity Ng, Sharon and Rajeev Batra (2017), “Regulatory Goals in a was manipulated through sentences unscrambling task (Zhang and Globalized World,” Journal of Consumer Psychology, 27(2), Khare 2009). Afterwards, participants read information about three 270-277. brands each of alarm clocks and microwaves: the target brand and Zhang, Yinlong and Adwait Khare (2009), “The Impact of two other comparison brands, which provided baseline price infor- Accessible Identities on the Evaluation of Global versus Local mation (Lalwani and Shavitt 2013). Participants were randomly as- Products,” Journal of Consumer Research, 36(3), 524-537.

Advances in Consumer Research 869 Volume 46, ©2018 The Impact of Price and Size Comparisons on Consumer Perception and Choice Jun Yao, Macquarie University, Australia Harmen Oppewal, Monash University, Australia Yongfu He, Monash University, Australia

EXTENDED ABSTRACT Study 2a and 2b aimed to test numerosity (Pandelaere, Briers, and In packaged-goods retailing, manufacturers and retailers seek to Lembregts 2011) as an alternative explanation for the PSC effect by set price levels and select package sizes such that their products are expressively varying the numerosity of package size and the numeros- optimally competitive. In doing so they can choose to differentiate ity of price. In study 2a, in the SC mode, the size difference was either on price or on product size. For example, suppose Kraft offers a 330 numerically larger (e.g., 1900 ml – 1800 ml = 100 ml) or smaller (e.g., ml bottle of salad dressing for $2.51, implying a unit cost of $0.76 1.9 liter – 1.8 liter = 0.1 liter) than the price difference in the PC mode per 100 ml. When a higher positioned competitor such as Praise aims (e.g., $8.54 – $8.09 = $0.45). In study 2b, prices are framed either in to offer its product at the higher unit cost of $0.84 per 100 ml, it dollars (e.g., $2.01 vs. $2.45) or in Indonesian Rupiahs (e.g., Rp20,100 can either offer a similar sized package at the higher price of $2.76 vs. Rp24,500) thus price differences are numerically either smaller or or resize its product to 300 ml and match Kraft’s price of $2.51. In larger than the size difference (e.g., 160g vs. 195g). Results showed both cases Praise’s unit cost stays at $0.84 as intended and the dif- that value differences were perceived as larger in the SC mode than in ference between the Kraft and Praise products can be deemed eco- the PC mode, this effect on perceived value difference did not interact nomically equivalent. We define the two situations as the two com- with the comparison mode thus numerosity can be excluded as an al- parison modes: price comparison (PC) mode and size comparison ternative account underlying the PSC effect. (SC) mode. Study 3 tested if computational ease (Thomas and Morwitz In the context of promotions, prior research has identified that, 2009) is a possible alternative explanation by manipulating the ease while a bonus pack and a price discount both represent a saving, con- of comparing price differences and size differences. Results showed sumers generally tend to prefer a bonus back over an economically that the ease of computation had neither a main effect nor an interac- equivalent price discount (Diamond and Sanyal 1990). The monetary tion with comparison mode thus computational ease is rejected as an cost being zero for a bonus pack makes that it is processed indepen- alternative account underlying the PSC effect. dently of its price and so is perceived as a pure gain. In contrast, a Study 4a reveled a downstream consequence of the PSC effect price discount is more likely to be integrated into the regular price on product choice. Results showed that the probability of choosing and is perceived as a reduction in loss (Chandran and Morwitz 2006). the low-end option was higher in the SC condition (i.e., the choice A pure gain looms larger than a reduction in loss on the value func- share of the larger package size option) than in the PC condition (i.e., tion, this results in a preference towards a bonus pack over a price choice share of the lower price option). The effect of comparison discount in most instances. mode on choice is mediated by perceived value difference. Study 4b Extending and generalizing the theorizing to our price-size com- demonstrated that the PSC effect on choice holds even in the pres- parison context, on the value function, a price difference elicits an in- ence of unit price information. cremental change (or a decremental change which is directionally op- The price or size comparison effect on consumer responses is of posite but magnitudinally the same as an incremental change) in loss direct and significant interest especially to marketers concerned with whereas a size difference elicits a pure change in loss or gain. Accord- inter-brand competition. Owing to consumers’ greater susceptibility ing to the principle of diminishing sensitivity in prospect theory, an to size differences, a more beneficial approach to the pricing ofa incremental change is further away from the reference point, resulting brand with a low-end positioning may be to carry the same price as in a lower sensitivity to the change than that of a pure change, which the competing brand but provide more product volume. On the other is adjacent to the reference point. In turn, due to their positions on the hand, for a brand with a high-end positioning a more appropriate value function, pure changes generally loom larger than economically strategy is to offer a package that is of equal size as the competing equivalent incremental changes. We therefore predict that consumers brand, avoiding any adverse feelings associated with size loss, and perceive a greater difference in value when two products differ in size instead differentiating the product by charging a higher price. (i.e., SC mode) than when they differ on price (i.e., PC mode). We call this the price size comparison (PSC) effect. REFERENCES A series of experimental studies examine how differences in Chandran, Sucharita and Vicki G. Morwitz (2006), “The Price price or size affect consumer value perceptions and choices when of “Free”-Dom: Consumer Sensitivity to Promotions with purchasing grocery products. Studies 1 aimed to provide initial evi- Negative Contextual Influences,”Journal of Consumer dence for the proposed PSC effect. The study comprised a 2 (mode: Research, 33 (3), 384-92. PC vs. SC) × 2 (magnitude of difference: small vs. large) between- Diamond, Willam D. and Abhijit Sanyal (1990), “The Effect of subjects design. Every participant saw a pair of unbranded biscuits Framing on the Choice of Supermarket Coupons,” in Advances either with a small (approximately 5%) or with a large (approxi- in Consumer Research, Vol. 17, ed. Marvin E. Goldberg, mately 18%) difference in the retail prices or sizes of the options. Gerald Gorn, and Richard W. Pollay, Provo, UT: ACR, 488-93 Participants evaluated the magnitude of the value difference between Pandelaere, Mario, Barbara Briers, and Christophe Lembregts the two product options on a seven-point bipolar scale with two items (2011), “How to Make a 29% Increase Look Bigger: The (r = .75) (i.e., “how do you rate the difference between these two Unit Effect in Option Comparisons,”Journal of Consumer products in terms of value: 1 = small difference between products, Research, 38 (2), 308-22. 7 = large difference between products; 1 = narrow value range, 7 = Thomas, Manoj and Vicki G. Morwitz (2009), “The Ease- wide value range”). Results showed that a comparison on size results of-Computation Effect: The Interplay of Metacognitive in a greater perceived value difference between product options than Experiences and Naive Theories in Judgments of Price a comparison on price. Differences,”Journal of Marketing Research, 46 (1), 81-91.

Advances in Consumer Research 870 Volume 46, ©2018 Feature a Benefactor or a Victim? How Charity Appeals with Different Protagonist Foci Affect Donation Behavior Bingqing (Miranda) Yin, University of Kansas, USA Jin Seok Pyone, University of Kansas, USA

EXTENDED ABSTRACT be more effective than a victim appeal and the effect is driven by Charities often emphasize victims’ plight in their appeals (e.g., moral elevation. an emaciated, hunger-stricken child featured in UNICEF’s ads) and In experiment 1B (N=81), we intend to replicate the effect by extensive donation research focuses on studying victim appeals. designing a new set of stimuli. Participants were randomly assigned However, victim appeals have their limitations. Factors such as do- to read either a benefactor appeal or a victim appeal. The charity pro- nors’ emotional exhaustion (Ein-Gar and Levontin 2013) or donors’ vides vision care to out-group recipients (African victims). Replicat- perceptions of justice (Lee, Winterich, and Ross 2014) tend to weak- ing the results from experiment 1A, participants who read the bene- en the effectiveness of victim appeals. Crucially, when charity recipi- factor appeal were more willing to donate than those who read the ents are out-group members, even charity appeals featuring an iden- victim appeal (p < .05, two-tailed). Mediation analysis based on10, tified victim lose their effectiveness (e.g., Kogut and Ritov 2007). 000 bootstraps (Hayes 2017, model 4) showed that the indirect effect In this paper, therefore, we consider featuring an alternative of moral elevation was significant (95% bootstrap CI: -1.36, -.40). protagonist in the charity appeal, especially when soliciting dona- These results provides further evidence that a benefactor appeal can tions for out-group members. Instead of a distressed victim, a charity promote charitable giving because of moral elevation. can feature a person who renders help to the needy to accomplish In experiment 2 (N=124), we examine the moderating effect of the charity’s mission―a benefactor. In practice, some charities have charity recipients’ group membership. This experiment employs a 2 already been featuring a benefactor in their appeals. For example, (charity appeal: benefactor vs. victim) X 2 (recipients’ group mem- United Way, in early 2017 launched a docu-series named “The Hero bership: in-group vs. out-group) and participants were randomly as- Effect,” in which each episode features an ordinary citizen having a signed into one of the four conditions. As predicted, results showed life-changing impact on an individual or a community. Yet, little is a significant interaction (p = .05, two-tailed). Pairwise comparisons known about whether the benefactor appeal is more effective than revealed that, when recipients were out-group members, participants the victim appeal in eliciting charitable donations, and if so, when who read the benefactor appeal were more willing to donate than and why. those who read the victim appeal (p = .03, two-tailed). On the other We suggest that while both a victim and a benefactor appeal can hand, when recipients were in-group members, both appeals elicited be useful in soliciting donations, these two appeals operate by evok- similar levels of donation intention (p = .59). ing different emotions. A victim appeal often induce negative emo- In experiment 3 (N=612), we included two more conditions tion, making people feel sorry, sad, or even upset about the victim’s where an out-group benefactor helps in-group/out-group recipients. distressful situation, in an attempt to trigger donors’ charitable ac- We used a 3 (charity appeal: in-group benefactor vs. victim vs. out- tions. In contrast, we posit that a benefactor appeal can evoke a spe- group benefactor) X 2 (recipients’ group membership: in-group vs. cific positive moral emotion—moral elevation, the warm, uplifting out-group) between-subjects design. As predicted, results showed feelings that people experience when witnessing others’ moral excel- a significant two-way interaction (p = .02, two-tailed). In pairwise lence (Haidt 2000), which trigger people’s desire to do a charitable comparisons, when the recipients were out-group members, dona- act for others. We further propose that the effectiveness of a benefac- tion likelihood was higher for the in-group benefactor appeal than tor appeal, relative to a victim appeal, depending on the charity re- the victim appeal and the out-group benefactor appeal (all ps< .03, cipients’ group membership. When recipients are in-group members, two-tailed), though both the in-group and out-group benefactor ap- people’s donation intentions are less likely to be influenced by the peals led to similar levels of moral elevation (p > .04). On the other charity appeal protagonist. However, when recipients are out-group hand, when the recipients were in-group members, no donation in- members, featuring a benefactor should be more effective as the tention difference was found among the three appeals (all ps > .18, benefactor appeal leads people to focus on their own desire to be a two-tailed). better person by helping others, not the victim’s distressful situation. In experiment 4 (N=166), we focus on the out-group recipients We further suggest that the group membership of the benefactor conditions and investigated the underlying mechanism of the bene- also influences the effectiveness of benefactor appeals. We specifi- factor’s group membership on donation. Participants were randomly cally theorize that this is because the benefactor’s group membership assigned to read one of the three charity appeals featuring a victim, influences the extent to which people act upon the emotion evoked an in-group benefactor, or an out-group benefactor. Results showed by the benefactor―moral elevation, which affects their donation be- that an out-group benefactor appeal, though elicited similar levels of havior. moral elevation as an in-group benefactor appeal (p > .10), led to do- We test our hypotheses in five experiments. In experiment 1A nation amount (p =.02). Mediation analysis revealed a significant in- (N=81), we test the main effect of protagonist focus on charitable direct effect of feeling versus reasoning based decision process (95% donation and the underlying mechanism. Participants were randomly bootstrap CI: .02. .18), such that people who read the out-group (vs. assigned to one of the two conditions—a charity appeal featuring in-group) benefactor appeal relied less on their feeling (vs. reason- either a benefactor or a victim. Results showed that participants ining), resulting in lower donation amount. dicated higher donation likelihood to the benefactor appeal than the Taken together, we develop a theoretical framework on when victim appeal (p < .05, two-tailed). and why a benefactor appeal elicits more donations. Five studies in- Using PROCESS (model 4; Hayes 2013) with 10,000 boot- volving both behavioral intention and real donation measures test strapped samples, the indirect effect of charity appeal focus on dona- and confirm the theoretical model. We contribute to research on char- tion amount was mediated by moral elevation (95% Bootstrap CI: itable donation, moral elevation and demonstrate to charities that a .45, 1.54). Overall, the results support that a benefactor appeal can benefactor appeal can be effective in promoting donations.

Advances in Consumer Research 871 Volume 46, ©2018 Charity Begins at Home: How Incremental Theory Enhances or Reduces Charitable Giving Alyssa Yoon, Korea University, South Korea Jongwon Park, Korea University, South Korea

EXTENDED ABSTRACT Finally, we investigated the moderating role of campaign goal People’s generosity declines with the self-other distance (Dovidio progress, both when perceived (study 3) and manipulated (study 4). et al. 1997; Jones and Rachlin 2006). While the negative impact of dis- In study 3, participants (n = 181) read a donation appeal with a goal tance on prosocial outcome is well understood, can the perception of amount but no progress indicator and then estimated the donation goal distance between self and others chronically differ? In this research, we progress. Theory priming did not affect the progress perception (Mentity = demonstrate that subscribing to incremental theory (vs. entity theory) 2.88 vs. Mincremental = 3.06, NS). As anticipated, implicit self-theories and leads to higher charitable intention, and such effect arises because the the perceived progress had a significant interaction (b = -.39, t = -2.93, p beneficiaries viewed as changeable, as in the lenses of incremental < .01). Floodlight analysis revealed two significant regions, suggesting theorists, feel closer to the self. that when perceived progress was high (vs. low), those who adopted the Individuals differ in the extent to which they view human quali- entity (vs. incremental) theory were more willing to donate. ties as malleable (Dweck, Chiu, and Hong 1995). While incremental In study 4 (n = 118), participants were assigned to either low (20%) theorists believe that people can change and improve who they are, or high (80%) campaign progress. Implicit theories and the progress ma- entity theorists view others to be consistent across time and situation. nipulation significantly interacted (b = -.43, t = -3.96, p < .001). Spot- Consequently, incremental theorists’ impressions of others consti- light analysis showed that the effect of progress on donation was positive tute the contextualized factors (e.g., situation and needs), whereas among entity theorists (b = .50, t = 2.04, p = .04), but negative among entity theorists seek to assess the unchanging, global traits (e.g., per- incremental theorists (b = -.87, t = -3.59, p < .001). Further, a mediated sonality and ability) of others (Molden, Plaks, and Dweck 2006). moderation analysis (Model 8; Hayes 2013) indicated that their primary Such low- (vs. high-) level focus is known to embody the construal drivers differ, supporting our theorizing that incremental theorists feel of proximate (vs. distant) others (Nussbaum, Trope, and Liberman closer to others and thus more committed to help (Fishbach et al. 2011). 2003; Rim, Uleman, and Trope 2009). In the similar vein, entity Specifically, we measured the perceived need for progress (e.g., “my theorists were shown to endorse the stereotypes more readily (Hong et participation is important”), and the perceived value of goal (e.g., “the al. 2004), and incremental theorists accord more individuating elements campaign goal will be reached”). We confirmed that the entity theorists’ to out-group members (Levy, Chiu, and Hong 2006). All evidence taken higher donation intention in the high-progress condition was mediated together suggests that compared to incremental theorists, entity theorists by the perceived value of goal (b = .33, SE = .12, CI95%: .12 to .61), but have a clearer demarcation between self and other, which is a traditional not by the need for progress (b = .26, SE = .16, CI95%: -.02 to .58). By hallmark of social distance (Liviatan, Trope, and Liberman 2008). contrast, the incremental theorists’ higher charitable intention in the low- We test the proposed mechanism in two ways, one via direct mea- progress condition was mediated by the need for progress (b = -.36, SE sures of social distance (study 2) and the other via moderation by cam- = .15, CI95%: -.67 to -.10), not by the goal value (b = .08, SE = .08, CI95%: paign goal progress (studies 3 and 4). Because whether one will pursue a -.03 to .28). collective goal (i.e., donation campaign) is an interactive effect of one’s Through four experiments, we found that incremental theorists are sense of closeness with its beneficiaries and, importantly, the perceived more likely to feel close to the victim and make donations as a result. goal progress (Fishbach, Henderson, and Koo 2011; Williams and Karau However, when donation campaign is highly progressed, the entity the- 1991), we expect a moderating role of donation goal progress in the link ory increases charitable giving. Further research replicating our findings between the donors’ implicit self-theories and their intention to contrib- in real donation campaign settings is underway. Meanwhile, our findings ute. Specifically, individuals who view the victims as remote are more urge that marketers and policy makers take into account consumers’ im- likely to question if the campaign is worth joining and thus become more plicit self-theories when seeking to encourage charitable behavior. motivated if the accumulated progress is high rather than low. However, ones who feel closer to victims and thus already consider the donation REFRENCES campaign as worthwhile will be more driven when the progress is un- Dovidio, John F, Kerry Kawakami, Craig Johnson, Brenda Johnson, satisfactory (Fishbach, Koo, and Finkelstein 2014; Williams and Karau and Adaiah Howard (1997), “On the Nature of Prejudice: 1991). Drawing on the motivation literature, we propose and observe a Automatic and Controlled Processes,” Journal of experimental reversed effect such that a high-progress campaign leads the entity theo- social psychology, 33 (5), 510-40. rists to become more charitable than their incremental counterparts. Dweck, Carol S, Chi-yue Chiu, and Ying-yi Hong (1995), “Implicit Study 1 (n =120) provided initial confirmation of the main effect of Theories and Their Role in Judgments and Reactions: A Word incremental theory on donation. Participants’ chronic incremental theory from Two Perspectives,” Psychological inquiry, 6 (4), 267-85. significantly increased their intention to donate to help Sierra Leone (b = Fishbach, Ayelet, Marlone D Henderson, and Minjung Koo (2011), .26, t = 2.36, p = .02). In study 2, participants (n = 178) were asked the “Pursuing Goals with Others: Group Identification and amount they were willing to give out of $100 to a Nepal earthquake re- Motivation Resulting from Things Done Versus Things Left lief. Perceived social distance was measured (“How distant/close do you Undone,” Journal of Experimental Psychology: General, 140 feel to people of Nepal?” and “Despite the fact that Nepal earthquake did (3), 520. not affect me, I can identify with the people in Nepal”; reverse-coded). Fishbach, Ayelet, Minjung Koo, and Stacey R Finkelstein (2014), Positive effect of incremental theory on donation was replicated (b = .08, “Motivation Resulting from Completed and Missing Actions,” t = 2.47, p = .01). Social distance predicted willingness to donate (b = in Advances in Experimental Social Psychology, Vol. 50: Elsevier, 257-307. .72, t = 7.23, CI95%: .53 to .92, p < .001), reducing the incremental theory effect on donation to nonsignificance (b = .05, t = .58, CI95%: -.12 to .22).

Hong, Ying-yi, Jill Coleman, Gloria Chan, Rosanna YM Wong, Molden, Daniel C, Jason E Plaks, and Carol S Dweck (2006), Chi-yue Chiu, Ian G Hansen, Sau-lai Lee, Yuk-yue Tong, ““Meaningful” Social Inferences: Effects of Implicit Theories and Ho-ying Fu (2004), “Predicting Intergroup Bias: The on Inferential Processes,” Journal of experimental social Interactive Effects of Implicit Theory and Social Identity,” psychology, 42 (6), 738-52. Personality and Social Psychology Bulletin, 30 (8), 1035-47. Nussbaum, Shiri, Yaacov Trope, and Nira Liberman (2003), Jones, Bryan and Howard Rachlin (2006), “Social Discounting,” “Creeping Dispositionism: The Temporal Dynamics of Psychological science, 17 (4), 283-86. Behavior Prediction,” Journal of personality and social Levy, Sheri R, Chi-yue Chiu, and Ying-yi Hong (2006), “Lay psychology, 84 (3), 485. Theories and Intergroup Relations,” Group Processes & Rim, SoYon, James S Uleman, and Yaacov Trope (2009), Intergroup Relations, 9 (1), 5-24. “Spontaneous Trait Inference and Construal Level Theory: Liviatan, Ido, Yaacov Trope, and Nira Liberman (2008), Psychological Distance Increases Nonconscious Trait “Interpersonal Similarity as a Social Distance Dimension: Thinking,” Journal of experimental social psychology, 45 (5), Implications for Perception of Others’ Actions,” Journal of 1088-97. experimental social psychology, 44 (5), 1256-69. Williams, Kipling D and Steven J Karau (1991), “Social Loafing and Social Compensation: The Effects of Expectations of Co-Worker Performance,” Journal of personality and social psychology, 61 (4), 570. Toward Optimal Symbolic Recovery: Comparing “Thank you” and “Sorry” in Addressing Service Delays Yanfen You, New Mexico State University, USA Xiaojing Yang, University of Wisconsin-Milwaukee, USA Lili Wang, Zhejiang University, China Xiaoyan Deng, Ohio State University, USA

EXTENDED ABSTRACT so after their self-regard was made salient. Some received informa- Service delay, or unexpected delay after a service has been tion that challenged participants’ self-serving comparative beliefs. scheduled (Taylor 1994), is the most common type of service failure As expected, estimates of personal risk factor standing among high (Goodwin and Ross 1992). Unlike product failure that usually re- self-regard salience/comparison information participants were less quires utilitarian recovery efforts to alleviate the “financial pain” felt correlated with similar estimates collected 2 months earlier, suggest- by consumers (Dunn and Dahl 2012researchers have found that con- ing the estimates had become less inductive. There was also tenta- sumers feel better about the product failure after complaining about tive evidence that, in this condition, self-esteem played a greater role it. In contrast, the authors show that when consumers are to blame in constructing these estimates. In Study 2, participants were led to for product failure, complaining has a \”detrimental effect on con- believe that their comparative standing on risk factors was better sumer reactions to the product. In this context, self-threat from the than it actually was, and in this case, they were more likely to use product failure is shown to motivate defensive processing in both the this information when making self-judgments about their personal content of complaints and the subsequent downstream product evalu- risk. Evidently, the extent to which people use diagnostic compari- ations. The authors establish the role of self-threat in product failure son information inductively when making self-inferences depends in two ways: (1), service delay is a typical process failure in which on the favorability of that information. (PsycINFO Database Record marketers could strategically employ symbolic recovery efforts to (c and therefore respond positively to those who boost their posi- ease consumers’ “psychological pain” (e.g., feeling of not being re- tive self-views (Colman and Olver 1978). As such, appreciation is spected). According to the sociometer theory of self-esteem, people’s more effective in promoting post-recovery satisfaction than apology. feelings about themselves are affected by the extent to which they be- Across six studies, we provide support for our theory and its process lieve other people have accepted or rejected them (Leary et al. 1995, explanations. 1998). As consumers perceive an unscheduled delay as a signal of Study 1 examined our hypothesis in a real-world delay situa- the marketer’s lack of attention to their interest and lack of effort tion. After completing lab experiments, participants were promised in protecting their welfare (Seiders and Berry 1998)financial institu- that they would receive their compensation on the same day. How- tions, and financial elites gain greater influence over economic policy ever, they were not contacted by the lab assistant until the following and economic outcomes. Financialization transforms the functioning day. Using a mobile payment app, the assistant greeted them with of economic systems at both the macro and micro levels. Its principal either “Thank you for waiting for such a long time!” or “Sorry for impacts are to (1, their self-esteem is threatened. Therefore, “what keeping you waiting for such a long time!” and distributed the com- to say” to customers after service delays is especially important for pensation. A follow-up survey link was sent to participants to collect marketers to repair consumers’ thwarted self-esteem and restore their feedback. Results revealed a higher response rate (91% vs. 78%) and satisfaction. more satisfaction (Mappreciation = 5.52 vs. Mapology = 5.11) in the appre- Despite research on speech acts suggesting that apology (e.g., ciation condition. “Sorry about the delay”) and appreciation (e.g., “Thank you for your Study 2 measured self-esteem in a one-way (symbolic recov- patience”) are equally logical marketer responses to service delays ery: appreciation vs. apology vs. control) between-subjects design. (Coulmas 1980), prior work has only discussed apology as a sym- Participants first read about a product delivery delay and were told bolic recovery strategy (e.g., Witrz and Mattila 2004). Apology has that they either received an apology, an appreciation or nothing from been shown to reduce victims’ unfairness perception, anger and ag- the online store. Recovery satisfaction, overall satisfaction, positive gression (e.g., Darby and Schlenker 1982) which are often manifes- WOM intention, and self-esteem were measured. Results revealed tations of the feelings of being disrespected (Dunn 2016; Seiders and significant effects on recovery satisfaction (Mcontrol = 3.03 vs. Ma-

Berry 1998). As such, apology is effective in repairing consumers’ pology = 4.40 vs. Mappreciation = 5.07), overall satisfaction, WOM, and thwarted self-esteem (Tyler, Degoey, and Smith 1996). self-esteem (Mcontrol = 3.10 vs. Mapology = 4.14 vs. Mappreciation = 4.80). Drawing from research on service recovery and the self, we ex- Furthermore, self-esteem mediated the effect of recovery strategy pect that appreciation goes beyond this reparative effect of apology on on recovery satisfaction (relative indirect effects: 1b = -.74, 95% CI self-esteem which is through acknowledging marketers’ mistake and -1.16 -.31; b2 = .46, 95% CI: .03 .94), overall satisfaction and recom- accountability. Appreciation shifts the focus from marketers to con- mendation intention. sumers by placing them in the benefactor position and highlighting Study 3 utilized in a 2 (self-esteem: low vs. high) × 2 (recov- their merits and contributions in the service delivery, thereby signal- ery strategy: apology vs. appreciation) between-subjects design. ing approval and acceptance. Based on the sociometer theory (Leary Self-esteem was manipulated by informing participants that their and Baumeister 2000), positive approvals and acceptance from oth- performance in an English word generation task was ranked in the ers boost self-esteem. The elevation in self-esteem in turn increases bottom (top) 10%. Then they read about a restaurant service delay customer satisfaction because based on the self-enhancement theory, situation in which the server either apologized for the delay by say- people are motivated to pursue positive self-views (Heine et al. 1999; ing “I’m sorry for keeping you waiting! I apologize,” or appreciated Schaumberg and Wiltermuth 2014)and their self-regard is salient, their patience by saying “Thank you for your patience! I appreciated they will be less likely to use the comparison information when judg- it.” Participants reported their satisfaction with the server, tipping ing their standing in the domain. Participants in Study 1 estimated likelihood and tip amount. Results showed significant interactions their standing on risk factors for two health problems, and some did of self-esteem and recovery strategy on all dependent variables (p’s

< .06). Specifically, in the low self-esteem condition, appreciation Goodwin, Cathy and Ivan Ross (1992), “Consumer Responses to led to higher satisfaction, tipping likelihood, and tipping amount (p’s Service Failures: Influence of Procedural and Interactional < .05). However, the superiority of appreciation disappeared in the Fairness Perceptions,” Journal of Business Research, 25(2), high self-esteem condition (p’s >.09). 149–63. Study 4 provided triangulating evidence for the underlying Heine, Steven J., Darrin R. Lehman, Hazel Rose Markus, and mechanism by utilizing narcissism as a moderator because it de- Shinobu Kitayama (1999), “Is There a Universal Need for termines the strength of consumers’ desire to be well regarded by Positive Self-Regard?,” Psychological Review1, 766–94. others. It featured a 2 (recovery strategy: apology vs. appreciation) Klein, William M P, Heather K. Blier, and Aura M. Janze (2001), × narcissism (continuous) design using the restaurant scenario as “Maintaining Positive Self-Evaluations: Reducing Attention the context. Narcissism was measured based on Ames et al. (2006). to Diagnostic but Unfavorable Social Comparison Information Floodlight analysis revealed that the superiority of appreciation (vs. When General Self-Regard Is Salient,” Motivation and apology) existed only among participants whose narcissism index Emotion, 25(1), 23–40. was higher than 3.47 (BJN = .59, p = .05).Study 5 examined the role Leary, Mark R. and Roy F. Baumeister (2000), “The Nature and of service delay severity and demonstrated that the superiority of Function of Self-Esteem: Sociometer Theory,” in Advances in appreciation disappeared when the delay was perceived to be minor. Experimental Social Psychology, 1–62. Study 6 showed that when consumers’ attention was drawn to ser- Leary, Mark R., Mark R. Leary, Ellen S. Tambor, Ellen S. Tambor, vice providers’ fault, apology was more effective than appreciation. Sonja K. Terdal, Sonja K. Terdal, Deborah L. Downs, and Our research connects research on service recovery, linguistic Deborah L. Downs (1995), “Self-Esteem as an Interpersonal framing, and self-concept to show how service providers can strate- Monitor: The Sociometer Hypothesis.,” Journal of Personality gically frame their response to increase customer recovery satisfac- and Social Psychology, 68(3), 518–30. tion. Leary, Mark R, Alison L Haupt, Kristine S Strausser, and Jason T Choket (1998), “Calibrating the Sociometer: The Relationship REFERENCES between Interpersonal Appraisals and State Self-Esteem,” Ames, Daniel R., Paul Rose, and Cameron P. Anderson (2006), Journal of Personality and Social Psychology, 74(5), 1290–99. “The NPI-16 as a Short Measure of Narcissism,” Journal of Schaumberg, Rebecca L. and Scott S. Wiltermuth (2014), “Desire Research in Personality, 40(4), 440–50. for a Positive Moral Self-Regard Exacerbates Escalation Colman, A. M. and K. R. Olver (1978), “Reactions to Flattery as of Commitment to Initiatives with Prosocial Aims,” a Function of Self-Esteem: Self-Enhancement and Cognitive Organizational Behavior and Human Decision Processes, Consistency Theories,” British Journal of Social and Clinical 123(2), 110–23. Psychology, 17(1), 25–29. Seiders, Katleen and Leonard Berry (1998), “Service Fairness : Coulmas, Florian (1980), “Poison to Your Soul: Thanks and What It Is and Why It Matters,” The academy of Management Apologies Contrastively Viewed.,” in Conversational Routine, Executive, 12(525), 8–20. Trier: University at Trier, Linguistic Agency., 69–91. Taylor, Shirley (1994), “Waiting for Service: The Relationship Darby, Bruce W. and Barry R. Schlenker (1982), “Children’s between Delays and Evaluations of Service,” Journal of Reactions to Apologies.,” Journal of Personality and Social Marketing, 58(2), 56–69. Psychology, 43(4), 742–53. Tyler, Tom, Peter Degoey, and Heather Smith (1996), Dunn, Lea and Darren W Dahl (2012), “Self-Threat and Product “Understanding Why the Justice of Group Procedures Matters: Failure: How Internal Attributions of Blame Affect Consumer A Test of the Psychological Dynamics of the Group-Value Complaining Behavior,” Journal of Marketing Research, Model.,” Journal of Personality and Social Psychology, 70(5), 49(5), 670–81. 913–30. Dunn, Louise S. (2016), “Bad Customer Service. How Does That Witrz, Jochen and Anna S. Mattila (2004), “Consumer Responses Make You Feel? Understanding The Psychology Behind to Compensation, Speed of Recovery and Apology after a Customer Service,” http://veterinarybusinessadvisors. Service Failure,” International Journal of Service Industry com/bad-customer-service-how-does-that-make-you-feel- Management, 15(2), 150–66. understanding-the-psychology-behind-customer-service/. Promoting Pi Day: Consumer Inferences About Special Day-Themed Promotions Daniel M. Zane, University of Miami, USA Kelly L. Haws, Vanderbilt University, USA Rebecca Walker Reczek, Ohio State University, USA

EXTENDED ABSTRACT Star Wars Lego item FREE.” In the special day-themed promotion Using marketing promotions to attract customers is so prevalent condition, the promotion was linked to “Star Wars Day” and the ac- that it takes something special for a given promotion to stand out (Kal- companying slogan, “May the Fourth Be With You.” In the standard wani and Yim 1992). In the Internet age, firms can ensure that a pro- promotion condition, the advertisement included no mention of this motion is distributed on a specific day and thus offer more promotions holiday or slogan. Participants then evaluated the promotion using around “special days” (e.g., a 31.4% discount for Pi Day on March the study 1 measure. A significant interaction emerged between the 14th) in addition to typical holiday promotions (e.g., standard Christ- promotion manipulation and participants’ familiarity with the holi- mas discounts). Are such special day-themed appeals effective? This day (β = 0.12, t(148) = 2.44, p < .05). Among participants who were research explores the effectiveness of special day-themed promotions familiar with the holiday (above 6.23; ~38% of participants), those and outlines important boundary conditions for their effectiveness. who viewed the special day promotion had significantly more posi- Research on marketplace metacognition demonstrates that con- tive evaluations than those who viewed the standard promotion. This sumers often think about and draw inferences from marketers’ ac- pattern was attenuated as familiarity declined and began to reverse tions and intentions (Wright 2002). For example, consumers reward at low levels of familiarity. Thus, special day-themed promotions are marketers with an increased willingness-to-pay for products when only effective when consumers can understand the marketer’s cre- they infer that marketers put extra effort into designing product dis- ativity and might actually backfire otherwise. plays (Morales 2005). Based on the metacognition literature more Study 4 establishes direct process evidence that inferences about generally, which shows that people are particularly likely to draw the marketer’s creativity drive consumers’ liking of special day- inferences about unexpected/atypical experiences (Schwarz 2004), themed promotions. We also demonstrate that positive evaluations of we argue that consumers are likely to draw inferences about atypical these promotions drive increased intentions to use them. All partici- special day-themed promotions compared to standard promotional pants viewed a 30% discount for a shoe store valid “today only.” Be- events. Specifically, we expect consumers to infer that that the mar- cause the study was run on June 1st (i.e., National Go Barefoot Day), keter is creative when they encounter a special day-themed promo- participants in the special day-themed promotion condition viewed a tion (subject to boundary conditions). Consumers then reward mar- National Go Barefoot Day sale with the tagline “You might not need keters for their creative effort with more positive evaluations of the shoes today, but we’ve got you covered for tomorrow!” Participants promotion and increased purchase intentions. in the traditional event-themed promotion condition viewed a Start In study 1, we conducted a field study with a small company of Summer sale with the same tagline. In the no event promotion that sells dog gear. Customers (n = 239) received one of two versions condition, participants saw a standard one-day sale not linked to any of a promotional email. In the special day-themed promotion condi- event at all with the same tagline. Participants reported their evaluation, they saw a 25% discount on dog gear in honor of the anniver- tions of the promotion, perceptions of the marketer’s creativity, and sary of the day the company rescued its mascot (a dog). The email intentions to use the promotion. in the standard promotion condition contained a 25% discount for a Participants in the special day-themed condition had signifi- standard one-day sale, with no mention of the special day. Custom- cantly more positive evaluations of the promotion (M = 5.04) than ers in the special day-themed promotion condition were significantly those in the traditional event-themed (M = 4.43) and no event promo- more likely to click on the “shop now” button compared to those in tion conditions (M = 4.50, t(297) = 2.82, p = .005). The same pat- the standard promotion condition (19.30% vs. 10.00%; χ2 = 4.16, p < tern resulted for participants’ perceptions of the marketer’s creativity. .05). Sales data showed a similar pattern. Finally, serial mediation models revealed that increased perceptions In study 2, participants in all three conditions imagined receiv- of the marketer’s creativity in the special day-themed promotion con- ing an email promotion for a clothing retailer on March 14th (Pi Day). dition led to more positive evaluations of the promotion, which then In the special day-themed promotion condition, participants saw a led to higher intentions of using the promotion to make a purchase, 31.4% discount for an “Ultimate Pi Day” one-day sale. In the stan- compared to both the traditional event-themed promotion condition dard smaller [larger] promotion condition, they saw a 30% [35%] and no event promotion condition. discount for a standard one-day sale. Participants then assessed their Together, these studies provide evidence that special day-themed liking of the promotion. Participants in the special day-themed pro- promotions boost participants’ evaluations and lead to increased us- motion condition had significantly more positive evaluations of the age intentions (study 1) even compared to standard promotions that promotion (M = 3.49) compared to participants both in the standard are more favorable financially (study 2), as long as participants un- smaller promotion condition (M = 2.29; t(168) = 5.30, p < .0001) derstand the marketer’s creativity (study 3). Ultimately, consumers and standard larger promotion condition (M = 2.14, t(186) = 5.88, p reward marketers for their creative effort in offering special day pro- < .0001). Thus, consumers appear to even like a special day-themed motions (study 4). promotion more than an objectively better promotion not attached to the special day. REFERENCES Study 3 extends these findings using a different special day and Kalwani, Manohar U., and Chi Kin Yim (1992), “Consumer Price different promotion held constant across conditions. Additionally, we and Promotion Expectations: An Experimental Study,” Journal explore an important boundary condition by measuring participants’ of Marketing Research, 29 (Feb), 90-100. familiarity with the holiday. All participants viewed a promotion for Morales, Andrea. C. (2005), “Giving Firms an “E” for Effort: Lego valid on May 4th only (the study was conducted in real-time Consumer Responses to High-effort Firms,” Journal of on the actual holiday): “Buy any 3 Star Wars Lego items, Get the 4th Consumer Research, 31 (March), 806-12.

Schwarz, Norbert (2004), “Metacognitive Experiences in Consumer Wright, Peter (2002), “Marketplace Metacognition and Social Judgment and Decision Making,” Journal of Consumer Intelligence,” Journal of Consumer Research, 28 (March), Psychology, 14 (4), 332-48. 677-82. Predicting Memory-Based Consumer Choices from Recall and Preferences Zhihao Zhang, University of California, Berkeley, USA Aniruddha Nrusimha, University of California, Berkeley, USA Ming Hsu, University of California, Berkeley, USA

EXTENDED ABSTRACT tions from different computational models shows that the two-stage Consumers often need to make choices when some or all of the model robustly outperformed models using preference or memory in- information relevant to the decision is not physically present. To deal formation alone in both categories (Figures 1D) (Fast food: memory with this, they need to retrieve the necessary information from their + preference R2 = 0.937; preference-only R2 = 0.455; memory-only own memory (Lynch, 1982), e.g. a list of brands within a category, R2 = 0.803) (Running shoes: memory + preference R2 = 0.993; pref- prior experiences with a brand, and knowledge about key attributes of erence-only R2 = 0.859; memory-only R2 = 0.593). Non-overlapping certain products. The processes mediating consumer recall, and their 95% confidence intervals indicated that the differences in prediction downstream effects on consumer choice, have been subject to intense performance between the memory + preference model and the other study by consumer researchers. An extensive literature in consumer two models were statistically significant (Figure 1D). research strongly indicates a positive role of memory in consider- These findings echo and extend the extensive prior literature on ation and choices. Brands that have higher awareness or accessibility the influence of memory on brand choice processes (Coates, Butler, are more likely to be chosen (Hoyer & Brown, 1990; Posavac, San- & Berry, 2006; Nedungadi, 1990; Posavac et al., 1997). Important- bonmatsu, & Fazio, 1997). Furthermore, increasing brand accessibil- ly, our study proposes and validates a simple mechanism by which ity or strengthening category-brand association boosts the likelihood higher accessibility of brand memory translates into advantages in of consideration and choice for a brand of interest (Nedungadi, 1990; memory-based consumer choices, which is independent with prefer- Posavac, Sanbonmatsu, Cronley, & Kardes, 2001). ence. In this mechanism, the key component mediating the effect of To date, however, it remains challenging to understand and memory is the intermediate step of internal menu generation in the predict how effects of marketing actions on memory are translated proposed two-stage model. The higher the accessibility of a brand, into actual behavior. In particular, what is the nature and force of the more likely it will be included in the internal menu. processes by which stronger brand memory leads to more choices This component in the two-stage model is closely related to of the brand? Could brand memory be a proxy for preference or a the substantial literature on consideration sets (Manzini & Mariotti, component of preference? Do brand memory and preference operate 2014; Shocker, Ben-Akiva, Boccara, & Nedungadi, 1991). In par- independently? Without such mechanistic insights, it is difficult for ticular, the second stage of our two-stage model that expresses the marketers to predict or validate consumer memory measures with problem as a preference-based choice from a given choice set (menu) actual purchase data, or to estimate the return on investment (ROI) resembles previous models in this literature (Manski, 1977; Roberts of an advertising campaign that is projected to increase brand aware- & Lattin, 1991; Shocker et al., 1991). The key conceptual differenc- ness (Rust, Lemon, & Zeithaml, 2004). es, however, lie in how the intermediate menus are generated. Most Accordingly, we sought to take a step toward addressing these extant studies on consideration sets did not focus on memory factors; questions by building on influential multi-stage decision-making instead, they modeled the formation of consideration sets in choice models (Manski, 1977; Manzini & Mariotti, 2014; Roberts & Lat- problems as a rational, voluntary choice based on a tradeoff between tin, 1991) that explicitly capture process-level insights into memory- the potential benefit and the cost of considering one more option based choices. Specifically, our model (hereafter called two-stage (Brown & Wildt, 1992; Hauser, 2014; Roberts & Lattin, 1991). In model) incorporates both recall and evaluation as sequential stages in contrast, our models describe menu generation as probabilistic pro- the decision-making process as follows. Prompted with the category cesses governed by the strength of category-brand associations, and cue, the agent is assumed to first construct a menu by retrieving a set hence the composition of the choice set is less of an active choice of relevant brands from memory, and the inclusion of a certain brand by the agent. It is worth pointing out that highlighting the role of is determined probabilistically by the strength of the category-brand memory does not necessarily contradict the possible existence of association. Next the agent makes a choice based on relative prefer- consideration sets generated from voluntary inclusion/exclusion of ences for the brands on the menu. options; rather, it is much more appropriate to view the two accounts To test this model, we collected independent datasets for brands as complementary. It will be fruitful in future work to combine both from two product categories (fast food restaurants and running shoes) aspects in a single model of memory-based choices. in order to define three measures: (1) brand accessibility, the percent- By linking brand memory and brand choice in a predictive age of participants recalling the brand of interest when prompted model, our approach presents both conceptual and methodological with a category cue; (2) brand preference, the percentage of partic- advances capable of estimating the impact of changing brand aware- ipants choosing the brand of interest in stimulus-based choices or ness and accessibility on market performance. Furthermore, it also SB-C (with a menu listing a large number of brands in a category); offers a flexible toolkit for brand managers to evaluate marketing and (3) memory-based brand choices or MB-C, the percentage of campaigns in a wide variety of competitive landscapes. participants choosing the brand of interest in memory-based choices (category cue only, no menu). We then used these measures as the REFERENCES input to our proposed model to compute predicted memory-based Brown, J. J., & Wildt, A. R. (1992). Consideration set measurement. choice probabilities for all brands. Journal of the Academy of Marketing Science, 20(3), 235-243. Predictions of MB-C probabilities for brands were generated Coates, S. L., Butler, L. T., & Berry, D. C. (2006). Implicit memory from the proposed two-stage model, as well as models using pref- and consumer choice: The mediating role of brand familiarity. erence or memory information alone. These predictions were then Applied Cognitive Psychology, 20(8), 1101-1116. doi:10.1002/ compared against the empirically measured MB-C probabilities in acp.1262 our sample (Figures 1A-C). A quantitative examination of predic-

Hauser, J. R. (2014). Consideration-set heuristics. Journal Posavac, S. S., Sanbonmatsu, D. M., Cronley, M. L., & Kardes, of Business Research, 67(8), 1688-1699. doi:10.1016/j. F. R. (2001). The effects of strengthening category-brand jbusres.2014.02.015 associations on consideration set composition and purchase Hoyer, W. D., & Brown, S. P. (1990). Effects of Brand Awareness intent in memory-based choice. Advances in Consumer on Choice for a Common, Repeat-Purchase Product. Research, Vol Xxviii, 28, 186-189. Journal of Consumer Research, 17(2), 141-148. doi:Doi Posavac, S. S., Sanbonmatsu, D. M., & Fazio, R. H. (1997). 10.1086/208544 Considering the best choice: Effects of the salience and Lynch, J. G. (1982). Memory and Attentional Factors in Consumer accessibility of alternatives on attitude-decision consistency. Choice - Concepts and Research Methods. Journal of Journal of Personality and Social Psychology, 72(2), 253-261. Consumer Research, 9(1), 18-37. doi:Doi 10.1086/208893 Roberts, J. H., & Lattin, J. M. (1991). Development and Manski, C. F. (1977). The structure of random utility models. Testing of a Model of Consideration Set Composition. Theory and decision, 8(3), 229-254. Journal of Marketing Research, 28(4), 429-440. doi:Doi Manzini, P., & Mariotti, M. (2014). Stochastic Choice and 10.2307/3172783 Consideration Sets. Econometrica, 82(3), 1153-1176. Rust, R. T., Lemon, K. N., & Zeithaml, V. A. (2004). Return on doi:10.3982/Ecta10575 marketing: Using customer equity to focus marketing strategy. Nedungadi, P. (1990). Recall and Consumer Consideration Sets Journal of Marketing, 68(1), 109-127. doi:DOI 10.1509/ - Influencing Choice without Altering Brand Evaluations. jmkg.68.1.109.24030 Journal of Consumer Research, 17(3), 263-276. doi:Doi Shocker, A. D., Ben-Akiva, M., Boccara, B., & Nedungadi, P. 10.1086/208556 (1991). Consideration set influences on consumer decision- making and choice: Issues, models, and suggestions. Marketing letters, 2(3), 181-197. Sizes are Gendered: Impact of Size Cues in Brand Names on Brand Stereotyping Kuangjie Zhang, Nanyang Technological University, Singapore Shaobo Li, Nanyang Technological University, Singapore Sharon Ng, Nanyang Technological University, Singapore

EXTENDED ABSTRACT Study 3 replicated previous findings by utilizing yet another What comes to your mind when you think of the brand name consumption context (Little Cafe vs. Grand Cafe). Importantly, in Mini Cooper or Grand Cherokee? Physical size would probably be this study we also measured the extent to which participants asso- among the first few associations that come to mind - Mini Cooper ciate the brand names with maleness (vs. femaleness). We showed is physically smaller, whereas Grand Cherokee is physically bigger. that participants associated the brand name containing small (vs. big) However, size cues in brand names also evoke other associations? size cue more strongly with femaleness (vs. maleness), and that such This research proposes that such size cues would trigger in- gender association mediated the effect of size cues in brand names on grained gender associations. Specifically, we argue that size cues of both warmth and competence perceptions. “smallness” (e.g., little, small) versus “bigness” (e.g., grand, big) in Study 4 ruled out an alternative explanation related to physi- brand names can activate female versus male gender associations, cal size. This study followed a 2 (semantic size: small vs. big) × 2 which subsequently affect the warmth and competence perceptions (physical size: small vs. big) between-subjects design. To manipulate of the brand. semantic size, we followed the same method used in previous stud- This possibility arises from anecdotal evidence and linguistics ies by using different brand names (Small Kitchen vs. Big Kitchen). research. With some exceptions, men are generally physically big- To manipulate physical size, we manipulated the visual size of the ger in size than women (Touraille and Gouyon 2008). Linguistically, target store on a map as either smaller or bigger than other stores. A people tend to generate descriptions such as big (vs. little) when they two-way ANOVA revealed only a significant main effect of semantic evaluate an object that has a masculine (vs. feminine) grammati- size cues on warmth and competence perceptions. In other words, cal gender in their native language (Aikhenvald 2016; Boroditsky, we found that physical size of the store did not affect warmth and Schmidt, and Phillips 2003). competence perceptions. These findings lead to the proposition that consumers have a Studies 5 and 6 explored the downstream consequence of size general mental association between size (small vs. big) and gender cues in brand names on brand evaluation depending the relative im- (female vs. male). As such, brand names involving size cues related portance of warmth and competence in the consumption context. to smallness (vs. bigness) would evoke female (vs. male) gender Study 5 examined the role of consumption goals (communal vs. associations, and this has important implications on how consum- agentic). Participants imagined they were looking for a restaurant ers perceive a brand. Specifically, it has been well-established that at which to meet either an old friend (communal goal) or a business women are stereotyped as warm whereas men are stereotyped as associate (agentic goal). A significant interaction between size cue competent (Fiske et al. 2002; White and Gardner 2009). Building on and consumption goal on purchase intention emerged. Participants this stream of research, we hypothesize that size cues in brand names indicated greater purchase intentions for the brand name with a size can evoke gender associations, which subsequently affect consum- cue of smallness (vs. bigness) when they had a communal goal (i.e., ers’ perceptions of warmth and competence of the target brand. meeting with a friend). Conversely, they indicated greater purchase Further, building on prior research (Aaker, Vohs, and Mogil- intentions for the brand name with size cue of bigness (vs. small- ner 2010; Dubois, Rucker, and Galinsky 2016), we hypothesize that ness) when they had an agentic goal (i.e., meeting with a business the downstream consequence of size cues in brand names on brand associate). evaluation depends on the relative importance of warmth and compe- Study 6 examined the role of brand gender positioning (femi- tence in the consumption context. We identify two situational factors, nine vs. masculine). Participants were told that the brand was a per- consumption goals and brand gender positioning, which interact with fume brand (feminine) or a cologne brand (masculine). A significant size cues in brand names to influence consumers’ brand evaluation. interaction between size cue and brand gender positioning on brand Seven studies were conducted to test the proposed hypotheses. evaluation emerged. Specifically, participants indicated more favor- Study 1 used a standard IAT to test the basic premise. In this IAT, able evaluations when a feminine brand (i.e., perfume) incorporated participants sorted either size cues (e.g., small, big) or person names the size cue of smallness versus bigness in the brand name. Con- (e.g., David, Mary). The results show that the average D-score was versely, participants indicated less favorable evaluations when a positive and significantly greater than zero, supporting our key prem- masculine brand (i.e., cologne) incorporated a size cue of smallness ise that participants indeed associate smallness with femininity and versus bigness in the brand name. bigness with masculinity. Taken together, we show that size cues of smallness (vs. big- Study 2A and 2B examined the effect of size cues on percep- ness) evoke femininity (vs. masculinity) gender associations and tions of warmth and competence of a brand among Chinese and US thereby increase the warmth (vs. competence) perceptions of brands. consumers. In study 2A, Chinese participants were randomly as- We further show that such gender associations evoked by the size signed to one of the two size cue (small vs. big) conditions. Spe- cues have important downstream consequences on consumers’ brand cifically, participants were presented with the same brand logo and evaluation, depending on the salience of consumption goals (com- information, but with different brand names (Kaiyuan Small Pavilion munal vs. agentic) and the brand’s gender positioning (feminine vs. vs. Kaiyuan Big Pavilion). Results showed that brand name with size masculine). Theoretical and managerial implications of this research cue of smallness (vs. bigness) was perceived to be higher in warmth are discussed. but lower in competence. In study 2B, we replicated these findings among US consumers using different brand names (Little Salon vs. Grand Salon).

Advances in Consumer Research 880 Volume 46, ©2018 Advances in Consumer Research (Volume 46) / 881 REFERENCES Fiske, Susan T., Amy J. C. Cuddy, Peter Glick, and Jun Xu Aaker, Jennifer, Kathleen D. Vohs, and Cassie Mogilner (2010), (2002), “A Model of (Often Mixed) Stereotype Content: “Nonprofits Are Seen As Warm and For-Profits as Competent: Competence and Warmth Respectively Follow from Perceived Firm Stereotypes Matter,” Journal of Consumer Research, 37 Status and Competition,” Journal of Personality and Social (2), 224-37. Psychology, 82 (6), 878-902. Aikhenvald, Alexandra Y. (2016), How Gender Shapes the World, Touraille, Priscille, and Pierre-Henri Gouyon (2008), “Why Are Oxford University Press. Women Smaller Than Men? When Anthropology Meets Boroditsky, Lera, Lauren A. Schmidt, and Webb Phillips (2003), Evolutionary Biology,” Nature Precedings, 713. “Sex, Syntax, and Semantics,” Language in Mind, 61-79. White, Judith B., and Wendi L. Gardner (2009), “Think Women, Dubois, David, Derek D. Rucker, and Adam D. Galinsky (2016), Think Warm: Stereotype Content Activation in Women with a “Dynamics of Communicator and Audience Power: The Salient Gender Identity, Using a Modified Stroop Task,” Sex Persuasiveness of Competence versus Warmth,” Journal of Roles, 60 (3), 247-60. Consumer Research, 43(1), 68-85. A Phenomenological Examination of Internet Addiction: Insights from Entanglement Theory Mohammadali Zolfagharian, Bowling Green State University, USA Atefeh Yazdanparast, University of Evansville, USA Reto Felix, University of Texas Rio Grande Valley, USA

EXTENDED ABSTRACT more sensitive approach capable of capturing the transitions and nu- Due to the scale, scope, and complexity of its impact, digital ances of IA at the level of lived experiences. The phenomenological technology has been referred to as the fourth industrial revolution insights came from an interpretive analysis of consumer narratives. (Schwab, 2017). Researchers have called for a better understand- The text was developed through in-depth interviews and personal es- ing of such impacts and raised questions about the possible conse- says in the Southwestern United States. Each narrative was first read quences of digital technology use for consumers (Turel, et al., 2017). as an isolated case to discover its own relevant meanings, and then Emerging research and practice commentary has started to point out compared against all other cases so as to identify the pervasive, com- the complex and often alarming ways in which digital technology mon, minority, unique, and deviant cases and instances. use affects individuals, organizations, and society, forming a field of The findings juxtapose three enabling themes representing In- inquiry referred to as the dark side of digital technology (Rosen et al., ternet dependence (immediacy, task multiplicity and multi-tasking, 2013; Turel et al., 2017). and intimacy and self-disclosure) with three themes that depict the Considering that Internet is the most important driver of the perils of Internet dependency (context agnosticity, automaticity, and digital revolution, it is not surprising that the dark side of digital corporal craving). Undeniably, Internet provides us with ways to find technology has been studied with reference to Internet and Internet important information instantly (i.e., immediacy), to be more effi- addiction (IA). Similar to other addictions, IA represents a compul- cient (i.e., multi-tasking), and to easily communicate with others and sive usage and repetitive engagement in a behavior that leads to un- develop close relationships (i.e., intimacy). However, we identify the desirable consequences for addicts as well as non-addicts (Lee et al., fine line between dependence and dependency and the path to entrap- 2014; Thomee et al., 2011). However, two important aspects of IA ment through forming dependences and dependencies. For instance, make it distinct from other addictions. First, whereas in the case of to some of our participants, Internet has become anthropomorphized addictions to drugs or cigarettes, the object of the addiction is rela- and is depicted as a friend who listens to them and someone who can tively concrete and tangible, in the case of IA, it is not clear to what be trusted. Such intimacy could in turn facilitate further Internet use exactly individuals are addicted. This ambiguity has prompted some and disclosure to the extent of getting addictive and. researchers to distinguish between addiction to Internet and addic- Internet is omnipresent and grabs our attention, no matter if at tion on Internet (e.g., Griffiths, 2010). While addiction to Internet re- work, at school, at home, or when being with family or friends; an lates to a generalized use of Internet, addiction on Internet represents aspect of Internet dependency that we termed “context agnosticity.” specific types of IA such as online gambling, online shopping, and Automaticity refers to the phenomenon that being connected to Inter- online pornography (Brand et al., 2014; Montag, 2017). net becomes an automatic process, thus skipping the typical first step Offering an alternative view, we posit that even specific types of of need recognition in decision-making. In other words, we connect IA represent addictions to underlying physiological, psychological, to Internet for being connected, rather than for a higher-level goal. and social reward systems. Thus, our approach dissolves the duality Finally, we find multiple accounts of corporal craving which our between being addicted to Internet and being addicted on Internet. participants expressed vividly as a “thirst,”, “hunger,” or “itch” for We introduce a doubly-mediated model of IA with the first mediation Internet. Consistent with entanglement theory, we find that the inter- layer consisting of Internet, and the second mediation layer consist- play between the enabling aspects of Internet (dependence) and the ing of specific activities (gambling, shopping, social media, etc.) on factors related to the dark side of digital technology (dependency) re- Internet facilitated by various digital platforms (e.g., smartphones). sults in consumer entrapment. We suggest that the multiple accounts We examine human-digital technology relationships at the level of of IA denial are largely motivated by the inherent nature of Internet lived experiences to provide a better understanding of how IA and its dependence versus dependency. Specifically, even though many of various forms are shaped. our informants acknowledged their excessive use of Internet, they Second, IA is substantially different from other addictions, be- frequently neglected the potential harm resulting from compulsive cause due to our reliance on Internet, it becomes increasingly dif- Internet use because they find it easy to point out how Internet im- ficult to function without it. Thus, we identify a constant tension proved their lives. Moreover, the widespread use of Internet and between relying on Internet and the risk of becoming addicted to changing societal norms as related to use of mobile devices make exit. We draw on entanglement theory (Hodder, 2012) and explore the cessive Internet use seem a generally accepted behavior as opposed entanglement of dependences and dependencies that lead to Internet to other forms of addiction in which the addict is not proud of his or entrapment (Hodder, 2014). Entanglement theory posits humans in her compulsive behavior. relation to material things and proposes a “dialectical relationship between dependence (the reliance of humans and things on each oth- REFERENCES er) and dependency (the constraints that humans and things place on Brand, M., Laier, C., and Young, K.S. (2014). Internet addiction: each other)” which potentially leads to entrapment (Hodder and Mol, coping styles, expectancies, and treatment implications. 2016, p. 1067). The state of entrapment describes a human condition Frontiers in Psychology, 5, 1-14. where individuals are caught in the interplay between the enabling Griffiths, M.D. (2010). Internet abuse and internet addiction in the nature of things and their undesired tendency to make individuals workplace. Journal of Workplace Learning, 22(7), 463-472. dependent users craving for more (Hodder, 2012). Hodder, I. (2012). Entangled: An archaeology of the relationships We employ a qualitative research approach based on phenome- between humans and things. Malden, MA: Wiley-Blackwell. nological interpretation (Hycner, 1985; Thompson, 1997) to pursue a

Hodder, I. (2014). The entanglements of humans and things: A Schwab, K. (2017). The fourth industrial revolution. Crown long-term view. New Literary History, 45, 19-36. Business, New York, NY. Hodder, I., and Mol, A. (2016). Network analysis and Thomee, S., Härenstam, A., and Hagberg, M. (2011). Mobile phone entanglement. Journal of Archaeological Method and use and stress, sleep disturbances, and symptoms of depression Theory, 23(4), 1066-1094. among young adults-a prospective cohort study. BMC public Hycner, R. H. (1985). Some guidelines for the phenomenological health, 11(1), 66. analysis of interview data. Human studies, 8(3), 279-303. Thompson, C. J. (1997), “Interpreting consumers: a hermeneutical Lee, Y. K., Chang, C. T., Lin, Y., and Cheng, Z. H. (2014). The dark framework for deriving marketing insights from the texts side of smartphone usage: Psychological traits, compulsive of consumers’ consumption stories,” Journal of marketing behavior and technostress. Computers in Human Behavior, 31, Research, 34(4), 438-455. 373-383. Turel, O., Soror, A. and Steelman, Z. (2017). The dark side of Montag, C., and Reuter, M. (2017). Internet addiction: information technology: Mini-track introduction. Proceedings Neuroscientific approaches and therapeutical implications of the 50th Hawaii International Conference on System including smartphone addiction, second edition. Cham, Sciences. Switzerland: Springer. Rosen, L.D., Whaling, K., Rab, S., Carrier, L.M. and Cheever, N.A. (2013). Is Facebook creating iDisorders? The link between clinical symptoms of Psychiatric disorders and technology use, attitude and anxiety. Computers in Human Behavior, 29, 1243-1254. Time-of-Day Effects on Consumers’ Social Media Engagement Ozum Zor, Rutgers University, USA Kihyun Hannah Kim, Rutgers University, USA Ashwani Monga, Rutgers University, USA

EXTENDED ABSTRACT We observe such a pattern across four studies. In study 1, we ex- Imagine that it is early in the morning and you have a couple of plored the factors urging Twitter usage and the attractiveness of dif- minutes to spend on Twitter. Which article would you “like” more: ferent topics in the morning and the evening hours. We found initial an article that guarantees a fun time with its catchy title or one that evidence that individuals are interested in both long-term and short- signals long-term benefits with its profound title? Would your pattern term benefits on Twitter and this interest may shift towards vice con- of “likes” change by the evening? Is it possible that the pattern de- tent through the end of the day. In study 2, we manipulated magazine pends on whether the information is perceived to be a virtue (i.e., not type (vice vs. virtue), and asked participants about their likelihood of tempting, but offering later benefits) or a vice (i.e., tempting, but with reading the articles tweeted by given magazines in the morning and few later benefits)? In four studies, including one using data from the evening. Likelihood to read vice (vs. virtue) magazines was low- Twitter, we demonstrate time-of-day effects such that the consump- er in the morning but preference in the evening shifted toward vice tion of vice (relative to virtue) increases as morning turns to evening. magazines. In study 3, we again manipulated magazine type (vice Engagement with social media continues to grow, and US firms vs. virtue), but measured the time at which participants responded to are expected to spend $67 billion on digital advertising by 2021 the survey (between 10 am and 6 pm). We asked participants about (Fisher 2017). Consequently, consumer behavior researchers have their likelihood of reading the articles tweeted by given magazines, started to explore the psychology that underlies engagement over and observed a significant interaction effect. Specifically, the reading social media (Barasch and Berger 2014; Toubia and Stephen 2013; likelihood shifts away from virtue towards vice as the day progresses. Wilcox and Stephen 2013). What remains unexplored, however, is Finally, in study 4, we collected more than two million tweets over a how social media engagement may change through the day, and if hundred twenty-five days, and then analyzed the unique tweets (after there is any asymmetry in engagement with different types of con- eliminating retweets and duplicates within a day). While controlling tent. We show that the time of day matters, and matters differently for for a variety of factors (e.g., total number of tweets by a magazine), vices versus virtues. we observed a significant interaction between time-of-day and mag- Our theorizing follows from the literature of self-control. Just as azine type. Specifically, social media engagement as measured by muscles are weakened after physical exertion, self-control weakens “likes” increases overall as the day progresses, but there is a statisti- after subsequent mental exertions (Muraven and Baumeister 2000). cally significant shift toward vices rather than virtues. Individuals are faced with multiple self-control challenges through We add to the self-control literature by revealing not only that the day, as they try to avoid activities and foods that may be tempt- depletion during a day influences the consumption of information ing, but are not good for them in the long run. Exerting self-control overall, but that such depletion results in asymmetry for vices ver- over and over again is depleting. Consequently, self-control is stron- sus virtues. Implications also arise for marketers who, depending on gest in the early hours of a day but weakens as the day progresses the nature of the social media content, may want to release different (Dai et al. 2015; Kouchaki and Smith 2014). What is not yet un- content at different times during the day in order to maximize the derstood is how self-control depletion may emerge for information engagement of consumers. consumption. One may simply expect depletion effects of the kind that have been observed earlier. That is, because consumers get tired REFERENCES through the day, they process less information because they are not Barasch, Alixandra and Jonah Berger (2014), “Broadcasting and left with adequate resources to read and understand more informa- Narrowcasting: How Audience Size Affects What People tion. We reveal a different result. Share,” Journal of Marketing Research, 51(3), 286–99. We argue that the effects of depletion may be different for dif- Dai, Hengchen, Katherine Milkman, David Hofmann, and Bradley ferent types of information. Specifically, we discover an asymmetry Staats (2015), “The Impact of Time at Work and Time Off in the consumption of vice versus virtue information. A relative vir- From Work on Rule Compliance: The Case of Hand Hygiene tue represents immediate costs with long-term benefits, but a relative in Health Care.,” Journal of Applied Psychology, 100(3), vice reflects an instantaneous pleasure with little delayed benefits and 846–62. perhaps even negative consequences (Khan, Dhar, and Wertenbroch Fisher, Lauren T. (2017), “US Digital Display Advertising 2018: 2004; Siddiqui, May, and Monga 2017; Wertenbroch 1998). Thus, an Seven Big Bets for Buyers and Sellers,” eMarketer Report article on the financial markets in Forbes could be considered virtue (December 21), https://www.emarketer.com/Report/US- information by many who find it hard to understand, but appreciate Digital-Display-Advertising-2018-Seven-Big-Bets-Buyers- the long-term benefit of becoming more knowledgeable on the topic. Sellers/2002198. However, salacious gossip in the Cosmopolitan could be considered Hayes, Andrew (2013), Introduction to Mediation, Moderation, and vice information by many who are tickled by the instantaneous plea- Conditional Process Analysis, New York: Guilford Press. sure, but know that there is no long-term benefit from it. We argue Hoch, Stephen and George Loewenstein (1991), “Time- that the self-control depletion arising at the end of the day is likely Inconsistent Preferences and Consumer Self-Control,” Journal to reduce the resources to process the Forbes article, but make one of Consumer Research, 17 (4), 492–507. more likely to succumb to the immediate gratification from reading Hofmann, Wilhelm, Kathleen D. Vohs, and Roy F. Baumeister the Cosmopolitan article. That is, self-control depletion through the (2012), “What People Desire, Feel Conflicted About, and Try day may shift consumption toward vice information and away from to Resist in Everyday Life,” Psychological Science, 23(6), virtue information. 582–88.

Johnson, Palmer O. and Jerzy Neyman (1936), “Tests of Certain Oster, Sharon. M and Fiona M. Scott Morton (2005), “Behavioral Linear Hypotheses and Their Application to Some Educational Biases Meet the Market: The Case of Magazine Subscription Problems,” Statistical Research Memoirs, 1 (January), 57–93. Prices,” B.E. Journal of Economic Analysis and Policy: Khan, Uzma and Ravi Dhar (2007), “Where There Is a Way, Is Advances in Economic Analysis and Policy, 5(1), 1–30. There a Will? The Effect of Future Choices on Self-Control,” Siddiqui, Rafay A., Frank May, and Ashwani Monga (2016), “Time Journal of Experimental Psychology: General, 136 (2), Window as a Self-Control Denominator: Shorter Windows 277–88. Shift Preference toward Virtues and Longer Windows toward Khan, Uzma, Ravi Dhar, and Klaus Wertenbroch (2005), “A Vices,” Journal of Consumer Research, 43 (6), 932-49. Behavioral Decision Theory Perspective on Hedonic and Spiller, Stephen A., Gavan J. Fitzsimons, John G. Lynch, and Gary Utilitarian Choice,” in Inside Consumption: Frontiers of H. McClelland (2013), “Spotlights, Floodlights, and the Magic Research on Consumer Motives, Goals, and Desires, ed. S. Number Zero: Simple Effects Tests in Moderated Regression,” (Ratti) Ratneshwar and David Glen Mick, London: Routledge, Journal of Marketing Research, 50(2), 277–88. 144–65. Toubia, Olivier and Andrew T. Stephen (2013), “Intrinsic vs. Kouchaki, Maryam and Isaac H. Smith (2014), “The Morning Image-Related Utility in Social Media: Why Do People Morality Effect: The Influence of Time of Day on Unethical Contribute Content to Twitter?,” Marketing Science, 32(3), Behavior,” Psychological Science, 25(1), 95–102. 368–92. Muraven, Mark and Roy F. Baumeister (2000), “Self Regulation Wertenbroch, Klaus (1998), “Consumption Self-Control by and Depletion of Limited Resources: Does Self Control Rationing Purchase Quantities of Virtue and Vice,” Marketing Resemble a Muscle?,” Psychological Bulletin, 126(2), 247–59. Science, 17(4), 317–37. Wilcox, Keith and Andrew T. Stephen (2013), “Are Close Friends the Enemy? Online Social Networks, Self-Esteem, and Self- Control,” Journal of Consumer Research, 40(1), 90–103. Charity Donors’ Response to Cause-Related Marketing: The Role of Attachment Styles Sondes Zouaghi, Nimec, IAE de Caen, France Aïda Mimouni Chaabane, Thema, University of Cergy-Pontoise, France

EXTENDED ABSTRACT (Frijda 2016), they will result in pro- (i.e., favorable actions such as Cause-related marketing (CM) is a type of sponsoring activity increasing one’s donations) or anti-charity (i.e., harmful actions such whereby firms, for a limited period of time, donate a certain amount as spreading negative word of mouth or revenge) behavioral inten- of the sales price of the firm’s product or revenue to a specific cause. tions. We hypothesize that positive emotions mediate the relationship While literature is prolific on how the firm (Mende, Bolton and Bit- between secure donors and pro-charity behavioral intentions (H1). ner 2013), the consumer (e.g., Krishna 2011), or, more marginally, We also hypothesize that anxious style of attachment is character- the cause (e.g., Lafferty and Golsmith 2005) benefit from CM, char- ized by ambivalence (emotions and behavior) (H2). Finally, as they ity donors responses are a neglected area. This topic is however im- are devoid of emotions, avoidant donors decrease their loyalty (H3). portant because the increased need for charitable giving necessitates 262 French donors were asked to choose a particular charity a large, strong, and stable portfolio of loyal donors who adhere to to which they made at least one donation (monetary, time or other). the charity’s strategy. Consequently, the present research focuses on Then, we measured the respondents’ AS, their familiarity with CM, the charity donors and not on consumers who buy CM offers. The and their attitude toward CM in general. In the next phase, we asked research question is thus the following: How do charity donors’ re- respondents to imagine that their charity was about to implement CM. spond to CM implemented by their charity? We then measured the mediating and dependant variables (positive As the literature is silent on the issue of donors’ response to emotions, negative emotions, pro-charity and anti-charity behavioral CM, we resort to an exploratory qualitative approach in Study 1 (22 intentions, with items presented randomly) and socio-demographics. semi-directed interviews). With no surprise, donors’ response to CM To test the effect of attachment style dimensions on pro- and implementation seemed to be influenced by their history with CM anti-charity behavior, we used model 4 from the Preacher-Hayes (i.e., general attitude, familiarity). More importantly, the analysis re- script for bootstrap analyses of indirect effects, with 5.000 resample vealed that the style of informant relationship with the charity, called (Preacher and Hayes 2004). The results provide support for H1 and attachment style (AS), seemed to affect their response. Some respon- H2 (p < 5%) and a partial support for H3 (p < 10%). dents talked about their charity in a very consistent and emotionally This research enriches the attachment style literature. First, in positive way and seemed to have a good level of involvement and line with previous work in marketing (e.g., Mende et al. 2013; Thom- commitment toward it. They seemed to consider CM as a possible son et al. 2012), the findings provide additional empirical support to funding source and even willing to propose adjustments to optimize the predictive power of AS. More precisely, AS have a significant its implementation for the charity. Another group of respondents direct and indirect effect on behavioral intentions, even when taking seemed to have an unemotional speech. They analyzed the CM with into account the donors’ previous history with CM and the type of a certain distance, and expressed doubts about the reason that drives their donation. The findings also highlight the theoretical relevance charities to participate to CM. Their reactions ranged from indiffer- of this concept: philanthropic relationships, as commercial (e.g., ent to negative. Finally, some respondents seemed to have high lev- Mende et al. 2013) and interpersonal relationships (e.g., Collins and els of expectations and recognition needs toward the charity. Their Read 1990), are characterized by secure and insecure AS. Second, speech appeared as very emotional but instable and paradoxical. On the present research offers a framework for explaining the - attach the one hand, they seemed highly involved in the charity and desired ment process based on emotions. Given the importance of these find- an exclusive relationship with it. On the other hand, they seemed to ings, charity managers are encouraged to embrace a bigger vision of feel insecure and unable to explore new ideas of fundraising oppor- CM impact by balancing its potential benefits in terms of notoriety tunities by fear of losing or damaging the relationship they entertain and image with the potential risk of upsetting and eventually losing with the charity. These respondents expressed rather negative emo- donors. In particular, they are encouraged to measure the AS of their tions and reactions (e.g., sadness, anger). donors in order to isolate donors segments that have the potential to Study 1 provides insights into donors’ relational profiles likely be more or less responsive to CM, and tailor CM implementation and to accept or reject CM. The key construct that emerged from our communication accordingly. Charity managers should also explain analysis in response to CM implementation by the charity is AS. AS the necessity of resorting to CM in order to build a positive general are conceptualized by two dimensions: avoidance and anxiety (Fral- attitude, especially among donors who volunteer their time. Further ey, Waller, and Brennan, 2000). A high level of avoidance indicates research should explore interventions that could blunt the negative poor perception and lack of trust in others. A high level of anxiety impact of CM on the anxious donors and test the generalizability of means lack of self-confidence and high need of reassurance. Based findings to other contexts of altruism. on this view, early research distinguishes between three AS: secure, avoidant and anxious (Collins and Read 1990; Hazan and Shaver REFERENCES 1987; Simpson 1990). In Study 2, we examine how (i.e. the mecha- Ainsworth, M.C., Waters, B.E., & Wall, S. (1978). Patterns of nisms under which) AS influence donors’ response to CM. attachment: A psychological study of the strange situation. AS are rooted in the first years of life where the emotional ef- Hillsdale, NJ: Lawrence Erlbaum. fects of relationships with the parental figure are experienced and Bowlby, J. (1969). Attachment and loss: Volume 1. attachment. activated at each interaction with others (Ainsworth et al. 1978; New York: Basic Books. Bowlby, 1969). We thus postulate that depending on their AS, donors Collins, N. L. & Read, S. J. (1990). Adult attachment, working will exhibit positive (e.g., happiness and confidence in the charity’s models, and relationship quality in dating couples. Journal of decision to resort to CM) or negative (e.g., disappointment by the Personality and Social Psychology, 58(4), 644–663. decision, anger and discontent) emotions toward CM implementation by the charity. As emotions represent motivational intentions

Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item Mende, M., Bolton R. N., & Bitner M. J. (2013). Decoding response theory analysis of self-report measures of adult customer-firm relationships: how attachment styles help attachment. Journal of Personality and Social Psychology, explain customers’ preferences for closeness, repurchase 78(2), 350–365. intentions, and changes in relationship breadth. Journal of Frijda N.H. (2016). The evolutionary emergence of what call Marketing Research, 50(1), 125-142. “emotions”. Cognition and emotions, 30(4), 609-620. Preacher, K. J. & Hayes, A. F. (2004). SPSS and SAS procedures Hazan, C. & Shaver, P. (1987). Romantic love conceptualized as for estimating indirect effects in simple mediation models. an attachment process. Journal of Personality and Social Behavior Research Methods, Instruments, & Computers, Psychology, 52(3), 511–524. 36(4), 717–731. Krishna, A. (2011). Can supporting a cause decrease donations Simpson, J. A. (1990), influence of attachment styles on romantic and happiness? The cause marketing paradox. Journal of relationships. Journal of Personality and Social Psychology, Consumer Psychology, 21(3), 338-345. 59(5), 971-980. Lafferty, B. A. & Goldsmith, R. E. (2005). Cause-brand alliances: Thomson, M., Whelan J, & Johnson A. R. (2012). Why brands Does the cause help the brand or does the brand help the should fear fearful consumers: How attachment style predicts cause? Journal of Business Research, 58(4), 423–429. retaliation. Journal of Consumer Psychology, 22(2), 289–298.

Videography

Alternative Worldviews on Human – Nonhuman Relations: The Turkish Case N. Alican Mecit, HEC Paris, France Tina M. Lowrey, HEC Paris, France

Literature foundations: Consumer research studies how we attribute meaning to commodities and bring them into our lives (Belk 1988, Miller 2008). Through singularization, commodities take on meaning and enter the universe of unique objects (Kopytoff 1986). Until they are recommodified, they remain less available for exchange (Epp and Price 2010). Today, in most cultures, the permanent constituents of the universe of singular objects are only humans as it is unthinkable to exchange humans for money (Kopytoff 1986). Humans, however, used to share this particular universe with animals who lived parallel lives to humans and roamed freely without any owner (Berger 2009). Today, animals, even historically the closest to us - cats and dogs - are reduced to a level at which they require humans to pull themselves out of the universe of commodities. For example, a cat is a de facto commodity until one pays its price and brings him or her to the universe of singular objects (Corrigan 1997). Intended contribution and findings: To explore an alternative worldview, we went to Istanbul, Turkey where cats and dogs still roam freely. We questioned the commodity status of animals and talked to the local people. We situated these animals in the universe of singular objects, deserving the same rights as humans. As the presence of animals in the universe of singular objects is a companionship offered to our loneliness as species, we also focused on the implications of the cohabitation with another kind in this conceptual universe. Method: The videography reports our observations of human – nonhuman relations in Istanbul, interviews with the local people, and the secondary sources that we relied on to triangulate our interim findings. Augmented Reality, Augmented Trust: How Augmented Reality Enhances Consumer Trust In Online Shopping Alberto Lopez, Tecnológico de Monterrey, Mexico Rachel Rodriguez, Tecnológico de Monterrey, Mexico Claudia Quintanilla, Tecnológico de Monterrey, Mexico Raquel Castaño, Tecnológico de Monterrey, Mexico

Intended Contribution to Knowledge: Online shopping has become a common practice among consumers worldwide (Nielsen, 2017). Since the transactions over the internet take place without personal contact, consumers are generally concerned about their purchases (Chen & Dhillon, 2003). It is therefore important for brands and companies to generate consumer trust in online contexts. On the other hand, Aug- mented Reality (AR) has emerged as a relevant interactive technology in the marketing environment. It is the ability to overlay the physical environment with virtual elements that can interact with the physical environment in real time. It consequently holds the potential to alter many consumer activities. Hence, there is a growing need to better understand its impact on consumer behavior (Javornik, 2016). In this research, we aim to study how the technology of AR affects consumer trust in an online shopping context. Literature Foundations: This study builds on consumer trust theory and adoption of new technologies. Research Method: Following an ethnographic approach, participants used AR in their homes as an opportunity to mix their personal spaces with virtual furniture within a mobile app that was developed for this purpose. Findings and implications: This research suggests that AR technology can serve as a coping strategy to either increase or decrease consumer trust in online contexts. We add to the literature by proposing that this new technology increases consumer trust by six motivators (problem-solving, benevolence, reduced psychological distance, interactivity, novelty, and enjoyment) and decreases consumer trust by two demotivators (need for touch concerns and feeling of intrusion). The authors also illustrate a few age differences regarding the adoption and usage of this new technology. Furthermore, it was found that AR also affects both, consumers and brands. Some marketing cases affected by the usage of AR are consumers’ expertise and satisfaction, brand evaluations and purchase intention.

Advances in Consumer Research 889 Volume 46, ©2018 890 / Videography Crossing Race and Markets: Introducing the Race in the Marketplace Research Network Kevin D Thomas, University of Texas at Austin, USA Sonya Grier, American University, USA Guillaume D Johnson, Université Paris-Dauphine, France

Both research and practice reveal that race plays a key ideological role in the functioning of consumption markets worldwide (Nopper 2011; Thomas 2017; Zerofsky 2016). However, scholarship on race-related issues in the context of marketplaces has generally been limited, overlooked or marginalized in the marketing/consumer research discipline. However, the tangibility, reality, and brutality of racial dynamics are increasingly “front-page news” and marketing scholarship increasingly emphasizes issues of inclusion and exclusion (Henderson and Williams 2013). As a result, the absence of scholarship- and scholars- who prioritize impactful work on race in marketing presents a challenge to move the discipline towards more inclusive scholarship, employment, and pedagogy. This film explains impediments to scholarship at the intersection of race and markets by examining the evolution of the RIM Research Network (RIM). Developed in 2017, RIM is a response to the marginalized status of race-related scholarship and scholars in the marketing domain. While marketing scholars and scholarship that focus on race and racism have long suffered a marginalized status, attempts to attract private sector dollars has resulted in large-scale shifts in institutional cultures and priorities which has served to exaggerate pre-existing hierarchies within research topics, disciplines, and methodolo- gies (Moleworth, Scullion, and Nixon 2010). Areas and methods most deemed able to benefit the bottom line of potential donors and strategic partners garner the most attention and praise, while critical race scholarship and other forms of research perceived as discordant with present- day marketplace functioning are further marginalized. The conceptual equivalent of consumers, these scholars find a dwindling marketplace in which to cultivate community, acquire funding, secure publication outlets, and gain a sense of legitimacy in the broader academic canon. In this marketized environment, research that focuses on diversity and multiculturalism often take precedence over critical examinations of marketplace activities that expressly investigate racism since the former tends to obscure or ignore systemic issues of power, oppression, and privilege, thereby leaving inherent marketplace inequities largely intact and unquestioned (Grier, Thomas, and Johnson 2017). RIM seeks to establish and legitimize a cohesive critical perspective which foregrounds the reality of power, privilege and oppression, question existing marketing strategies and link them to an overall intersectional framework that can promote inclusive, just and liberatory marketplaces. This film chronicles RIM’s quest to counteract the current academic environment by innovating a transdisciplinary approach to the understanding of race in the marketplace that will produce a space within academia wherein RIM-related scholars and scholarship can thrive. Don’t Troll Me Bro: A Study of Griefing in Video Games Elana Harnish, Ohio University, USA Jacob Lee Hiler, Ohio University, USA

Intended Contribution to Knowledge: The purpose of this study is to examine consumer griefing and toxicity in the context of video games, and seeks to uncover the antecedent, define the process, and discuss the outcomes of the phenomenon. Literature Foundations: This videography contributes to research in consumer culture theory by exploring an underexplored but timely topic, consumer griefing and toxicity, in the context of video games. Research Method: A combination of videographic, netnographic, and experimental methods were used in this study. Five semi-structured interviews were conducted with gamers and approximately 50 hours of griefing footage was collected via YouTube and Twitch. Additionally, a field experiment was conducted through 46 matches of a popular MOBA game over approximately 20 hours. This research was filmed in accordance with the videography practices set forth by Belk & Kozinets (2005). Findings and implications: This research uncovers the antecedents, process, and outcomes of consumer griefing and toxicity in video games, and demonstrates that toxicity has a tangible effect only players win rate in competitive games. From a Culinary Phantasm to an Eudaimonic Well-Being: Exploring The Experience of Amateur Cooking Classes Virginie Brégeon de Saint-Quentin, Ferrandi Paris, France Ophélie Mugel, Chaire SDSC AgroParisTech, Université Paris Est, France

Intended contribution to knowledge : This research, Consumer Culture Theory and Grounded Theory driven, focuses on the dimensions of the « learning by doing » experience of amateur cooking classes. Literature foundations : This exploration refers, a posteriori, to an eudemonic approach of well-being (an individual fulfillment relating to self-actualization leading to a series of meaningful experiences). It echoes with the three psychological needs of Self-Determination Theory (the needs for competence, autonomy and relatedness) and the optimal experience of flow. Research Method : Through an interpretive approach, we conducted ethnographical observations, recorded films, took photographs and interviewed amateur cooks and professional Chefs in five cooking classes, three in France and two in Japan. Findings and implications : Thanks to an in depth analysis, we designed a framework for cooking class experience and noticed six dimensions : (1) Expertise, as the desire to reproduce Chef’s technical gestures and collect expert information about products and recipes ; (2) Immersion, as strategies of imitation and roleplay ; (3) Convivialité (and commensality), as the pleasure of sharing ; (4) Identity, as a self pursuit and affirmation and cultural heritage ; (5) Sensoriality, as five senses synesthesia ; (6) « Naturality », concerning authentic, seasonal or healthy products and sustainable practices. Also, in a Transformative Consumer Research perspective, public policy and healthcare stakeholders are encouraged to promote cooking competences in order to improve consumers’ food well-being. These dimensions can also be manipulated to enhance motivation and satisfaction of cooking school trainees as well as for raising cooking classes profitability and for developing cooking class concepts adapted to the new generation of millennials worldwide. Advances in Consumer Research (Volume 46) / 891 Meat the Needs: Ahold Delhaize Sustainable Retailing Model Darrell Eugene Bartholomew, Pennsylvania State University, USA Maggie M Mehalko, Pennsylvania State University, USA

Intended Contribution to Knowledge. Zero waste initiatives are having a real impact on reducing food waste of edible and inedible parts of food. There are good zero waste practices for diverting food away from landfills to be used to generate energy such as in a food digester or to be used as compost. Better practices help to divert food for human consumption. Many grocers do a lot of good things for the food banks and their local community partners such as shelters and soup kitchens to relieve hunger. The purpose of this paper is to illustrate an innovative zero waste food rescue program that can be used as an example of best practices in sustainable retailing. Literature Foundations. Recent research has focused on this intersection of sustainability, marketing and public policy which are all part of green retail strategies (Iyer and Reczek 2017). Food waste prevention is a goal of most grocers, food service companies, farmers, and food manufacturers (Block et al. 2016). Perceived risk to brand (Murray and Schlacter 1990) and challenges of food safety and handling procedures as well as consumer behavior and perception of inventory were all issues that Ahold Delhaize had to address when the program was piloted. Research Method.Using a case methodology Giant Foods meat rescue program will illustrate this innovative program from the perspective of the decision makers that created the ‘Meat the Needs’ Program. Findings and Implications. The impact of this innovative program has effectively altered the supply chains at six major food banks in the Northeastern U.S. challenging traditional methods of food banking and creating several new opportunities for food rescue programs involving ugly produce, fruit rescue, and milk rescue in addition to the original meat rescue program. Tattoo: A Perspective Beyond Estethics Luana C. Moraes, Universidade de Sao Paulo, Brazil Gabriela L. Pinheiro, Universidade de Sao Paulo, Brazil Nathalia S. Arthur, Universidade de Sao Paulo, Brazil Eliani C. Flores, Universidad Catolica del Peru, Peru Jose Mauro C. Hernandez, Centro Universitário FEI, Brazil

Over the past 20 years, tattoos have become extremely popular in Western countries, spreading among several audiences and becoming a very popular form of body modification. Through the centuries, body modification has been an important part of human´s life as an expression of the self. Until recently, tattoos were exclusive to marginalized groups such as sailors, prostitutes, and criminals, and portraying a tattoo was motive for being prejudiced. It was not until the 60’s that tattoos assumed a new role in our society. First, as an expression of self-identification of youth movements and lately as a cultural phenomenon of body embellishment. Today, tattoos are increasingly seen as a form of art adorning the human canvas and, as such, tattoos compete in the body adornment industry. Our videography sets out to uncover the meanings associated with the tattooing consumption experience. In our journey, we found that the tattooed body is seen by their owners as a canvas or as a media. When the body serves as a canvas, the tattoo is an expression of art whose pure objective is the body embellishment; the symbolic meaning attached to the tattoo can be extremely important or not important at all – in this case, the tattooing is just the art for the art’s sake. When the body serves as a media, the tattoo expresses the cultural, religious, philosophic, gender, sociological, and political ideology of the owner and the defining aspect of the tattoo is its meaning. Interestingly, due to its perenniality, the tattoo is re-signified as the individual matures and a canvas can be transformed into a media and vice versa – the art becomes an ideology and the ideology becomes an art. Teaching Consumer Resistance in Jamaica: Subvertising in Action Michelle Renee Nelson, University of Illinois at Urbana-Champaign, USA Yanyun (Mia) Wang, University of Illinois at Urbana-Champaign, USA Kathy Tian, University of Illinois at Urbana-Champaign, USA Gail Ferguson, University of Illinois at Urbana-Champaign, USA Rachel Powell, CDC Foundation, USA Candace Wray, University of West Indies, Jamaica

Our videography features a ‘subvertising competition’ among adolescents and mothers in Jamaica, as part of an intervention to build persuasion knowledge by teaching consumer resistance through creativity. Although all Jamaicans are inundated by U.S. media, those who internalize U.S. culture via “remote acculturation” are especially vulnerable to negative health habits promoted by U.S. advertising (Ferguson et al. 2017). To address this problem, we developed the J(amaican and) U(nited) S(tates) Media? Programme (2-session workshop teaching nutrition and persuasion knowledge). The film allows viewers to see the intervention (Belk et al. 2017); the critiques are performance acts of consumer power. Through art, poetry and song, families give voice to their skepticism. Literature/Contribution. We focus on family as an important consumer socialization force (e.g, Moore, Wilkie and Desrochers 2017) to build persuasion knowledge (PK; Friestad and Wright 1994). Not much is known about PK in developing countries or how to teach consumer resistance. Subvertisements simultaneously allow audiences to critically evaluate media content and to construct subversive narratives, often through parody (Harold 2004). We taught families to ask: What information is missing? What is the ad really saying? Then, families critiqued, created, and competed. Method. 92 teen-mother dyads in Kingston were selected based on higher remote acculturation and unhealthy eating. Dyads were randomly assigned to control or intervention groups. Food intake and media literacy/PK were measured at multiple points. Focus groups about the subvertisements were held. Findings.

Overall, the intervention groups scored higher than control group in media literacy (PK) and fruit consumption. Focus groups revealed knowledge of health consequences and persuasion. This pragmatic, substantive videography (Belk et al. 2017) shows the promise of transdisciplinary, theory-driven interventions to change real-world behavior (Lynch et al. 2012). The narrative follows a problem-solution format with the ‘process’ and ‘performances’ highlighted and findings embedded (Cayla and Arnould 2013). 892 / Videography Time and Space for Robots and AI Marat Bakpayev, University of Minnesota Duluth, USA

The specific gap in the knowledge our videography seeks to address is the influence of robot/artificial intelligence (AI) technology representation on consumer-level timeflow. We study how robot/AI time-space imaginations are presented in news. We aim to use evocative power of moving images (Hietanen, Rokka, and Schouten 2014). The videography contributes to study of spatial and temporal perspectives in consumer research (Castilhos, Dolbec, and Veresiu 2017; Robinson 2018; Woermann and Rokka 2015). Woermann and Rokka (2015) identified the elements that shape consumer experiences on micro-level. Our study shows how temporality and spatiality are presented by technology chronotopes on meso-level. Drawing from literary theory, we examine discourse on robots and AI. Focusing on representation of time and space in news, we contextualize Bakhtin’s theory of chronotope (Bakhtin 1981). Chronotope is “the intrinsic connectedness of temporal and spatial relationships that are artistically expressed in literature” (84). We identify four main chronotopes in the narratives on these technologies: Fantasies of the Past, Present Timeflow, Human-Master Future, and Robot-Master Future. Fantasies of the Past are rich with references to historical mythology of machines and failed expectations of AI. Present Timeflow portrays various relationships that exist now – robots/AI born in research labs and science-fiction films move to our roads, streets, homes, hospitals, schools, offices, restaurants, and so on. Future is the most dominant element of the storytelling. We have Human-Master Future and Robot-Master Future - two central ideological representations of utopia and dystopia. Human-Master may incorporate the use of robots/AI, and this chronotope calls for “robot-enhanced humanity” with robots serving humans. Robot-Master Future is a more predominant rhetoric. That chronotope assumes “robot apocalypse.” This is where machines are presented as “demons” and “supervillains” that most likely will “destroy humanity.” Workshop Summaries

Workshop Better Marketing for a Better World Jonah Berger, University of Pennsylvania, USA Jonathan Zev Berman, London Business School, UK Darren Dahl, University of British Columbia, Canada Markus Giesler, York University, Canada Rebecca Hamilton, Georgetown University, USA Gita Venkataramani Johar, Columbia University, USA John Lynch, University of Colorado, USA Andrea Morales, Arizona State University, USA

In this session leading consumer behavior researchers will discuss what research in their area would look like using the lens of “Better Marketing for a Better World.”

Workshop Trust in Doubt: Co-Chair’s Invited Panel Adam Berinsky, Massachusetts Institute of Technology, USA John Gray, MentionMapp.com, Canada Andre Spicer, City University of London, UK

In a special invited session panel members from political science management and the online research business each provide their perspectives on the post-trust world its effects and that way that they are seeking to understand it. This session intends to provide consumer researchers with alternative perspectives on post-trust phenomena that will help broaden and enrich our collective view of our changing social environment.

Workshop Conducting Consumer-Relevant Research Jeffrey Inman, University of Pittsburgh, USA Margaret C. Campbell, University of Colorado, USA Amna Kirmani, University of Maryland, USA Linda L Price, University of Oregon, USA

This workshop will include perspectives on how to conduct conceptually based consumer-relevant research that offers practical insights.

Forum Using Multi-Methods in Behavioral Pricing Research Haipeng Chen, University of Kentucky, USA David Hardesty, University of Kentucky, USA Akshay Rao, University of Minnesota, USA Lisa Bolton, Pennsylvania State University, USA

Behavioral pricing has been shaped by and is helping to shape the nascent trend towards cross-disciplinary multi-method research in consumer behavior. The proposed roundtable aims to bring together a group of scholars in the area of behavioral pricing within and outside of marketing to discuss the unique opportunities and challenges researchers in this area face with the multi-method approach. The roundtable together with the January 2021 JACR special issue and a mini-conference (to be sponsored by the University of Kentucky) aims at spurring additional interest and enthusiasm in research using multi-methods among consumer researchers interested in behavioral pricing and beyond.

Forum What’s Your Construct? Exploring the Different Definitions and Operationalizations of Scarcity Kelly Goldsmith, Vanderbilt University, USA Rebecca Hamilton, Georgetown University, USA Caroline Roux, Concordia University, Canada Meng Zhu, Johns Hopkins University, USA

This forum aims to encourage constructive discussions around the different definitions and operationalizations of scarcity in order to 1) start delimiting the boundaries of the construct 2) overcome some of the challenges for advancing this literature given the differences between various scarcity-related manipulations and 3) stimulate future collaborations.

Forum Promoting Well-being and Combating Harassment in the Academy Ekant Veer, University of Canterbury, New Zealand Zeynep Arsel, Concordia University, Canada June Cotte, Ivey Business School, Canada Jenna Drenten, Loyola University Chicago, USA Markus Geisler, York University, Canada Lauren Gurrieri, RMIT University, Australia Julie L. Ozanne, University of Melbourne, Australia Nicholas Pendarvis, California State University Los Angeles, USA Andrea Prothero, University College Dublin, Ireland Minita Sanghvi, Skidmore College, USA Rajiv Vaidyanathan, University of Minnesota Duluth, USA Stacy Wood, North Carolina State University. USA

This roundtable has been proposed to address a growing concern over well-being and safety in academia. This roundtable will discuss issues of harassment experienced by academics in order to better advocate for policies systems and a culture that will help protect all members of the academic community whilst carrying out their teaching learning administrative and research duties. This may be in their classrooms in the field at conferences and/or online. The roundtable will offer an opportunity for those attending to discuss broad issues associated with harassment and discuss different means to promote a safe and welcoming academic community free from harassment of any form. The feedback from this roundtable will be later collated by the roundtable organizers and distributed to attendees so as to provide guidance for university ad- ministrators (such as participants’ Deans department chairs etc). The collated feedback will also be used to help inform ways in which ACR’s newly formed anti-harassment working committee can work to support its members. The roundtable aims to bring a sense of empowerment and encouragement to those in attendance to combat these issues. To this end it should be noted that the purpose of the roundtable is to share broad ideas to combat harassment and promote well-being and safety. It is NOT a place to developed finalized policy or an opportunity to air specific grievances or make accusations against individuals. This will be strictly enforced by the roundtable chair.

I8. How Food Images on Social Media Influence Online Reactions

Annika Abell, University of South Florida, USA Dipayan Biswas, University of South Florida, USA

The results of four studies including two field experiments show that an image of a female gets fewer positive online reactions and is evaluated less favorably when she is pictured next to an unhealthy (vs. a healthy) item; these effects get reversed for males.

D11 . A Hidden Cost of Advocating: Attitude Depolarization After Recommending

Ravini Savindya Abeywickrama, University of Melbourne, Australia Gergely Nyilasy, University of Melbourne, Australia Simon M. Laham, University of Melbourne, Australia

The consequences of word-of-mouth recommendations on advocators themselves have largely been neglected in consumer research. The current work demonstrates paradoxical effects of advocacy on consumer attitudes by illustrating reduced-level product evaluations following a consumer word-of-mouth episode. The results demonstrate a self-persuasion theory-based depolarization effect mediated by metacognitive processing.

C1 . Promoting Subjective Preferences in Simple Choices During Sleep

Sizhi Ai, First Affiliated Hospital of Xinxiang Medical University, China Yunlu Yin, University of Hong Kong, China Yu Chen, Peking University, China Lin Lu, Peking University, China Lusha Zhu, Peking University, China Jie Shi, Peking University, China

Combining target memory reactivation method with a food evaluation/choice task we find that presenting verbal cueing during sleep but not wakefulness shifted food preferences and decisions in simple choice pointing to a potential for nudging human behavior with subtle manipulation of memory processing during sleep.

R3 . In Experiences We Trust: From Brand-centric Experiences to Experience-centric Brands

Melissa Archpru Akaka, University of Denver, USA Ana Babic Rosario, University of Denver, USA Gia Nardini, University of Denver, USA

Experiences are replacing brands as the most sought-after market offerings. In the context of American Yoga we investigate the role of brands in the assemblage of extended consumption experiences. We find that consumers’ shifting institutional logics influence their polyga- mous relationships with brands and guide the reassemblage of experiences over time.

G8. How Does Pronunciation Difficulty of Brand Names Influence Consumer Responses? The Role of Self-Construal

Gunben Ceren Aksu, Rutgers University, USA Yeni Zhou, Rutgers University, USA Alokparna (Sonia) Monga, Rutgers University, USA

We find that when a brand name is difficult-to-pronounce consumers with independent (vs. interdependent) self-construal respond more favorably. However when a brand name is easy-to-pronounce consumers with interdependent (vs. independent) self-construal respond more favorably. Further we show that differences in desire for uniqueness is the underlying mechanism.

Pia Annette Albinsson, Appalachian State University, USA B. Yasanthi Perera, Brock University, Canada Bidisha Burman, University of Mary Washington, USA Lubna Nafees, Appalachian State University, USA

Businesses are increasingly developing collaborative consumption-based (CC) models. To capitalize on opportunities they must understand what drives and deters consumers’ CC participation. This research examines how attitudes trust sustainability materialism risk propensity and generosity affect consumers’ CC usage. We find that attitude trust sustainability and generosity predict CC usage.

A4 . Persuasive Messaging in the Modern Media Landscape: Character Assassinations in Social Network’s Echo Chambers

Niek Althuizen, ESSEC Business School, France Sourjo Mukherjee, ESSEC Business School, France Negin Latifi Kasani, ESSEC Business School, France

In an empirical study we investigated if disinformation from unverified sources through social media affects people’s opinions about characters who played a role in the US 2016 presidential election and its aftermath. The results show that opinions are malleable and character boost might be more effective in altering perceived source integrity.

R6. The Anatomy of a Rival: The Influence of Inequity and Resentment on Rival Brands

Diego Alvarado-Karste, University of North Texas, USA Blair Kidwell, University of North Texas, USA

The authors examine whether brand rivalries can be formed and why they influence consumer’s perceptions of brands and subsequent intentions to purchase. We propose a new model that incorporates the emotional side of rivalries beyond current research to demonstrate that inequity and resentment are the drivers of these unique relationships.

A1 . Trusting and Acting on Chance Online

Shivaun Anderberg, University of Sydney, Australia Ellen Garbarino, University of Sydney, Australia

Gambling and games of chance have become more common in the online environment present in mobile games advergames and sales promotions. This study explores how the nature of the stimuli used to visually depict probabilistic components online affects people’s trust in them being truly random and subsequent engagement behavior.

K10. The Acronym Effect: Acronym and Buzzword Use Lowers Consumer Persuasion

Sumitra Auschaitrakul, University of the Thai Chamber of Commerce, Thailand Dan King, University of Texas Rio Grande Valley, USA Ashesh Mukherjee, McGill University, Canada

Many organizations use well-known acronyms such as CSR to represent full referents such as corporate social responsibility which itself is a shortened referent to represent performing actions that benefit the community. We show that shortening referents has the surprising consequence of lowering perceptions of genuine emotions underpinning social responsibility. Advances in Consumer Research (Volume 46) / 899 J14. You Reflect Me: Narcissistic Consumers Prefer Anthropomorphized Arrogant Brands

Norah Awad, Hongik University, South Korea Nara Youn, Hongik University, South Korea

Narcissism is trait commonly associated with arrogance and impacts people’s interactions with others; prior research has shown that narcissists love other narcissists like they love their own reflections. Our research examines narcissists’ attitudes toward anthropomorphized brands that have arrogant personalities and the role of self-congruence in this interaction.

E11. Influence of Ethical Beliefs and Trust on Purchase Decisions: The Moderating Effect of Involvement

Marija Banovic, Aarhus University, Denmark Athanasios Krystallis, Athens University of Economics and Business, Greece

We test the moderating effect of involvement on the relationship between ethical beliefs trust and purchase decisions. The higher the ethical beliefs and trust in product’s ability to deliver the value the higher the likelihood of product being-bought and vice-versa. Trust holds more relevance for high- than low-involved consumers.

Q11. The Effect of Message Ephemerality on Information Processing

Uri Barnea, University of Pennsylvania, USA Robert Meyer, University of Pennsylvania, USA Gideon Nave, University of Pennsylvania, USA

Ephemeral communication - the transmission of messages which can be accessed only a limited number of times - is ubiquitous in social interactions (e.g. verbal conversations Snapchat messaging). In three studies we find that message ephemerality increases attention improves memory recall and leads to longer observation time.

P13 . Self-Selected versus Fixed Price Bundling: The Effect of Bundle Type on Perceived Quality

Burcak Bas, Bocconi University, Italy Gulen Sarial Abi, Bocconi University, Italy

This research explores the effect of bundle type on perceived quality. A lab experiment shows that self-selected (versus fixed) price bundling increases perceived quality and psychological ownership is the mechanism underlying this effect.

Q5 . Conceptualizing the Digital Experience in Luxury

Wided Batat, American University Beirut, Lebanon

This research introduces a theory of digital luxury experience conceptualizing what it is and identifying its drivers. This is of importance to luxury consumption as it offers a conceptual framework to understand digital luxury an empirical phenomenon which has gained promi- nence in the last 10 years and has remained poorly understood.

F5 . Alternative Food Consumption (AFC) Adoption and Low SES Youth Food Well-Being: From Precontemplation to Maintenance

Wided Batat, American University Beirut, Lebanon

This research will exemplify the overall mission of marketing and public policy researchers in four ways. First we empirically show how allocentric and idiocentric factors of AFC adoption affect low SES young consumer well-being according to different stages of change. 900 / Working Papers J15. The Deliberation Effect on the Judgment and Choice of Anthropomorphized Products

Juliana M. Batista, EAESP Fundação Getúlio Vargas, Brazil Jose Mauro C. Hernandez, Centro Universitário FEI, Brazil

The study investigated the anthropomorphized products’ effect on purchase intention and choice and the moderating role of deliberation and rational personality. The first study demonstrates that personifying a product leads to higher purchase intention through product’s anthropomorphization. Studies two and three show that deliberating inhibits the anthropomorphization process.

N10. How Does It Make You Feel? Emotional Reasoning and Consumer Decisions

Andrea Rochelle Bennett, University of North Texas, USA Blair Kidwell, University of North Texas, USA Jonathan Hasford, University of Central Florida, USA David Hardesty, University of Kentucky, USA Molly Burchett, University of Kentucky, USA

The authors examine how consumers’ think about their emotions and the favorable outcomes that result. The conscious contemplation of emotions in decision making (emotional reasoning) was found to improve consumer choice quality beyond the effects of other types of emotional processing including trust-in-feelings and affect-as-information.

I4. Pink Tax: Are Some Marketing Practices Discriminatory?

Andrea Rochelle Bennett, University of North Texas, USA Audhesh Paswan, University of North Texas, USA Kate Goins, University of North Texas, USA

Are some marketing mix elements inherently unethical? For example via the “pink tax” products aimed at women are priced higher than similar products aimed at men. While segmentation targeting and differentiation are legal can they become unethical? This study explores this phenomenon using the pink tax as the research context.

L13. The Recipient Effect on Consumers’ Preference for Products Displayed in Different Horizontal Locations

Sheng Bi, Washington State University, USA Nik Nikolov, Washington State University, USA Julio Sevilla, University of Georgia, USA

This research proposes a recipient effect on products displayed in different horizontal locations. The congruence between recipient and product position increases information processing fluency leading to favorable product attitudes. This effect is mitigated by making consumers conceptualize time vertically or by changing purchasing scenarios from buying for others to selves.

F2. Can Stricter Ethical Standards Increase Tolerance for Ethical Misconduct?

Olya Bullard, University of Winnipeg, Canada Sara Penner, University of Manitoba, Canada Kelley Main, University of Manitoba, Canada

Can stricter ethical standards increase tolerance for ethical misconduct? The current research begins to answer this question. Relying on the anchoring and adjustment heuristic we demonstrate that stricter ethical standards can increase tolerance for ethical misconduct by lowering the category mean on which judgments of transgression severity are anchored. Advances in Consumer Research (Volume 46) / 901 L4. Attentional Breadth Moderates the Effect of Store Environments on Product Evaluation

Oliver B. Büttner, University of Duisburg-Essen, Germany Benjamin G. Serfas, University of Duisburg-Essen, Germany Daria Euler, University of Duisburg-Essen, Germany Mathias Clemens Streicher, University of Innsbruck, Austria

In an eye-tracking experiment we primed attentional breadth (broad vs. narrow) and assessed the influence of context valence (attractive vs. unattractive store environments) on product evaluation. Participants liked products more when they were presented in attractive (vs. unattractive) store environments. This difference was more pronounced with broad (vs. narrow) attention.

B6. A Study About the Moderator Effect of the Information Trust in the Relationships Between the Users’ Participation in Virtual Communities and the Benefits Obtained

Sara Campo, Autonomous University of Madrid, Spain Jano Jiménez, Autonomous University of Madrid, Spain Natalia Rubio, Universidad Autónoma of Madrid, Spain Nieves Villaseñor, Universidad Autónoma of Madrid, Spain Mªjesus Yague, Universidad Autónoma of Madrid, Spain

This study compares the role of information trust in virtual communities domain. The participants´ level of trust in the community contents reinforces their hedonic and social benefits.Those who trust to a lesser extent in the information shared feel more expert as a result of their participation in the virtual community.

N9. Effects of Awe on Consumers’ Preferences for Bounded Brand Logos

Fei Cao, Renmin University of China, China Xia Wang, Renmin University of China, China

Two studies probe and demonstrate that experiencing awe a positive emotion characterized by perceived vastness and need for accommodation compared with neutral emotion enhances consumers’ preferences for bounded brand logos and this effect is probably driven by loss of control. The effect is stronger for chronic promotion-focused (versus prevention-focused) consumers.

D3 . Social Exclusion and WOM about Past versus Future Experiences

Melis Ceylan, Koc University, Turkey Ezgi Akpinar, Koc University, Turkey Selin Atalay, Frankfurt School of Finance and Management, Germany

This research explores how social exclusion affects WOM. The authors demonstrate that socially excluded individuals compared to their non-excluded counterparts are more likely to talk about personal events that happened in the past to help others be prepared for a similar experience and consequently strengthen their social connections with them.

K4 . Movie Reviews and their Sentiments: Evidence of a Bandwagon Effect in Individualistic Cultures

Subimal Chatterjee, SUNY Binghamton, USA Ning Fu, SUNY Binghamton, USA Qi Wang, SUNY Binghamton, USA

We analyze how the Chinese and American audience rate a common set of 173 movies released in both countries and find a bandwagon effect in the American audience in that their ratings mimic the positive sentiments of the review they read last before posting their own. 902 / Working Papers H11. Not for Me: Identity Needs and Consumer Interest in Different Types of Co-creation

Lagnajita Chatterjee, University of Illinois at Chicago, USA David Gal, University of Illinois at Chicago, USA

This research investigates the influence of identity needs on choices between different product designing routes (e.g. self-customization crowdsourcing). Two experimental studies explore the influence of uniqueness and assimilation needs of individual consumers on selection of different product-designing formats. In future the role of moderating factors like expertise will be explored.

H6. The Influence of Anthropomorphized Service Toolkit on Consumer Satisfaction in Service Coproduction

Rocky Peng Chen, Hong Kong Baptist University, China Kimmy Chan, Hong Kong Baptist University, China

As technology evolves anthropomorphized entities become increasingly pervasive in the service context like service coproduction. This research finds that anthropomorphized service toolkit can have either positive or negative impacts on consumer satisfaction in service coproduction depending on whether the coproduction task is self-expressive or not.

N14. The Bright Side of Sadness: How Mood Affects Goal Initiation

Yunqing Chen, Chinese University of Hong Kong, China Leilei Gao, Chinese University of Hong Kong, China

How does mood affect people’s likelihood to initiate a long-term goal pursuit? Two studies show that compared to happy individuals sad individuals have stronger preference for self-discontinuity which leads to greater motivation to start pursuing a long-term goal.

P9 . Gift Budget Adherence and Price Discounts

Yuna Choe, Texas A&M University, USA Christina Kan, Texas A&M University, USA

This research suggests that people are more likely to spend extra money to adhere to their budgets when the budgeted purchase is sold at a discount if that budget is for a gift than if it is for the self.

J1. The Effect of Identity Abstractness on Information Processing Styles

Woojin Choi, University of Seoul, South Korea Min Jung Kim, Manhattan College, USA HyukJin Kwon, University of Seoul, South Korea Jiyun Kang, Texas State University, USA

This study suggests that how a social identity is construed (i.e. identity abstractness) would influence one’s information processing style. Specifically we predict that when a social identity with high (vs. low) identity abstractness is activated a person is likely to process information more abstractly (vs. concretely).

I7 . Male Spokespeople: Antecedents and Consequences of Social Comparison

Hsuan-Yi Chou, National Sun Yat-sen University, Taiwan Xing-Yu (Marcos) Chu, Nanjing University, China Chieh-Wen Cheng, National Sun Yat-sen University, Taiwan

This study demonstrated that highly attractive male spokespeople revealing muscles or gazing directly at viewers evoked upward comparisons and stronger envy emotions from male consumers. Stronger relative benign envy facilitates advertising effects. Female image and beauty type moderate the formation of envy. Results contribute to social comparison and envy theories. Advances in Consumer Research (Volume 46) / 903 O1. Choice, Rejection, and Context Effects

Shih-Chieh Chuang, National Chung Cheng University, Taiwan Yin-Hui Cheng, National Taichung University of Education, Taiwan

Most research on the context effect (compromise effect and attraction effect) has focused on choice-making tasks rather than rejection tasks. The data reported for two experiments show that the context effect is alleviated when decision-making tasks are performed in a selection scenario rather than a rejection scenario.

P4 . Consistent Price Endings Increase Consumers Perceptions of Cheapness

Shih-Chieh Chuang, National Chung Cheng University, Taiwan Yin-Hui Cheng, National Taichung University of Education, Taiwan

Should marketers make the last digit of a sale price consistent with the last digit of the original price? Across multiple studies including field studies and a variety of product categories this research shows price perception and purchase intention are related to consistency between the sale price and original price

Q9. Free or Fee? Consumers’ Decision to Pay for the Premium Version of a Music Streaming Service Rather than Using its Free Version

Sebastian Danckwerts, Heinrich-Heine-University, Germany Peter Kenning, Heinrich-Heine-University, Germany

This research examines factors that influence consumers’ likelihood of subscribing to the chargeable premium version of a music streaming service rather than using its free basic version. The results show that both the similarity between the two versions’ functionality and the perceived quality of personalized content significantly affect consumers’ decision.

A5 . Trusting the Tweeting President: Inside the Donald’s reality: Gaslighting, pschometrics and social media

Dianne Dean, University of Hull, UK Fiona Walkley, Hull University Business School, UK Robin Croft, Brunel University, UK

Research examines how Donald Trump’s unconventional campaign. Using cultural branding theory this research critically evaluates Donald Trump’s propagandist narrative. For success it needed to be believable erode the credibility of opposing candidates; create a reality that emphasizes the lost values of the American Dream and apportion blame to a significant ‘other’.

F3 . The Dark Side of Happy Brands: A Case Study of Newport Cigarette Advertising

Timothy Dewhirst, University of Guelph, Canada Wonkyong Beth Lee, Western University, Canada

This qualitative study uses the tobacco industry’s consumer research made public from litigation to examine how happiness appears particularly applicable to hedonic consumption such as smoking even though products and brands pertaining to hedonic consumption may not ultimately be beneficial or healthy to consumers. 904 / Working Papers D4. How the Product Positioning Affect the Influence of Online Consumer Reviews on Consumer Intentions

Helison Bertoli Alves Dias, Federal University of Paraná, Brazil José Carlos Korelo, Federal University of Paraná, Brazil Danielle Mantovani, Federal University of Paraná, Brazil Cecilia Souto Maior, Federal University of Paraná, Brazil Paulo Henrique Muller Prado, Federal University of Paraná, Brazil

Research in marketing indicates that different review types lead consumers to distinct intentions. However an important element of this relationship has not received enough attention: the reviewed product positioning. This paper shows that product positioning influences such relationship. Further this interaction between products and reviews is explained by reviews diagnosticity.

F7 . Mere Packaging and Consumer Choice

Tim Philipp Doering, University of Michigan, USA Katherine Burson, University of Michigan, USA Andrew D Gershoff, University of Texas at Austin, USA

Despite retailer and consumer efforts packaging still makes up the biggest proportion of municipal solid waste. Two studies show that though consumers prefer packaged goods they are willing to buy unpackaged goods significantly more often if they are in control of unpack- aging the product.

I6. How Does Runner’s World Shape a Runner’s World? Understanding Representations of the “Ideal” Female Body in Fitness Advertising

Carly Drake, University of Calgary, Canada Scott Radford, University of Calgary, Canada

The female body is salient at the intersection of sport the marketplace and media. In this critical review we interpret representations of the female body in fitness advertisements targeting female recreational endurance runners. We argue that they offer conflicting – and often unhealthy – directives on how to achieve the “ideal” body.

O7. Helpful Mental Shortcuts or a Shortcut to Bias? Two Perspectives on Heuristics and One New Direction for Consumer Research

Carly Drake, University of Calgary, Canada Mehdi Mourali, University of Calgary, Canada

Consumer research tends to follow a perspective on heuristics that argues heuristics bias and mislead those who use them. However a second perspective posits that heuristics may be more efficient and accurate than complex computations despite using less information. We explore what this perspective can offer future consumer research.

Q6 . Online Social Status Predicts Subjective Well-being: a Two Population Study

Rui Du, University of Hawaii, USA Miao Hu, University of Hawaii, USA

Through three studies across two populations we identified a novel construct: online social status. Online social status is positively related to consumers’ subjective well-being (Study 1 and Study 2). Additionally only when online and offline identities were highly overlapped online social status is related to subjective well-being (Study 3). Advances in Consumer Research (Volume 46) / 905 L14. Christmas Decorations in September – What Happened to Halloween? The Effect of Prospective Event Markers on Time Perceptions and Attitudes Towards Promotions

Chaumanix Dutton, University of Southern California, USA Kristin Diehl, University of Southern California, USA

We propose that reminders of future events such as holiday promotions trigger prospective event markers which affect time perceptions and attitudes towards the promotion. In three studies we show that event-related markers decrease time perceptions rendering early holiday promotions more acceptable while unrelated event markers have opposing effects.

G5 . The Phenomenon of Brand Noise and Related Consumer Preferences in the Luxury Industry

Daria Erkhova, University of Bern, Switzerland Elena Ehrensperger, University of Bern, Switzerland Harley Krohmer, University of Bern, Switzerland Wayne Hoyer, University of Texas at Austin, USA John Zhang, University of Pennsylvania, USA

This paper introduces the concept of brand noise as a new comprehensive approach to “loud” and “quiet” luxury branding. Further we empirically show that consumer preferences toward loud and quiet goods are not always driven by social status but also by personality traits such as extraversion and openness to experience.

M13 . Keep Consistency in Good Old Days: The Effect of Nostalgia on Consumers’ Consistency Seeking Behavior

Yafeng Fan, Tsinghua University, China Jing Jiang, Renmin University of China

The present research explored how nostalgia affects consumer’s consistency seeking behavior. Three studies demonstrated that nostalgic feelings increase consumers’ consistency seeking behavior in consumption. The results showed that nostalgia (vs. control) condition participants higher self-continuity preference which increases their consistency seeking behavior in subsequent consumption situation.

M9 . Exploring Historical Nostalgia and its Relevance to Consumer Research

Matthew Farmer, University of Arizona, USA Caleb Warren, University of Arizona, USA

Historical nostalgia a sentimental longing for a time outside of one’s own personal past appears to draw a comparison between present-day values and past values making it a potentially distinct construct from personal nostalgia. We report a pilot study that supports this claim and develop a program of promising research.

I5 . Take That Mirror Away From me! Clothing Consumption by the Elderly and the Self-identity of the Young

Daniela Ferreira, Federal University of Rio de Janeiro, Brazil

This study focuses on the impact of elderly’s clothing consumption on the self-identity of the young in the Brazilian context. Using netnography and interviews and building from Terror Management Theory we found that moral condemnation of the elderly represents youngers’ fear to challenge their self-identity and privileged status in society. 906 / Working Papers N12. Untangling Different Envy Episodes and their Effects on Brand Attitude

Kirla C Ferreira, EAESP-FGV, Brazil & City University of London, UK Delane Botelho, EAESP-FGV, Brazil Suzana Valente Battistella-Lima, EAESP-FGV, Brazil

An envy episode develops in various directions depending on individual and contextual factors and its effect on brand attitude varies accordingly. Based on three studies we suggest that brand attitude decreases linearly across envy episodes (from no envy benign envy malicious envy by luck to malicious envy by deceit).

H3. Does the Style Looks More Expensive? The Effect of Visual Complexity on Luxury Perception of Art Infused Products

Cheng Gao, Nanjing University, China Chunqu Xiao, Nanjing University, China Kaiyuan Xi, Nanjing University, China Hong Zhu, Nanjing University, China

Experiments results demonstrate that consumers perceive products infused with complex visual arts are more luxury. This is because complex arts tend to capture consumers’ attention and broaden their pupil size which in return is explained by arousal of that stimuli. Stimuli’s artistic level is found to moderate the mechanism.

B9. The Power of Self-Effacing Brand Messages: Building Trust and Increasing Brand Attitudes

Tessa Garcia-Collart, Florida International University, USA Jessica Rixom, University of Nevada, Reno, USA

Self-enhancing messages convey information in a boastful and arrogant manner while self-effacing messages do so in a more modest and humble way. We propose that compared to self-enhancing self-effacing marketing appeals will increase consumers’ brand attitudes by lowering consumer skepticism and increasing brand trust.

M2. Of Expectations and Experiences: The Moderating Effect of Valanced Expectations on Enjoyment of a Positive versus Negative Experience

Brian Gillespie, University of New Mexico, USA Molly McGehee, University of New Mexico, USA

Some extant literature argues that expectations drive enjoyment of experiences while other literature argues that expectations only comparatively influence enjoyment. The current work addresses this discrepancy and demonstrates that while positive expectations do result in favorable enjoyment of experiences negative expectations result in the comparative process suggested by others.

A7 . Credible Critters: Source and Message Expectancy Violation and Influence on Perceived Trustworthiness and Credibility

Justin Graeber, University of Texas at Austin, USA

Trustworthiness and credibility are essential traits of effective leaders journalists and spokespersons. This paper explores whether pairing formal and informal messages with different sources can yield a positive expectancy violation that builds trust reinforces credibility and challenges previous biases against traditionally formal sources. Advances in Consumer Research (Volume 46) / 907 H7 . Too Risky to Be Luxurious: Stigmatized Luxury Product Attributes Can Weaken or Increase Social Risk to Determine Conspicuous Consumption

Jerry Lewis Grimes, Grenoble Ecole de Management, France Yan Meng, Grenoble Ecole de Management, France

There are circumstances when conspicuous luxury consumption becomes socially stigmatized and this overt display of luxury is actually harmful to the consumer. Two studies show that stigmatized product attributes (i.e. antisocial or distasteful) generally reduce purchase intentions however low social visibility can undermine social stigma increasing luxury product purchase intentions.

L2 . Wish List Thinking: The Role of Psychological Ownership in Consumer Likelihood to Purchase or Remove a Product from an Online Wish List

Christopher Groening, Kent State University, USA Jennifer Wiggins, Kent State University, USA Iman Raoofpanah, Kent State University, USA

Psychological ownership of products on a consumer’s wish list increases the likelihood of purchase and decreases the likelihood of removal from the wish list. This effect becomes curvilinear when the product has been on the wish list a long time or the wish list is publicly viewable.

C4 . The role of Attachment to a Human Brand in Improving Eating Habits

Amélie Guèvremont, Université du Québec à Montréal, Canada

This research studies behavioural change by demonstrating the power of a human brand in influencing eating habits (through a culinary blog). A netnography and eleven interviews identify four categories of behavioural changes and substantiate the role of attachment. Two attachment antecedents are identified: brand-self connection and satisfaction of individual needs.

L7. The Joy of Shopping: Reconciling Mixed Effects of Positive Emotions on Shopping Behavior

Kelley Gullo, Duke University, USA Duncan Simester, Massachusetts Institute of Technology, USA Gavan Fitzsimons, Duke University, USA

The current work develops a framework to identify when positive emotions lead to increased shopping behavior. A field experiment with over a million transactions and a controlled lab study provide initial evidence that the environment in which a positive emotion is felt shapes the behavioral outcomes of that emotion.

E6. The Effect of Crowding Perception on Helping Behavior —Is Squeeze Warmer than Isolation?

Qingqing Guo, Shanghai Jiao Tong University, China

Based on 7 studies (991 participants) we investigated crowding perception on helping behavior. Consumers in crowds have a stronger willingness to help others as they tend to experience a self-other overlap and stronger social connection as well as empathy in serial in crowds. The composition of crowds moderated the effect. 908 / Working Papers N1 . The Experiential Advantage in Eudaimonic Well-being – An Experimental Assessment

Aditya Gupta, University of Nebraska-Lincoln, USA James Gentry, University of Nebraska-Lincoln, USA

Despite the documented hedonic superiority of experiential purchases to material purchases a focus on their relative well-being implications remains unexplored. Given that the very idea of well-being is changing the authors report results from two experimental studies that evaluate the experiential advantage in terms of a new framework – eudaimonic well-being.

N2 . The Devil Wears FAKE Prada: Dual Envy Theory Explains Why Consumers Intend to Purchase Non-Deceptive Luxury Counterfeits

Tanvi Gupta, Indian Institute of Management Bangalore, India Preeti Krishnan Lyndem, Indian Institute of Management Bangalore, India

Applying the dual envy theory to understand why consumers deliberately intend to purchase non-deceptive luxury counterfeits we posit that benign enviers intend to purchase such counterfeits when they cannot afford the original whereas malicious enviers are driven to purchase such counterfeits by a desire to punish the unfair original brand.

J8 . Exchange with The Rich, Concern with The Poor: The Effects of Social Class on Consumer Response to Brand Relationship

Bing Han, Shanghai Jiao Tong University, China Liangyan Wang, Shanghai Jiao Tong University, China

This article proposes that consumer-brand relationship position (exchange vs. communal) should be built in accordance with the social class of consumers. Through three experiments in manufacturing and service industries the authors find that a matching effect between brand relationship and social class can enhance brand evaluation and brand attitude.

R13 . Brand Humanization: Applying Two Dimensions of Humanness to Brand

Mycah L Harrold, Washington State University, USA Andrew Perkins, Washington State University, USA

We create a measure of brand humanness that unlike past measures of anthropomorphism is easily applied to a variety of study para- digms. We situate brand humanness along two independent dimensions: characteristics descriptive of basic human nature and characteristics that identify ways in which humans are unique from other species.

H1. How Anthropomorphized Roles Influence Consumers’ Attitude Towards Innovative Products

Yuanqiong He, Huazhong University of Science and Technology, China Zhou Qi, Huazhong University of Science and Technology, China

This research explores effects of anthropomorphism strategy for innovative products. The results indicated that friend role is more suitable for ergonomic innovative product while servant role is more suitable for feature innovative product. Meanwhile perceived product innovativeness by consumers will moderate this effects. This study enriches the field of anthropomorphism. Advances in Consumer Research (Volume 46) / 909 K12 . Use Language to Change People’s Mind: The Persuasive Power of Online Marketing Communications

Xun He, Katholieke University Leuven, Belgium Barbara Briers, Vlerick Business School, Belgium Luk Warlop, Norwegian School of Management, Norway

Marketing and economies depend mainly on persuasive success. This study investigates which and how linguistic features influence a target’s actual persuasiveness of messages in the setting of online marketing communications.

N6 . Not Myself: The Impact of Secret-Keeping on Consumer Choice Regret

Dongjin He, Hong Kong Polytechic University, China Yuwei Jiang, Hong Kong Polytechic University, China

This research investigates how the experience of secret-keeping affects consumers’ perception of their previous choice. Three studies reveal that the experience of secret-keeping increases consumers’ regret for the consumption choices they made previously and show that this effect is driven by the perceived low intrinsic motivation for the chosen option.

C3 . Using Goal Theory to Promote Habit Formation During and After a Bike-to-Work Campaign

Bettina Rebekka Höchli, University of Bern, Switzerland Claude Messner, University of Bern, Switzerland Adrian Brügger, University of Bern, Switzerland

Health interventions are often ineffective because people discontinue the targeted behavior after the intervention period. Forming habits could maintain the behavior beyond the intervention. We show that combining a superordinate goal with a subordinate goal can produce stronger habits than focusing on a subordinate goal alone.

Q13 . Liquid Consumption From Another Perspective: The Case of “Investomers”

Carina Hoffmann, Heinrich-Heine-University, Germany Lasse Meißner, Heinrich-Heine-University, Germany Peter Kenning, Heinrich-Heine-University, Germany

To extend recent theorizing on liquid consumption and to add a new perspective to this field we investigate the case of investors turning into consumers or reverse (“investomer”). We assume commitment and trust as “coupling” factors between these two categories and find support for this assumption.

O12. When do People Waste Time? Testing a Mechanism for Parkinson’s Law.

Holly S Howe, Duke University, USA Tanya Chartrand, Duke University, USA

Parkinson’s law states that work expands to fill the time available for its completion. In a series of studies we demonstrate this law and identify a possible mechanism. Specifically we demonstrate that people who allot too much time to complete a task fail to use time-saving strategies.

Q8 . Avatars, Consumers and Possession in Online Gaming

Feihong Hu, Lancaster University, UK Xin Zhao, Lancaster University, UK Chihling Liu, Lancaster University, UK

This study seeks to re-examine the relationship between consumers and their possessions. We observed online gamer interactions and conducted depth-interviews with Chinese online gamers. Initial findings have generated rich insights into how consumers experience and manage the relationships between their sense of self avatars and virtual possessions. 910 / Working Papers O10. Individual Differences in Consumers’ Need For Cognition and Affect: A Neuromarketing Study Using Voxel-Based Morphometry

Jianping Huang, Tsinghua University, China Yang Sun, Tsinghua University, China Jie Sui, University of Bath, UK Xiaoang Wan, Tsinghua University, China

We revealed the anatomical variations in consumers’ need for cognition (NFC) and need for affect (NFA) in a neuromarketing study. The right lingual gyrus was associated with NFC whereas the right inferior occipital gyrus was linked to NFA.

L8 . Recover the Unrecoverable: How Co-Recovery Shifts Consumers ‘Attribution Following a Failed Recovery

Bo Huang, HEC Montreal, Canada Yany Grégoire, HEC Montreal, Canada Matthew Philp, HEC Montreal, Canada

Exploring the impact of co-recovery on consumers ‘attribution the present research finds that when consumers are highly participative in a failed recovery episode (i.e. co-recovery) they blame themselves more than the firm. However this effect is reversed when they participate less and disappeared when they do not participate at all.

M7. The Mixed Effects of Nostalgia on Consumer Switching Behavior

Zhongqiang (Tak) Huang, University of Hong Kong, China Xun (Irene) Huang, Nanyang Technological University, Singapore Yuwei Jiang, Hong Kong Polytechic University, China

This research shows that contrary to conventional wisdom nostalgia can increase consumer switching behavior. However this effect only occurs when consumers focus on the consumption outcome and it is reversed when consumers are led to focus on the process of consumption. These findings provide a nuanced understanding of nostalgia marketing.

D5. Bragging about Effort? Personal Effort Decreases Word-of-Mouth

Jiexian (Chloe) Huang, Hong Kong Polytechic University, China Yuwei Jiang, Hong Kong Polytechic University, China

Three experiments revealed consumers were less likely to share a positive achievement to others through word-of-mouth if they spent personal effort in pursuing it than if no effort was involved in the pursuit. We further found the effect was dismissed when consumers’ interpersonal warmth was assured by other contextual cues.

O9. The Role of Numerical Identification in Customer Reaction toward Service Failure

Jiexian (Chloe) Huang, Hong Kong Polytechic University, China Yuwei Jiang, Hong Kong Polytechic University, China

This research investigates the consequence of assigning numbers to customers (e.g. room numbers) in a service context. Two experiments show that after being identified by a number customers would be more tolerant of service failure. This effect is found to be driven by the self-dehumanization process triggered by numerical identification.

A3 . Why People Still Do Not Trust Algorithmic Advice in Decision Making

Jaewon Hwang, Sejong University, South Korea Dong Il Lee, Sejong University, South Korea

The current research examines why people don’t trust on algorithmic advice in decision making and verifies the underlying mechanism of this effect. The present study shows that people trust advice from algorithms less than advice from people and low trust in the algorithm is caused by the low attribution externality. Advances in Consumer Research (Volume 46) / 911 C2 . The Bad Taste of Healthy Food Discounts

Iina Ikonen, Vrije Universiteit Amsterdam, The Netherlands Aylin Aydinli, Vrije Universiteit Amsterdam, The Netherlands Peeter Verlegh, Vrije Universiteit Amsterdam, The Netherlands

Price promotions have often been suggested as a way to increase healthy consumption but research suggests they may be less effective in healthy food categories. Through four experiments we offer a potential explanation: they lower the pre-purchase tastiness perceptions of healthy foods by increasing reliance on the unhealthy=tasty intuition.

K8. Framing Matters. How Comparisons to Ideal and Anti-Ideal Reference Points Affect Brand Evaluations

Magdalena Zyta Jablonska, SWPS University of Social Sciences and Humanities, Poland Andrzej Falkowski, SWPS University of Social Sciences and Humanities, Poland

We test how comparisons to ideal and anti-ideal reference points affect brand evaluations in political and consumer context. The results support the negativity effect showing that negative comparisons have a stronger effect than their positive counterparts.

R14. Are Lonely Consumers Loyal Consumers? Loneliness Breadth and Depth

Eunyoung Jang, Oklahoma State University, USA Zachary Arens, Oklahoma State University, USA

People feel loneliness either due to inadequate closeness in their relationships or inadequate number of relationships. Across two studies we show that lonely people who lack deep relationship exhibit higher brand loyalty whereas lonely people who lack broad relationships choose a larger variety of brands.

G10. The Effects of Self-Construal on Evaluations of Brand Logo Colors

Eunmi Jeon, Sungkyunkwan University, South Korea Myungwoo Nam, Georgia Tech, USA

The present research investigates the effectiveness of different types of color combinations on brand evaluations. In this research we propose that consumers’ self-view influences the effectiveness of analogous versus complementary color combinations. We suggest that consumers’ self-views would influence effectiveness of analogous versus complementary color combinations used in brand logos.

G3. Warm or Cold? The Effect of Color Temperature of Logo on Evaluation of For-Profits and Nonprofits

Eunmi Jeon, Sungkyunkwan University, South Korea Myungwoo Nam, Sungkyunkwan University, South Korea

The current research examines the fit effect between color temperature and organization type and how it influences product evaluations. We suggest that product evaluations of for-profits were higher when the logo color was in cooler colors compared to nonprofits.

J12. The Influence of Pet-Ownership on Consumer Behavior

Lei Jia, Ohio State University, USA Xiaojing Yang, University of Wisconsin - Milwaukee, USA Yuwei Jiang, Hong Kong Polytechic University, China

Across four studies we show that pet ownership influence consumers’ regulatory focus such that dog (vs. cat) owners are more promotion- (vs. prevention-) oriented. Their different regulatory orientations can carry over to influence their financial decisions attitudes toward ads and purchase intentions. 912 / Working Papers Q7 . Desire in Performed Consumption: Examining the Case of Korean Beauty Vlogging

Marie-Eve Jodoin, HEC Montreal, Canada Marie-Agnès Parmentier, HEC Montreal, Canada

This research project examines the processes of territorializing and sustaining desire on YouTube in the context of Korean beauty vlogging. It finds that in consuming the performed consumption of vloggers material desire social desire and desire for otherness are territorialized on the network and are sustained outside of bodily-reality.

E9 . “Power Distance, Social Aspiration, and Fair Trade Products” – the Interaction Effect of Power Distance Belief and Status Motivation on Fair Trade Product Consumption

Sunghee Jun, Seoul National University, South Korea Libby Youngjin Chun, Seoul National University, South Korea Kiwan Park, Seoul National University, South Korea

Research on power distance belief shows that people with high (vs. low) PDB engage in prosocial behaviors more. Marrying these research results we conducted one preliminary field study and lab experiments to demonstrate that when PDB is high high status motive individuals are more likely to purchase fair-trade products.

M12. From the Occult to Mainstream – Tracing Commodification of the Spiritual in the Context of Alternative Spiritualities

Richard Kedzior, Bucknell University, USA

In this paper we focus on examining various institutional logics that facilitated the evolution of a sacred object into a marketplace phenomenon.

G12. The Effect of (Non-)appetizing Ambient Scents on Consumers’ Affinity Toward Vices and Virtues in the Retail Environment

Corinne M Kelley, Florida State University, USA Anders Gustafsson, Karlstad University, Sweden Poja Shams, Karlstad University, Sweden Martin Mende, Florida State University, USA Maura Scott, Florida State University, USA

Appetizing (non-appetizing) ambient scents decrease consumers’ personal control and increase their affinity toward vice (virtuous) offerings. Appetizing scents increase affinity toward vice offerings via increased visual attention enhancing recall of vice products desire for vice products and store attitude. Effects are tested in three field studies at three different retailers.

F6. Can CSR Save a Firm From a Crisis? A Role of Gratitude in the Buffering Effect of CSR on Consumer Vindictive Behavior.

Junghyun Kim, NEOMA Business School, France Taehoon Park, University of South Carolina, USA Myungsuh Lim, Sangji University, South Korea

This research examines the buffering effect of corporate social responsibility (CSR) on consumers vindictive behaviors during a product-harm crisis. Three studies demonstrate how CSR plays a critical role in reducing consumers intention to engage in vindictive behaviors (e.g. negative word-of-mouth) and propose a mediating role of consumer gratitude. Advances in Consumer Research (Volume 46) / 913 R8 . Brand Perceptions and Consumer Support in the Face of a Transgression: Warmth Over Competence

Summer Hyoyeon Kim, University of Kansas, USA Jessica Li, University of Kansas, USA Jenny Olson, Indiana University, USA Shailendra Pratap Jain, University of Washington, USA

We examined the role of brand perception in consumer judgment and support following a brand transgression. Consumers blame warm brands less for the same product failure. They are also more likely to support warm brands following a moral transgression. We explain this link though the process of moral decoupling.

B3. The Effect of Temporal Distance on Online Reviews’ Recommendation Power: The Role of Spontaneous Retrieval and Perceived Trust

Kyu Ree Kim, Seoul National University, South Korea Wujin Chu, Seoul National University, South Korea

We challenge lay belief that reviews on recent experience are more powerful. We examined that reviews written on the experiences of distant past have stronger perceived recommendation power. We also examined the mediation effect of perceived trust and the effect of retrieval process (spontaneous vs. deliberate) on perceived trust.

P7 . Consumer Evaluations of Sale Prices: The Role of the Spatial Representation of Time

Yaeeun Kim, Temple University, USA Joydeep Srivastava, Temple University, USA

Based on the ease of the subtraction task we tested whether value perception is influenced by the interaction between sale price displays and temporal information. The findings revealed that participants perceived the price offer as more valuable with time information. The studies contribute to the pricing and fluency literature streams.

P5 . Can(Can’t) Control, thus Try to Save (Earn): The Joint Effect of Perceived Control and Financial Deprivation on Financial Decisions

Min Jung Kim, Manhattan College, USA

This research suggests that one’s preference for earning vs. saving will depend on their personal control and financial deprivation. We predict that individuals who perceive low personal control are more likely to prefer earning to saving but such a tendency can be reversed when their financial deprivation is considered.

I12. The Effect of Susceptibility-Induced Threat in the Preventative Communication

Moon-Yong Kim, Hankuk University of Foreign Studies, South Korea

This research investigates gender differences in responding to the preventative communication by varying the level of susceptibility-induced threat. Specifically it is predicted that higher level of susceptibility is more likely to have a negative effect on persuasion for females (vs. males) due to their maladaptive responses to the threatening message.

G11. Sensory Placebo Effects: The Role of Sensory Signaling in Enhancing Marketing Placebos and Consumer Outcomes

Dan King, University of Texas Rio Grande Valley, USA Sumitra Auschaitrakul, University of the Thai Chamber of Commerce, Thailand

Pairing a placebo expectation with a sensory stimulus (“sensory placebo”) elicits significantly stronger placebo effects in both cognitive and physical domains. Participants given a sensory placebo solved more word puzzles and were better at lifting weights and the outperfor- mance over standard placebos was mediated by phantom sensations of placebo energization. 914 / Working Papers L11 . Consumer Search Mode Produces Unintended Marketing Consequences

Dan King, University of Texas Rio Grande Valley, USA Sumitra Auschaitrakul, University of the Thai Chamber of Commerce, Thailand

Recently marketers have migrated consumers from written to voice search assuming that this produces positive outcomes. We show that this elicits unintended negative outcomes such as delays in willingness to purchase products and lowered spending. We show that differences in mindsets when people use their voice account for this effect.

D10 . It’s Meant for Me: When Serendipity Increases Word-of-Mouth

Colleen Patricia Kirk, New York Institute of Technology, USA Joann Peck, University of Wisconsin - Madison, USA Claire Hart, University of South Hampton, UK Constantine Sedikides, University of South Hampton, UK

High narcissistic consumers are more likely than low to spread word-of-mouth about serendipitous events. We explain this effect in terms of two distinct paths motivated by narcissistic admiration and narcissistic rivalry. We further offer empirical evidence for the role of psychological ownership motivation in driving word-of-mouth.

J3 . You Think I’m Yours but, Trust Me, I’m Not: How Consumers Value Dogs and Cats

Colleen Patricia Kirk, New York Institute of Technology, USA Samantha Renee Kirk, Boston College, USA

Two experiments reveal that dog owners are willing to pay more for life-saving surgery and related expenses than cat owners. This effect is sequentially mediated by perceived behavioral control psychological ownership and emotional attachment. This first examination of psychological ownership of living creatures offers implications for pet owners and marketers.

J13 . The Good and Bad of a Delayed Indulgence: Effects on Self-Perceptions and Purchase Satisfaction

Argiro Kliamenakis, Concordia University, Canada Kamila Sobol, Concordia University, Canada

We show that the mere act of delaying an indulgent purchase can evoke the same level of perceived self-control as not indulging at all –a level that is higher than in instances of immediate indulgence. Paradoxically the bolstered perceptions of self-control resulting from a delayed indulgence subsequently reduce purchase satisfaction.

K5. Advertising Organizational Culture as a Selling Tactic for For-Profit Organizations

Dejun Tony Kong, University of Houston, USA Maria Ng, University of Houston, USA

By conducting four experiments we found that advertising organizational culture can be an effective selling tactic for for-profit organizations particularly in a stereotypically fun industry. Specifically advertising fun (vs. neutral) organizational culture appeared to enhance consumers’ perception of organizational warmth which elevated their positive anticipated emotions and subsequent purchase intention.

O11 . Have Less, Compromise Less: How the Perception of Resource Scarcity Influences Compromise Decisions

Kate Kooi, University of Miami, USA Caglar Irmak, University of Miami, USA

The compromise effect has been shown to be robust however the current research seeks to illuminate a circumstance under which it is less predictive by demonstrating that under the perception of resource scarcity consumers are less susceptible to the compromise effect because of their proclivity for status-seeking and risky choices. Advances in Consumer Research (Volume 46) / 915 R1. How Consumers Deal With Brand Failure-An Individual Differences Approach

Melika Kordrostami, California State University-San Bernardino, USA Elika Kordrostami, Rowan University, USA

The impact of individual differences (regulatory focus and attachment style) on consumers’ reactions at the time of brand failure is studied. Our three studies showed that higher promotion leads to higher positive and lower negative behaviors. Furthermore this effect is moderated by individuals’ attachment styles.

E8. Perceptions of Out-Group Members: The Effects of Language Abstraction

Afra Koulaei, University of South-Eastern Norway, Norway Daniela Cristian, City University of London, UK

One way to reduce the discrimination towards the members of stigmatized groups is to describe their situations or behaviors in a more concrete (vs. abstract) form. Concrete descriptions lead people to attribute the homelessness to the situational cause (as opposed to individual) which in turn leads to greater helping behavior.

J11. The Myth of Return – Success or Failure? Consumer Identity and Belonging in the Case of Repatriate Migrants

Sonja N. Kralj, University of Augsburg, Germany Michael Paul, University of Augsburg, Germany

This poster introduces a new migration pattern into research on consumer migration and acculturation. Investigating structural individual and relational factors we stress the dynamic nature of how consumption identity and belonging are related in the case of consumers having returned from the diaspora to the ancestral homeland after generations.

K11. Effects of Emotional vs. Rational Thinking on Consumer Responses to Verbal Precision

Ann Kronrod, University of Massachusetts, USA Guang-Xin Xie, University of Massachusetts Boston, USA Shai Danziger, Tel Aviv University, Israel

We define verbal precision and examine consumer responses to more and less verbally precise communication. Results suggest that emotionally (vs. rationally) oriented recipients are more sensitive to verbally precise messages. For these consumers higher verbal precision increases perceived conversational cooperativeness of the source and results in more positive product attitudes.

Q4 . The Notion of Self-Optimization in Context of Self-Tracking and Beyond

Agnieszka Krzeminska, Leuphana University Lüneburg, Germany Dorthe Brogård Kristensen, University of Southern Denmark, Denmark

The aim of this paper is to analyse the metaphors and ideologies of optimization with a special consideration of the notion of self-optimization which is being prevalently used in descriptions of goals of people using self tracking devices or apps (e.g. Lupton 2014 Ruckenstein 2014 Selke 2016).

I11 . Self-Presentation in the Mating Market: The Influence of Gender and Sexual Orientation on Profiles on Tinder and Grindr

Chaim Kuhnreich, Concordia University, Canada Lilian Carvalho, FGV/EAESP, Brazil Gad Saad, Concordia University, Canada

Examining how people signal potential mates on Tinder and Grindr males are more likely to use conspicuous consumption in their profiles when compared to women. However in the homosexual community males are actually more likely to display sexually suggestive pictures and signs of physical fitness as opposed to conspicuous consumption. 916 / Working Papers N4 . Induction of Construal-Level Mindset via Surprise and the Follow-up Effect on Consumer Evaluations and Judgments

Atul A Kulkarni, University of Missouri, USA Joëlle Vanhamme, EDHEC Business School, France

We find that an experience of surprise may influence response to subsequently presented stimuli in the same fashion as abstract/concrete processing would do. Three studies show that when primed with positive (negative) surprise participants evaluated promotionally (preven- tionally) framed ad messages more favorably and leaned toward desirability (feasibility) laden shopping options.

F4 . Social Support First, Money Later: Perceived Economic Mobility Increases Happiness When Perceived Social Support Opens the Door

Yong Ju Kwon, Seoul National University, South Korea Sara Kim, University of Hong Kong, China Youjae Yi, Seoul National University, South Korea

Our research shows that perceived economic mobility (PEM) has a positive effect on happiness. More importantly perceived social support moderates the effect of PEM on happiness: people with high PEM are happier than those with low PEM only when they meet a certain level of perceived social support.

B4 . Prestige is the Truth: Luxury Branding Drives for Fairness Perception of Non-physical Attribute Based Dynamic Pricing

Edward Yuhang Lai, Virginia Tech, USA Cindy Yoonjoung Heo, Ecole hôtelière de Lausanne, Switzerland

Dynamic pricing is widely adopted in experience products. However consumers may not perceive different pricing policies as equally fair. An experiment finds evidence that consumers feel ambiguity in judging differential pricing policy based on non-physical (vs. physical) attribute and perceive such policy as fairer for luxury brands (vs. economy brands).

M8 . Nostalgia Increases Healthy Attitudes and Behaviors

Jannine Lasaleta, Yeshiva University, USA Carolina O. C. Werle, Grenoble Ecole de Management, France Amanda Pruski Yamim, Grenoble Ecole de Management, France

This research tests nostalgia as a motivation force in healthful food consumption. Across three studies we find support for the hypothesis that nostalgia increases positive attitudes and behaviors toward healthy food but decreases them toward unhealthy food. Moderated mediation evidence demonstrated this effect is driven by nostalgia-evoked social support.

C9 . Filling the Expectations: How Packaging Sustainability Influences Consumers’ Inference of Product Attributes

Olga Lavrusheva, Aalto University, Finland Alexei Gloukhovtsev, Aalto University, Finland Kristina Wittkowski, Aalto University, Finland Tomas Falk, Aalto University, Finland Pekka Mattila, Aalto University, Finland

Our study aims to investigate the effect of sustainable packaging materials on the perceptions of taste quality and health-related benefits of food products and whether this effect would then facilitate consumers’ purchase intention. Advances in Consumer Research (Volume 46) / 917 C10. Beyond Self-control: A Field Exploration of the Interactive Effect between Cue-induced and Prospective Decision Making on Long-term Weight Loss

Wanyu Li, McGill University, Canada Laurette Dube, McGill University, Canada Yu Ma, McGill University, Canada

This research goes beyond the episode-level self-control and examines how the cue-driven tendency interacts with the endophenotype underlying prospective judgment in shaping the longer-term dynamics of weight management.

N13. Smaller Self but Larger Tips? The Effect of Awe on Consumers’ Tipping Intention

Ran Li, Chinese University of Hong Kong, China

Two experiments showed that feeling awe can systematically affect consumers’ tipping intention. Specifically this awe-tipping relationship is driven by two countervailing processes: feeling awe makes consumers feel closer to the server (which increases the tendency of tipping) but simultaneously decrease their power distance belief (which decreases the tendency of tipping).

H2. Influencing Consumer Response to Products with High Styling: The Role of Mindsets

Ying-Ching Lin, National Chengchi Uniersity, Taiwan Angela Chang, Northeastern University, USA

This research investigates the conditions under which consumer response to products with a high level of styling will more likely be positive. Empirical studies demonstrate that consumers evaluate a high-styling (vs. low-styling) product more favorably when they are in an abstract rather than a concrete mindset.

L10. How Physical Distance and Power Distance Belief Affect Salesperson Evaluations and Purchase Intentions

Chia-Wei Joy Lin, University of Tennessee at Chattanooga, USA Saerom Lee, University of Texas at San Antonio, USA Bingxuan Guo, University of Texas at San Antonio, USA

Consumers with lower (vs. higher) PDB report lower evaluations of the salesperson who stays physically close to (vs. far from) them and subsequently have lower purchase intentions in the shopping context. This is because lower PDB consumers are more likely to infer disrespect from the physical closeness of the salesperson.

K9 . Measuring Internet Slang Style in Advertisement: Scale Development and Validation

Shixiong Liu, Shenzhen University, China Yi Wu, Tsinghua University, China Wu Gong, Shenzhen University, China

Drawing on linguistics and stylistics theories this paper introduces and conceptualizes the construct of Internet slang style (ISS) to describe characteristics of Internet slang. A dimensional ISS framework consisting of four dimensions (amiability overtness candor and harshness) is proposed. The corresponding scale was developed and validated across five studies. 918 / Working Papers A6 . “Alexa, let’s make a trade”: Search Behavior, Trust, and Privacy with Voice-Activated Assistants

Weizi Liu, University of Illinois at Urbana-Champaign, USA David William Ross, University of Illinois at Urbana-Champaign, USA Kieshana M. Williams-Beeler, University of Illinois at Urbana-Champaign, USA Yoonah Lee, University of Illinois at Urbana-Champaign, USA Michelle Renee Nelson, University of Illinois at Urbana-Champaign, USA

Through 16 interviews with owners of voice-activated assistants (VAA) we explore user trust and search practices as they occur in voice interactions with their device. We uncover preferences for visual elements and less cognitive search tasks when seeking information as well as cautious trust in the device’s maintenance of privacy.

J7 . Alienation from Ourselves, Alienation from Our Products: A Carry-over Effect of Self-alienation on Self-possession Connection

(Joyce) Jingshi Liu, Hong Kong University of Science and Technology, China Amy Dalton, Hong Kong University of Science and Technology, China

When consumers feel alienated from the self they also feel alienated from their products which is an important part of the self. Moreover this effect only applies to products that are relevant to the self and is attenuated among consumers with an interdependent (vs. independent) mindset.

E3. Having Power, Giving More? The Effect of Psychological Power on Consumers’ Charitable Giving of Time

Wumei Liu, Lanzhou University, China

Will high power increase consumers’ charitable giving of time? Existing research has not addressed this research question yet. We propose that high (vs. low) power will increase consumers’ illusory control over time and consequently their readiness to donate when encountering volunteering appears. The boundary condition is also examined.

J4 . A Large Pack of Toilet Paper is Bad for Me: Self-control and Consumers’ Responses to Product Quantity

(Joyce) Jingshi Liu, Hong Kong University of Science and Technology, China Keith Wilcox, Columbia University, USA Amy Dalton, Hong Kong University of Science and Technology, China

We examine the association between product quantity and self-control using utilitarian products a product domain that does not generally threaten self-control. We find a bidirectional effect between product quantity and self-control whereby exposures to large product quantity decrease self-control and activation of self-control decreases evaluation of product quantity.

G13 . Odor Priming and Product Preferences: When Smells Regulate Preferences for Semantically-Congruent Products and Brands

Ramona De Luca, EAESP-FGV, Brazil Delane Botelho, EAESP-FGV, Brazil

This article extends the idea that olfactory stimuli influence consumer preferences even when they are perceived unconsciously. Two laboratory experiments demonstrate that incidental exposure to odors unconsciously activate a mental representation when they are matched for valence and when are related to the intended semantic category thus regulating product preferences. Advances in Consumer Research (Volume 46) / 919 D1 . When Intention to Share on Social Media Increases Variety-Seeking: The Role of Self-Enhancement

Jingjing Ma, Peking University, China David Dubois, INSEAD, France Fei Jin, Peking University, China

This research investigates how calls to sharing online affect decisions. It shows that consumers would seek variety to self-enhance when they intent to share with others. This effect is stronger when the consumption is non-conspicuous and when consumers are non-experts. Six studies using lab and field data support these effects.

I10. Can Body Size Influence the Judgments of Warmth and Competence?

Trang Thanh Mai, University of Manitoba, Canada Olya Bullard, University of Winnipeg, Canada Luming Wang, University of Manitoba, Canada

We propose that the effect of body size on social evaluation extends beyond the association between thinness and beauty. Specifically we posit that body size has implications for warmth and competence judgments made by observers: a thin (vs. heavy) body signals competence whereas a heavy (vs. thin) body signals warmth.

I2. Can Skinnier Body Figure Signal Higher Self-Control, Integrity, and Social Status?

Trang Thanh Mai, University of Manitoba, Canada Luming Wang, University of Manitoba, Canada Olya Bullard, University of Winnipeg, Canada

Currently conventional wisdom as well as much of social psychology literature equates thinness with beauty and attractiveness in West- ern societies. We propose that a thin body signals valuable characteristics—self-control and integrity—that lead to the perception of social status because they represent social capital in a Western society.

E12 . Green versus Premium Choice and Feelings of Pride

Cecilia Souto Maior, Federal University of Paraná, Brazil Danielle Mantovani, Federal University of Paraná, Brazil Rafael Demczuk, Federal University of Paraná, Brazil

We showed that choosing a green product over a premium option increases (vs. decreases) authentic (vs. hubristic) pride perceptions. These same effects are found when participants need to justify their choice in public. When participants are aware that their decision is private the choice does not differently impact pride facets.

F12 . Matching Green Advertising Strategies and Brand Positioning to Improve Brand Evaluation

Danielle Mantovani, Federal University of Paraná, Brazil Victoria Vilasanti, Federal University of Paraná, Brazil Marcelo Vinhal Nepomuceno, HEC Montreal, Canada Cecilia Souto Maior, Federal University of Paraná, Brazil

This study shows that brand evaluations are highest when luxury positioning is used with green demarketing and when fast fashion is used with environmentalism. Ad believability mediated this effect. Environmentalism and demarketing have an equal positive effect on self- brand connection and brand attitudes of notoriously green brands. 920 / Working Papers G1 . Enchantment through Retro Product Consumption in a Digital World

Varala Maraj, City University of London, UK Fleura Bardhi, City University of London, UK Caroline Wiertz, City University of London, UK

This study examines the role of materiality in how consumers are managing the experience of disenchantment resulting from dematerialized digital consumption practices. We show that consumption of retro products revivals of outdated technology objects complements digital consumption by providing historical atmospheric value nostalgic time value and a greater perceived authenticity.

G2. You Run When Time Flies: Time Metaphors Affect Inferences from the Speed of Time

N. Alican Mecit, HEC Paris, France L. J. Shrum, HEC Paris, France Tina M. Lowrey, HEC Paris, France

In two lab experiments we show that the subjective experience of time serves as a cue that people use to infer their speed. When time is perceived to have passed quickly depending on the time perspective people speed up and suffer from cognitive trade-offs such as accuracy problems and impulsivity.

G9. The Voice From Afar: How Reverberation Affects Consumer Cognition

Johann Melzner, New York University, USA Jochim Hansen, University of Salzburg, Austria

How do acoustic parameters in the human voice affect consumers? Drawing on literature in acoustics and construal level theory this research proposes that reverberation an intuitive cue in auditory distance perception affects consumers’ inferences about both spatial and non-spatial distances and increases construal level with consequences for judgments and decision-making.

E13. Rooting for Rocky or Apollo? Underdog Narratives and Crowdfunding Success

Hua (Meg) Meng, Longwood University, USA César Zamudio, Kent State University, USA Yiru Wang, Kent State University, USA

This research investigates how underdog narratives in crowdfunding pitches influence crowdfunding success. Results show that backers are more willing to pledge when reading underdog (vs. top dog and neutral) narratives because they are more narratively transported and activate a higher level of empathy and altruism.

F9. Protection against Deception: The Moderating Effects of Knowledge Calibration on Consumer Responses to Ambiguous Advertisement Information

Joel Alan Mohr, Queens University, Canada Peter A. Dacin, Queens University, Canada

A common characteristic of deceptive persuasion tactics is information ambiguity. This research demonstrates that whether consumers perceive high (vs. low) ambiguity advertisements as more deceptive depends on whether their topic knowledge is sufficiently calibrated. While calibrated knowledge provides a safeguard against deception miscalibrated knowledge biases persuasive information elaboration.

F13 . A Story of Waste: Trust, Symbolic Adoption & Sustainable Disposal

Marwa Gad Mohsen, Babson College, USA

This working paper reflects on behavioral challenges in relation to pro-environmental waste management. A proposed combined theori- zation of symbolic adoption and Construal Level Theory (CLT) with the mediating effect of consumer trust aims at building a potential model exploring consumer likelihood to adopt ethical disposal practices. Advances in Consumer Research (Volume 46) / 921 B7 . Conceptualizing Brand Arrogance and Its Impact on Consumer Trust

Sampoorna Nandi, University of Connecticut, USA Robin A. Coulter, University of Connecticut, USA

Our research addresses the under-researched construct of brand arrogance. We conceptualize and define brand arrogance and explicate brand-related factors that feed consumer perceptions of brand arrogance. We explore the impact of brand arrogance on consumer trust and propose that perceived product quality brand use and consumer arrogance moderate this relationship.

M4 . How Consumption Experiences Create Value

Gia Nardini, University of Denver, USA Melissa Archpru Akaka, University of Denver, USA Deborah MacInnis, University of Southern California, USA Richard J Lutz, University of Florida, USA

Consumption experiences are the heart of consumer behavior. However research in this area varies widely. We integrate prior literature to develop a conceptual framework of consumption experience. Our framework organizes the fragmented literature on consumption experience and identifies avenues for future research in investigating how consumption experiences drive value creation.

J6 . Cozying up to the Kardashians: A Theory for Consumers’ Affinity towards Celebrity Gossip

Jayant Nasa, Indian School of Business, India Tanuka Ghoshal, Baruch College, USA Raj Raghunathan, University of Texas at Austin, USA

The present research examines celebrity gossip consumption using the lens of compensatory consumption. Two experiments show that people exhibit an increased propensity to consume celebrity gossip after experiencing a psychological threat and such consumption has a restorative effect on their self-worth. These effects are moderated by the consumers’ trait self-esteem.

I3. Hormonal Effects on Materialism and the Moderating Role of Intrasexual Competition

Marcelo Vinhal Nepomuceno, HEC Montreal, Canada Cristina Maria de Aguiar Pastore, Pontifical Catholic University of Paraná - PUCPR, Brazil Eric Stenstrom, Miami University, Ohio, USA

Across two studies we demonstrate how digit ratio (2D:4D) an indicator of prenatal exposure to estrogen and testosterone is predictive of materialism. Furthermore we show how intrasexual competition the rivalry with same-sex others over access to romantic partners moderates the association between digit ratio and materialistic tendencies.

A8. Do You Accept The Terms And Conditions? The Role Of Trust And Hedonic Content On Self-Disclosure To Apps

Carla Freitas Silveira Netto, Federal University of Rio Grande do Sul, Brazil Simoni F Rohden, Federal University of Rio Grande do Sul, Brazil Marina de Wallau Lugoch, Federal University of Rio Grande do Sul, Brazil Natalia Englert, Federal University of Rio Grande do Sul, Brazil Valentina Ortiz Ubal, Federal University of Rio Grande do Sul, Brazil

Online consumers disclose personal data in exchange for mobile apps. The role of trust endorsement familiarity risk perception privacy concerns and hedonic content of the app were analyzed in this working paper. Results from three different experiments suggested the positive effect of trust and negative effect of hedonic content. 922 / Working Papers K2. Influence of Attentional Breadth on Processing and Memory of Brand Advertisements

Nicolas Noack, University of Duisburg-Essen, Germany Lynn Brinkmann, University of Duisburg-Essen, Germany Oliver B. Büttner, University of Duisburg-Essen, Germany

This research examined the influence of attentional breadth on ad perception and brand recall. Eye-tracking data reveals increased gaze towards the ad background for broad attention. Additionally broad attention enhanced brand recall. The findings suggest that ad background acts as an additional memory cue.

P1. Constructed Preferences in Time-Money Tradeoffs: Evidence for Greater Violation of Procedural Invariance for Time as Opposed to Money Elicitations

Nazli Gurdamar Okutur, London Business School, UK Jonathan Zev Berman, London Business School, UK

When are preferences for time-money tradeoffs likely to be constructed? We argue that due to the ambiguous value of time consumers tend to construct their preferences and exhibit failures of procedural invariance more when their preferences are elicited based on the time (vs. money) dimension of the tradeoff.

A10. Opting Opt-in or Out? Effects of Defaults on Perceived Control and Valuation of Personal Data

Iris van Ooijen, University of Twente, The Netherlands

This work hypothesizes and finds that opt-in (compared opt-out) choice designs with regards to online data disclosure requests enhance individuals’ experienced control over personal data and their data valuation in turn. By doing so it emphasizes the importance of ‘privacy by design’. Follow-up studies are currently being conducted.

K7 . Responses to Online Behavioral Advertising Disclosures: Effects of Disclosure Source Trustworthiness and Message Type on Advertising Outcomes

Iris van Ooijen, University of Twente, The Netherlands

This work demonstrates that for a less trusted online platform (Facebook) the source of online behavioral advertising (OBA) disclosures affects consumer responses to personalized ads. Specifically OBA disclosures only increase advertising effectiveness when the source is a trusted third party and when the message is framed as a notice (vs. warning).

D7 . ‘That’s (Not) My Business’: Examining Behavior, Interactions and Implications of Consumer Brand Advocates and Brand Adversaries in Social Media

Marcus Opitz, University of Vienna, Austria Sabine Einwiller, University of Vienna, Austria

Word-of-mouth and related advocacy and adversary behavior disseminated by consumers in an online environment are among the most trusted forms of brand communication. This study offers a first systematic content analysis of the communicative behavior of advocates and adversaries and the factors influencing it within the context of brands’ Facebook profiles.

G7 . The Presence of Dividing Line Decrease Perceived Quantity

Jun Ouyang, Xiamen University, China Yanli Jia, Xiamen University, China Zhaoyang Guo, Xiamen University, China

Three studies showed that participants reported smaller quantity estimates when shown a set of items separated with a dividing line than when shown them as a whole. We attributed the results to that the dividing line might attract participants’ attentions from the items thus leading them to underestimate items’ quantity. Advances in Consumer Research (Volume 46) / 923 M5. The More Expensive a Gift Is, the More It Is Appreciated? The Effect of Gift Price on Recipients’ Appreciation

Jooyoung Park, Peking University, China Mengshu Chen, Tencent Holdings Limited, China

Two studies show that a gift with a higher or lower price than the recipient expects decreases their appreciation. Our findings suggest that recipients perceive givers as being less thoughtful when the gift price is lower than expected whereas they feel burden to repay when the gift price is higher.

P10 . Omission Bias in the Gain vs . Loss Domain

Jen H. Park, Stanford University, USA

This paper examines whether individuals exhibit greater omission bias in a loss-framed risky choice (lose $10000 or lose $0) compared to the gain frame (earn $0 or earn $10000). While prior studies reveal mixed results this research introduces anticipated regret as a moderator and uses the Monty Hall problem.

P14 . Financial Behavior Among Young Adult Consumers: The Influence of Self-determination and Financial Psychology

Heejung Park, University of Wyoming, USA

Research of the consumer financial behavior has been investigated in various ways. Better financial behaviors are essential for improving quality of life. Specifically young adult consumers have particular difficulty making decisions involving accruing debt. Therefore this paper deals with the financial behavior of young adult consumers through self-determination theory.

N8. Effect of Awe on Collectable Consumer Experience

Eujin Park, Washington State University, USA Andrew Perkins, Washington State University, USA Betsy Howlett, Washington State University, USA

It has been known that awe is caused by encountering an unexpected event and stimulates explanation-seeking via a need for cognitive accommodation (Valdesolo Shtulman and Baron 2017). Collectable experiences are often unusual and unconventional (Keinan and Kivetz 2010). Hence we hypothesize that awe increases individual’s preference for collectable consumer experience.

P12 . Disclosure of Project Risk in Crowdfunding

Jooyoung Park, Peking University, China Keongtae Kim, Chinese University of Hong Kong, China

This research examines whether and how risk disclosure of crowdfunding projects influences crowdfunders’ project perceptions and funding decisions with a particular focus on the contrast between technology and non-technology projects. We find that crowdfunders respond to the risk disclosure of projects but differently between technology projects and non-technology projects.

R9. The Asymmetric Effects Of Attitude Toward The Brand (Symbolic Vs. Functional) Upon Recommendation System (Artificial Intelligence Vs. Human)

Kiwan Park, Seoul National University, South Korea Yaeri Kim, Seoul National University, South Korea Seojin Stacey Lee, Seoul National University, South Korea

We investigated how the recommendation system types (AI vs. Human) affect brand attitude depending on the brand image (Symbolic vs. Functional). We found consumers are reluctant to accept a recommendation from AI where in symbolic brand (Study1). This effect was further explained by uncanny-feeling toward the AI recommendation system (Study2). 924 / Working Papers D2 . When a Negative Review Can Help the Company: The Role of the Unfairness and Empathy

Maria Alice Pasdiora, Federal University of Rio Grande do Sul, Brazil Cristiane Pizzutti, Federal University of Rio Grande do Sul, Brazil Natalia Englert, Federal University of Rio Grande do Sul, Brazil

This work extends literature on the influence of justice perception on consumer behavior by showing that consumers sensitize with injustice perpetrated on companies by other consumers. We demonstrate that injustice episodes trigger feelings of empathy and consequently lead witnesses to present favorable attitudes and more purchase intention towards the company.

Q10 . Social Media Agency: Exploring the Role of Social Media Structures in Shaping Consumers’ Identity Projects

Gabrielle Patry-Beaudoin, Queens University, Canada Jay Handelman, Queens University, Canada

Drawing on the mediatization theory (Hjarvard 2008) this study analyzes the role of social media in shaping consumers’ online identity projects. Our qualitative empirical investigation contributes to the consumer identity literature by examining how three YouTube structures create unique tensions for consumers that affect their identity projects.

J5 . Buy Better, Buy Less: Future Self-Continuity and Construal Level Affect Investment in Sustainable Consumer Products

Rebecca Peng, Northeastern University, USA Daniele Mathras, Northeastern University, USA Katherine Loveland, Xavier University, USA

Results of three studies reveal that consumers who believe their future selves will be similar (vs. dissimilar) to their current selves are most likely to embrace a “buy it for life” mindset. Specifically higher future self-continuity promotes higher level construal and ultimately ratings for brands that sell long-lasting products.

Q3 . Academic Branding on Social Media: Examining the Influence of Personal Brands on the Professor-Student Relationships

Edita Petrylaite, Northumbria University, UK Darija Aleksic, University of Ljubljana, Slovenia

With the increasing interaction on social media and its proliferated impact on personal image professionals including university lecturers communicate on digital channels for work and personal purposes. This research aims to find out how personal branding using the same social media account influences professors’ relationships with students as university clients.

P6 . Marginal Cost Consideration

Ethan Pew, Stony Brook University, USA Hyunhwan Lee, University of Miami, USA

A large-scale study (N=12003) examines how price presentation influences choice. Six experimental conditions present absolute and marginal prices (across 30 price levels) for a flight selection task. Highlighting savings accruing from choosing the lower price option results in greater choice share for the higher priced option particularly as prices increase. Advances in Consumer Research (Volume 46) / 925 K6 . Persuading the Moral Consumer: Matching Messages to Attitude Basis

Aviva Philipp-Muller, Ohio State University, USA Andrew Luttrell, Ball State University, USA Richard Petty, Ohio State University, USA

We explored the persuasive effects of moral versus practical counter-attitudinal messages on individuals with varying levels of moral conviction toward recycling. Across two studies when pro-recycling moral conviction was high the moral (vs. practical) anti-recycling message was more persuasive but the effects reversed when moral conviction was low.

F10 . Food Waste: On the Normalization of Structural Violence

Andreas Plank, Privatuniversität Schloss Seeburg, Austria

While more than one billion tons of palatable food are thrown away every year (FAO 2013b) about 900 million people are undernour- ished worldwide (FAO 2013b). This article tries to answer the research question: How do international organizations (i.e. organizations dealing with the phenomenon food waste institutionally) frame food waste?

B1 . Dynamic Pricing in Stationary Retailing - The Role of Consumer’s Trust

Maximilian Clemens Pohst, Heinrich-Heine-University, Germany Caspar Krampe, Heinrich-Heine-University, Germany Peter Kenning, Heinrich-Heine-University, Germany

The integration and application of electronic shelf labels (ESL) enables retailers to implement dynamic pricing practices. Therefore this research work investigates how the use of ESL is perceived by consumers. Preliminary findings revealed that consumer’s trust positively affects the perception of ESL as well as perceived price fairness and loyalty.

R4 . Human Brands and Their Consumers: How Consumers Reform Brand Understandings Following Critical Incidents

Kimberley Mosher Preiksaitis, Siena College, USA

Human brands offer consumers cultural material that differs from traditional brands. When human brands encounter critical incidents consumers face the tension of separating the celebrity from the person to understand their relationship with the brand. This paper offers insights of three ways consumers reform brand understandings following critical incidents.

I9 . From Childhood Toys to Grownup Choices: Understanding the Gendered Appeal of Violent Media

Martin A. Pyle, Ryerson University, Canada

Why do mean seem to enjoy violent media more than women? This research addresses this question by examining how childhood toys influence perceptions of media as adults. This research contributes to our understanding of the appeal of violent media by demonstrating a cohesive link between socialized values and media choice.

R5. Autonomy or Enjoyment? The Contingent Nature of Brand Ritual

Yaxuan Ran, Zhongnan University of Economics and Law, Vietnam Echo Wen Wan, University of Hong Kong, China

Brand rituals can affect consumers’ feelings and experience. Adding to this literature the present research demonstrates the contingent nature of brand ritual by showing that brand ritualization can increase enjoyment but simultaneously decrease autonomy both of which increase WTP. Furthermore brand personality moderates the relative salience of the two perception-processes. 926 / Working Papers G4 . That’s So Sweet: Baby Cuteness Semantically Activates Sweetness to Increase Sweet Food Preference

Shaheer Ahmed Rizvi, University of Alberta, Canada Sarah G Moore, University of Alberta, Canada Paul Richard Messinger, University of Alberta, Canada

We examine the semantic association between baby (kindchenschema) cuteness and sweetness and how kindchenschema cuteness increases preference for sweet foods. Kindchenshcema cuteness semantically activates the construct of sweetness thereby increasing preference for sweeter foods. Through moderation we show that kindchenschema and not whimsical cuteness produces this increased desire for sweets.

L5. Understanding the Components and Effects of the Omnichannel Seamless Experience

Paula Rodríguez-Torrico, Universidad de Burgos, Spain Lauren Trabold, Manhattan College, USA Sonia San-Martín, University of Burgos, Spain Rebeca San José, University of Valladolid, Spain

This paper aims to investigate empirically the concept of Omnichannel Seamless Experience (OSE) and its impact on Social Media En- gagement (SME). Results confirm the multidimensionality of the OSE namely consistency freedom in channel selection and synchronization which directly impact on consumer satisfaction with the omnichannel interaction and indirectly on SME.

M1. How Rewarding is Your Rewards Program? Experiential vs. Material Rewards

Ayalla Ruvio, Michigan State University, USA Farnoosh Khodakarami, Michigan State University, USA Clay Voorhees, Michigan State University, USA

The research on loyalty rewards programs focuses on material rewards. In four studies we explore consumers’ preference for experiential (vs. material) rewards. They view experiential (vs. material) rewards as having greater expected utility which in turn increases their behavioral intentions and engagement with the loyalty program.

D9 . Consumption Closure as a Driver of Positive Word of Mouth

Christina Saenger, Youngstown State University, USA Veronica Thomas, Towson University, USA

Consumption closure affects consumers’ cognitive processing and WOM. Manipulating closure through packaging consumers who experience closure (single-serving package) as opposed to those who do not (equivalent amount from multi-serving package) provide more positive WOM due to increased attitude certainty. Marketers may benefit by encouraging reviews of single-serve packages.

E10 . Sustainable Initiatives: Cultural Identity, Regulatory Focus, and Construal Perspective

Ekaterina Salnikova, Aarhus University, Denmark Yuliya Strizhakova, Rutgers University, USA Klaus G Grunert, Aarhus University, Denmark

We integrate cultural identity regulatory focus and construal theory perspectives to examine effectiveness of sustainable product initiatives. Across five studies we demonstrate strong moderating effects of cultural identity in consumer responses to sustainability framing. Globally-oriented consumers are more engaged with sustainability framed in promotion and more distant terms. Advances in Consumer Research (Volume 46) / 927 P8 . Understanding Financial Literacy: A Meta-Analysis of the Antecedents, Consequents and Moderators

Fernando De Oliveira Santini, Universidade Do Vale Do Rio Dos Sinos, Brazil Frederike Monika Budiner Mette, ESPM, Brazil Mateus Canniatti Ponchio, ESPM, Brazil Wagner Junior Ladeira, Universidade Do Vale Do Rio Dos Sinos, Brazil

A meta-analysis was conducted to evaluate the antecedents consequents and moderators of financial literacy. We obtained 13 significant antecedents of the financial literacy construct and four significant consequent constructs. The observations reflected the moderating effect of culture and the level of economic development of the country of origin.

E5 . Volunteer Motivations for Direct versus Indirect Service

Abigail Schneider, Regis University, USA Eric Hamerman, Iona College, USA

Nonprofit organizations rely heavily on volunteers yet recruitment and retention of volunteers is a major challenge. We explore the different motivations for volunteer tasks finding that different drivers of behavior (categorized by the Volunteer Functions Inventory) correspond differently to indirect and direct service activities that require differing levels of skill.

N3 . Emotion Regulation and Memory for Negative Emotion Ads

Sandra Segal, Ben Gurion University, Israel Hila Riemer, Ben Gurion University, Israel Shai Danziger, Tel Aviv University, Israel Gal Sheppes, Tel Aviv University, Israel

People may employ emotion regulation to process negative emotion stimuli. We find that when people use emotion regulation they regulate high-arousal negative ads using distraction and low-arousal negative ads using reappraisal resulting in better memory for low-arousal ads. When they do not use emotion regulation people better remember high-arousal ads.

E7 . Pronouns in Fundraising Appeals – The Impact of I vs . S/He on Donations

Amir Sepehri, Western University, Canada Rod Duclos, Western University, Canada Hamid Elahi, Western University, Canada

Donors evaluate online fundraisings partly on their linguistic characteristics. Extending prior research showing negative (positive) emotions increase (decrease) donations we show 1) self-related (other-related) pronouns result in lower (higher) donations and 2) the negative (positive) effect of self-related (other-related) pronouns is stronger when negative (positive) emotions increase.

E2 . Donation versus Adoption: How the Mode of Helping Moderates the Effect of Emotions on Helping

Ziqi Shang, Renmin University of China, China Xiuping Li, National University of Singapore, Singapore Aradhna Krishna, University of Michigan, USA

Ads asking for donations depict help-recipients’ sad or happy emotions. We examine if the effect of help-recipients’ emotion on observers’ helping intention varies depending on the helping mode. We find that the decision to adopt(donate) makes people more self-focused(other-focused) and more(less) likely to help when seeing happy (vs. sad) help-recipients. 928 / Working Papers J10 . Transnational Consumer Lifestyle

zahra Sharifonnasabi, Queen Mary University of London, UK Fleura Bardhi, City University of London, UK

In a world of increasing global mobilities this study investigates the nature of consumption when individuals inhabit multiple countries simultaneously. First four meanings of place have emerged in our findings each representing specific meanings relationships and consumption practices. Second we introduce the notion of transnational consumption.

M6 . Approaching Negative Experience

Liang Shen, Shanghai Jiao Tong University, China Fengyan Cai, Shanghai Jiao Tong University, China Ying Yu, Huazhong Agricultural University, China

Because approaching unfamiliar negative experiences could dissipate one’s uncertainty of the environment we propose that people are more curious about and hence more likely to choose a negative experience over a positive experience. Three experiments provide preliminary evidence for this effect.

M10 . I Need a Hero: How Loneliness Interacts with the Symbolic Meaning of Products to Affect Consumer Attitude

Sirajul Arefin Shibly, SUNY Binghamton, USA Jinfeng Jiao, SUNY Binghamton, USA

While extensive literature exists on symbolic meaning of products and on loneliness how the interaction of these two concepts can affect attitude development towards a product has not been assessed. The current paper demonstrated that lonely people have a more positive attitude towards products that bear a superhero symbolic meaning.

Q12. Exploring Effects of Perceived Authenticity of Instagram Models on Aad and Buying Intentions

Heather Shoenberger, University of Oregon, USA Eunjin Kim, Southern Methodist University, USA

This study suggests that consumers’ ad evaluations and purchase intention are highest when a plus size not-airbrushed model is featured on a brand’s Instagram account. More importantly this study confirms that perceived authenticity plays a significant mediating role for the effects of model-image modification and body size on ad evaluations.

A11. When Political Neutrality Backfires

Ike Silver, University of Pennsylvania, USA Alex Shaw, University of Chicago, USA Rob Kurzban, University of Pennsylvania, USA

Politicians and public figures often refrain from taking sides on hot-button social issues to avoid alienating potential supporters. Two studies (n=751) find that this strategy can backfire: Neutrality is sometimes interpreted as indicating a strong strategically concealed opinion and can lead to diminished trustworthiness and voting support in such cases.

E4 . Doing Good for Nothing: Motive Inferences from the Probabilistic Profits of Prosociality

Ike Silver, University of Pennsylvania, USA Jackie Silverman, University of Pennsylvania, USA

We explore how the probability of profiting from prosocial behavior impacts perceived motive. Eight studies (N=3557) demonstrate that agents whose prosocial actions may return large rewards but may also return nothing are evaluated more positively than those whose prosocial actions yield small certain rewards of equal expected value. Advances in Consumer Research (Volume 46) / 929 M3 . #FOMO: How the Fear of Missing Out Drives Consumer Purchase Decisions

Michelle van Solt, Florida International University, USA Jessica Rixom, University of Nevada, Reno, USA Kimberly Taylor, Florida International University, USA

This research focuses on consumers’ purchase decisions motivated by FOMO. Two studies demonstrate that purchase intent is higher when participants’ imagine that their close social groups will attend an effect mediated by FOMO. Furthermore the type of experience (ordinary extraordinary) moderates the relationship of social groups (close distant) on FOMO.

N5 . Mixed Feelings, Mixed Baskets: How Emotions of Pride and Guilt Drive the Relative Healthiness of Sequential Food Choices

Julia Storch, University of Groningen, The Netherlands Koert van Ittersum, University of Groningen, The Netherlands Jing Wan, University of Groningen, The Netherlands

Although healthy diets start with healthy shopping baskets previous research primarily investigated factors influencing the healthiness of isolated food purchases. Instead we propose that dependencies exist between the healthiness of shoppers’ sequential choices. We address this research gap and investigate whether emotions experienced while shopping underlie these dependencies.

R7. How and Why Life Transition Influences Brand Extension Evaluation

Lei Su, Hong Kong Baptist University, China Alokparna (Sonia) Monga, Rutgers University, USA Yuwei Jiang, Hong Kong Polytechic University, China

Life transition is a ubiquitous aspect of life and occurs as people transcend various life stages. Four studies showed that life transition resulted in higher perceived fit and more favorable brand extension evaluation. This effect is mediated by self-concept ambiguity and consequently dialectical thinking.

I13 . Ambient Light, Gender, and Creativity

Courtney Szocs, Louisiana State University, USA Franziska Metz, EBS, USA Dipayan Biswas, University of South Florida, USA

The findings of our experiment demonstrate that females are more creative with bright (vs. dim) ambient light while males are more creative with dim (vs. bright) ambient light. The findings of our research also help in resolving the inconsistency in the prior literature for effects of ambient light on creativity.

O8 . Valuation and Allocation of Bought Time

Eisa Sahabeh Tabrizi, University of Southeast Norway Marit Engeset, University of Southeast Norway Luk Warlop, Norwegian School of Management, Norway

We show when consumers buy time by outsourcing their tasks they may be willing to pay less for the service and are more likely to find the price unfair. We propose this is driven by higher domain self-efficacy. We also document consumers’ tendency for allocating bought time more to hedonic activities. 930 / Working Papers C11 . More of a Bad Thing: How Consumers Ignore Pollutant Levels in Healthiness Assessment

Aner Tal, Ono Academic College, Israel Yaniv Gvili, Ono Academic College, Israel Moty Amar, Ono Academic College, Israel

Consumers appear to consider only the presence rather than the quantity of pollutants in food in assessing healthiness. The current studies demonstrate this with assessment of healthiness of food containing pesticides as well as demonstrate effective debiasing by providing a standard of assessment.

O13 . Pain of Loss: How Losing in a Promotional Competition Influences Consumer Attitude

Arash Talebi, ESSEC Business School, France Sonja Prokopec, ESSEC Business School, France Ayse Onculer, ESSEC Business School, France

This research shows that the increase in brand attitude of the winners of an uncertain promotion comes at the cost of a greater decrease in the attitude of the losers; in turn that leads to an overall decrease in brand attitude. We show that the solution lies in unresolved uncertainty.

L6. The Influence of Social Exclusion on Consumers’ Perceptions of and Responses to Consumer-Dense Retail Environments

Veronica Thomas, Towson University, USA Christina Saenger, Youngstown State University, USA

Social exclusion affects consumers’ perceptions of and responses to consumer-dense retail environments. Socially excluded (v. included) consumers exhibit increased crowd-size estimates perceive higher neutral and positive (but equal negative) crowding and prefer crowded retailers. Socially excluded consumers’ heightened affiliation motivation increases crowding perceptions enhancing intentions to spend time and money.

C5. Krabby Patties, Kelp Chips, or KitKats?: Exploring the Depictions of Food Featured in Children’s Television Shows

Kathy Tian, University of Illinois at Urbana-Champaign, USA Regina Ahn, University of Illinois at Urbana-Champaign, USA Michelle Renee Nelson, University of Illinois at Urbana-Champaign, USA

Time spent watching TV may contribute to unhealthy eating yet most studies only review food advertising. We examine food and beverage references within 64.5 hours of children’s television programs. Our results show prevalence of un-branded and less healthy food and beverages primarily as snacks. Ramifications for child obesity are discussed.

C8. Can Packaging Imagery Fill Your Stomach? Effects of Product Image Location on Flavor Richness, Consumption Quantity, and Subsequent Choice

Taku Togawa, Chiba University of Commerce, Japan Jaewoo Park, Musashi University, Japan Hiroaki Ishii, Seikei University, Japan Xiaoyan Deng, Ohio State University, USA

Through three studies we found that an image placed at the bottom of a food package increased consumers’ perception of flavor richness (Study 1) decelerated their food consumption (Study 2) and led them to choose a healthy snack after consumption (Study 3). Advances in Consumer Research (Volume 46) / 931 F11 . Anti-Consumption for Sustainability: The Environmental Impact of Anti-Consumption Lifestyles, Environmentally Concerned Individuals and Ethical Consumers

Laurie Touchette, HEC Montreal, Canada Marcelo Vinhal Nepomuceno, HEC Montreal, Canada

We compare the environmental impact of anti-consumption lifestyles ethical consumption and environmental concern. Environmental impact is lowest for tightwadism (i.e. an anti-consumption lifestyle) unrelated with environmental concern and highest for ethical consumption. Such findings suggest that resisting consumption offers a viable and effective way towards sustainable consumption.

Q2. Why do Kids Love Watching Unboxing Videos? Understanding The Motivations of Children to Consume Unboxing Toy Videos

Teresa Trevino, Universidad de Monterrey, Mexico Mariela Coronel, Universidad de Monterrey, Mexico Valeria Martínez, Universidad de Monterrey, Mexico Ivanna Martínez, Universidad de Monterrey, Mexico Daniela Kuri, Universidad de Monterrey, Mexico

Unboxing videos are videos that shows the opening gathering and exhibit of products across social media platforms such as YouTube. This paper contributes to the literature of consumer behavior by shedding light on the motivations of children to consume unboxing videos and its effects on brand engagement with toy brands.

A2. A Window to the Brand’s Soul: How Models’ Eyes Affect Consumers’ Attitudes

Maria Giulia Trupia, IESE Business School, Spain Martina Cossu, Bocconi University, Italy Zachary Estes, Bocconi University, Italy

Three studies tested the impact of pupil dilation on consumers’ ad attitudes. Models whose pupils are constricted (versus dilated) improved ad attitudes (Study 1). Models’ perceived attractiveness and trustworthiness mediate the effect (Study 2). Moreover evidence of pupil mimicry was found but it did not affect ads’ evaluations (Study 3).

B5. Money Doesn’t Buy Happiness, But What About Buying Trust? The Effectiveness of Financial Compensation in Restoring Trust After Double Deviation

Valentina Ortiz Ubal, Federal University of Rio Grande do Sul, Brazil Cristiane Pizzutti, Federal University of Rio Grande do Sul, Brazil Katja Gelbrich, Catholic University Eichstätt-Ingolstadt, Germany

Situations in which a financial compensation is effective to rebuild trust after a double deviation have been neglected by marketing literature. We show that financial compensation could be more effective than non-financial tactics (apology promise) in recovering trust after double deviation when the initial failure causes a financial loss.

O6 . Be Aware of Your Suspicion: When “Being Suspicious” Ironically Leads to Suboptimal Judgment- and Decision-Making

Julie Verstraeten, Ghent University, Belgium Tina Tessitore, INSEEC Business School, France Maggie Geuens, Ghent University, Belgium

Contrary to the belief that “being suspicious” benefits judgment- and decision-making four studies demonstrate the opposite to hold true. Suspicion rendered judgments and decisions suboptimal for money and food in an unrelated context. As suspicion is ever-prevailing in nowadays’ marketplace these findings highlight relevant and important consequences for consumer welfare. 932 / Working Papers O2. The Streaking Star Effect: Why People Want Individual Winning Streaks to Continue More than Group Streaks

Jesse Walker, Cornell University, USA Thomas Gilovich, Cornell University, USA

In seven studies participants exhibited a greater desire for streaks of individual success to continue than identical streaks of success by groups. Fairness or concern about the other competitors and the experience of awe inspired by an individual streak mediate this effect. This phenomenon has downstream consequences for consumer behavior.

P2 . The Upside of Myopic Loss Aversion

Daniel Wall, Carnegie Mellon University, USA Gretchen Chapman, Carnegie Mellon University, USA

Broad brackets lead to increased risk taking which is normative without reinvestment but can be disastrous with reinvestment. In our study narrow bracketed participants are more loss averse—a better decision—yielding higher expected log wealth. Myopia can be good.

R12 . Brand Primes Can Satiate (Important) Consumer Goals

Darlene Walsh, Concordia University, Canada Chunxiang Huang, Concordia University, Canada

In Study 1 we show that exposure to a brand can satiate a previously primed brand-related goal. However the satiating effect only occurs when the goal is important to the individual. In Study 2 we show that perceived goal progress can in part explain this effect.

E1. Effects of Recipients’ Emotional Expressions on Donors’ Preference for Helping with Development versus Survival

Xue Wang, University of Hong Kong, China He (Michael) Jia, University of Hong Kong, China Sara Kim, University of Hong Kong, China

Across three studies the current research shows that emotional expressions of recipients influence donors’ relative preference for developmental helping versus survival helping to the recipients through altering donors’ optimistic belief about the recipients.

J9. The Beautified Me is Me: How Interdependence Increase Usage of Beauty App

Qin Wang, Arizona State University, USA Andrea Morales, Arizona State University, USA Adriana Samper, Arizona State University, USA

This research investigates when people are more likely to use photo-editing beauty apps and why. Two studies show that people with an interdependent (vs. independent) self-construal have higher intentions to use beauty apps because they see larger overlaps between their beautified selves and their true selves.

D6. How to Boast Appropriately When Word of Mouth Flows Internationally?

Xingyu Wang, Huazhong University of Science and Technology, China Yaping Chang, Huazhong University of Science and Technology, China Jun Yan, Huazhong University of Science and Technology, China

The current research explores the interaction effect between “Prophase Word-of-Mouth” and “Boastful Communication Strategy” on consumers’ purchase intention. Through two experiments we found when PWOM valence is high managers should use a high-level “Boastful Communication Strategy” when PWOM valence is low managers should use a low-level boasting strategy. Advances in Consumer Research (Volume 46) / 933 C7. The Visually Simple = Healthy Intuition and Its Effects on Food Choices

Yan Wang, Renmin University of China, China Jing Jiang, Renmin University of China, China

This research proposes an outcome discounting relationship between visual simplicity of packages and food healthiness perception. Across three experiments we find that consumers tend to perceive the food healthier when it is packed visually simple (vs. complex) and thereby more likely to choose the food.

F1 . Reach out in the Darkness: How Unfair Treatments Shape Social Connection Motivation

Yijie Wang, Hong Kong Polytechic University, China Yuwei Jiang, Hong Kong Polytechic University, China Mandy Mantian Hu, Chinese University of Hong Kong, China Ninghua Zhong, Tongji University, China

Individuals who were unfairly treated exhibited a higher need to belong and exert more effort to connect with acquaintances and strangers than fairly treated peers. Furthermore this effect was mediated by the feeling of resource deprivation after experiencing injustice and was more salient among individuals with relatively low self-esteem.

C6. How Does Unsatisfied Curiosity Stir Our Craving For Food?

Chen Wang, Drexel University, USA

Curiosity cues abound where consumers make their food choices. However little is known about how curiosity influences consumers’ subsequent food decisions. This research explores the impact of incidental curiosity on consumers’ food choice. We propose multiple outcomes determined by the food options available and the nature of curiosity.

H9 . Market Emergence: the Alignment Process of Entrepreneurs’ Socio Cognition and Consumers’ Perception of the Market

Hao Wang, University of South Florida, USA

Drawing on social constructive perspective this study bridges entrepreneurial decision-making and legitimacy underlying the micro-foundation of the market creation. I argue that the extent to which the market is legitimate depends on the alignment between entrepre- neur’s socio cognition of the market and consumers’ perception of the market.

L9 . To Save Face or Follow My Heart: Salesperson’s Inquiries of In-Group Identity on Consumers’ Purchase

Lingru Wei, Tencent Holdings Limited, China Jooyoung Park, Peking University, China

This research shows that the salience of in-group identity (e.g. a student of a prestigious school) leads to purchase due to enhanced self-presentation concern especially when customers sense the possibility of violating social norms. Our finding suggests that simple conversation between salespersons and consumers can change sales.

N7. Emotion Or Information? Effects of Online Social Support on Customer Engagement

Chuang Wei, Tsinghua University, China Maggie Wenjing Liu, Tsinghua University, China Qichao Zhu, Tsinghua University, China

Short Abstract: This paper investigates the interactive effect and the underlying mechanism of social support and source on customer engagement.. We conducted a study to show that consumer engagement in online brand communities is affected by the support type (informational vs. emotional) and the effect is moderated by source of support. 934 / Working Papers O4. Will Winning Always Encourage Risk Taking? The Effect of Competition Results and the Closeness of These Results

Beixi Wen, Renmin University of China, China En-Chung Chang, Renmin University of China, China Chunya Xie, Renmin University of China, China

In this research we found that winning a competition or hearing news that a familiar team has won a game will increase people’s risk preference. However a close result can reverse this effect; that is compared with narrow-winners narrow-losers will be more willing to take risks.

B2 . The Prevention Oriented Chameleon: Mimicry in a Prevention Orientation Leads to More Brand Trust

Judith Willberger, Technical University of Munich, Germany Gavan Fitzsimons, Duke University, USA

Will customers trust a brand and its products more when their trust and security needs are fulfilled? This study shows that customers trust a brand and its products more when their trust need is fulfilled by being mimicked and their security need by being aware of the brand’s prevention orientation.

L1. The Effects of Cultural Syndromes on Customers’ Responses to Service Failures: A Perspective-Flexibility-Based Mechanism

Vincent Chi Wong, Lingnan University, China Robert Wyer Jr., University of Cincinnati, USA

Customers with a collectivistic (vs. individualistic) orientation or a long-term (vs. short-term) orientation are likely to attribute a service failure more to the service provider’s contextual factors and less to the service provider’s dispositional factors. These effects are mediated by the flexibility of perspectives customers take when making a judgment.

H8. The Beneficial Side of Haze: Air Pollution Promotes Innovation and Creativity

Yi Wu, Tsinghua University, China Yifan Chen, Tsinghua University, China Yuhuang Zheng, Tsinghua University, China

a counterintuitive hypothesis that air pollution promotes innovation and creativity. Two archival studies (Study 1-2) found that air pollution predicted innovation at country- and company-level. Study 3a-3b experimentally established causal connection between air pollution and creativity. Study 4 replicated hypothesis in real weather context and proved arousal’s mediating role.

P11 . A Price Premium on A Trivial but Weak Preferred Attribute Increase Choice: The Roles of Scarcity, Arousal and Perceived Risk

Yueyan Wu, Hunan University, China Chunyan Xie, Western Norway University of Applied Sciences, Norway Zhi Yang, Hunan University, China Luluo Peng, Hunan University, China

We add to extant research on trivial attribute by addressing a“small price premium effect”on choice and providing a psychological mechanism. Results showed that a small price premium on a trivial attribute increases choice share of weak preferred options. Large price premium eliminates this kind of effect. Underlying mechanisms were explored. Advances in Consumer Research (Volume 46) / 935 F8 . Dual Routes for Consumer Responses to Corporate Social Responsibility: The Role of Positive Moral Emotions, Attitudes, and Empathy

Chunyan Xie, Western Norway University of Applied Sciences, Norway Richard P. Bagozzi, University of Michigan, USA

We adds to extant CSR research by investigating both cognitive and affective processes underlying consumer reactions to CSR. Results showed that both positive moral emotions and positive attitudes mediate the impact of CSR actions on consumer positive responses toward the company. Further empathy regulates the elicitation of positive moral emotions.

K1 . The Impact of Moral Violation and Advertising Appeals on Brand Attitude

Chunya Xie, Renmin University of China, China En-Chung Chang, Renmin University of China, China Beixi Wen, Renmin University of China, China

This study found that moral violators possess a more positive attitude towards brand advertised in two-sided (versus one-sided) message appeals while moral violation witness possess a more positive attitude towards brand advertised in one-sided (versus two-sided) appeals.

B10 . Wearing V Neck, Getting More Trust: An Evolutionary Psychology Approach to Examine the Effect of Collar Style on Trust

Jialiang Xu, University of Manitoba, Canada Fang Wan, University of Manitoba, Canada Chenbo Zhong, University of Toronto, Canada

“What a strange power there is in clothing” — Isaac Bashevis Singer. Three experiments findings show that one’s collar styles (V neck vs. turtleneck) can affect others’ trust of him or her.

H5. Being Looked Up or Being Looked Down? The Divergent Effects of Camera Angle on Nonanthropomorphized and Anthropomorphized Product in Ads

Changchun Xuan, Xiamen University, China Rui Chen, Xiamen University, China Jing Jiang, Suzhou Broadcasting System, China Bin Yu, Tsingtao Brewery Company Limited, China Jingjia Peng, Icartoon Culture & Media Corporation ltd, China Shengdong Lin, Xiamen University, China

This paper finds that for anthropomorphized product being presented in the downward-looking angle lead to higher product evaluation compared with being presented in upward-looking angle. On contrast for nonanthropomorphized product the effect of camera angle is reversed. The two different mechanisms underlying the effect of camera angle were also tested.

Algorithm Attraction versus Aversion: The Role of the Perceived Self-Efficacy of the Decision Maker

Gizem Yalcin, Erasmus University Rotterdam, The Netherlands Anne-Kathrin Klesse, Erasmus University Rotterdam, The Netherlands Darren Dahl, University of British Columbia, Canada

Nowadays algorithms are used to generate recommendations in numerous areas including ones that are pure matter of taste. Across four studies we demonstrate that consumers value the same recommendation differently depending on the framing of its source—an algorithm versus human expert—and their own perceived level of self-efficacy. 936 / Working Papers G6 . Brightness Increases More Positive Views of Humanity and Prosocial Behavior of People Low in Moral Identity Internalization

Jun Yan, University of Manitoba, Canada Luke Zhu, University of Manitoba, Canada Fang Wan, University of Manitoba, Canada

This paper tests an interactive effect of ambient lighting and Moral Identity Internalization (MII) on prosocial behavior. Individuals low (vs. high) in MII are more likely to help others in bright than dark environment because the brightness increases their positive views of humanity.

A9 . I know It’s not real, but I like it!

Junxian Yang, Singapore University of Social Sciences, Singapore Yue Wang, Singapore University of Social Sciences, Singapore Jufinnie Lim, Singapore University of Social Sciences, Singapore Yu-chen Hung, Singapore University of Social Sciences, Singapore

Questionable practices in online advertising have triggered cautiousness among consumers when they encountered information on social media. While consumers do not completely trust it our studies demonstrate that they remain influenced. We further explore why people fail to discount information on social media even when they question its authenticity.

O3. The Effect of Numeric Information on Product Evaluation

Zhen Yang, Drexel University, USA Yanliu Huang, Drexel University, USA Dengfeng Yan, University of Texas at San Antonio, USA

This research proposes that precise number will polarize product evaluation making the judgment of positive product information more positive and negative information more negative. It happens because processing precise number is more arousing and the heightened arousal leads to evaluative extremity. Three studies were conducted to test our predictions.

O5. The Effect of Synchrony on Non-Human Objects Involved in the Synchronous Performance

Xiaoyin Ye, Xiamen University, China Jun Ye, Xiamen University, China

Drawing from an embodied cognition framework this study examines the main effect and the mediating mechanism of synchrony on product evaluation. Four experiments were conducted and our findings indicate that exposure to synchrony experience activates the concept of harmony which has a spillover effect that positive product evaluation.

K13. When Does Humor Increase Sharing?

John Yi, University of Arizona, USA Caleb Warren, University of Arizona, USA

Are consumers more likely to share content (stories social media posts etc.) when it is humorous? Not always. Although adding humor increases the likelihood that consumers share negative content (e.g. embarrassing stories complaints etc.) adding humor does not increase the likelihood of sharing positive content (e.g. happy stories praise etc.). Advances in Consumer Research (Volume 46) / 937 L12 . Should I Stay or Should I Go: When Our Companies Have Eyes for Other Consumers

Na Ri Yoon, Indiana University, USA Jenny Olson, Indiana University, USA Adam Duhachek, Indiana University, USA

To attract new customers companies may lose existing customers. Our research examines how consumers’ attachment styles influence the likelihood of switching companies. When new customers receive exclusive benefits existing customers’ attachment anxiety predicts switching. Attachment avoidance predicts greater switching regardless of benefit recipient. Perceived betrayal and firm commitment underlie switching.

H10 . No Pain, No Out-of-Box Thinking: An Examination of the Effects of Self-Threat on Creativity

Huan You, University of Manitoba, Canada Fang Wan, University of Manitoba, Canada Luke Zhu, University of Manitoba, Canada Haiyang Yang, Johns Hopkins University, USA

Two experiments examined the effect of self threat on creativity. Findings suggest that type of threat matters; a threat from a dynamic domain enhances divergent creativity and a threat from fixed domain enhances convergent creativity.

D12 . Future Decisions and Temporal Contiguity Cues: When Absence of Temporal Contiguity Cues Increases Online Reviews’ Persuasiveness .

Francesco Zanibellato, Ca’ Foscari University, Venice, Italy

Past research shows that temporal contiguity cues make online reviews more persuasive. However rooting in Construal Level Theory we show that when consumers face a distant-future consumption timeframe they are more persuaded by online reviews without temporal contiguity cues.

L3 . Categorizing Engagement Behaviors from the Perspective of Customer Resources

Xianfang Zeng, University of Calgary, Canada James Agarwal, University of Calgary, Canada Mehdi Mourali, University of Calgary, Canada

From the perspective of customer resources the authors propose a typology of customer engagement behaviors (CEBs). This perspective well aligns with the service-dominant (S-D) logic (Vargo and Lusch 2004) that has evolved into an important paradigm in the marketing area.

P3 . Cash Costs You: The Pain of Holding

J Zenkic, University of Melbourne, Australia Kobe Millet, Vrije Universiteit Amsterdam, The Netherlands Nicole Mead, University of Melbourne, Australia

Cash can have an associated psychological cost: the pain of holding. This pain is aversive varies with the physical properties of money – such as the differences in weight and bulk between coins and notes - and may in turn encourage needless spending.

H12. Does Economic Development Influence Consumer Innovativeness?

Fuchun Zhan, University of Wisconsin - Madison, USA Nancy Wong, University of Wisconsin - Madison, USA Julie Anne Lee, University of Western Australia, Australia

This research explores the relationship between income frugality and consumer innovativeness. Results show that income positively affects innate innovativeness but only for the higher income countries while an indirect effect through frugality is found in lower income countries. Frugality has no direct effect on intention of innovative behavior. 938 / Working Papers I1. Blaming Him or Them? A Study on Attribution Behavior

Chun Zhang, University of Dayton, USA Michel Laroche, Concordia University, Canada Yaoqi Li, Sun Yat-Sen University, China

The current work examines the gender difference in attribution behavior. Results show that males are more likely to blame individuals while females are more likely to blame groups which can be explained by relational and collective interdependent self-construal. Furthermore similarity and service failure magnitude moderate the gender effect.

D8 . Why Employees Communicate Positive eWOM on Social Networking Sites: Motivations and Moderators

Jing Zhang, Huazhong University of Science and Technology School of Management, China Ya Zhang, Huazhong University of Science and Technology School of Management, China

This paper aims at exploring what motivate employees to voluntarily create positive eWOM on SNS. Its theoretical contributions and managerial implications are connecting eWOM and OCB literature from perspective of internal employees distinguishing from employees’ knowledge sharing research elaborating the way to motivate employees to create positive eWOM for managers.

M11. Motivated to Be Moral: The Effect of Nostalgia on Consumers’ Recycling Behavior

Xiadan Zhang, Renmin University of China, China Xiushuang Gong, Jiangnan University, China Jing Jiang, Renmin University of China, China

Prior research on nostalgia has primarily focused on its psychological and social functions. However little has been done about whether and how nostalgia affects individuals’ pro-environmental behaviors. Through two studies we demonstrate that nostalgia can increase consumers’ intention to recycle and this effect is mediated by moral purity.

J2. Consistence vs. Variety: The Effect of Temporal Orientation on Variety Seeking

Yuan Zhang, Xiamen University, China Shaoqing Zhang, Quanzhou Normal University, China

The research examines the effect of temporal orientation on variety seeking and demonstrates that people in past orientation chose less variety than those in future orientation (Study 1) and preference for consistence was the underlying mechanism (Study 2). However the effect disappeared in self-irrelatively condition (Study 3).

K3. Goal or Knowledge? Exploring the Nature of Culture and its Consequential Effect

Xiaohua Zhao, Tsinghua University, China Fang Wan, University of Manitoba, Canada Antonios Stamatogiannakis, IE Business School, IE University, Spain Haiyang Yang, Johns Hopkins University, USA

Prior research has largely focused on how cultural values get activated but has not distinguished whether cultural values are semantic knowledge or goals. As the salience of them tends to fluctuate and subsequently impact judgments and behaviors we proposed an approach to distinguish culture values between semantic knowledge and goals.

R11. The Influence of Brand Rituals on Perceived Brand Authenticity

Lijing Zheng, University of Hong Kong, China Echo Wen Wan, University of Hong Kong, China Zhongqiang (Tak) Huang, University of Hong Kong, China

Built on literature of ritual and brand authenticity we predicted that consumers will perceive greater brand authenticity from brand rituals and that this effect is driven by perceived symbolism in the brand. Three studies provide initial evidences for our hypotheses. This research provides new insights into the influence of brand rituals. Advances in Consumer Research (Volume 46) / 939 R10. Emotional-Transference or Exclusivity? an Emotional Attachment Approach to Brand Extension for Cultural and Creative Products

Wu Zhiyan, Shanghai University of International Business and Economics, China Luo Jifeng, Shanghai Jiao Tong University, China Liu Xin, Shanghai University of International Business and Economics, China

This research uses three studies to examine how brand fit and self-construal moderate the relationship between brand attachment (emotional-transference versus exclusive) and the brand extension effect and how online social interaction moderates the moderating effect of self-construal on the relationship between brand attachment and the brand extension effect.

H4. Anthropomorphism Moderates the Effect of Ownership on Self Perceptions

Qiang Zhou, University of Texas at San Antonio, USA Dengfeng Yan, University of Texas at San Antonio, USA

Previous literature has shown that consumers often perceive themselves to have the characteristics of the products that they own. We proposed and found that this effect is mitigated by anthropomorphism because individuals are less likely to include a product into the self-category once it is anthropomorphized.

R2 . Brand-to-Brand Communications: How Consumers React to Flattery Between Brands

Lingrui Zhou, Duke University, USA Katherine Crain, Duke University, USA Keisha Cutright, Duke University, USA

Should brands go out of their way to talk positively about their competitors? We find that brand-to-brand flattery changes the consumer’s perception of how a brand values its relationships globally which influences brand trust and subsequently brand evaluations. 940 / Working Papers DataBlitz

Always Trust in Your Friends? Cross-cultural Effects of Review Source and Incentives on Trustworthiness

Dionysius Ang, Leeds University Business School, UK

Will culture affect how consumers perceive incentivized reviews from friends? Amongst Taiwanese participants incentivized reviews were deemed as breach of trust. Thus incentivized reviews from friends were deemed less trustworthy than incentivized reviews from strangers. Amongst Americans reviews from friends were more trustworthy than reviews from strangers regardless of incentive.

Consuming Time-Space Imaginations: Bakhtin’s Chronotope on Robots and Artificial Intelligence

Marat Bakpayev, University of Minnesota Duluth, USA Alima Yesmukanova, KIMEP University, Kazakhstan

We examine developing discourse on two controversial phenomena - robots and Artificial Intelligence. Drawing from literary theory we focus on the representation of time-space in news articles. We contextualize Bakhtin’s theory of chronotope identifying how temporality and spatiality presented by technology narratives.

Teaching Old Dog New Tricks… and Old Bottles New Jeans . The Role of Implicit Theories in the Evaluation of Recycled Products

Alessandro Biraglia, University of Leeds, UK J. Josko Brakus, University of Leeds, UK Lucia Mannetti, Sapienza University of Rome, Italy Ambra Brizi, Sapienza University of Rome, Italy

Across three studies involving real choice implicit associations and purchase intention we show how incremental theorists prefer recycled products more than entity theorists. Entity theorists can nevertheless be persuaded in buying more recycled products if they attribute the merit of the transformation to the internal characteristics of the materials.

Understanding the Role of Gifts in Managing Marriage and Family Relations: The Case of the Male Phoenix in China

Jia Cong, Lancaster University, UK Xin Zhao, Lancaster University, UK Chihling Liu, Lancaster University, UK

This paper seeks to understand the role of gifts in managing an inter-class marriage from the perspective of the-male-phoenix in contemporary China. Through fieldwork our initial findings show that gift plays a significant role for male-rural-migrants to manage class disparity and maintain harmonious relationships with urban families they marry into.

What a Waste: Effects of Food Consumption Choices and Traffic Light Labelling on Disposal

Marit Drijfhout, University of Groningen, The Netherlands Jenny van Doorn, University of Groningen, The Netherlands Koert van Ittersum, University of Groningen, The Netherlands

The large amount of food wasted by consumers suggests a mismatch between consumers’ planned consumption decisions and their actual consumption behavior. We show that food waste occurs when consumers impulsively prefer unhealthy vice foods over healthy virtue foods initially chosen and that traffic lights aggravate rather than mitigate food waste.

Xiaomeng Fan, Northwestern University, USA Galen V. Bodenhausen, Northwestern University, USA

This research shows that evaluative conditioning involves both associative and relational processing. Relational processing can be promoted when people can allocate their cognitive resources in evaluations.

“It’s Not You, It’s Me”: How Corporate Social Responsibility Decreases Customer Citizenship Behavior

Sofia Batista Ferraz, EAESP-FGV, Brazil Andres Rodriguez Veloso, University of São Paulo, Brazil Diogo Hildebrand, Baruch College, USA

This paper analyzes how CSR activities influence helping behaviors among customers. Across two studies we demonstrate that participants faced with an opportunity to satiate self-defining goals through CSR had a reduced propensity to help other customers. Besides we propose entitativity as a driver of customer helping behavior.

‘Family Tech-Support’: Consequences for Family Assemblages and Non-Purchase Decision Technology Adoption

Pao Franco, University of Melbourne, Australia

Consumer technology adoption research frames tech products entering into consumers’ lives as purchase decisions (e.g. computers wearables). What about situations that are not purchase decisions? This research explores technology adoption in non-decision situations (e.g. shared-at-home gifts) through a qualitative investigation of families and ‘family tech-support’ – utilizing an assemblage approach.

Is CSR for Sale? Investigating the Effects of Acquisition of Socially Responsible Brands on CSR Perceptions

Bianca Grohmann, Concordia University, Canada Argiro Kliamenakis, Concordia University, Canada H. Onur Bodur, Concordia University, Canada

This research examines whether firms can increase consumers’ CSR perceptions by acquiring socially responsible brands. Two experiments demonstrate that (1) firms are more likely to increase CSR perceptions through acquisition of a functional (vs. symbolic) brand and (2) for symbolic brands new brand development (vs. acquisition) enhances CSR perceptions.

A Penny for Self-disgust: The Effects of Favorable Review Reward on Consumers Behavior

Qingqing Guo, Shanghai Jiao Tong University, China Liangyan Wang, Shanghai Jiao Tong University, China Bing Han, Shanghai Jiao Tong University, China

Through three experiments this study shows that a monetary reward has a backfire effect on consumer behavior because it leads receiving consumers to self-disgust for the referral. This effect is attenuated for consumer who has an exchange relationship norm while sharpened for consumers who has a communal relationship norm.

Less Time, More Procrastination? The Impact of Time Pressure on Task Initiation

Jing Jiang, Renmin University of China, China Alisa Yinghao Wu, Columbia University, USA

When under time pressure individuals are more likely to postpone the initiation of tasks—whether these are routine errands or consumption tasks. The rationale is that time pressure impairs an individual’s momentary ability to focus on tasks and leads to subsequent postpone- ment of task initiation. Advances in Consumer Research (Volume 46) / 943 The Effects of Glossy Versus Matte Imagery on Consumers’ Decision Making

Yoonho Jin, INSEAD, Singapore Amitava Chattopadhyay, INSEAD, Singapore

Three studies collectively suggest that the use of glossy imagery in advertising because of its embodied meaning of easygoingness could increase consumers’ optimistic perception of target consumption thereby enhancing purchase intention. The studies also show that priming either the danger of slipperiness or optimistic orientation may attenuate the effects.

When Humans Consume Humanlike Animals: Anthropomorphism, Power, and Cruelty-free Consumption

Ji Myoung Danny Kim, University at Buffalo, USA Sunyee Yoon, University at Buffalo, USA

The interaction of anthropomorphism of animals and consumers’ sense of power explains preference and attitude towards cruelty-free products. The research suggests low-power individuals engage in prosocial behavior selectively depending on perceived similarity with the beneficiary.

When CSR Becomes a Liability for Firms in Crises: Effects on Perceived Hypocrisy and Consumer Forgiveness

Argiro Kliamenakis, Concordia University, Canada H. Onur Bodur, Concordia University, Canada

This research examines circumstances under which positive CSR reputation amplifies negative consumer responses to firm crises. Al- though a positive CSR reputation becomes a liability for firms experiencing values-related crises it may still offer a shielding effect but only for ambiguous crises (i.e. crises that may or may not constitute transgressions).

Morality Matters in the Marketplace: The Influence of Morally Based Attitudes on Consumer Purchase Intentions

Andrew Luttrell, Ball State University, USA Jacob Teeny, Ohio State University, USA Richard Petty, Ohio State University, USA

In social psychology morally based attitudes have been shown to strongly predict attitude-consistent behaviors. However whether this basis predicts equally well in consumer research remains unanswered. Across five studies we show that morally based attitudes are indeed a reliable predictor of consumer purchase intentions even above other important factors.

Tackling Over-Consumption: How Proximal Depictions of Unhealthy Food Products Influence the Consumption Behavior

Sumit Malik, IE Business School, IE University, Spain Eda Sayin, IE Business School, IE University, Spain Kriti Jain, IE Business School, IE University, Spain

Do spatially proximal depictions of unhealthy (but not healthy) food products within an advertisement image increase the desire to eat? Across a pilot and three main studies we find support for the influence of proximal ad depictions on food consumption outcomes. These results have implications for the phenomena of over-eating. 944 / Working Papers Stating the Obvious: How “Ugly” Labels Can Increase the Desirability of Odd-Shaped Produce

Siddhanth Mookerjee, University of British Columbia, Canada Yann Cornil, University of British Columbia, Canada Joey Hoegg, University of British Columbia, Canada

Food producers and retailers throw away large amounts of food that fail to meet beauty standards. Our research demonstrates that using the words “ugly” as part of an advertising message increases the desirability of odd-shaped produce that would otherwise go to waste. Per- ceived brand credibility mediates this effect.

The Influence of Goal Specificity on Planning Effectiveness: A Meta-Analysis

Ekaterina Napolova, University of Technology Sydney, Australia Francois Carrillat, University of Technology Sydney, Australia Leona Tam, University of Technology Sydney, Australia

This meta-analytic study aims to investigate the effectiveness of various planning strategies depending on the level of goal specificity. The paper presents the largest meta-analysis on the implementation intentions (planning) effectiveness in the health domain and provides important insights for researchers consumers practitioners and policy makers.

Thank You = Trust Me: When Gratitude Expressions Help Promote New Products

Maria Ng, University of Houston, USA Dejun Tony Kong, University of Houston, USA Vanessa Patrick, University of Houston, USA

Drawing upon risk and trust theories we find in four experiments that in one-off (transactional) new product promotion interactions a salesperson’s gratitude (vs. neutral) expression could elicit consumers’ felt appreciation and then influence their product evaluation via enhanced trust and reduced psychological risk. These positive reactions drive consumers’ purchase intentions.

How Matte Product Surface Enhances Perceived Durability

Taehoon Park, University of South Carolina, USA Junghan Kim, Singapore Management University, Singapore

Prior literature has shown that different perceptual qualities can be inferred from object surface features. Extending prior work we propose reflected light on surface (glossy vs. matte) as an inferential cue that can shape product quality. Across three studies we show that matte (vs. glossy) product surface enhances perceived durability.

Influence of Visual Crowding and Space Between Products on Consumer Choice

Ana Scekic, HEC Paris, France Selin Atalay, Frankfurt School of Finance and Management, Germany Cathy Liu Yang, HEC Paris, France Peter Ebbes, HEC Paris, France

We investigate the effect of assortment size and of spacing between products on consumer choice. We provide initial evidence that in large (i.e. crowded) assortments distancing the products on the shelf from one another leads to greater perceived variety in the assortment and to a choice experience perceived as easier. Advances in Consumer Research (Volume 46) / 945 The Interaction Effect of Food Variety and Simulation of Eating on Consumers’ Calorie Estimation

Liang Shen, Shanghai Jiao Tong University, China Fengyan Cai, Shanghai Jiao Tong University, China Ying Yu, Huazhong Agricultural University, China

Given the inconsistent findings about the relationship between food variety and calorie estimation this research proposes simulation of eating as one moderator. Specifically we propose that when a simulation of eating is (not) stimulated food variety decreases (increases) consumers’ calorie estimation. Three experiments demonstrated this proposition and its underlying mechanism.

Gaze Reflects Loss Aversion

Feng Sheng, University of Pennsylvania, USA Arjun Ramakrishnan, University of Pennsylvania, USA Darsol Seok, University of Pennsylvania, USA Puti Cen, University of Pennsylvania, USA Michael Platt, University of Pennsylvania, USA

By modeling individuals’ gaze traces on potential gains and losses while they decided whether to accept or reject gambles in an eye-tracking experiment we demonstrated that loss aversion reflected the overweighting of information about potential losses relative to potential gains sampled during visual inspection of gambles.

Memory-Based Models of Predicting Inferences about Brand Quality

Yvetta Simonyan, University of Bath, UK Dan Goldstein, Microsoft Research, USA

How are consumers’ inferences about product quality related to information in memory? By modelling inferences as well as confidence in inferences via belief distributions this paper aims at explaining some unsolved phenomena regarding the relationship between quality perceptions on one side and recognition and other memory information on the other.

Digital Storytelling and Post-Trust Online Sperm Marketing

Jennifer Takhar, Institut Supérieur de Gestion, Paris, France Laetitia Mimoun, HEC Paris, France

This paper examines extended sperm donor profiles and blogs from two international sperm bank websites and reveals the post-trust digital storytelling strategies they use to convince prospective buyers. We also identify three dimensions of post-trust digital marketing: over- determined authenticity consumer fantasy and loss of distinction.

When Waste Costs: The Influence of Price on Consumers’ Perceived Waste and Purchase Intention of an Excessive Amount of Product

Tao Tao, Hong Kong Baptist University, China Robert Wyer Jr., University of Cincinnati, USA

This paper suggests that price can serve as an indicator of waste perception and help to decrease excessive purchase behavior. Two studies supported our proposition that controlling the price as the same consumer’s purchase intention of an excessive offer over an adequate one is lower when the price is higher. 946 / Working Papers Paper Box or Plastic bag? Structural Package Design Elements Affect Health Perception and Consumption.

Joyce De Temmerman, Ghent University, Belgium Iris Vermeir, Ghent University, Belgium Hendrik Slabbinck, Ghent University, Belgium

This paper explores whether structural package design elements bias health perception and consumption. We propose that material (paper vs. plastic) and format (box vs. bag) can lead to higher health perceptions of the package and the packaged product and hence increase consumption.

Prices in Red: When a Red Price Becomes a Stop Sign

Hongjun Ye, Drexel University, USA Siddharth Bhatt, Drexel University, USA Rajneesh Suri, Drexel University, USA

“Sale” and “stop” are both referential meanings attached to the color red. This study shows that prices presented in red in store flyers influence consumers memory of prices and their evaluation of the deal of not just the focal product but also the products surrounding the red price.

Safety or Luxury: The Effect of Competitiveness on Consumer Preference in Social Crowding

Lijun Zhang, Nanyang Technological University, Singapore Yee Ling, Elaine Chan, Nanyang Technological University, Singapore

Prior research suggests that people prefer safety products in crowded environments. This research contributes to the literature by examining the moderating effect of competitiveness. While people low in competitiveness prefer safety products people high in competitiveness prefer status consumption over safety consumption in crowded environments. Author Index

Bagozzi, Richard P...... 462, 935 Bodenhausen, Galen V...... 498, 942 A Bai, Yuxin ...... 465 Bodur, H. Onur ...... 816, 942, 943 Aaker, Jennifer ...... 303 Bajde, Domen ...... 231 Böhm, Robert ...... 620 Abbasi, Amir Zaib ...... 817 Bakpayev, Marat ...... 892, 941 Bolderdijk, Jan Willem ...... 654 Abell, Annika ...... 897 Banerjee, Syagnik ...... 110 Bolton, Lisa ...... 895 Abeywickrama, Ravini Savindya . . . . .897 Banker, Sachin ...... 226 Bonezzi, Andrea ...... 63 Abi, Gulen Sarial ...... 899 Banovic, Marija ...... 899 Bonsu, Samuel K...... 563 Achar, Chethana ...... 309, 451 Barasch, Alixandra . . .28, 209, 220, 283, 357 Bos, Maarten ...... 63 Affonso, Felipe Marinelli ...... 453 Barasz, Kate ...... 199, 536 Botelho, Delane ...... 906, 918 Agarwal, James ...... 937 Barauskaite, Dovile ...... 541 Bothello, Joel ...... 333 Aggarwal, Pankaj ...... 835 Barauskaitė, Dovilė ...... 467 Botti, Simona ...... 775 Agrawal, Nidhi ...... 309, 451, 794 Barbera, Michael ...... 469 Boven, Leaf Van ...... 706 Ahluwalia, Rohini ...... 732 Barboza, Renata Andreoni ...... 369 Boztug, Yasemin ...... 610 Ahmed, Sohel ...... 817 Bardhi, Fleura ...... 214, 920, 928 Brace-Govan, Jan ...... 636 Ahn, Regina ...... 930 Barnea, Uri ...... 899 Bradlow, Eric T...... 204 Ai, Sizhi ...... 897 Barnes, Aaron J...... 314 Brakus, J. Josko ...... 941 Akaka, Melissa Archpru . . . . .98, 897, 921 Barnett, Samuel B...... 374 Brasel, S. Adam ...... 68 Akpinar, Ezgi ...... 523, 901 Barnhart, Michelle ...... 333 Breaz, Monique ...... 610 Aksit, Mina Seraj ...... 458 Barnier, Virginie de ...... 420 Brick, Danielle J...... 144 Aksu, Gunben Ceren ...... 897 Barros, Lucia S. G...... 471 Briers, Barbara ...... 909 Albinsson, Pia Annette ...... 898 Bart, Yakov ...... 666 Brinberg, David ...... 581 Aleksic, Darija ...... 924 Bartholomew, Darrell Eugene ...... 891 Brinkmann, Lynn ...... 922 Alemany-Oliver, Mathieu ...... 730 Bas, Burcak ...... 899 Brito, Eliane Zamith ...... 214, 276 Allard, Thomas ...... 328, 340 Baskin, Ernest ...... 39, 509 Brizi, Ambra ...... 941 Allen, BJ ...... 671 Batat, Wided ...... 899, 899 Brogliato, Marcelo Salhab ...... 471 Almaguer, Jacob ...... 769 Batista, Juliana M...... 900 Brooks, Alison Wood ...... 28, 677 Althuizen, Niek ...... 898 Battistella-Lima, Suzana Valente . . . . .906 Brooks, Gillian ...... 489 Alvarado-Karste, Diego ...... 898 Bayer, Rabia ...... 647 Brucks, Merrie ...... 138 Amar, Moty ...... 455, 930 Bechler, Christopher ...... 252 Brügger, Adrian ...... 909 Amir, On ...... 220, 242, 357 Bedi, Suneal ...... 476 Brunk, Katja H...... 402 Anderberg, Shivaun ...... 898 Bei, Lien-Ti ...... 378 Bucchia, Céline Del ...... 491 Anderson, Eric T...... 328 Belk, Russell ...... 33, 603 Buechel, Eva C...... 39, 357 Andonopoulos, Vicki ...... 469 Belk, Russell W...... 557 Bullard, Olya . . . . .493, 495, 900, 919, 919 Andrade, Eduardo ...... 608 Bellezza, Silvia ...... 165 Burchett, Molly ...... 900 Andrade, Eduardo B...... 471, 825 Benartzi, Shlomo ...... 775 Burkhardt, Jesse ...... 658 Ang, Dionysius ...... 941 Bennett, Andrea Rochelle ...... 900, 900 Burlat, Claire ...... 491 Aquino, Karl ...... 340 Bennett, Aronté Marie ...... 393 Burman, Bidisha ...... 898 Arens, Zachary ...... 911 Benoit, Ilgım Dara ...... 479 Burson, Katherine ...... 904 Argo, Jennifer ...... 209, 839 Berger, Jonah . . .264, 283, 363, 736, 819, 893 Buschgens, Mark ...... 382 Arias, Robert ...... 231 Bergh, Bram Van den ...... 805 Büttner, Oliver ...... 803 Ariely, Dan ...... 252, 318, 328 Berinsky, Adam ...... 893 Büttner, Oliver B...... 901, 922 Arnold, Eric ...... 352 Berman, Jonathan ...... 28, 283 Aron-Wisnewsky, Judith ...... 258 Berman, Jonathan Z...... 57 Arora, Poonam ...... 583 Berman, Jonathan Zev ...... 893, 922 C Arora, Sandeep ...... 149 Bernadic, Ursa ...... 664 Cadario, Romain ...... 497 Arsel, Zeynep ...... 895 Bertilsson, Jon ...... 692 Cai, Fengyan ...... 45, 928, 945 Arthur, Nathalia S...... 891 Bezençon, Valéry ...... 481 Cai, Fengyan (Cindy) ...... 498 Ashworth, Laurence ...... 763 Bharadwaj, Anupama Mukund ...... 483 Caldwell, Marylouise ...... 499 Askegaard, Søren ...... 50 Bharti, Mehak ...... 485 Campbell, Colin ...... 288 Atakan, S. Sinem ...... 458 Bhatia, Sudeep ...... 204, 204 Campbell, Margaret C...... 340, 893 Atalay, A. Selin ...... 773 Bhatnagar, Kushagra ...... 121 Campero, Andrés ...... 374 Atalay, Selin ...... 901, 944 Bhatt, Siddharth ...... 946 Campo, Sara ...... 901 Atik, Deniz ...... 460, 503 Bhattacharjee, Amit ...... 57, 57 Cannon, Christopher ...... 165 Atlas, Stephen ...... 23 Bi, Sheng ...... 900 Cao, Fei ...... 901 Auschaitrakul, Sumitra . . . . .898, 913, 914 Biliciler-Unal, Gunes ...... 68 Caprioli, Sara ...... 237 Awad, Norah ...... 899 Binay, Itir ...... 521 Cardoso, Flavia ...... 276 Ay, Ufuk ...... 643 Biraghi, Silvia ...... 413 Carpenter, Gregory ...... 50 Aydinli, Aylin ...... 911 Biraglia, Alessandro ...... 941 Carrillat, Francois ...... 944 Biswas, Dipayan ...... 487, 897, 929 Carter, Erin Percival ...... 57 Bitterly, Brad ...... 28 Carvalho, Lilian ...... 915 B Blair, Sean ...... 74, 149 Castaño, Raquel ...... 170, 889 Bagchi, Rajesh . . . . . 45, 45, 110, 160, 293 Blanchard, Simon J...... 857 Castelo, Noah ...... 63, 177, 183 Bagozzi, Benjamin E...... 462 Bockenholt, Ulf ...... 530 Advances in Consumer Research 947 Volume 46, ©2018 948 / Author Index

Catapano, Rhia ...... 303 Coronel, Mariela ...... 931 Dube, Laurette ...... 917 Catlin, Jesse R...... 501 Cossu, Martina ...... 931 Dubois, David ...... 165, 919 Cavusoglu, Lena ...... 503 Costello, John P...... 104, 517 Duclos, Rod ...... 293, 340, 927 Celsi, Mary ...... 288 Cotte, June ...... 719, 895 DuFault, Beth Leavenworth ...... 50 Cen, Puti ...... 945 Coulter, Robin A...... 846, 921 Duhachek, Adam ...... 183, 937 Cerf, Moran ...... 374 Crain, Katherine ...... 939 Duke, Kristen ...... 242, 357 Ceylan, Melis ...... 901 Crain, Katherine M...... 363 Duncan, Shannon ...... 530 Chaabane, Aïda Mimouni ...... 886 Crener-Ricard, Sandrine ...... 571 Dunn, Elizabeth W...... 303 Chabert, Michele ...... 258 Cristian, Daniela ...... 519, 915 Dunn, Lea ...... 483 Chae, Boyoun Grace ...... 686 Critcher, Clayton R...... 28, 199 Dunn, Timothy ...... 23 Chakrabarti, Avik ...... 462 Croft, Robin ...... 903 Duque, Lola C...... 827 Chan, Cindy ...... 677 Crolic, Cammy ...... 80 Durante, Kristina ...... 819 Chan, Kimmy ...... 902 Cruz, Angela Gracia B...... 521, 636 Dutton, Chaumanix ...... 905 Chandon, Pierre ...... 80, 258, 497 Cui, Shiliang ...... 605 Chang, Angela ...... 917 Cunha Jr., Marcus ...... 507 Chang, Chia-Chi ...... 863 Cuny, Caroline ...... 653 E Chang, Chun-Tuan ...... 505, 675 Curry, David J...... 86, 92 Ebbes, Peter ...... 773, 944 Chang, En-Chung ...... 934, 935 Cutright, Keisha ...... 939 Egan, Daniel ...... 23 Chang, Hui-Wen ...... 675 Cutright, Keisha M...... 303 Ehrensperger, Elena ...... 905 Chang, Yaping ...... 932 Ein-Gar, Danit ...... 532 Chapman, Gretchen ...... 932 Einwiller, Sabine ...... 922 Chartrand, Tanya ...... 909 D Elahi, Hamid ...... 927 Chater, Nick ...... 575 Dacin, Peter A...... 920 Elder, Ryan S...... 188 Chatterjee, Lagnajita ...... 902 Dadanlar, Hazel H...... 769 Elliot, Steve ...... 499 Chatterjee, Subimal ...... 702, 901 Dahl, Darren ...... 810, 893, 935 Elshiewy, Ossama ...... 534 Chattopadhyay, Amitava . . . . 618, 829, 943 Dai, Hengchen ...... 657 Engeler, Isabelle ...... 536 Cheema, Amar ...... 220 Dallolio, Adriana Schneider ...... 214 Engeset, Marit ...... 929 Chen, Fangyuan ...... 587 Dalton, Amy ...... 687, 918, 918 Englert, Natalia ...... 921, 924 Chen, Haipeng ...... 895 Danckwerts, Sebastian ...... 903 Epley, Nicholas ...... 283 Chen, Jiemiao ...... 861 Daniels, Michelle ...... 177 Erkhova, Daria ...... 905 Chen, Johnny ...... 573 Dannewald, Till ...... 610 Erner, Carsten ...... 833 Chen, Mengshu ...... 923 Danziger, Shai ...... 915, 927 Ertimur, Burçak ...... 98, 288 Chen, Qihui ...... 165 Davidai, Shai ...... 357 Estes, Zachary ...... 787, 803, 931 Chen, Rocky Peng ...... 902 Dean, Dianne ...... 903 Etkin, Jordan ...... 188 Chen, Rui ...... 507, 935 DeBerry-Spence, Benét ...... 50 Euler, Daria ...... 901 Chen, Stephanie Y...... 237 Dellande, Stephanie ...... 288 Evangelidis, Ioannis ...... 199, 538 Chen, Steven ...... 98 Delvicchio, Devon ...... 844 Evers, Ellen R. K...... 318 Chen, Yifan ...... 934 Demczuk, Rafael ...... 919 Chen, Yu ...... 897 Demir, Kivilcim Dogerlioglu ...... 523 Chen, Yunqing ...... 363, 902 Deng, Qian ...... 86 F Cheng, Andong ...... 509 Deng, Qian (Claire) ...... 92 Cheng, Chieh-Wen ...... 902 Falk, Tomas ...... 916 Deng, Xiaoyan ...... 144, 874, 930 Falkowski, Andrzej ...... 911 Cheng, Li-Keng ...... 378 DePaoli, Alexander ...... 357 Cheng, Yin-Hui ...... 903, 903 Fan, Linying ...... 110 Desiraju, Shweta R...... 323 Fan, Tingting ...... 539 Cherup, Abigail Nappier ...... 510 DeVoe, Sanford E...... 104 Chevtchouk, Yanina ...... 758 Fan, Xiaomeng ...... 498, 942 Dewhirst, Timothy ...... 903 Fan, Xiucheng ...... 529 Cho, Young Joo ...... 512 Dhaliwal, Nathan ...... 115 Choe, Yuna ...... 247, 902 Fan, Yafeng ...... 567, 905 Dias, Helison Bertoli Alves ...... 904 Farmer, Matthew ...... 905 Choi, Incheol ...... 220 Diehl, Kristin ...... 905 Choi, Woojin ...... 902 Faro, David ...... 775 Dietvorst, Berkeley J...... 525 Feemster, Kristen ...... 309 Chou, Hsuan-Yi ...... 902 Ding, William ...... 527 Chu, Charlene ...... 514 Felix, Reto ...... 882 Ding, Ying ...... 567 Fennis, Bob ...... 519 Chu, Wujin ...... 913 Ding, Yu ...... 138, 154, 188 Chu, Xing-Yu (Marcos) ...... 505, 902 Fennis, Bob M...... 467, 541 Doering, Tim Philipp ...... 904 Ferguson, Gail ...... 891 Chuang, Shih-Chieh ...... 903, 903 Dolbec, Pierre-Yann ...... 98, 333 Chun, Libby Youngjin ...... 912 Ferguson, Melissa ...... 800 Dommer, Sara Loughran ...... 237, 270 Fernbach, Philip ...... 115, 298 Chung, Emily ...... 214 Dong, Ping ...... 194 Chung, Jaeyeon ...... 194 Fernbach, Philip M...... 23, 23, 670 Donnelly, Grant E...... 104, 323 Ferraro, Rosellina ...... 177 Chung, Myungjin ...... 515 Doorn, Jenny van ...... 183, 941 Cian, Luca ...... 68 Ferraz, Sofia Batista ...... 942 Drake, Carly ...... 904, 904 Ferreira, Daniela ...... 905 Clair, Julian K Saint ...... 386 Drenten, Jenna ...... 231, 895 Clément, Karine ...... 258 Ferreira, Kirla C ...... 906 Drijfhout, Marit ...... 941 Ferreira, Marcia Christina ...... 214 Cocker, Hayley ...... 465 Drozdova, Natalia ...... 391 Coleman, Nicole Verrochi ...... 74, 270 Fertig, Andrew ...... 298 Drummond, Caitlin ...... 115 Figueiredo, Bernardo ...... 382, 543 Cong, Jia ...... 941 Du, Jiangang ...... 529 Connell, Paul M...... 309 Finkelstein, Stacey R...... 309 Du, Rui ...... 904 Fırat, A. Fuat ...... 460 Conroy, Denise M ...... 398 Duan, Xingyu ...... 842 Cornil, Yann ...... 80, 258, 340, 944 Fischer, Eileen ...... 98, 170 Duani, Nofar ...... 209 Fischhoff, Baruch ...... 115 Advances in Consumer Research (Volume 46) / 949

Fishbach, Ayelet ...... 154, 209, 318 Gollnhofer, Johanna ...... 352 Hart, Claire ...... 914 Fisher, Robert J...... 264 Gong, Wu ...... 917 Hartmann, Benjamin Julien ...... 402 Fitzsimons, Gavan ...... 127, 907, 934 Gong, Xiushuang ...... 567, 938 Hasford, Jonathan ...... 900 Fitzsimons, Gavan J...... 133, 144, 790 Gonzales, Gabriel E...... 569 Häubl, Gerald ...... 33, 220, 242 Flores, Eliani C...... 891 Goodman, Joe ...... 86, 92 Haws, Kelly L...... 127, 876 Foos, Adrienne E...... 545 Goodman, Joseph K...... 220 He, Dongjin ...... 144, 587, 909 Fox, Craig R...... 833 Goor, Dafna ...... 571 He, Sharlene ...... 328, 585 Franco, Pao ...... 942 Gorn, Gerald J...... 144, 798 He, Xun ...... 909 Frankin, Drew ...... 694 Goswami, Indranil ...... 194 He, Yongfu ...... 870 Frethey-Bentham, Catherine ...... 469 Gounder, Babu ...... 160 He, Yuanqiong ...... 908 Fu, Ning ...... 901 Graeber, Justin ...... 906 Held, Johanna ...... 487 Fuchs, Christoph ...... 63, 237 Granulo, Armin ...... 63 Hellén, Katarina ...... 591 Fuschillo, Gregorio ...... 692 Gray, John ...... 893 Hemetsberger, Andrea ...... 352 Grégoire, Yany ...... 910 Henderson, Marlone D...... 630, 633 Grewal, Dhruv ...... 183 Henry, Paul ...... 499 G Grewal, Lauren ...... 127 Heo, Cindy Yoonjoung ...... 916 Gabriel, Kelly ...... 393 Grier, Sonya ...... 50, 890 Hernandez, Jose Mauro C...... 891, 900 Gai, Phyliss J...... 547, 549 Griffin, Dale ...... 328 Hershfield, Hal ...... 160 Gal, David ...... 902 Grillo, Tito L. H...... 23 Hershfield, Hal E...... 237, 303 Gallo, Iñigo ...... 551 Grimes, Jerry Lewis ...... 907 Heuvinck, Nico ...... 593 Gambetti, Rossella ...... 413 Grimm, Pamela E...... 170 Hildebrand, Christian ...... 33 Gan, Caixia ...... 398 Griskevicius, Vladas ...... 165 Hildebrand, Diogo ...... 80, 942 Gan, Chin Chuan ...... 817 Groening, Christopher ...... 907 Hilden, Thomas ...... 33 Gao, Cheng ...... 906 Grohmann, Bianca ...... 942 Hiler, Jacob Lee ...... 890 Gao, Huachao ...... 553 Grunert, Klaus G ...... 926 Hingston, Sean T...... 849 Gao, Leilei ...... 363, 539, 902 Gu, Yangjie ...... 80 Hmurovic, Jillian ...... 127 Garbarino, Ellen ...... 898 Guèvremont, Amélie ...... 907 Hoang, Chi ...... 594, 596 Garbinsky, Emily ...... 323, 346 Guhl, Daniel ...... 534 Höchli, Bettina Rebekka ...... 909 Garcia-Collart, Tessa ...... 906 Gullo, Kelley ...... 144, 907 Hoegg, JoAndrea ...... 483 Garcia-Rada, Ximena ...... 346 Gunasti, Kunter ...... 133 Hoegg, Joey ...... 810, 944 Garcia-Ramon, Lorena ...... 769 Guo, Bingxuan ...... 917 Hoffman, Donna ...... 33 Garg, Nitika ...... 80 Guo, Qingqing ...... 907, 942 Hoffmann, Carina ...... 909 Gauri, Dinesh ...... 837 Guo, Wenxia ...... 133 Hong, Jennifer Seokhwa ...... 293 Gaustad, Tarje ...... 133 Guo, Zhaoyang ...... 922 Howard, Chuck ...... 247 Gaviria, Pilar Rojas ...... 276 Gupta, Aditya ...... 908 Howe, Holly S ...... 909 Gaviria, Pilar Silveira Rojas ...... 231 Gupta, Tanvi ...... 908 Howlett, Betsy ...... 923 Geisler, Markus ...... 895 Gurel-Atay, Eda ...... 573 Hoyer, Wayne ...... 905 Gelbrich, Katja ...... 931 Gurhan-Canli, Zeynep ...... 647 Hoyer, Wayne D...... 170 Gentry, James ...... 908 Gurney, Nikolos ...... 575 Hsee, Christopher K...... 199, 258 Georgiev, Nikolay ...... 555 Gurrieri, Lauren ...... 895 Hsieh, Meng-Hua ...... 598 Gershoff, Andrew D ...... 904 Gustafsson, Anders ...... 912 Hsu, Ming ...... 204, 878 Gershon, Rachel ...... 57 Gvili, Yaniv ...... 455, 930 Hu, Feihong ...... 909 Geuens, Maggie ...... 931 Hu, Mandy Mantian ...... 933 Ghosh, Anastasiya Pocheptsova . 252, 252, 742 Hu, Miao ...... 904 Ghoshal, Tanuka ...... 557, 921 H Huang, Bo ...... 910 Giambastiani, Gaia ...... 561 Habib, Rishad ...... 340 Huang, Chunxiang ...... 932 Gibbert, Michael ...... 750 Hachefa, Celia ...... 80 Huang, Guei-hua Flora ...... 675 Giesler, Markus ...... 288, 333, 402, 893 Hadi, Rhonda ...... 80 Huang, Jianping ...... 910 Gillespie, Brian ...... 906 Hagen, Linda ...... 127 Huang, Jiexian (Chloe) ...... 910, 910 Gillingham, Kenneth ...... 658 Hagerty, Serena ...... 577 Huang, Liang ...... 252, 600 Gilly, Mary ...... 288 Hagtvedt, Henrik ...... 68 Huang, Szu-chi ...... 188, 252, 808 Gilovich, Thomas ...... 551, 932 Hall, Matthew J...... 579 Huang, Xun (Irene) ...... 910 Gineikiene, Justina ...... 541 Hamby, Anne ...... 581 Huang, Yanliu ...... 936 Gineikienė, Justina ...... 467 Hamerman, Eric ...... 927 Huang, Zhongqiang (Tak) ...... 910, 938 Gino, Francesca ...... 39, 616 Hamilton, Rebecca ...... 893, 895 Huff, Aimee Dinnin ...... 333 Girardin, Florent ...... 481, 679 Hamilton, Ryan ...... 831 Hui, Miachel K...... 529 Gladstone, Joe ...... 346 Han, Bing ...... 908, 942 Humayun, Mariam ...... 603 Gloukhovtsev, Alexei ...... 916 Han, Eunjoo ...... 633, 640 Humphreys, Ashlee ...... 50 Gneezy, Ayelet ...... 363 Han, Hyerin ...... 138 Hung, Yu-chen ...... 936 Godefroit-Winkel, Delphine ...... 563 Han, Minju ...... 133 Hutchinson, J. Wesley ...... 204 Godfrey, D. Matthew ...... 121 Han, Yegyu ...... 110 Hwang, Jaewon ...... 867, 910 Goedegebure, Robert ...... 565 Handelman, Jay ...... 924 Hydock, Chris ...... 149, 149, 605 Goins, Kate ...... 900 Hansen, Jochim ...... 920 Hyodo, Jamie D...... 579 Goldenberg, Jacob ...... 188, 752 Hardesty, David ...... 895, 900 Goldsmith, Kelly ...... 270, 671, 895 Hardisty, David ...... 226, 247 Goldstein, Dan ...... 945 Hardisty, David J...... 45, 115, 328, 583 I Goldszmidt, Rafael ...... 825 Harnish, Elana ...... 890 Ibrahim, Nahid ...... 220 Goldzsmidt, Rafael ...... 608 Harrold, Mycah L ...... 908 II, Americus Reed ...... 859 950 / Author Index

Ikonen, Iina ...... 911 Kahle, Lynn R...... 573 Koukova, Nevena T...... 660, 756 Inbar, Yoel ...... 115, 318 Kähr, Andrea ...... 170 Koulaei, Afra ...... 915 Ince, Elise Chandon ...... 742 Kan, Christina ...... 247, 298, 902 Kowalczyk, Christine ...... 651 Inman, J. Jeffrey ...... 487 Kang, In-Hye ...... 177, 638 Kozinets, Robert ...... 413 Inman, Jeffrey ...... 893 Kang, Jiyun ...... 902 Kralj, Sonja N...... 915 Irmak, Caglar ...... 133, 914 Kapitan, Sommer ...... 640 Kramarczyk, Justyna ...... 730 Isaac, Mathew ...... 45 Kaplan-Oz, Begum ...... 734 Kramer, Thomas ...... 80, 133 Isaac, Mathew S...... 45 Karababa, Eminegül ...... 643 Krampe, Caspar ...... 925 Ishii, Hiroaki ...... 930 Karataş, Mustafa ...... 647 Krantz, David H...... 583 Isisag, Anil ...... 333 Karmarkar, Uma R...... 318 Krause, Rebecca J...... 662 Islam, Towhidul ...... 606 Kasani, Negin Latifi ...... 898 Krefeld-Schwalb, Antonia ...... 664 Islam, Zayed Bin ...... 606 Kedzior, Richard ...... 912 Krishna, Aradhna . . . 68, 154, 596, 787, 927 İşlek, Mahmut Sami ...... 643 Keeling, Debbie Isobel ...... 545 Kristensen, Dorthe Brogård . . . . . 50, 915 Ittersum, Koert van ...... 654, 929, 941 Keeling, Kathleen ...... 545 Kristofferson, Kirk ...... 177, 177, 309 Keh, Hean Tat ...... 861 Krohmer, Harley ...... 170, 905 Keinan, Anat ...... 571 Kronrod, Ann ...... 666, 915 J Keller, Punam A...... 10 Krystallis, Athanasios ...... 899 Jablonska, Magdalena Zyta ...... 911 Kelley, Corinne M ...... 912 Krzeminska, Agnieszka ...... 915 Jacob, Jorge ...... 608 Kenning, Peter ...... 903, 909, 925 Ksendzova, Masha ...... 323, 328 Jahn, Steffen ...... 610 Ketron, Seth ...... 649, 651 Kuchmaner, Christina ...... 170 Jain, Gaurav ...... 314 Khamitov, Mansur . . . .86, 92, 293, 340, 754 Kuhnreich, Chaim ...... 915 Jain, Kriti ...... 943 Khan, Shanze ...... 98 Kulkarni, Atul A ...... 916 Jain, Shailendra P...... 45 Khenfer, Jamel ...... 653 Kumar, Amit ...... 283 Jain, Shailendra Pratap ...... 913 Khodakarami, Farnoosh ...... 926 Kumar, Ankita ...... 50 Jampol, Lily ...... 551 Kidwell, Blair ...... 626, 628, 898, 900 Kunreuther, Howard ...... 583 Janakiraman, Narayanan ...... 869 Kim, Aekyoung ...... 819 Kuri, Daniela ...... 931 Jang, Eunyoung ...... 911 Kim, Eunjin ...... 928 Kurzban, Rob ...... 928 Janiszewski, Chris ...... 453 Kim, Hye-young ...... 160 Kwon, HyukJin ...... 902 Jaspers, Esther ...... 612 Kim, JeeHye (Christine) ...... 829 Kwon, JaeHwan ...... 314 Jeon, Eunmi ...... 911, 911 Kim, Ji Myoung Danny ...... 943 Kwon, Yong Ju ...... 916 Jhang, Ji Hoon ...... 23 Kim, Junghan ...... 944 Kyung, Ellie ...... 328 Jia, He (Michael) ...... 932 Kim, Junghyun ...... 912 Kyung, Ellie J...... 188, 188 Jia, Jayson S...... 614 Kim, Keongtae ...... 923 Jia, Lei ...... 144, 911 Kim, Kihyun Hannah ...... 884 Jia, Miaolei (Liam) ...... 154 Kim, Kyu Ree ...... 913 L Jia, Yanli ...... 922 Kim, Min Jung ...... 902, 913 Laan, Nynke van der ...... 565 Jiang, Jing ...... 905, 933, 935, 938, 942 Kim, Moon-Yong ...... 913 Ladeira, Wagner Junior ...... 927 Jiang, Li ...... 616 Kim, Nicole ...... 110, 209 Lafreniere, Katherine C...... 264 Jiang, Yuwei .110, 144, 587, 798, 909, 910, 910, Kim, Sara ...... 916, 932 Laham, Simon M...... 897 910, 911, 929, 933 Kim, Seo Young ...... 183 Lai, Edward Yuhang ...... 916 Jiao, Jinfeng ...... 928 Kim, Soo ...... 74 Lala, Vishal ...... 777 Jifeng, Luo ...... 939 Kim, Summer Hyoyeon ...... 913 Lalwani, Ashok K...... 844, 869 Jiménez, Jano ...... 901 Kim, Tae Woo ...... 183 Lamberton, Cait ...... 127 Jin, Fei ...... 919 Kim, Tami ...... 199 Lancelot, Caroline ...... 491 Jin, Yoonho ...... 618, 943 Kim, Yaeeun ...... 913 Landgraf, Polina ...... 668 Jodoin, Marie-Eve ...... 912 Kim, Yaeri ...... 923 Lane, Kristen E...... 138 Joerling, Moritz ...... 620 Kim, Yoon Duk ...... 264, 264 Lange, Christine ...... 80 Johar, Gita ...... 194 King, Dan ...... 898, 913, 914 Langhe, Bart de ...... 670 Johar, Gita V...... 138, 154 Kirk, Colleen Patricia ...... 914, 914 Laroche, Michel ...... 938 Johar, Gita Venkataramani ...... 893 Kirk, Samantha Renee ...... 914 Larson, Jeff ...... 671 John, Deborah Roedder ...... 6, 144 Kirmani, Amna ...... 110, 138, 638, 893 Lasaleta, Jannine ...... 916 John, Elicia ...... 160 Kleppe, Ingeborg Astrid ...... 231 Lavrusheva, Olga ...... 916 Johnson, Allison R...... 849 Klesse, Anne-Kathrin ...... 549, 935 Leban, Marina ...... 673, 674 Johnson, Eric J...... 258, 530 Kliamenakis, Argiro ...... 914, 942, 943 LeBoeuf, Robyn ...... 39 Johnson, Guillaume D ...... 890 Knöferle, Klemens ...... 594, 596 Lee, Byung ...... 258 Johnson, Samuel G. B...... 624 Knoll, Melissa ...... 247 Lee, Dong Il ...... 910 Joireman, Jeff ...... 149 Koch, Jan Andre ...... 654 Lee, Hsiao-Ching ...... 675 Jones, Niusha ...... 626, 628 Koetz, Clara ...... 656 Lee, Hyunhwan ...... 924 José, Rebeca San ...... 926 Kogler, Stephanie ...... 406 Lee, Julie Anne ...... 937 Jun, Sunghee ...... 912 Kong, Dejun Tony ...... 914, 944 Lee, Kelly Kiyeon ...... 226, 677 Jung, Haesung Annie ...... 630, 633 Koningsbruggen, Guido M. van . . . . . 541 Lee, Leonard ...... 808 Jung, Minah ...... 104 Koo, Minjung ...... 657, 867 Lee, Michael SW ...... 398 Jung, Minah H...... 28, 199, 363 Kooi, Kate ...... 914 Lee, Saerom ...... 640, 917 Jung, SungJin ...... 165 Kopalle, Praveen K...... 658 Lee, Seojin Stacey ...... 923 Kordrostami, Elika ...... 915 Lee, Wonkyong Beth ...... 903 Kordrostami, Melika ...... 915 Lee, Yoonah ...... 918 K Korelo, José Carlos ...... 904 Lee, Yu-kang ...... 675 Kouchaki, Maryam ...... 616 Kadomskaia, Veronika ...... 636 Lefkowitz, Jaclyn ...... 298 Advances in Consumer Research (Volume 46) / 951

Lehmann, Don ...... 63 Mai, Ke Michael ...... 657 Morris, Joshua I...... 252 Leliveld, Marijke ...... 177 Mai, Trang Thanh ...... 919, 919 Morwitz, Vicki ...... 293 Lemarié, Linda ...... 679 Main, Kelley ...... 177, 900 Morwitz, Vicki G...... 717 Leung, Eugina ...... 33 Main, Kelley J...... 495 Mourali, Mehdi ...... 904, 937 Levav, Jonathan ...... 110 Maior, Cecilia Souto ...... 904, 919, 919 Mourey, James A...... 188, 704 Levontin, Liat ...... 715 Makkar, Marian ...... 694 Moussa, Anthony ...... 420 Lewis, Joshua ...... 194, 681 Makov, Tamar ...... 177 Mrkva, Kellen ...... 706 Li, Charis ...... 298 Malik, Sumit ...... 943 Mugel, Ophélie ...... 890 Li, Jessica ...... 913 Malkoc, Selin A...... 104, 226 Mühlbacher, Hans ...... 352 Li, Ran ...... 917 Mandal, Sudipta ...... 696 Mukherjee, Ashesh ...... 898 Li, Shaobo ...... 880 Mandel, Naomi ...... 74 Mukherjee, Sourjo ...... 898 Li, Wanyu ...... 917 Mannetti, Lucia ...... 941 Mukherjee, Sudipta ...... 708, 711 Li, Xi ...... 719 Mantovani, Danielle ...... 904, 919, 919 Mukhopadhyay, Anirban ...... 687 Li, Xiaolin ...... 110 Manuel, Rommel J...... 501 Mulier, Lana ...... 713 Li, Xilin ...... 199 Maraj, Varala ...... 920 Munichor, Nira ...... 715 Li, Xinlong ...... 247 Martin, Kelly D...... 149 Munz, Kurt P...... 220, 717 Li, Xiuping ...... 154, 927 Martínez, Ivanna ...... 931 Li, Yanyan ...... 194 Martínez, Valeria ...... 931 Li, Yaoqi ...... 938 Matherly, Ted ...... 765 N Li, Yi ...... 593 Mathras, Daniele ...... 924 Nafees, Lubna ...... 898 Lieberman, Alicea J...... 357 Mattila, Pekka ...... 916 Nam, Myungwoo ...... 911, 911 Light, Alysson E...... 270 May, Frank ...... 711, 792 Nandi, Sampoorna ...... 921 Light, Nicholas ...... 115 Mayrhofer, Mira ...... 138 Napolova, Ekaterina ...... 944 Lim, Jufinnie ...... 936 Mazar, Nina ...... 252 Nardini, Gia ...... 897, 921 Lim, Myungsuh ...... 912 Mažar, Nina ...... 298 Nasa, Jayant ...... 921 Lin, Chia-Wei Joy ...... 917 McFerran, Brent ...... 209, 340 Nave, Gideon ...... 899 Lin, Shengdong ...... 935 McGehee, Molly ...... 906 Nayakankuppam, Dhananjay ...... 314 Lin, Ying-Ching ...... 917 McGraw, A. Peter ...... 57 Nelson, Leif D...... 199, 242, 363 Ling, Yee ...... 946 Mead, Nicole ...... 110, 323, 937 Nelson, Michelle Renee . . . . 891, 918, 930 Lisjak, Monika ...... 74, 74 Mecit, N. Alican ...... 698, 889, 920 Nelson, Russ ...... 288 Liu, (Joyce) Jingshi ...... 687, 918, 918 Mehalko, Maggie M ...... 891 Nenkov, Gergana Y...... 323 Liu, Chihling ...... 909, 941 Mehta, Ravi ...... 839 Nenkova, Ani ...... 264 Liu, Maggie Wenjing . . . . . 682, 684, 933 Meißner, Lasse ...... 909 Nepomuceno, Marcelo Vinhal . .919, 921, 931 Liu, Mengmeng ...... 686 Meloy, Margaret G...... 569, 857 Netto, Carla Freitas Silveira ...... 921 Liu, Shixiong ...... 917 Melumad, Shiri ...... 264 Netzer, Oded ...... 194 Liu, Weizi ...... 918 Melzner, Johann ...... 920 Newman, George ...... 133 Liu, Wumei ...... 918 Mende, Martin ...... 183, 912 Newman, George E...... 177 Loewenstein, George ...... 575 Meng, Hua (Meg) ...... 920 Newman, Kevin ...... 781 Logg, Jennifer M...... 63 Meng, Yan ...... 907 Ng, Maria ...... 914, 944 Longoni, Chiara ...... 63, 68, 293 Messinger, Paul Richard . . . . . 86, 92, 926 Ng, Sharon ...... 880 Lopes, Nuno Jose ...... 690 Messner, Claude ...... 909 Nguyen, Peter ...... 86, 92, 719 Lopez, Alberto ...... 170, 889 Mette, Frederike Monika Budiner . . . . .927 Nickerson, Dionne ...... 722 Loveland, Katherine ...... 924 Metz, Franziska ...... 929 Nicklaus, Sophie ...... 80 Lowe, Michael ...... 722 Meyer, Robert ...... 264, 264, 899 Nikolov, Nik ...... 900 Lowrey, Tina M...... 698, 889, 920 Miceli, Gaetano “Nino” ...... 700 Nikolova, Hristina ...... 346, 724 Lu, Joy ...... 204 Mijovic-Prelec, Danica ...... 293 Noack, Nicolas ...... 922 Lu, Lin ...... 897 Miller, Elizabeth G...... 479, 734 Nordgren, Loran F...... 264 Luca, Ramona De ...... 918 Millet, Kobe ...... 937 Northey, Gavin ...... 469 Lucarelli, Andrea ...... 352, 692 Mimoun, Laetitia ...... 214, 945 Norton, David A...... 846 Lugoch, Marina de Wallau ...... 921 Minson, Julia A...... 63 Norton, Michael I. . .28, 39, 104, 323, 346, 346, Lunardo, Renaud ...... 481 Mittal, Vikas ...... 553 577 Lusch, Robert F...... 121 Mochon, Daniel ...... 252 Noseworthy, Theodore J...... 814 Luttrell, Andrew ...... 925, 943 Mogilner, Cassie ...... 303, 303 Novak, Thomas ...... 33 Lutz, Richard J ...... 921 Mohr, Joel Alan ...... 920 Novemsky, Nathan ...... 39 Lutz, Richard J...... 14 Mohsen, Marwa Gad ...... 920 Nrusimha, Aniruddha ...... 204, 878 Lynch Jr., John G...... 440 Monga, Alokparna (Sonia) . . . . . 897, 929 Nunes, Joseph C...... 561 Lynch, John ...... 23, 893 Monga, Ashwani ...... 884 Nyffenegger, Bettina ...... 170 Lynch, John G...... 298 Mookerjee, Siddhanth ...... 944 Nyilasy, Gergely ...... 897 Lyndem, Preeti Krishnan ...... 908 Mookherjee, Satadruta ...... 702 Moon, Alice ...... 199 Moore, Don A...... 63 O M Moore, Sarah G ...... 926 O’Connor, Marcus ...... 499 Ma, Jingjing ...... 919 Moore, Sarah G...... 264 Moraes, Luana C...... 891 O’Donnell, Michael ...... 28 Ma, Yu ...... 917 Oh, Hyewon ...... 220 Maciel, Andre F...... 510 Morales, Andrea . . . . . 127, 177, 893, 932 Morales, Andrea C...... 74 Oh, Travis Tae ...... 726, 728 MacInnis, Deborah ...... 921 Okan, Mehmet ...... 523 Maehle, Natalie ...... 231 Morewedge, Carey K. 39, 63, 226, 293, 328, 357 Morrin, Maureen ...... 686 Okutur, Nazli Gurdamar ...... 922 Maglio, Sam J...... 298 Olivola, Christopher ...... 204, 226 952 / Author Index

Olson, Jenny ...... 346, 913, 937 Phipps, Marcus ...... 121 Reed, Americus ...... 309 Olson, Jenny G...... 346 Piepenbrink, Anke ...... 656 Rehman, Varisha ...... 785 Olson, Nicholas J...... 732 Pieters, Rik ...... 612 Reiff, Joseph S...... 237 Onculer, Ayse ...... 930 Pinheiro, Gabriela L...... 891 Reimann, Martin ...... 170 Ooijen, Iris van ...... 922, 922 Piskorski, Mikolaj ...... 489 Reinholtz, Nicholas ...... 247 Opitz, Marcus ...... 922 Pizzetti, Marta ...... 750 Reit, Emily ...... 199 Oppewal, Harmen ...... 870 Pizzutti, Cristiane ...... 924, 931 Reutskaja, Elena ...... 690 Ordabayeva, Nailya ...... 237, 571 Plank, Andreas ...... 925 Rick, Scott ...... 346 Ordanini, Andrea ...... 561 Plassmann, Hilke ...... 258 Riemer, Hila ...... 927 Osborne, Matthew ...... 298 Platt, Michael ...... 789, 945 Rifkin, Jacqueline R...... 303, 363 Osselaer, Stijn M.J. van ...... 538 Plennert, Matthias ...... 674 Risen, Jane L...... 39 Oswald, Laura R...... 422 Pohst, Maximilian Clemens ...... 925 Rixom, Jessica ...... 906, 929 Otnes, Cele ...... 231, 231 Poitou, Christine ...... 258 Rizvi, Shaheer Ahmed ...... 926 Ouyang, Jun ...... 922 Polman, Evan ...... 160, 298 Robinson, Aaron ...... 374 Ozanne, Julie L...... 895 Pomerance, Justin ...... 247 Robinson, Thomas Derek ...... 823 Ponchio, Mateus Canniatti ...... 927 Robitaille, Nicole ...... 763 Porto, Nilton ...... 23 Rocklage, Matthew D...... 264 P Poupis, Lauren ...... 270 Rodas, Maria ...... 802 Packard, Grant ...... 736 Pow, Jessica ...... 346 Rodas, Maria A...... 144 Paharia, Neeru . . . . .28, 149, 149, 149, 165 Powell, Emily ...... 104 Rodrigues, Max ...... 624 Pai, Jieun ...... 104 Powell, Rachel ...... 891 Rodriguez, Alejandra ...... 765 Paik, Sung-Hee W...... 738 Prado, Paulo Henrique Muller ...... 904 Rodriguez, Rachel ...... 889 Paluch, Stefanie ...... 620 Praxmarer-Carus, Sandra ...... 425 Rodríguez-Torrico, Paula ...... 926 Pancer, Ethan ...... 814 Preiksaitis, Kimberley Mosher ...... 925 Rohden, Simoni F ...... 921 Pandelaere, Mario ...... 593, 708 Prelec, Drazen ...... 293 Rosario, Ana Babic ...... 897 Paolacci, Gabriele ...... 346, 740 Price, Basil Arnould ...... 214 Rosario, Ana Babić ...... 823 Park, Eujin ...... 923 Price, Linda L ...... 214, 893 Rose, Chris ...... 374 Park, Heejung ...... 923 Price, Linda L...... 121, 276, 276 Rosenthal, Benjamin ...... 276 Park, Jaewoo ...... 930 Prokopec, Sonja ...... 930 Ross, David William ...... 918 Park, Jane ...... 792 Proserpio, Davide ...... 220 Ross, William T...... 846 Park, Jen H...... 242, 923 Prothero, Andrea ...... 895 Roux, Caroline ...... 895 Park, Jongwon ...... 512, 872 Pundak, Chen ...... 752 Rozin, Paul ...... 115 Park, Jooyoung ...... 923, 923, 933 Puntoni, Stefano ...... 63, 547 Ruan, Bowen ...... 160 Park, Kiwan ...... 912, 923 Putnam-Farr, Nell ...... 252 Rubin, Dan ...... 80 Park, Sang Kyu ...... 745 Puzakova, Marina ...... 754 Rubio, Natalia ...... 901 Park, Taehoon ...... 742, 912, 944 Puzakova, Marina A...... 756 Rucker, Derek ...... 74, 226 Parker, Jeffrey R...... 237 Pyle, Martin A...... 758, 925 Rucker, Derek D...... 165, 264, 328, 662 Parmentier, Marie-Agnès ...... 912 Pyone, Jin Seok ...... 871 Russell, Cristel Antonia ...... 491 Pasdiora, Maria Alice ...... 924 Russmann, Janna ...... 293 Pastore, Cristina Maria de Aguiar . . . . .921 Ruvio, Ayalla ...... 926 Paswan, Audhesh ...... 900 Q Patall, Erika ...... 633 Qi, Zhou ...... 908 Patrick, Vanessa ...... 944 Qiu, Ruyi ...... 759 S Patrick, Vanessa M...... 68 Quintanilla, Claudia ...... 889 Saad, Gad ...... 915 Patry-Beaudoin, Gabrielle ...... 924 Quoidbach, Jordi ...... 237, 303 Sääksjärvi, Maria ...... 591 Pattabhiramaiah, Adithya ...... 722 Saccardo, Silvia ...... 363 Paul, Iman ...... 237 Saenger, Christina ...... 926, 930 Paul, Michael ...... 915 R Sahay, Arvind ...... 696 Pearson, Matthew ...... 194 Radford, Scott ...... 904 Saini, Ritesh ...... 515 Peck, Joann ...... 914 Raghunathan, Raj ...... 837, 921 Saint-Quentin, Virginie Brégeon de . . . .890 Peer, Eyal ...... 104 Raghunathan, Rajagopal ...... 68 Salisbury, Linda Court ...... 323 Pellandini-Simányi, Léna ...... 748 Rahman, Kaleel ...... 382 Salnikova, Ekaterina ...... 926 Peñaloza, Lisa ...... 563 Rahmani, Vahid ...... 761 Samper, Adriana ...... 127, 309, 932 Pendarvis, Nicholas ...... 895 Raimondo, Maria Antonietta ...... 700 Sample, Kevin L...... 68, 127 Peng, Jingjia ...... 935 Rajagopal, Ananya ...... 430 Samuelsen, Bendik ...... 133 Peng, Luluo ...... 934 Ramakrishnan, Arjun ...... 945 Sanghvi, Minita ...... 895 Peng, Rebecca ...... 924 Rampullo, Alberto ...... 551 San-Martín, Sonia ...... 926 Penner, Sara ...... 495, 900 Ran, Yaxuan ...... 925 Santana, Shelle ...... 28 Perera, B. Yasanthi ...... 898 Rank-Christman, Tracy ...... 270 Santini, Fernando De Oliveira ...... 927 Perfecto, Hannah ...... 242 Rao, Akshay ...... 895 Saravade, Swapnil ...... 769 Perkins, Andrew ...... 527, 908, 923 Rao, Akshay R...... 138 Sarvary, Miklos ...... 183 Petrylaite, Edita ...... 924 Raoofpanah, Iman ...... 907 Savary, Jennifer ...... 270, 600 Petty, Richard ...... 925, 943 Rath, Suzanne ...... 763 Sayin, Eda ...... 943 Pew, Ethan ...... 924 Ratner, Rebecca ...... 209 Scaraboto, Daiane . . . . . 98, 214, 276, 543 Pham, Chi ...... 80 Ratner, Rebecca K...... 39 Scekic, Ana ...... 773, 944 Pham, Michel Tuan ...... 726 Read, Daniel ...... 226 Schanbacher, Anja ...... 775 Philipp-Muller, Aviva ...... 925 Reczek, Rebecca Walker . . . . 127, 517, 876 Schau, Hope Jensen ...... 98, 288 Philp, Matthew ...... 910 Reeck, Crystal ...... 258 Scheibehenne, Benjamin ...... 664 Advances in Consumer Research (Volume 46) / 953

Schindler, Robert M...... 777 Slot, Johanna ...... 569 Thomas, Manoj ...... 328 Schiro, Julie ...... 670 Sluis, Helen van der ...... 309 Thomas, Tandy Chalmers ...... 276 Schmidt, Liane ...... 258 Small, Deborah ...... 194, 283 Thomas, Veronica ...... 926, 930 Schmitt, Bernd ...... 183, 183 Smith, Andrew ...... 758 Thompson, Craig ...... 333 Schneider, Abigail ...... 927 Smith, Rosanna ...... 57 Thomson, Matthew ...... 86, 92, 849 Scholz, Agnes ...... 664 So, Jane ...... 794 Tian, Kathy ...... 891, 930 Schöps, Jonathan D...... 434 Sobol, Kamila ...... 796, 914 Tijssen, Irene ...... 565 Schroeder, Juliana ...... 39, 283, 283 Sok, Pharo ...... 333 Tillotson, Jack ...... 121 Schroll, Roland ...... 779 Solt, Michelle van ...... 929 Ting, Ding Hooi ...... 817 Schultz, Ainslie ...... 781 Soman, Dilip ...... 298 To, Ngoc (Rita) ...... 68 Schulz, Michael ...... 783 Song, Camilla (Eunyoung) ...... 154 Togawa, Taku ...... 930 Schwartz, Camille ...... 80 Song, Camilla Eunyoung ...... 657 Tok, Dickson ...... 505 Schwartz, Janet ...... 288 Song, Jiaqi ...... 798 Torres, Lez Trujillo ...... 50, 50 Schweitzer, Maurice ...... 28 Spann, Martin ...... 242 Touchette, Laurie ...... 931 Scopelliti, Irene ...... 700 Spears, Nancy ...... 649 Touré-Tillery, Rima ...... 154, 154, 270 Scott, Maura ...... 183, 912 Spicer, Andre ...... 893 Toyoki, Sammy ...... 121 Scott, Rebecca ...... 121 Spiller, Stephen ...... 318 Trabold, Lauren ...... 926 Scott, Sydney E...... 115 Sprott, David ...... 149, 527 Tran, An ...... 440 Sedikides, Constantine ...... 914 Sreekumar, Arun ...... 231 Trevino, Teresa ...... 931 Segal, Sandra ...... 927 Srivastava, Joydeep ...... 660, 913 Trijp, Hans van ...... 565 Selvanayagam, Karthik ...... 785 Stamatogiannakis, Antonios . . . . .668, 938 Tripathi, Sanjeev ...... 696 Sen, Berna Basar and Sankar ...... 473 Steffel, Mary ...... 346 Trupia, Maria Giulia ...... 931 Seo, Eunjin ...... 633 Steinhart, Yael ...... 539, 752 Tsai, Chun-Chen ...... 505 Seo, Yuri ...... 521 Stenstrom, Eric ...... 921 Tsai, Claire I...... 194 Seok, Darsol ...... 945 Sternthal, Brian ...... 585 Tu, Yanping ...... 154, 209, 298 Sepehri, Amir ...... 583, 927 Stich, Lucas ...... 242 Tuckett, David ...... 624 Septianto, Felix ...... 469 Stillman, Paul Edgar ...... 800 Tully, Stephanie ...... 220 Serfas, Benjamin G...... 901 Stoner, Jennifer L...... 802 Sevilla, Julio ...... 45, 293, 900 Storch, Julia ...... 929 Sezer, Ovul ...... 28 Stornelli, Jason ...... 462 U Shaddy, Franklin ...... 209, 318 Streicher, Mathias ...... 803 Ubal, Valentina Ortiz ...... 921, 931 Shah, Avni ...... 298 Streicher, Mathias Clemens ...... 901 Uduehi, Esther ...... 309 Shah, Avni M...... 247 Strizhakova, Yuliya ...... 926 Ülkü, Sezer ...... 605 Shams, Poja ...... 912 Stuppy, Anika ...... 805 Ülkümen, Gülden ...... 833 Shang, Ziqi ...... 927 Su, Lei ...... 929 Ulu, Sevincgul ...... 819 Shani-Feinstein, Yael ...... 188 Suh, Wang Suk ...... 573 Unnava, H. Rao ...... 314 Shanks, Ilana ...... 183 Suher, Jacob ...... 808 Urminsky, Oleg . . 33, 160, 194, 237, 258, 323 Sharif, Marissa A...... 303 Sui, Jie ...... 910 Uzunoğlu, Ebru ...... 460 Sharifonnasabi, zahra ...... 928 Sun, Daniel ...... 811 Sharma, Varun ...... 787 Sun, Jennifer ...... 165 Sharon ...... 485 Sun, Sijie ...... 869 V Shavitt, Sharon ...... 314 Sun, Yang ...... 910 Shaw, Alex ...... 928 Sun, Yixia ...... 810 Vadakkepatt, Gautham G...... 149 Shechter, Steven ...... 45 Sundar, Aparna ...... 738 Vaidyanathan, Rajiv ...... 895 Shen, Liang ...... 928, 945 Suri, Rajneesh ...... 946 Vallen, Beth ...... 309 Sheng, Feng ...... 789, 945 Sussman, Abigail ...... 23, 247, 323 VanBergen, Noah ...... 133, 821 Shenhav, Amitai ...... 318 Swift, Sam ...... 23 Vanhamme, Joëlle ...... 916 Shepherd, Steven ...... 790 Szocs, Courtney ...... 929 Vanhuele, Marc ...... 555 Sheppes, Gal ...... 927 Veer, Ekant ...... 895 Sherman, Gary D...... 309 Velloso, Luciana ...... 170 Shi, Jie ...... 897 T Veloso, Andres Rodriguez ...... 942 Shibly, Sirajul Arefin ...... 928 Veludo-de-Oliveira, Tânia ...... 369 Tabrizi, Eisa Sahabeh ...... 929 Verbeke, Willem ...... 462 Shim, Yoonji ...... 74 Taillon, Brian ...... 651 Shiv, Baba ...... 614 Veresiu, Ela ...... 288, 333, 823 Takhar, Jennifer ...... 945 Verlegh, Peeter ...... 911 Shoenberger, Heather ...... 928 Tal, Aner ...... 455, 930 Shrivastava, Sunaina ...... 314 Vermeir, Iris ...... 713, 946 Talebi, Arash ...... 930 Verstraeten, Julie ...... 931 Shrum, L. J...... 698, 920 Tam, Leona ...... 944 Shu, Suzanne ...... 160, 514 Vieites, Yan ...... 608, 825 Tannenbaum, David ...... 833 Vilasanti, Victoria ...... 919 Siddiqui, Rafay A...... 792 Tanner, Robin J...... 160 Silver, Ike ...... 928, 928 Villaseñor, Nieves ...... 901 Tao, Tao ...... 945 Viswanadham, Ratnalekha ...... 258 Silverman, Jackie ...... 928 Taylor, Jeanette ...... 777 Simester, Duncan ...... 907 Vlahos, Aphrodite ...... 121 Taylor, Kimberly ...... 929 Vock, Marlene ...... 340 Simmons, Joseph ...... 681 Taylor, Nükhet ...... 814 Simonsohn, Uri ...... 525 Voorhees, Clay ...... 926 Teeny, Jacob ...... 943 Vosgerau, Joachim ...... 104 Simonson, Itamar ...... 242 Temmerman, Joyce De ...... 946 Simonyan, Yvetta ...... 945 Voyer, Benjamin ...... 673 Tessitore, Tina ...... 931 Voyer, Peter ...... 444 Singh, Amit Surendra ...... 314 Tezer, Ali ...... 816 Slabbinck, Hendrik ...... 713, 946 Vries, Eline L.E. De ...... 827 Thomas, Kevin D ...... 890 Vu, Tiffany ...... 346, 462 954 / Author Index

Wiggins, Jennifer ...... 170, 907 Yesmukanova, Alima ...... 941 W Wilcox, Keith ...... 728, 853, 918 Yi, John ...... 936 Wadhwa, Monica ...... 829 Willberger, Judith ...... 934 Yi, Youjae ...... 916 Waisman, Rory ...... 220 Williams, Elanor F...... 346 Yin, Bingqing (Miranda) ...... 871 Waldman, Ari Ezra ...... 704 Williams, Patricia A...... 57 Yin, Yunlu ...... 897 Walker, Jesse ...... 932 Williams, Patti ...... 74 Yip, Jeremy A...... 677 Walkley, Fiona ...... 903 Williams-Beeler, Kieshana M...... 918 Yoon, Alyssa ...... 872 Wall, Daniel ...... 932 Wilroy, Gretchen R...... 857 Yoon, Haewon ...... 226 Wallach, Karen ...... 831 Wilson, Anne ...... 28 Yoon, Heeyoung ...... 28 Wallpach, Silvia von ...... 352 Wilson, Anne V...... 104 Yoon, Na Ri ...... 937 Walsh, Darlene ...... 932 Wittkowski, Kristina ...... 916 Yoon, Sunyee ...... 943 Walters, Daniel ...... 833 Wolkenstoerfer, Stefan ...... 425 Yoon, Y. Rin ...... 512 Wan, Echo Wen ...... 925, 938 Wong, Nancy ...... 937 You, Huan ...... 937 Wan, Fang ...... 859, 935, 936, 937, 938 Wong, Vincent Chi ...... 934 You, Yanfen ...... 841, 874 Wan, Jing ...... 835, 929 Wood, Stacy ...... 1, 895 Youn, Nara ...... 899 Wan, Xiaoang ...... 759, 910 Wray, Candace ...... 891 Yu, Bin ...... 935 Wang, Chen ...... 839, 933 Wright, Scott ...... 781 Yu, Ying ...... 928, 945 Wang, Hao ...... 933 Wu, Alisa Yinghao ...... 942 Yucel-Aybat, Ozge ...... 598 Wang, Jessie J...... 844 Wu, Freeman ...... 127 Wang, Juan ...... 606 Wu, Yi ...... 917, 934 Wang, Liangyan ...... 908, 942 Wu, Yuechen ...... 39 Z Wang, Lili ...... 841, 874 Wu, Yueyan ...... 934 Zamudio, César ...... 920 Wang, Luming ...... 919, 919 Wyer Jr., Robert ...... 934, 945 Zane, Daniel M...... 876 Wang, Qi ...... 901 Zanibellato, Francesco ...... 937 Wang, Qin ...... 74, 932 Zant, Alex Van ...... 283 Wang, Qizhou ...... 846 X Zayer, Linda Tuncay ...... 231, 231 Wang, TzuShuo Ryan ...... 138 Xi, Kaiyuan ...... 906 Zeng, Xianfang ...... 937 Wang, Wangshuai ...... 837 Xiao, Chunqu ...... 842, 906 Zenkic, J ...... 937 Wang, Wenbo ...... 829 Xiao, Jing Jian ...... 23 Zhan, Fuchun ...... 937 Wang, Xia ...... 901 Xie, Chunya ...... 934, 935 Zhang, Chun ...... 938 Wang, Xian ...... 684 Xie, Chunyan ...... 934, 935 Zhang, Jiao ...... 39, 738 Wang, Xin ...... 86, 86, 842 Xie, Guang-Xin ...... 915 Zhang, Jing ...... 938 Wang, Xin (Shane) ...... 92, 92, 719 Xie, Jieru ...... 45 Zhang, John ...... 905 Wang, Xingyu ...... 932 Xin, Liu ...... 939 Zhang, Kuangjie ...... 865, 880 Wang, Xue ...... 932 Xiong, Ji (Jill) ...... 154 Zhang, Lijun ...... 946 Wang, Xuehua ...... 810 Xu, Jialiang ...... 935 Zhang, Shaoqing ...... 938 Wang, Yajin ...... 110, 110, 165 Xuan, Changchun ...... 935 Zhang, Xiadan ...... 938 Wang, Yan ...... 933 Zhang, Ya ...... 938 Wang, Yanyun (Mia) ...... 891 Zhang, Yan ...... 363 Wang, Yijie ...... 933 Zhang, Yinlong ...... 553 Wang, Yiru ...... 920 Y Zhang, Yuan ...... 938 Wang, Yitong ...... 501 Yague, Mªjesus ...... 901 Zhang, Zhihao ...... 204, 878 Wang, Yue ...... 936 Yalcin, Gizem ...... 57, 740, 935 Zhao, Jiaying ...... 115 Ward, Adrian ...... 209 Yamim, Amanda Pruski ...... 916 Zhao, Min ...... 323, 835 Ward, Adrian F...... 23, 68, 340 Yan, Dengfeng ...... 936, 939 Zhao, Xiaohua ...... 938 Ward, Morgan ...... 831 Yan, Jun ...... 932, 936 Zhao, Xin ...... 465, 909, 941 Warlop, Luk . . .133, 519, 594, 596, 909, 929 Yan, Jun (Wendy) ...... 859 Zheng, Lijing ...... 938 Warren, Caleb ...... 905, 936 Yan, Li ...... 861 Zheng, Qiuying ...... 529 Warren, Nathan B...... 276 Yang, Adelle ...... 160 Zheng, Yuhuang ...... 682, 934 Warren, Samantha ...... 121 Yang, Adelle X...... 258, 867 Zhiyan, Wu ...... 939 Watson, Jared ...... 138 Yang, Cathy Liu ...... 773, 944 Zhong, Chenbo ...... 935 Weber, T.J...... 149 Yang, Chun-Ming ...... 841, 863 Zhong, Ninghua ...... 933 Wegerer, Philipp K...... 352 Yang, Haiyang ...... 668, 865, 937, 938 Zhou, Lingrui ...... 939 Wei, Chuang ...... 933 Yang, Junxian ...... 936 Zhou, Qiang ...... 939 Wei, Lingru ...... 933 Yang, Lu ...... 682 Zhou, Yeni ...... 897 Wei, Sarah ...... 258 Yang, Xiaojing ...... 144, 874, 911 Zhu, Hong ...... 842, 906 Weingarten, Evan ...... 57, 86, 92, 357 Yang, Yang ...... 80, 226, 745 Zhu, Luke ...... 936, 937 Weiss, Liad ...... 847 Yang, Zhen ...... 936 Zhu, Lusha ...... 897 Wen, Beixi ...... 934, 935 Yang, Zheshuai ...... 363 Zhu, Meng ...... 357, 895 Werle, Carolina O. C...... 916 Yang, Zhi ...... 934 Zhu, Qichao ...... 684, 933 Whelan, Jodie ...... 849 Yang, Zhilin ...... 133 Zhu, Rui (Juliet) ...... 839 Whillans, Ashley V...... 104, 303, 346 Yang, Zhiyong ...... 869 Zilberman, Ronen ...... 374 White, Katherine ...... 74, 177 Yao, Jun ...... 870 Zolfagharian, Mohammadali . . . . .769, 882 Wider, Serena ...... 352 Yazdanparast, Atefeh ...... 882 Zor, Ozum ...... 884 Wiener, Hillary J.D...... 851 Ye, Hongjun ...... 946 Zouaghi, Sondes ...... 886 Wiener, Josh ...... 851 Ye, Jun ...... 936 Zucco Jr, César ...... 471 Wiertz, Caroline ...... 920 Ye, Xiaoyin ...... 936 Association For Consumer Research

$99.00 ISBN 978-0-915552-25-2 59900>

9 780915 552252