Brands as Rivals: Consumer Pursuit of Distinctiveness and the Role of Brand Anthropomorphism

Brands as Rivals: Consumer Pursuit of Distinctiveness and the Role of Brand Anthropomorphism Abstract Although past research has shown that anthropomorphism enhances consumers’ attraction to a brand when social-connectedness or effectance motives are active, the current research demonstrates that anthropomorphizing a brand becomes a detrimental marketing strategy when consumers’ distinctiveness motives are salient. Four studies show that anthropomorphizing a brand positioned to be distinctive diminishes consumers’ sense of agency in identity expression. As a result, when distinctiveness goals are salient, consumers are less likely to evaluate anthropomorphized (vs. nonanthropomorphized) brands favorably and are less likely to choose them to express distinctiveness. This negative effect of brand anthropomorphism, however, is contingent on the brand’s positioning strategy—brand-as-supporter (supporting consumers’ desires to be different) versus brand-as-agent (communicating unique brand features instead of focusing on consumers’ needs) versus brand-as-controller (limiting consumers’ freedom in expressing distinctiveness). Our results demonstrate that an anthropomorphized brand-as-supporter enhances consumers’ sense of agency in identity expression, compared to both an anthropomorphized brand-as-agent and an anthropomorphized brand-as-controller. In turn, enhancing or thwarting consumers’ sense of agency in expressing their differences from others drives the differential impact of anthropomorphizing a brand positioned to be distinctive. brand anthropomorphism, distinctiveness motive, agency in identity expression Prior research demonstrates that people often seek to differentiate themselves from others (Snyder and Fromkin 1980). The need to express one’s distinctiveness is reflected in people’s unique product choices (Snyder 1992), intentions to recustomize the product (White and Argo 2011), and preferences for unique shopping venues (Burns and Warren 1995). In other words, distinctiveness motives are predictive of people’s consumption behaviors because products are seen as their self-extensions and as signals of their identity (Snyder and Fromkin 1980). However, missing from prior work is an understanding of the effects of the innate features or positioning strategies of brands, such as imbuing brands with humanlike features that are used by managers to help consumers signal their distinctiveness. Recent industry trends suggest that in attempts to find unique positioning strategies, many managers turn to the tactic of anthropomorphizing their brands (i.e., imbuing brands with humanlike features; Aggarwal and McGill 2012). For example, the Swatch brand introduced a collection of watches that activate various human schemas ranging from a Blue Wild Face to the hot-pink Lady Spy. Similarly, the Hammacher Schlemmer Institute retails a variety of unique products that are often designed to project humanlike features (e.g., an infrared supine sauna with a giant face). One possible reason for the popularity of this approach is that it might increase consumer connection to these unique products (Ahn, Kim, and Aggarwal 2014). In fact, people who lack social connectedness prefer anthropomorphized (vs. nonanthropomorphized) brands to help compensate for that gap (Chen, Wan, and Levy 2017). However, when purchasing unique brands, consumers pursue the goal of being distinct from others. The source of this distinctiveness motive is the desire to be different through signaling one’s unique identity (Vignoles 2009). When brands become important for identity signaling, factors that inhibit consumers’ need for self-expression should negatively affect brand evaluations. If anthropomorphized brands are perceived as being mindful (Puzakova and Kwak 2017), can anthropomorphizing a brand designed to uniquely express one’s identity become a detrimental brand strategy? With this research, we attempt to answer this question and make three important contributions. Our key contribution is to show that under salient distinctiveness motives, brand anthropomorphism has a detrimental effect on evaluations of distinctive brands. As our theoretical foundation, we rely on prior work showing that anthropomorphized brands are viewed as intentional agents that are able to control others’ actions (Epley and Waytz 2009). Consequently, anthropomorphized distinctive brands retain their own identities and, being seen as agentic, are perceived as reducing consumers’ sense of agency in self-expression, thereby leading to less favorable brand evaluations. Thus, our contribution lies in demonstrating the novel process through which brand anthropomorphism affects consumer preference under distinctiveness motives (i.e., reducing consumers’ agency in identity expression). This is in contrast to prior work showing positive effects of anthropomorphism when consumers’ belongingness (Chen et al. 2017) or effectance (Waytz et al. 2010b) goals are salient. Next, we establish important boundary conditions of the negative effect of brand anthropomorphism: the effect will occur only when a brand is positioned to be distinctive and the product usage context is public. This is because a distinctiveness motive enhances consumer preference for brands that are able to satisfy this goal, and because the signaling value of distinctive brands is higher in a public context (Tian, Bearden, and Hunter 2001). Finally, we contribute to prior work by identifying a theoretically relevant moderator (i.e., positioning a brand as supporting vs. controlling consumers’ sense of agency in self-expression). Because people develop strong ties to others who enhance their sense of agency (Thomson 2006), we show that positioning an anthropomorphized distinctive brand as a supporter to consumers increases their perceived sense of agency in identity expression and brand evaluations, compared to the situation when an anthropomorphized brand is positioned either as a controller or as an independent agent (an entity focused on its own features vs. consumers’ needs). The findings reveal an interesting nuance by indicating that distinctiveness motives enhance consumers’ sensitivity to cues that signal a potential threat to their sense of agency, and positioning a distinctive brand as an agent can make consumers view the brand as a less controllable entity, and hence, as less desirable. Overall, unlike prior work showing that anthropomorphism has positive effects under salient belongingness or effectance goals, our work demonstrates that anthropomorphism has negative effects on preferences for distinctive brands under distinctiveness motives. Four studies provide converging evidence in support of our theory; implications for managers and future research are discussed. CONSUMERS’ DISTINCTIVENESS MOTIVES AND PRODUCT CONSUMPTION Distinctiveness Motivation and Identity Expression Prior work defines distinctiveness motivation as the desire to differentiate oneself from others in appearance, qualities, traits, and abilities (Vignoles et al. 2006; White and Argo 2011). According to the uniqueness theory (Snyder and Fromkin 1980), people seek to be unique because they learn over time that they are different from others, and this learned self-image drives an active search for ways to differentiate from others. Prior work also shows that possessions are extensions of the self (Belk 1988) and hence can be used for self-expression (White and Argo 2011). Thus, people express their distinctiveness through consumption of products that are scarce, unique, customized, or odd (Snyder 1992). For example, Tian et al. (2001) find that people with a high need for uniqueness prefer products that are less popular and publicly visible. This finding highlights the crucial role of publicly visible products in satisfying people’s distinctiveness motives. Most pertinent to our work is research delineating how people motivated by distinctiveness diverge from others in their choices as they communicate their identity to others (Berger and Heath 2007). Importantly, past work emphasizes that the distinctiveness motive is paramount to identity construction (Vignoles 2009). Hence, people tend to make unique choices in product domains that have strong signaling value (e.g., car models; Berger and Heath 2007). In this regard, White and Argo (2011) show that when being mimicked by similar others, consumers high in distinctiveness motive respond with choices that differentiate them (e.g., recustomize the product). Similarly, consumers who are high in distinctiveness buy unique products to obtain a favorable evaluation from others (Tian et al. 2001). Overall, we adopt the view that consumers motivated by distinctiveness make choices to self-express and to ensure that others understand who they are. Distinctiveness Motivation and Agency in Identity Expression We further build on prior research showing that identity expression is motivated by another fundamental drive: desire for agency (Baumeister 1998). In general, agency refers to an individual’s need to feel that one’s actions are self-endorsed, as opposed to controlled and pressured by others (Deci and Ryan 2000). Prior research shows that when consumers express their identity, a certain type of marketing message (e.g., directly defining self-expression) can suggest that forces external to the self are determining one’s behavior. In other words, such messages diminish the extent to which identity-expressing choices are perceived to be determined by the self, thus limiting consumer perceptions of agency and reducing evaluations of products being promoted with these messages (Bhattacharjee, Berger, and Menon 2014). Similarly, Mochon, Norton, and Ariely (2012) show that a feeling of control over one’s environment is the driving force behind consumers’ self-expressive tasks (e.g., assembling furniture). Overall, this line of work emphasizes the importance of one’s agency in identity expression. Here, we merge two streams of research: the first shows that the distinctiveness motive is driven by the desire to communicate one’s difference to others, and the second emphasizes the importance of one’s own agency in identity expression. Because the latter is a crucial force behind the motive to be distinct, we argue that a brand positioning strategy that reduces consumers’ sense of agency in identity expression should decrease consumers’ brand evaluations. BRAND ANTHROPOMORPHISM Brand anthropomorphism is the process of imbuing brands with humanlike features that enhance consumers’ attributions of mind to the brands (Puzakova and Kwak 2017). Marketers encourage consumers to see products in humanlike forms, either by creating humanlike characters (e.g., the California Raisins) or through visual representations on product packages (e.g., Capri-Sun). Brands are also given names that signal social roles and familial relationships (e.g., Uncle Ben’s). These tactics are employed to enhance consumer–brand connections and to entice consumers to choose these brands over others (Fournier 1998). Prior work shows that anthropomorphism influences product evaluations (Aggarwal and McGill 2007), emotional responses (Delbaere, McQuarrie, and Phillips 2011), and intentions to replace a product (Chandler and Schwarz 2010). However, simply imbuing products with humanlike traits does not always improve evaluations. The attributes of the consumer, the nature of the context, and brand features impact the valence of the anthropomorphism influence (Maeng and Aggarwal 2018; Puzakova and Kwak 2017). Importantly, past research shows that salient consumer motivations influence the direction of anthropomorphism effects (Kim and McGill 2011). In this regard, prior work identifies two key consumer motivations—social connectedness and effectance—that lead to positive downstream consequences of anthropomorphism (Epley, Waytz, and Cacioppo 2007). For example, people who desire to connect with others tend to prefer anthropomorphized (vs. nonanthropomorphized) products (Chen et al. 2017). In turn, effectance motivation that is driven by the desire to explain and predict an agent’s action also enhances a tendency to humanize (Waytz et al. 2010b). Here, we theorize that the workings of brand anthropomorphism under high distinctiveness motives go beyond the consequences of this phenomenon under social-connectedness or effectance goals. One could argue that a distinctiveness motive is the opposite of a social-connectedness goal in that it leads to a lowered desire for social interactions, which could potentially explain the predicted negative reactions to anthropomorphized brands. However, prior research points out that a distinctiveness goal should be seen as a desire to establish a difference from others rather than to increase social separateness (Tian et al. 2001). For example, Vignoles (2009) distinguishes between the two motives by illustrating a distinctiveness motive as the difference from others through the presentation of unique geometrical figures (e.g., six squares vs. one circle), and the social-separateness goal via the spatial distancing of one circle from the cluster of other circles. To that end, Baumeister and Sommer (1997) posit that if people were to pursue social separateness by being distinctive, it would be most beneficial to cultivate bad unique traits; however, people tend to differentiate themselves on choices that positively distinguish them from others. Marketing scholars also argue that the desire to be different drives the choice of unique products, whereas the motive for social separateness leads to product choices based on internal tastes, which may or may not be unique (Tian et al. 2001). Thus, past work concludes that social connectedness and distinctiveness motives have different underlying forces. Although both distinctiveness and effectance motives are driven by a desire to control, we argue that the fundamental source of this desire varies. The primary drive of the distinctiveness motive is to have agency (or control) over one’s ability to signal identity, whereas the major force of effectance motivation is to understand and predict an agent that is unpredictable in order to more effectively interact with it (Epley et al. 2007). For example, Waytz et al. (2010b) show that the effectance motivation is triggered only when an object behaves randomly or unexpectedly, and giving it a human intent makes the random behavior seem more understandable. In fact, other work shows that effectance motivation leads to a greater preference for anthropomorphized entities (Chen, Sengupta, and Adaval 2013). In contrast, we predict that consumers’ evaluations of an anthropomorphized (vs. nonanthropomorphized) distinctive brand decrease under distinctiveness motives because a brand that is anthropomorphized is seen as having humanlike agency, which then reduces consumers’ own sense of agency. Overall, adopting the view that the distinctiveness motive is driven by the desire for difference rather than social separateness or efficacy, we elaborate on the role of brand anthropomorphism next. BRAND ANTHROPOMORPHISM UNDER DISTINCTIVENESS MOTIVES As documented by prior work (Tian et al. 2001; White and Argo 2011), people motivated by distinctiveness are more attracted to products that are positioned to be distinctive (i.e., products with unique features or that are advertised to be different) because these products satisfy consumers’ differentiation drives compared to nondistinctively positioned products. However, as discussed earlier, consumers with high distinctiveness motives are also concerned with maintaining a sense of agency in self-expression (Bhattacharjee et al. 2014). Here, we propose that humanizing a distinctive brand decreases consumers’ perceived agency in identity expression, thus counteracting the increased preference for a distinctive brand under high distinctiveness goals. This contention is borne out by several streams of research. First, research on self-representation suggests a hierarchy of self-allied objects (McClelland 1951). Studies involving the task of sorting items on the continuum of self to not-self show that possessions are viewed as closer to the self than other people (Prelinger 1959). Second, the reason why products are more self-defining than other humans is that people exert greater control over external objects than over other people (Belk 1988). Pertinent to this, nonhuman entities lack “free will” (Epley and Waytz 2009) and could be controlled and subsumed as “self” by consumers, thus giving consumers more agency in self-expression. In contrast, because other people retain their own identities, this “meaning” or identity of other people cannot be fully controlled by an individual. Since anthropomorphized entities are also viewed as mindful (Epley and Waytz 2009), they are perceived as less likely to be controlled by others and thus are evaluated less positively. It is important to note that we expect this negative effect to occur only when a brand is distinctive and when a distinctiveness goal is salient. This is because a nondistinctive brand is unable to satisfy consumers’ differentiation goals (Tian et al. 2001), and because consumers low in distinctiveness motive are generally less sensitive to cues that reduce their sense of agency in self-expression. This last premise is consistent with prior work showing that when no specific goals are salient, consumers’ preference for anthropomorphized versus nonanthropomorphized brands is no different (Waytz et al. 2010a). For example, Chen et al. (2017) find no effect of product anthropomorphism in the low-social-connectedness condition, whereas Hur, Koo, and Hoffman (2015) show that anthropomorphizing a tempting product does not reduce consumers’ experiences of conflict when a product is goal-irrelevant. H1:When the distinctiveness motive is salient, consumers will develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) distinctive brand. In contrast, we expect no differences between an anthropomorphized and a nonanthropomorphized distinctive brand when the distinctiveness motive is not salient. H2:When the distinctiveness motive is salient, the negative impact of an anthropomorphized (vs. nonanthropomorphized) distinctive brand on brand attitudes will be driven by consumers’ perception of reduced agency in identity expression. Next, given that the motive to communicate one’s distinctiveness is stronger when a product is public (White and Dahl 2006), we argue that the product usage context will moderate the core negative effect. Prior work shows that consumers motivated to be different are less willing to generate positive word of mouth for a public product because a product that becomes popular will lose its exclusivity (Cheema and Kaikati 2010). Dissociative group influence on product choice is also heightened in a public context (White and Dahl 2007). When consumers strive to signal their image through brands, the signaling role of brand personality is restricted to public contexts (Swaminathan, Stilley, and Ahluwalia 2009). Overall, because public products facilitate self-expression (Tian et al. 2001), we expect that the negative effect of brand anthropomorphism under a high distinctiveness motive will emerge only for products used publicly (vs. privately). H3: In the public usage context, consumers will develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) distinctive brand under high distinctiveness motives, whereas this effect will be attenuated in the private usage context. In contrast, we expect no differences between an anthropomorphized and a nonanthropomorphized distinctive brand when the distinctiveness motive is not salient, either in public or private usage contexts. CAN AN ANTHROPOMORPHIZED BRAND BE AGENCY-SUPPORTIVE? Prior work demonstrates that people’s sense of agency depends on contexts and agency-supportive or agency-controlling social roles (Williams et al. 2002). In particular, agency-supportive social contexts emphasize opportunities for free choice and encouragement of personal initiative (Gagne, Ryan, and Bargmann 2003). For example, agency-supportive physicians emphasize freedom of choice rather than forcing a change (Williams et al. 2002). In turn, this supportive style enhances one’s own sense of agency in a goal pursuit and increases satisfaction with the process (Williams et al. 2002). In contrast, agency-controlling contexts where someone else is presumably in control makes people perceive that their actions are a consequence of imposition by others (Deci and Ryan 2008). In turn, these contexts result in a lower sense of agency and decreased satisfaction with the goal-pursuit process (Williams et al. 2002). Here, we argue that positioning a distinctive brand as being agency-supportive may reverse the effect predicted in hypothesis 1. Prior work shows that people are attracted to others who support their sense of agency (La Guardia et al. 2000). For example, the extent to which a human brand (e.g., a celebrity) provides consumers with a sense of agency determines their intensity of brand attachment (Thomson 2006). Thus, positioning an anthropomorphized distinctive brand as a supporter is unlikely to impinge upon consumers’ sense of agency in self-expression. As a result, consumers are likely to evaluate an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-supporter more favorably. Next, we propose that a brand could also be positioned as limiting consumers’ perceived agency in self-expression (i.e., brand-as-controller). Because people develop more distant relationships with others who thwart their agency (La Guardia et al. 2000), we expect that an anthropomorphized (vs. nonanthropomorphized) brand-as-controller should lead to less favorable consumer attitudes toward the brand. Similarly, we predict that consumers are likely to evaluate an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-agent (i.e., a brand positioned as an independent entity focusing on its own features without emphasizing consumer needs) less favorably. Anthropomorphizing a brand increases consumer perception of it as having its own unique identity and as being capable of controlling others’ actions (Epley and Waytz 2009). Because consumers high in distinctiveness are sensitive to cues that thwart their agency in self-expression, they are likely to be sensitive to cues indicating that the brand cannot be fully controlled. Thus, a distinctive anthropomorphized brand-as-agent would signal to consumers that they cannot use it to communicate who they are to others, much like they are unable to use others’ identities to express their own. Finally, we expect that these effects of brand anthropomorphism will occur only under high distinctiveness motives. Consumers with low distinctiveness motives are less concerned with having agency in self-expression and are unlikely to be influenced by social entities (including anthropomorphized brands) that enhance or reduce their sense of perceived agency. H4: When distinctiveness motives are salient, consumers will have more favorable evaluations of an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-supporter, whereas they will exhibit less favorable evaluations of an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-agent and brand-as-controller. In contrast, when distinctiveness motives are low, we expect no differences between anthropomorphized and nonanthropomorphized distinctive brands positioned as a supporter, a controller, or an agent. Four studies test hypotheses 1–4. Study 1 provides evidence for hypothesis 1 and examines the moderating role of consumers’ need for agency. Study 2 shows that hypothesis 1 holds only under high distinctiveness motives and only for a distinctive brand; it also demonstrates the mediating role of agency in identity expression and rules out alternative explanations (i.e., a brand’s ability to provide social distance, perceived personal space violation). Study 3 establishes the moderating role of a product usage context (private vs. public) and rules out the role of perceived efficacy. Finally, using a moderation approach, study 4 shows that positioning an anthropomorphized distinctive brand as supporter (vs. controller or agent) enhances consumers’ sense of agency, thus reversing the effect predicted in hypothesis 1. STUDY 1: THE MODERATING ROLE OF AGENCY The main goal of study 1 is to test hypothesis 1. To activate distinctiveness motive, study 1 employs a subtle manipulation—namely, the visual presentation of either unique or homogenous geometrical shapes. Prior research shows that exposure to unique stimuli that have no a priori meaning, such as arrays of geometrical shapes, enhances the appeal of unique products (Maimaran and Wheeler 2008). Accordingly, the exposure to unique visual arrays is expected to activate a distinctiveness motive and reduce consumer attraction to anthropomorphized (vs. nonanthropomorphized) brands. Next, hypothesis 2 predicts the mediating role of agency in self-expression. Thus, the second goal of study 1 is to show the underlying mechanism by examining individual differences in the need for agency (Clee and Wicklund 1980). We expect that the negative effect of anthropomorphizing a distinctive brand will occur only among individuals motivated to be different who are high in the need for agency and not for those low in need for agency. Participants, Design, and Procedure Three hundred eighty US participants (average age = 36 years old, 43% female) were recruited from Amazon Mechanical Turk (MTurk) and paid $1.00. To enhance the quality of data (Hauser and Schwarz 2016), only MTurk workers with high reputation (i.e., 98% Human Intelligence Task [HIT] approval rate, 5,000 approved HITs) were eligible to complete this task. Across all studies, we employed three data exclusion criteria; that is, we omitted data of those participants who 1) were not diligent in reading and following our instructions, and hence failed the attention check (Oppenheimer, Meyvis, and Davidenko 2009), 2) were extreme outliers (i.e., 2.5 deviations from the mean; Meyvis and van Osselaer 2017), and 3) correctly guessed the true research hypothesis being examined (Sawyer 1975). We believe that it is prudent to use these criteria, as failure to account for these factors has been shown to increase noise and reduce experimental power. For study 1, to identify nondiligent respondents, we used a single-item self-reported measure of attentiveness. None of the MTurk workers selected the lowest value on the attention scale, there were also no outliers, and none of the participants correctly guessed the true research question being examined; thus, we retained the full sample of 380 participants. We employed a 2 (distinctiveness motivation: uniqueness vs. homogeneity visual arrays) × 2 (anthropomorphized vs. nonanthropomorphized brand) between-subjects factorial design; we also included a scale to measure individual differences in need for agency. After exposure to a disclosure form, participants learned that they would take part in two ostensibly unrelated studies. The first was introduced as a “visual perception” study. Respondents were told that the researchers were interested in the effect of figure arrangement on visual perception and accuracy. Consistent with Maimaran and Wheeler (2008), respondents viewed five pictures and indicated the number of squares and circles they saw in each. Each image consisted of different geometrical shapes and conveyed either uniqueness (e.g., OOOOΔOO) or homogeneity (e.g., OOOOOOO). A separate pretest (n = 82) showed that priming respondents with the unique visual (vs. homogeneity) arrays increased their desire for unique and unusual products (4.45 vs. 3.68; F(1, 80) = 6.74, p = .011, η2 = .08), measured on four items (α = .92) used in prior work (Ruvio, Shoham, and Brencic 2008). After the priming task, participants were invited to take part in a “product evaluation” study. They were asked to imagine that they were looking for a new pair of sunglasses and viewed one of two versions of an advertisement (i.e., showing an anthropomorphized or a nonanthropomorphized brand) for a fictitious brand of sunglasses, Safiro. Following prior work (Aggarwal and McGill 2007), we manipulated brand anthropomorphism using verbal (i.e., first- vs. third-person) communication (figure 1). The brand was positioned as emphasizing its distinctive features. The results of the pretest confirmed the effectiveness of brand anthropomorphism and brand distinctiveness manipulations (see web appendix W1 for details). After viewing the ad, participants rated their brand attitudes, responded to the need-for-agency scale, and answered demographic questions. Finally, they reported how much attention they paid to the survey (1 = “not too much attention,” 7 = “paid a lot of attention”) and responded to the demand probe. FIGURE 1 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM IN STUDY 1 FIGURE 1 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM IN STUDY 1 Measures We measured brand attitude with four items (e.g., 1 = “unfavorable, unpleasant, bad, dislike,” 7 = “favorable, pleasant, good, like”; α = .98). Following prior work (Bhattacharjee et al. 2014), participants’ need for agency was operationalized as reactance to compliance and resistance to influence from others and measured with the six-item reactance scale, adopted from Hong and Faedda (1996). A factor analysis with a varimax rotation revealed a one-factor structure, with factor loadings of all items being higher than .64. Thus, we combined all items in one scale (α = .87). The details of all measures are presented in web appendix W2. Results We performed a 2 (distinctiveness motivation: uniqueness vs. homogeneity visual arrays) × 2 (anthropomorphized vs. nonanthropomorphized brand) ANOVA with brand attitude as the dependent variable. There were no significant main effects of either brand anthropomorphism or distinctiveness motivation (ps > .19). However, consistent with hypothesis 1, we found a significant interaction effect between distinctiveness motivation and brand anthropomorphism (F(1, 376) = 5.33, p = .022, ηp2 = .01). We followed up with an analysis of planned contrasts. To elaborate, when distinctiveness motivation was high, consumers developed less favorable brand attitudes when the brand was anthropomorphized (M = 3.67, SD = 1.60) than when it was nonanthropomorphized (M = 4.15, SD = 1.63; F(1, 376) = 3.27, p = .071, ηp2 = .01), but this effect was only marginally significant. In contrast, there was no significant effect of brand anthropomorphism in the homogeneity condition (M = 3.85, SD = 1.78 vs. M = 3.51, SD = 1.74; F(1, 376) = 2.07, p = .151). Furthermore, before examining the role of consumers’ need for agency, we ran a 2 (distinctiveness motivation: uniqueness vs. homogeneity visual arrays) × 2 (anthropomorphized vs. nonanthropomorphized brand) ANOVA with need for agency as the dependent variable to ensure that our manipulation of distinctiveness motive and brand anthropomorphism did not impact the need for agency. The results revealed no main or interaction effects of manipulated variables on need for agency (ps > .60). Next, to test for the moderating effect of consumers’ need for agency, we conducted a moderation analysis (model 3; Hayes 2013) with distinctiveness motivation, brand anthropomorphism, consumers’ need for agency, and two- and three-way interactions among them as the independent variables, and brand attitude as the dependent variable. The results revealed a significant two-way interaction between brand anthropomorphism and distinctiveness motivation (b = –.20, t(372) = – 2.28, p = .023). This effect was qualified by a marginally significant three-way interaction among brand anthropomorphism, distinctiveness motivation, and consumers’ need for agency (b = –.16, t(372) = – 1.84, p = .066). There were no other significant effects (ps > .11). We further performed a floodlight analysis in the high distinctiveness condition using the Johnson-Neyman technique (Spiller et al. 2013). The results revealed a significant negative effect of brand anthropomorphism on brand attitudes among participants with values of need for agency above 3.62 (BJN = –.25, SE = .13, p = .05), with this effect being nonsignificant when need for agency was below 3.62. Discussion Overall, study 1 shows the negative effect of brand anthropomorphism on attitudes toward a brand positioned to be distinctive when distinctiveness motivation is salient. In addition, study 1 provides preliminary evidence of the underlying process by highlighting the moderating role of consumers’ need for agency; the results show that under high distinctiveness motivation, consumers who are high on need for agency develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) brand positioned to be distinctive, whereas no such effect emerges for consumers low in need for agency. STUDY 2: THE ROLE OF BRAND DISTINCTIVENESS Study 2 builds on study 1 in a number of ways. First, study 2 investigates two important boundary conditions: 1) whether the effect is specific only to consumers with high distinctiveness concerns, and 2) whether the effect holds only for brands that are positioned distinctively (vs. nondistinctively). Because a high distinctiveness motive attracts consumers to products that satisfy this goal (Tian et al. 2001), and because consumers are unlikely to be concerned with agency in identity expression under low distinctiveness motives, we expect the hypothesized negative effect of brand anthropomorphism to emerge only when a brand is positioned to be distinctive and only for consumers with a high desire to differentiate themselves. Second, study 2 measures consumers’ agency in identity expression to more directly examine the role of this factor. The third goal of study 2 is to rule out alternative explanations. Prior work shows that humanizing an entity enhances its ability to provide a sense of social connectedness (Epley et al. 2008b). Although our conceptualization of distinctiveness led to our predictions regarding the mediating role of agency in identity expression, it is important to empirically rule out an alternative explanation that an anthropomorphized (vs. nonanthropomorphized) distinctive brand provides a reduced sense of social separateness, thereby lowering brand evaluations of consumers high in distinctiveness motives. Furthermore, prior research by Xu, Shen, and Wyer (2012) finds that physical proximity to other people in a crowded context increases the motivation to seek uniqueness. Thus, as another alternative explanation, it is possible that having a distinctiveness motive may increase consumers’ goal to avoid violation of personal space by increasing their desire to stay away from other people and anthropomorphized entities. Study 2 directly tests these two potential alternative explanations. Participants, Design, and Procedure Four hundred twenty workers who met a 98% HIT approval rate and 5,000 approved HITs as the qualification criteria were recruited from Amazon MTurk. They provided their consent by agreeing to a disclosure and received a $1.00 financial compensation for their participation. We applied the same three data exclusion criteria as in study 1. However, to identify inattentive respondents we used an instructional manipulation check (IMC) (“Instead of entering your MTurk ID number, please enter number six”; Oppenheimer et al. 2009). Prior work identifies that a self-reported measure of attentiveness (as was used in study 1) is less reliable than an IMC because participants may not want to admit their negligence (Oppenheimer et al. 2009). Overall, we omitted the data of those participants (n = 9) who failed the IMC included at the end of the study and the data of extreme outliers (n = 9), leaving a sample of 402 participants (average age = 34 years old, 36% female). Similar to study 1, none of the participants were able to accurately guess the research hypothesis being examined. We employed a 2 (distinctiveness motivation: uniqueness visual arrays vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (brand positioning: distinctive vs. nondistinctive) between-subjects design. The procedure and manipulation of distinctiveness motive were similar to that of study 1. In the control condition, respondents immediately proceeded to the product-evaluation study. All participants were presented with one of four versions of an advertisement (figure 2) for a fictitious brand of watch, Tesmer. After viewing the ad, participants indicated their brand attitudes, responded to questions that measured potential mediating variables, and completed a manipulation check on distinctiveness motivation. The presentation order of mediating variables was randomized to avoid a carryover effect from one construct to the other. Neither the effect of presentation order nor its interaction with any other variables was significant (ps > .11). At the end of the study, participants responded to an IMC question, filled out demographic variables, and responded to the demand probe question. FIGURE 2 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND DISTINCTIVENESS IN STUDY 2 FIGURE 2 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND DISTINCTIVENESS IN STUDY 2 Anthropomorphism and Brand Distinctiveness Manipulation Following prior research (Aggarwal and McGill 2007), we manipulated brand anthropomorphism using first-person (vs. third-person) communication (figure 2). We carefully selected the stimulus for the watch product image since we intended its appearance to be ambiguous with respect to anthropomorphism, such that it could resemble a human face when accompanied by anthropomorphized ad copy but not necessarily be seen as a human face if not prompted by other cues in the ad. We manipulated brand distinctiveness by modifying the ad copy. In the distinctive brand condition, the ad copy was oriented toward distinguishing the brand from other watches. In the nondistinctive brand condition, the message emphasized typical product features and suitability for everyday use (figure 2). The results of the pretest confirmed the effectiveness of brand anthropomorphism and brand distinctiveness manipulations (web appendix W1). Measures We measured brand attitude (α = .96) and desire for unique products (α = .95) with the same items used in study 1. Following prior work (Bellezza, Gino, and Keinan 2014; Bhattacharjee et al. 2014), we assessed participants’ sense of agency in identity expression with two items (e.g., “Owning a Tesmer watch would allow me to freely express who I am”; 1 = “strongly disagree,” 7 = “strongly agree”; r = .89, p < .001; web appendix W2). The brand’s ability to provide social distance was measured with two items (e.g., “Owning a Tesmer watch would allow me to be distant from others”; 1 = “strongly disagree,” 7 = “strongly agree”; r = .80, p < .001). We constructed the questions by adopting items from previous work that delineates social separateness (Vignoles, Chryssochoou, and Breakwell 2002) and modifying them for our context. Finally, the violation of personal space was measured with three items on a Likert-type scale (e.g., “I would feel my personal space is violated”; α = .94), adopted from Harrell, Hutt, and Anderson (1980). The construct correlation matrix is presented in web appendix W3. Results Manipulation Check. A manipulation check on the distinctiveness motive—a 2 (unique visual array priming vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (distinctive vs. nondistinctive brand) ANOVA with the desire for unique products as the dependent variable—showed that participants viewing the uniqueness visual array (vs. control) reported greater desire for unique products (3.76 vs. 3.38; F(1, 392) = 5.05, p = .025, ηp2 = .01), with the main or interaction effects of other variables being nonsignificant (ps > .13). Hypothesis Tests. We performed a 2 (distinctiveness motivation: unique visual arrays vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (distinctive vs. nondistinctive brand) ANOVA with brand attitude as the dependent variable. There was a significant effect of brand anthropomorphism (4.90 vs. 5.18; F(1, 394) = 4.51, p = .034, ηp2 = .01) and a significant interaction between brand distinctiveness and brand anthropomorphism (F(1, 394) = 4.92, p = .027, ηp2 = .01). In support of hypothesis 1, we also found that when the brand was positioned to be distinctive, there was a significant interaction between brand anthropomorphism and distinctiveness motivation (F(1, 394) = 4.28, p = .039, ηp2 = .01). No other main or interaction effects were significant (ps > .14). We followed up with an analysis of planned contrasts. First, confirming our prediction, respondents with high distinctiveness motive developed more favorable attitudes toward a nonanthropomorphized distinctive (vs. nondistinctive) brand (5.62 vs. 5.04; F(1, 394) = 4.79, p = .029, ηp2 = .01; table 1). Because anthropomorphizing a brand counteracts the increased preference for a distinctive brand under high distinctiveness goals, the planned contrast tests comparing an anthropomorphized and a nonanthropomorphized brand in the high distinctiveness condition revealed that anthropomorphizing a distinctive brand decreased respondents’ brand attitudes (4.67 vs. 5.62; F(1, 394) = 14.82, p < .001, ηp2 = .04). Finally, because a nondistinctive brand is unable to satisfy consumers’ differentiation goals, the planned contrasts in the condition of high distinctiveness motive and when a brand was nondistinctive showed no significant effects of anthropomorphism (5.06 vs. 5.04; F(1, 394) = .005, p = .94). We also found no effect of brand anthropomorphism on brand attitudes in the low distinctiveness condition (distinctive brand: 4.85 vs. 5.04; F(1, 394) = .63, p = .48; nondistinctive brand: 5.03 vs. 5.03; F < .001, p = .98). TABLE 1 CELL MEANS IN STUDY 2   Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)    Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a significant difference of the cell mean from (b). TABLE 1 CELL MEANS IN STUDY 2   Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)    Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a significant difference of the cell mean from (b). Process Analysis. First, the results of a 2 (distinctiveness motivation: unique visual arrays vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (distinctive vs. nondistinctive brand) ANOVA with agency in identity expression as the dependent variable confirmed that anthropomorphizing a distinctive brand reduced consumers’ perceived agency in identity expression in the high distinctiveness condition (3.68 vs. 4.39; F(1, 394) = 5.26, p = .022, ηp2 = .01). This effect was nonsignificant in all other conditions (p > .27; table 1). Next, we conducted a moderated mediation analysis using a bootstrapping approach (model 12; Hayes 2013) with distinctiveness motivation, brand anthropomorphism, brand distinctiveness, and all two- and three-way interactions among them as the independent variables, brand attitude as the dependent variable, and agency in identity expression as the mediating variable. Results revealed that when distinctiveness motive was high and the brand was distinctive, there was a significant effect of brand anthropomorphism on brand attitude through consumers’ agency in identity expression (a point estimate of the effect = –.36; 95% CI [–.68, –.07]). There were no significant effects in other conditions (web appendix W4). Overall, the results confirmed hypothesis 2: the negative effect of anthropomorphism on brand attitude is driven by the perception of an anthropomorphized distinctive brand reducing consumers’ sense of agency in identity expression. Ruling Out Alternative Explanations. To further examine the underlying process, we used a bootstrapping approach (model 12; Hayes 2013) to test the regression model with brand anthropomorphism, brand distinctiveness, distinctiveness motivation, and all two-way and three-way interactions among them as the independent variables; consumers’ agency in identity expression, a brand’s ability to provide social distance, and perceived personal space violation as three mediators; and brand attitude as the dependent variable. Consistent with our expectations, we found that in the high distinctiveness condition and when a brand was positioned distinctively, consumers’ agency in identity expression mediated the effect of brand anthropomorphism on brand attitudes (a point estimate of the effect = –.37; 95% CI [–.70, –.05]). In the same condition, the indirect paths for the other two mediators were nonsignificant (the brand’s ability to provide social distance: a point estimate of the effect = .009; 95% CI [–.01, .08]; perceived personal space violation: a point estimate of the effect = .03; 95% CI [–.01, .11]). All other paths for all three mediators in other conditions were nonsignificant (details are presented in web appendix W4). Next, although the results of the parallel mediation test reveal a mediating role of agency in self-expression, it is possible that brand anthropomorphism reduces consumers’ perceptions of distance, thus simultaneously increasing perceived personal space violation and, as a result, leading to a diminished sense of agency. To ensure that agency in identity expression mediates over and above the other two process variables, in the condition of high distinctiveness motive and when a brand is positioned distinctively, we conducted a serial mediation analysis (i.e., a brand’s ability to provide social distance → perceived personal space violation → agency in identity expression). The only significant path from brand anthropomorphism to brand attitude emerged for agency in identity expression (a point estimate of the effect = –.27; 95% CI [–.59, –.01]; web appendix W4). Discussion Study 2 replicates the findings of study 1 with a different product (i.e., a watch) and establishes two important boundary conditions: the negative impact of brand anthropomorphism emerges only for brands that are likely to signal consumers’ distinctiveness and when consumers’ distinctiveness concerns are high. Furthermore, study 2 shows that consumers’ perceptions of lowered agency in identity expression drive the negative impact of brand anthropomorphism when distinctiveness motives are salient. The nonsignificant indirect paths through both a brand’s ability to provide social distance and perceived personal space violation rule out alternative process explanations and further enhance our confidence in the role of agency in identity expression. STUDY 3: THE ROLE OF PRIVATE VERSUS PUBLIC USAGE CONTEXT The major goal of study 3 is to examine the premise that the negative effect of brand anthropomorphism will hold only in a public (vs. private) product usage context because public products are visible to others and are important for self-expression (Tian et al. 2001), as laid out in hypothesis 3. The second goal of study 3 is to empirically rule out the alternative explanation that the distinctiveness motive leads to effects similar to those triggered by effectance motivation. However, as noted earlier, we propose that distinctiveness motive results in a greater need for agency in identity expression, while effectance leads to a greater need to make sense of an agent’s unpredictability. If distinctiveness motivation were really a subset of effectance motivation, then it should increase consumers’ desire to understand and predict the performance of the brand (Epley et al. 2007). If this is the case, then one possible reason why anthropomorphizing a distinctive brand reduces brand attitudes when distinctiveness goals are salient is that anthropomorphizing an entity that is already unique and different (e.g., a distinctive brand) can potentially reduce a consumer’s ability to comprehend it, thereby challenging consumers’ effectance motivation. Hence, study 3 aims to empirically rule out the potential mediating role of consumers’ perceived efficacy. In study 3 we manipulate the distinctiveness motive utilizing an alternative procedure. We asked participants to elaborate on the importance of distinctiveness (Cheema and Kaikati 2010). Finally, study 3 examines the impact of brand anthropomorphism on an additional dependent variable that is closer to an actual purchasing behavior—namely, brand consideration. Participants, Design, and Procedure Study 3 used a 2 (distinctiveness motivation: high vs. low) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (product usage context: public vs. private) between-subjects design. Three hundred forty-two undergraduate business students at a private US East Coast university participated in study 3 in exchange for extra credit. Following the procedure of studies 1 and 2, we omitted the data of extreme outliers (n = 4) and of one respondent who reported an awareness of the study purpose. There was no attention check in this study. Overall, our final sample consisted of 337 participants (57% female). Upon arrival at a laboratory room with the maximum capacity of 27 students, participants were provided with the consent form and were told that they would take part in two unrelated studies. The first was introduced as examining the use of language in different situations, whereas the second study involved brand evaluations. Consistent with Cheema and Kaikati (2010), participants in the high distinctiveness condition elaborated on the importance of distinctiveness (being different from others). In the low distinctiveness (control) condition, respondents elaborated on the importance of a neutral activity (eating three times a day). A separate pretest (n = 168) with participants from the same population as in the main study confirmed that elaborating on distinctiveness (vs. on the importance of eating three times a day) increased participants’ desire for unique and unusual products (4.80 vs. 4.30; F(1, 166) = 9.57, p = .002, ηp2 = .06), measured with the same four items (α = .83) as used in studies 1 and 2. After the elaboration exercise, approximately half of the participants were asked to imagine shopping for an electric outdoor bike (a public usage context), and the rest were asked to imagine shopping for an in-home exercise electric bike (a private usage context) called Zidley. Next, they viewed one of four versions of an advertisement where we manipulated brand anthropomorphism as well as product features that made the product more appropriate for public or private usage. Brand anthropomorphism was manipulated via a combination of visual and verbal elements (Aggarwal and McGill 2007; figure 3). The results of the pretest showed that our manipulations were successful (web appendix W1). After viewing the ads, participants provided their ratings on brand attitude and brand consideration and responded to questions measuring agency in identity expression and perceived efficacy. The survey ended with demographic questions and a demand probe. FIGURE 3 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND PRODUCT USAGE CONTEXT IN STUDY 3 FIGURE 3 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND PRODUCT USAGE CONTEXT IN STUDY 3 Measures Brand attitude (α = .93) and agency in identity expression (r = .84, p < .001) were measured with the same items as in study 2. As a measure of brand consideration (Puzakova, Kwak, and Rocereto 2013), respondents provided their agreement with two statements (e.g., “I can imagine buying a Zidley bike”; 1 = “strongly disagree,” 7 = “strongly agree”; r = .81, p < .001; web appendix W2). Perceived efficacy was measured with two items adopted from prior work (Waytz et al. 2010b) and modified for our context (e.g., “Please indicate the extent to which you feel you can understand the Zidley brand”; 1 = “not at all,” 7 = “very much”; r = .33, p < .001). The construct correlation matrix is presented in web appendix W3. Results Hypothesis Tests. We ran two ANOVA tests (one with brand attitude and the other with brand consideration as the dependent variables) that included brand anthropomorphism, usage context, and distinctiveness motivation as the independent variables. The results revealed a significant interaction effect between distinctiveness motive and brand anthropomorphism for brand attitude (F(1, 329) = 3.92, p = .049, ηp2 = .01) and a marginally significant effect for brand consideration (F(1, 329) = 2.97, p = .086, ηp2 = .01). In addition, the findings showed a significant interaction effect between brand distinctiveness and brand anthropomorphism on brand attitude when the product usage context was public (F(1, 329) = 4.55, p = .034, ηp2 = .01), with no interaction effect when the context was private (F(1, 329) = .54, p = .462). Finally, the results revealed that when the distinctiveness motive was high, there was a marginally significant interaction between brand anthropomorphism and product usage context for brand attitude (F(1, 329) = 3.49, p = .063, ηp2 = .01); however, this effect became nonsignificant in the control condition (F(1, 329) = .22, p = .64). No other main or interaction effects were significant (ps > .11). Further planned contrasts showed that in the high distinctiveness condition and when the product usage context was public, participants had less favorable brand attitudes (4.42 vs. 4.91; F(1, 329) = 4.20, p = .041, ηp2 = .01) and reported lower brand consideration (3.40 vs. 3.95; F(1, 329) = 3.44, p = .065, ηp2 = .01) when the brand was anthropomorphized (vs. nonanthropomorphized). Consistent with a theoretical prediction that a high distinctiveness motive enhances attraction to brands positioned to be distinctive, we found that respondents exhibited more favorable attitudes toward a nonanthropomorphized brand in the distinctiveness (vs. control) condition only when the usage context was public (4.91 vs. 4.41; F(1, 329) = 4.49, p = .091, ηp2 = .01). This effect was nonsignificant for brand consideration; however, the mean difference was in the predicted direction (3.95 vs. 3.50; F(1, 329) = 1.51, p = .22). Interestingly, a marginally significant difference emerged in brand attitudes between the anthropomorphized and nonanthropomorphized brand in the control condition and when the product usage context was private (4.65 vs. 4.14; F(1, 329) = 2.89, p = .09, ηp2 = .009). We address this finding in the discussion section. Finally, we found no significant differences between the means across all other conditions for both dependent variables (ps > .19; table 2). TABLE 2 CELL MEANS IN STUDY 3   Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)    Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (**)indicates a marginally significant difference (p < .10) between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a marginally significant difference of the cell mean from (b); (a1)indicates a significant difference of the cell mean from (b1); (c)indicates a significant difference of the cell mean from (b). TABLE 2 CELL MEANS IN STUDY 3   Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)    Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (**)indicates a marginally significant difference (p < .10) between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a marginally significant difference of the cell mean from (b); (a1)indicates a significant difference of the cell mean from (b1); (c)indicates a significant difference of the cell mean from (b). Process Analysis. First, consistent with our expectations, we found that when the usage context was public and the distinctiveness motive was high, brand anthropomorphism attenuated participants’ agency in identity expression compared to a nonanthropomorphized brand condition (2.87 vs. 3.42; F(1, 329) = 4.28, p = .039, ηp2 = .01), whereas we observed no significant effects of anthropomorphism on perceived agency in other conditions (ps > .12). Next, we conducted two moderated mediation analyses (one for brand attitude and one for brand consideration as the dependent variables) using a bootstrapping approach (model 12; Hayes 2013) with brand anthropomorphism as the independent variable, product usage context and distinctiveness motivation as two moderators, and agency in identity expression as the mediating variable. The results revealed that when the distinctiveness motive was high and the product was public, there was a significant effect of brand anthropomorphism on brand attitude (a point estimate of the effect = –.21; 95% CI [–.43, –.01]) and on brand consideration (a point estimate of the effect = –.29; 95% CI [–.57, –.04]) through perceived agency in identity expression. There were no other significant indirect paths through agency in identity expression (web appendix W4). Ruling Out the Role of Perceived Efficacy. First, the results of a confirmatory factor analysis provided evidence for discriminant validity between agency in identity expression and perceived efficacy: their squared correlation of .14 was lower than both the average variance extracted for agency in identity expression (.84) and the average variance extracted for perceived efficacy (.34). Second, to further examine the underlying process of the effect of brand anthropomorphism, we used a bootstrapping approach (model 12; Hayes 2013) to test the regression model with brand anthropomorphism, product usage context, distinctiveness motive, and all two-way and three-way interactions among them as the independent variables; consumers’ agency in identity expression and perceived efficacy as two mediators; and brand attitude as the dependent variable. Consistent with our expectations, we found that in the high distinctiveness condition and in the public usage context, consumers’ agency in identity expression mediated the effect of brand anthropomorphism on brand attitudes (a point estimate of the effect = –.18; 95% CI [–.39, –.02]). In the same condition, the indirect path for perceived efficacy was nonsignificant (a point estimate of the effect = –.04; 95% CI [–.17, .08]). All other paths for the two mediators in other conditions were nonsignificant (web appendix W4). The outcomes for brand consideration were similar (web appendix W4). Discussion The results of study 3 are consistent with our theoretical prediction that when the distinctiveness motivation is activated, consumers evaluate an anthropomorphized (vs. nonanthropomorphized) brand positioned to be distinctive less positively; however, the negative effect of brand anthropomorphism occurs only for public (vs. private) usage contexts. These results are in accord with prior work showing that a signaling of uniqueness through choice of distinctive products occurs when products are public (Tian et al. 2001). The findings of study 3 further rule out the role of perceived efficacy in driving the negative effect of anthropomorphizing a distinctive brand under high distinctiveness goals. These outcomes are not surprising given that effectance motivation arises primarily in response to entities that are unexpected (Waytz et al. 2010b), and anthropomorphizing a distinctive brand does not decrease consumers’ perceived efficacy. Interestingly, the findings of study 3 also show that consumers develop more favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) brand in a private usage context in the control condition. Our speculation is that elaborating on the importance of eating three times a day (which was intended to be a control) might have inadvertently activated participants’ health goals, and an in-home exercise bike was perceived as related to this goal. Because attributions of mental capacities to technological agents enhance trust that they will perform their intended functions competently (Waytz, Heafner, and Epley 2014), it is possible that participants viewed an electric exercise bike as being able to assist them with their goal more effectively. In study 4, we directly test the possibility that when a distinctive brand is positioned as supporting consumer agency in achieving distinctiveness, the negative effect of brand anthropomorphism under high distinctiveness concerns reverses. STUDY 4: BRAND AS SUPPORTER, AGENT, OR CONTROLLER Agency-supportive social roles enhance consumers’ agency in achieving a goal (Williams et al. 2002). Hence, we expect the negative effect of brand anthropomorphism on consumer evaluations under high distinctiveness motivation to be reversed when the anthropomorphized brand is positioned as supporting consumers’ agency in identity expression, thereby providing further evidence for the underlying role of consumers’ agency in self-expression. Consequently, in study 4 we examine the moderating effect of anthropomorphizing a brand-as-supporter (vs. -controller and -agent) on consumer responses. Participants, Design, and Procedure Participants were recruited on Amazon MTurk and received a $1 monetary compensation. The HIT was available to workers with high reputation (98% of HIT approval rate and 5,000 approved HITs). To identify inattentive workers, we included a prescreen criterion based on an IMC (i.e., “Please indicate number ‘six’ in the text box below instead of ‘Today’s date’”; Oppenheimer et al. 2009), similar to that used in study 2. Five hundred forty-five MTurk workers passed the IMC question and were allowed to participate in the study. As in study 3, we omitted the data of extreme outliers (n = 5), thus leaving a sample of 540 participants (average age = 35; 39% female). None of the participants could accurately guess the true nature of the research hypothesis being examined. We employed a 2 (distinctiveness motivation: high vs. low) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) between-subjects factorial design. First, we manipulated high versus low distinctiveness motivation in a manner similar to that of study 3. In the high distinctiveness condition, participants were asked to elaborate on the importance of distinctiveness (being different from others). In the low distinctiveness (control) condition, respondents wrote about an innocuous topic (US geography; Berger and Shiv 2011). Next, participants were instructed to imagine that they were considering buying a new pair of sneakers and that, while searching for the shoes, they were exposed to an ad for the Geono (hypothetical) brand. Subsequently, they viewed one of six versions of the advertisement (three brand roles × two brand anthropomorphism conditions). Brand anthropomorphism was manipulated similar to that of previous studies (figure 4). We manipulated brand roles by modifying the ad copy, relying on prior research about autonomy-supportive and autonomy-controlling social roles. Prior work shows that central features of agency support are encouraging individuals to make their own choices, promoting a sense of ownership over one’s behaviors, emphasizing self-direction, and providing goal-related advice intended as information rather than direct pressure (Gagne et al. 2003; Williams et al. 2002). Building on this prior work, in the brand-as-supporter condition, we positioned the brand as encouraging consumers to stay true to their own unique style and emphasizing consumers’ own choices and behaviors, while providing options and supporting these behaviors (figure 4). In contrast to agency-supportive styles, in a controlling social role individuals generally direct others to behave in a certain way, fail to elicit others’ perspectives, provide no options, and use assertive language such as “must” and “need to” (Sheldon and Filak 2008; Silva et al. 2010). Relying on this work, in the brand-as-controller condition, we positioned the brand as overtly controlling consumers’ behaviors and choices in expressing their distinctiveness (figure 4). Finally, in the brand-as-agent condition, the ad copy emphasized the brand’s features and its distinctiveness, without focusing on the consumer. The results of the pretest revealed the effectiveness of our manipulations of brand anthropomorphism and the three brand roles (web appendix W1). After viewing the ad, participants reported their brand attitudes and agency in identity expression as the key dependent variables, and distinctiveness motivation as the manipulation-check measure. To ensure that our manipulation of brand role did not inadvertently influence perceived effectiveness of the brand or the impact of the brand in fulfilling distinctiveness goals, we measured perceived brand effectiveness. The study ended with the demographic information section and the demand probe. FIGURE 4 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND ROLES IN STUDY 4 FIGURE 4 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND ROLES IN STUDY 4 Measures Brand attitude (α = .97), agency in identity expression (r = .88, p < .001), and manipulation check on consumer distinctiveness motivation (α = .94) were measured with the same items as in studies 1–3. As a measure of brand effectiveness, we asked participants to indicate how certain they felt that the Geono brand would make them distinct from others (1 = “not at all certain,” 7 = “very certain”). Results Distinctiveness Motivation Manipulation Check and Brand Effectiveness. The results of a 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) × 2 (distinctiveness motivation: high vs. low) ANOVA revealed that consumers had greater desire for unique and unusual products in the high (vs. low) distinctiveness condition (4.24 vs. 3.80; F(1, 524) = 9.47, p = .002, ηp2 = .02), with all other effects being nonsignificant (ps > .27). Next, the results of a 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) × 2 (distinctiveness motivation: high vs. low) ANOVA test with brand effectiveness as the dependent variable showed no significant main or interaction effects (ps > .11). Importantly, the contrast tests showed no significant differences in perceived brand effectiveness among brand roles (Msupporter = 4.46 vs. Mcontroller = 4.32 vs. Magent = 4.50; ps > .31). Hypothesis Tests. A 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) × 2 (distinctiveness motivation: high vs. low) ANOVA with brand attitude as the dependent variable showed a significant effect of distinctiveness motivation (Mhigh = 4.42 vs. Mlow = 3.64; F(1, 528) = 26.10, p < .001, ηp2 = .05) and a two-way interaction between brand anthropomorphism and brand role (F(1, 528) = 5.29, p = .005, ηp2 = .02), with all other effects being nonsignificant (ps > .16). Further contrast analysis showed that participants had more favorable attitudes toward a nonanthropomorphized brand when their distinctiveness motives were high (vs. low) (brand-as-controller: 4.72 vs. 3.56; F(1, 528) = 9.18, p = .003, ηp2 = .02; brand-as-agent: 4.52 vs. 3.67; F(1, 528) = 5.65, p = .018, ηp2 = .01; brand-as-supporter: 4.19 vs. 3.49; F(1, 528) = 3.45, p = .064, ηp2 = .006). Next, within the high distinctiveness condition, results revealed that participants developed less favorable brand attitudes toward an anthropomorphized (vs. nonanthropomorphized) brand in the brand-as-agent condition (3.73 vs. 4.52; F(1, 528) = 4.68, p = .033, ηp2 = .01). In contrast, in the brand-as-supporter condition, consumers had more favorable brand attitudes (5.18 vs. 4.19; F(1, 528) = 6.83, p = .009, ηp2 = .01) when the brand was anthropomorphized (vs. nonanthropomorphized). Furthermore, although the mean difference was in the expected direction in the brand-as-controller condition, the effect of brand anthropomorphism was nonsignificant (4.16 vs. 4.72; F(1, 528) = 1.98, p = .16). However, the results also showed that participants reported more favorable attitudes toward an anthropomorphized brand-as-supporter compared to both an anthropomorphized brand-as-agent and a brand-as-controller (Msupporter = 5.18 vs. Mcontroller = 4.16 vs. Magent = 3.73; ps < .01), whereas there were no significant differences among brand roles in the nonanthropomorphized brand condition (ps > .16). Finally, the contrast tests within the low distinctiveness condition showed no significant effects of brand anthropomorphism on brand attitudes (ps > .25). Overall, hypothesis 4 was partially supported (table 3). TABLE 3 CELL MEANS IN STUDY 4   Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)    Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; (*) indicates a significant difference between anthropomorphized and nonanthropomorphized brand conditions; (**) indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition at p = .071; pairs of cell means (b1–b2), (c1–c2), (d1–d2), (h1–h2), (i1–i2), (g1–g2), and (L1–L2) are significantly different at p < .05; a pair of cell means (a1–a2) is marginally significantly different at p = .064; (b1) is significantly different from both (f) and (g) at p < .05; (i1) is significantly different from both (k1) and (m1) at p < .05. TABLE 3 CELL MEANS IN STUDY 4   Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)    Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; (*) indicates a significant difference between anthropomorphized and nonanthropomorphized brand conditions; (**) indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition at p = .071; pairs of cell means (b1–b2), (c1–c2), (d1–d2), (h1–h2), (i1–i2), (g1–g2), and (L1–L2) are significantly different at p < .05; a pair of cell means (a1–a2) is marginally significantly different at p = .064; (b1) is significantly different from both (f) and (g) at p < .05; (i1) is significantly different from both (k1) and (m1) at p < .05. Process Analysis. We employed a moderated mediation analysis with a multicategory moderator variable (brand role) using a sequential coding for three brand-role conditions (model 12; Hayes 2013). Results showed that in the high distinctiveness condition, there was a significant negative indirect effect of brand anthropomorphism on brand attitude through agency in identity expression when the brand was positioned as an agent (a point estimate of the effect = –.53; 95% CI [–1.08, –.01]) or as a controller (a point estimate of the effect = –.62; 95% CI [–1.24, –.06]). In contrast, consumers perceived a greater agency in self-expression when a brand-as-supporter was anthropomorphized (vs. nonanthropomorphized), which in turn led to a more positive effect of brand anthropomorphism on brand attitudes (a point estimate of the effect = .55; 95% CI [.01, 1.08]). There were no significant indirect paths in other conditions (web appendix W4). Further analysis within brand anthropomorphism revealed that in the high distinctiveness condition, positioning an anthropomorphized brand as supporter (vs. agent) increased perceived agency in identity expression, which led to more positive brand attitudes (a point estimate of the effect = .83; 95% CI [.31, 1.34]). Similarly, greater agency in identity expression explained more favorable attitudes toward an anthropomorphized brand-as-supporter compared to a controller (a point estimate of the effect = .41; 95% CI [.15, 1.69]). Discussion Results of study 4 partially support hypothesis 4. When the brand is positioned as an agent, in keeping with hypothesis 4 consumers with salient distinctiveness motives develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) distinctive brand, and view it as reducing their agency in identity expression. In contrast, brand-as-supporter increases the extent to which consumers view the brand as enhancing their agency in identity expression, thus leading to more favorable evaluations of an anthropomorphized (vs. nonanthropomorphized) brand. In the brand-as-controller condition, we found only a directionally lower attitude toward an anthropomorphized (vs. nonanthropomorphized) brand. However, our results are consistent with our theorizing that consumers perceive an anthropomorphized brand-as-supporter more favorably than both an anthropomorphized brand-as-agent and a brand-as-controller. These outcomes dovetail well with prior work showing that people develop stronger attachments to others who satisfy (vs. thwart) their agency needs (Thomson 2006) and contribute to this work by establishing the key role of positioning an anthropomorphized distinctive brand-as-supporter under consumers’ heightened concerns to be different. Overall, study 4 provides evidence for the mediating mechanism of agency in identity expression using a moderation approach (through manipulating different brand roles). GENERAL DISCUSSION This research provides converging evidence for our theory that anthropomorphism has a detrimental effect on evaluations of a distinctive brand under high distinctiveness motives. This finding contributes to prior work establishing the role of salient consumer goals in the direction of anthropomorphism effects (Kim and McGill 2011; Waytz et al. 2010b). In combination, the results of studies 1–4 show convergent evidence that brand anthropomorphism negatively affects perceived sense of agency in identity expression, through the moderating roles of consumers’ need for agency (study 1) and positioning the brand as agency-supporting versus agency-controlling (study 4), and through direct measurements of perceived agency in identity expression (studies 2–4). Study 2 also rules out an alternative explanation related to the potential impact of an anthropomorphized brand’s ability to provide social connectedness and a perceived violation of personal space when a distinctiveness motive is salient, whereas study 3 further rules out the role of perceived efficacy. In addition, studies 2 and 3 shed light on boundary conditions by establishing that the core negative effect of anthropomorphizing a brand holds only when consumers’ distinctiveness goals are heightened, only when a brand is positioned to be distinctive, and only when a product is publicly consumed. Importantly, our findings provide novel insights into consumer reactions to positioning a brand as supporting or reducing consumers’ sense of agency in self-expression. That is, study 4 demonstrates that when consumers’ distinctiveness concerns are high, an anthropomorphized brand-as-supporter enhances consumers’ sense of agency in identity expression, and thus brand evaluations, compared to both an anthropomorphized brand-as-agent and an anthropomorphized brand-as-controller. Theoretical and Managerial Implications Our work contributes to the literature on the role of products and brands in expressing the self (Belk 1988). Prior research shows that people infer the identities of others from both their possessions (Escalas and Bettman 2005) and the friendships that they develop (Cialdini and De Nicholas 1989). However, when practitioners imbue brands with humanlike qualities, consumers perceive these brands in a manner similar to how they perceive other humans (Kwak, Puzakova, and Rocereto 2015). As a result, our work provides crucial insights regarding which entities (i.e., anthropomorphized vs. nonanthropomorphized) consumers choose to self-express when seeking to be different from others. That is, by bringing together diverse research streams on signaling power of brands (Ahuvia 2005) and identity expression as the major driver of distinctiveness (Vignoles 2011), and by highlighting the key role of agency in self-expression (Bhattacharjee et al. 2014), the current work is the first to shed light on the extent to which consumers would prefer to use a nonanthropomorphized versus an anthropomorphized distinctive brand to signal their uniqueness. Our results are consistent with prior work showing that the intention to signal one’s identity through distinctive products is attenuated in private conditions (Berger and Heath 2007), and we add to this work by revealing that the negative effect of anthropomorphizing a distinctive brand dissipates when a product is used privately. Our research also contributes to anthropomorphism theory by unveiling counterintuitive outcomes of the influence of distinctiveness goals on the valence of the anthropomorphism effects. The current work gives insights into the unique process of brand anthropomorphism through reduced consumer agency in identity expression, as opposed to being the instrument for social connectedness or effectance motivation, as shown in prior work (Epley et al. 2008b). This understanding is crucial from a theoretical standpoint because knowledge about the role of unique goals in consumer reactions to anthropomorphized entities leads to significant insights about the downstream effects of anthropomorphism. This research further contributes to prior work highlighting consumers’ and marketers’ actions that can decrease the uniqueness of a product. For example, Cheema and Kaikati (2010) show that positive word of mouth about a unique brand decreases its perceived uniqueness. Our work adds novel insights by showing that anthropomorphizing a distinctive brand decreases its attractiveness under distinctiveness concerns. This conclusion has an important managerial implication because the decreased ability of a brand to confer distinctiveness on the consumer could result in lower firm profits. Our findings thus suggest that the managers of uniquely positioned brands or otherwise unique products should exercise caution when employing a brand anthropomorphism strategy. While brand anthropomorphism could be successful with certain consumer groups (e.g., with the assimilation goal), it might backfire with other consumers who seek distinctiveness. For example, if a woman buys upscale “Wink Face” shoes from the Alice and Olivia brand with the goal to assimilate with others at a sophisticated social function, humanlike designs of shoes may be appealing. In contrast, if a man buys a watch to signal his distinctiveness, the anthropomorphized look of the Blue Wild Face Swatch will likely be less attractive due to a decreased sense of agency in identity expression. Extensions and Directions for Future Research Several interesting research extensions are possible. The optimal distinctiveness theory states that the forces of distinctiveness and assimilation are interconnected (Brewer 1991). Thus, people tend to choose brands that simultaneously signal difference from one group and similarity to another group (Chan, Berger, and Van Boven 2012). Our work examined the goal of difference from others; however, the effects of brand anthropomorphism under the simultaneous goals of being distinct and similar to others could be a fruitful area for future research. Interestingly, given that distinctiveness in the form of difference from others has negative connotations in collectivistic cultures (Vignoles et al. 2002), our findings raise the question of whether our hypothesized effects would hold across cultures. In addition, prior work shows that some cultures (e.g., Italy, Australia) define human nature more in terms of warm traits and emotions (vs. mental states; Bain et al. 2012). Hence, it is possible that in these cultures, the negative effect of humanizing a brand on agency in identity expression may not occur. Furthermore, prior research shows that consumers strategically choose unique products to cope with social exclusion (Wan, Xu, and Ding 2014). Will social exclusion then lead to lower evaluations of anthropomorphized (vs. nonanthropomorphized) unique brands because consumers are motivated to be different? If so, such an effect could present a boundary condition for prior research showing that consumers are attracted to anthropomorphized brands when they feel excluded (Epley et al. 2008a). Future studies might explore in greater detail the specific circumstances when social exclusion may change evaluations of anthropomorphized brands. In conclusion, this research establishes a systematic impact that consumers’ differentiation motives have on the evaluations of anthropomorphized (vs. nonanthropomorphized) distinctive brands. Results suggest that anthropomorphizing unique brands hampers consumers’ perceived sense of agency in identity expression. When consumers desire uniqueness, they dislike humanness. DATA COLLECTION INFORMATION Data for studies 1, 2, and 4 were collected on Amazon Mechanical Turk by the first author. The data collection periods are as follows: study 1 in summer 2015, study 2 in fall 2016, and study 4 in spring 2017. The first author collected data for study 3 at Lehigh University in fall 2017. The first author conducted primary analyses, and both authors reviewed and discussed the results. The authors thank the editor, the associate editor, the three anonymous reviewers, and the trainee reviewer for their excellent guidance throughout the review process. Financial support from the College of Business and Economics, Lehigh University, and the Social Sciences and Humanities Research Council of Canada is gratefully acknowledged. Supplementary materials are included in the web appendix accompanying the online version of this article. References Aggarwal Pankaj, McGill Ann L. ( 2007), “ Is That Car Smiling at Me? Schema Congruity as a Basis for Evaluating Anthropomorphized Products,” Journal of Consumer Research , 34 4, 468– 79. Google Scholar CrossRef Search ADS   Aggarwal Pankaj, McGill Ann L. ( 2012), “ When Brands Seem Human, Do Humans Act Like Brands? Automatic Behavioral Priming Effects of Brand Anthropomorphism,” Journal of Consumer Research , 39 2, 307– 23. Google Scholar CrossRef Search ADS   Ahn Hee-Kyung, Kim Hae Joo, Aggarwal Pankaj ( 2014), “ Helping Fellow Beings: Anthropomorphized Social Causes and the Role of Anticipatory Guilt,” Psychological Science , 25 1, 224– 9. Google Scholar CrossRef Search ADS   Ahuvia Aaron C. ( 2005), “ Beyond the Extended Self: Loved Objects and Consumers’ Identity Narratives,” Journal of Consumer Research , 32 1, 171– 84. Google Scholar CrossRef Search ADS   Bain Paul, Vaes Jeroen, Kashima Yoshihisa, Haslam Nick, Guan Yanjun ( 2012), “ Folk Conceptions of Humanness: Beliefs about Distinctive and Core Human Characteristics in Australia, Italy, and China,” Journal of Cross-Cultural Psychology , 43 1, 53– 8. Google Scholar CrossRef Search ADS   Baumeister Roy F. ( 1998), “The Self,” in The Handbook of Social Psychology , ed. Gilbert Daniel T., Fiske Susan T., Lindzey Gardner, New York: McGraw-Hill, 680– 740. Baumeister Roy F., Sommer Kristin L. ( 1997), “ What Do Men Want? Gender Differences and Two Spheres of Belongingness: Comment on Cross and Madson,” Psychological Bulletin , 122 1, 38– 44. Google Scholar CrossRef Search ADS   Belk Russell W. ( 1988), “ Possessions and the Extended Self,” Journal of Consumer Research , 15 2, 139– 68. Google Scholar CrossRef Search ADS   Bellezza Silvia, Gino Francesca, Keinan Anat ( 2014), “ The Red Sneakers Effect: Inferring Status and Competence from Signals of Nonconformity,” Journal of Consumer Research , 41 1, 35– 54. Google Scholar CrossRef Search ADS   Berger Jonah, Heath Chip ( 2007), “ Where Consumers Diverge from Others: Identity Signaling and Product Domains,” Journal of Consumer Research , 34 2, 121– 34. Google Scholar CrossRef Search ADS   Berger Jonah, Shiv Baba ( 2011), “ Food, Sex, and the Hunger for Distinction,” Journal of Consumer Psychology , 21 4, 464– 72. Google Scholar CrossRef Search ADS   Bhattacharjee Amit, Berger Jonah, Menon Geeta ( 2014), “ When Identity Marketing Backfires: Consumer Agency in Identity Expression,” Journal of Consumer Research , 41 2, 294– 309. Google Scholar CrossRef Search ADS   Brewer Marilynn B. ( 1991), “ The Social Self: On Being the Same and Different at the Same Time,” Personality and Social Psychology Bulletin , 17 5, 475– 82. Google Scholar CrossRef Search ADS   Burns David J., Warren Homer B. ( 1995), “ Need for Uniqueness: Shopping Mall Preference and Choice Activity,” International Journal of Retail & Distribution Management , 23 12, 4– 12. Google Scholar CrossRef Search ADS   Chan Cindy, Berger Jonah, Van Boven Leaf ( 2012), “ Identifiable but Not Identical: Combining Social Identity and Uniqueness Motives in Choice,” Journal of Consumer Research , 39 3, 561– 73. Google Scholar CrossRef Search ADS   Chandler Jesse, Schwarz Norbert ( 2010), “ Use Does Not Wear Ragged the Fabric of Friendship: Thinking of Objects as Alive Makes People Less Willing to Replace Them,” Journal of Consumer Psychology , 20 2, 138– 45. Google Scholar CrossRef Search ADS   Cheema Amar, Kaikati Andrew M. ( 2010), “ The Effect of Need for Uniqueness on Word of Mouth,” Journal of Marketing Research , 47 3, 553– 63. Google Scholar CrossRef Search ADS   Chen Fangyuan, Sengupta Jaideep, Adaval Rashmi ( 2013), “The Effects of Anthropomorphization on Brand Personality Perceptions: A Motivational Account,” in NA—Advances in Consumer Research , Vol. 41, ed. Botti Simona, Labroo Aparna, Duluth, MN: Association for Consumer Research, 27– 8. Chen Rocky Peng, Wan Echo Wen, Levy Eric ( 2017), “ The Effect of Social Exclusion on Consumer Preference for Anthropomorphized Brands,” Journal of Consumer Psychology , 27 1, 23– 34. Google Scholar CrossRef Search ADS   Cialdini Robert B., De Nicholas Maralou E. ( 1989), “ Self-Presentation by Association,” Journal of Personality and Social Psychology , 57 4, 626– 31. Google Scholar CrossRef Search ADS   Clee Mona A., Wicklund Robert A. ( 1980), “ Consumer Behavior and Psychological Reactance,” Journal of Consumer Research , 6 4, 389– 405. Google Scholar CrossRef Search ADS   Deci Edward L., Ryan Richard M. ( 2000), “ The ‘What’ and ‘Why’ of Goal Pursuits: Human Needs and the Self-Determination of Behavior,” Psychological Inquiry , 11 4, 227– 68. Google Scholar CrossRef Search ADS   Deci Edward L., Ryan Richard M. ( 2008), “ Facilitating Optimal Motivation and Psychological Well-Being across Life’s Domains,” Canadian Psychology , 49 1, 14– 23. Google Scholar CrossRef Search ADS   Delbaere Marjorie, McQuarrie Edward F., Phillips Barbara J. ( 2011), “ Personification in Advertising: Using a Visual Metaphor to Trigger Anthropomorphism,” Journal of Advertising , 40 1, 121– 30. Google Scholar CrossRef Search ADS   Epley Nicholas, Akalis Scott, Waytz Adam, Cacioppo John T. ( 2008a), “ Creating Social Connection through Inferential Reproduction: Loneliness and Perceived Agency in Gadgets, Gods, and Greyhounds,” Psychological Science , 19 2, 114– 20. Google Scholar CrossRef Search ADS   Epley Nicholas, Waytz Adam ( 2009), “Mind Perception,” in The Handbook of Social Psychology , 5th edition, ed. Fiske Susan T., Gilbert Daniel T., Lindzey Gardner, New York: Wiley, 498– 541. Epley Nicholas, Waytz Adam, Akalis Scott, Cacioppo John T. ( 2008b), “ When We Need a Human: Motivational Determinants of Anthropomorphism,” Social Cognition , 26 2, 143– 55. Google Scholar CrossRef Search ADS   Epley Nicholas, Waytz Adam, Cacioppo John T. ( 2007), “ On Seeing Human: A Three-Factor Theory of Anthropomorphism,” Psychological Review , 114 4, 864– 86. Google Scholar CrossRef Search ADS   Escalas Jennifer Edson, Bettman James R. ( 2005), “ Self-Construal, Reference Groups, and Brand Meaning,” Journal of Consumer Research , 32 3, 378– 89. Google Scholar CrossRef Search ADS   Fournier Susan ( 1998), “ Consumers and Their Brands: Developing Relationship Theory in Consumer Research,” Journal of Consumer Research , 24 4, 343– 73. Google Scholar CrossRef Search ADS   Gagné Marylène, Ryan Richard M., Bargmann Kelly ( 2003), “ Autonomy Support and Need Satisfaction in the Motivation and Well-Being of Gymnasts,” Journal of Applied Sport Psychology , 15 4, 372– 90. Google Scholar CrossRef Search ADS   Harrell Gilbert D., Hutt Michael D., Anderson James C. ( 1980), “ Path Analysis of Buyer Behavior under Conditions of Crowding,” Journal of Marketing Research , 17 1, 45– 51. Google Scholar CrossRef Search ADS   Hauser David J., Schwarz Norbert ( 2016), “ Attentive Turkers: MTurk Participants Perform Better on Online Attention Checks than Do Subject Pool Participants,” Behavior Research Methods , 48 1, 400– 7. Google Scholar CrossRef Search ADS   Hayes Andrew F. ( 2013), Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach , New York: Guilford. Hong Sung-Mook, Faedda Salvatora ( 1996), “ Refinement of the Hong Psychological Reactance Scale,” Educational and Psychological Measurement , 56 1, 173– 82. Google Scholar CrossRef Search ADS   Hur Julia D., Koo Minjung, Hoffman Wilhelm ( 2015), “ When Temptations Come Alive: How Anthropomorphism Undermines Self-Control,” Journal of Consumer Research , 42 2, 340– 58. Kim Hyeongmin (Christian), Kramer Thomas ( 2015), “ Do Materialists Prefer the ‘Brand-as-Servant’? The Interactive Effect of Anthropomorphized Brand Roles and Materialism on Consumer Responses,” Journal of Consumer Research , 42 2, 284– 99. Google Scholar CrossRef Search ADS   Kim Sara, McGill Ann L. ( 2011), “ Gaming with Mr. Slot or Gaming the Slot Machine? Power, Anthropomorphism, and Risk Perception,” Journal of Consumer Research , 38 1, 94– 107. Google Scholar CrossRef Search ADS   Kwak Hyokjin, Puzakova Marina, Rocereto Joseph F. ( 2015), “ Better Not Smile at the Price: The Differential Role of Brand Anthropomorphization on Perceived Price Fairness,” Journal of Marketing , 79 4, 56– 76. Google Scholar CrossRef Search ADS   La Guardia Jennifer G., Ryan Richard M., Couchman Charles E., Deci Edward L. ( 2000), “ Within-Person Variation in Security of Attachment: A Self-Determination Theory Perspective on Attachment, Need-Fulfillment and Well-Being,” Personality and Social Psychology Bulletin , 79 3, 367– 84. Google Scholar CrossRef Search ADS   Lynn Michael, Harris Judy ( 1997), “ The Desire for Unique Consumer Products: A New Individual Differences Scale,” Psychology and Marketing , 14 6, 601– 16. Google Scholar CrossRef Search ADS   Maeng Ahreum, Aggarwal Pankaj ( 2018), “ Facing Dominance: Anthropomorphism and the Effect of Product Face Ratio on Consumer Preference,” Journal of Consumer Research , 44 5, 1104– 22. Google Scholar CrossRef Search ADS   Maimaran Michal, Wheeler S. Christian ( 2008), “ Circles, Squares, and Choice: The Effect of Shape Arrays on Uniqueness and Variety Seeking,” Journal of Marketing Research , 45 6, 731– 40. Google Scholar CrossRef Search ADS   McClelland David ( 1951), Personality , New York: Holt, Rinehart, & Winston. Google Scholar CrossRef Search ADS   Meyvis Tom, Van Osselaer Stijn M. J. ( 2017), “ Increasing the Power of Your Study by Increasing the Effect Size,” Journal of Consumer Research , 44 5, 1157– 73. Mochon Daniel, Norton Michael I., Ariely Dan ( 2012), “ Bolstering and Restoring Feelings of Competence via the IKEA Effect,” International Journal of Research in Marketing , 29 4, 363– 9. Google Scholar CrossRef Search ADS   Oppenheimer Daniel M., Meyvis Tom, Davidenko Nicolas ( 2009), “ Instructional Manipulation Checks: Detecting Satisficing to Increase Statistical Power,” Journal of Experimental Social Psychology , 45 4, 867– 72. Google Scholar CrossRef Search ADS   Parish Steven M. ( 1991), “ The Sacred Mind: Newar Cultural Representations of Mental Life and the Production of Moral Consciousness,” Ethos: Journal of the Society for Psychological Anthropology , 19 3, 313– 51. Prelinger Ernst ( 1959), “ Extension and Structure of the Self,” Journal of Psychology , 47 1, 13– 23. Google Scholar CrossRef Search ADS   Puzakova Marina, Kwak Hyokjin ( 2017), “ Should Anthropomorphized Brands Engage Customers? The Impact of Social Crowding on Brand Preferences,” Journal of Marketing , 81 6, 99– 115. Google Scholar CrossRef Search ADS   Puzakova Marina, Kwak Hyokjin, Rocereto Joseph F. ( 2013), “ When Humanizing Brands Goes Wrong: The Detrimental Effect of Brand Anthropomorphization amid Product Wrongdoings,” Journal of Marketing , 77 3, 81– 100. Google Scholar CrossRef Search ADS   Ruvio Ayalla, Shoham Aviv, Makovec Brenčič Maja ( 2008), “ Consumers’ Need for Uniqueness: Short-Form Scale Development and Cross-Cultural Validation,” International Marketing Review , 25 1, 33– 53. Google Scholar CrossRef Search ADS   Sawyer Alan G. ( 1975), “ Demand Artifacts in Laboratory Experiments in Consumer Research,” Journal of Consumer Research , 1 4, 20– 30. Google Scholar CrossRef Search ADS   Sheldon Kennon M., Filak Vincent ( 2008), “ Manipulating Autonomy, Competence, and Relatedness Support in a Game-Learning Context: New Evidence That All Three Needs Matter,” British Journal of Social Psychology , 47 2, 267– 83. Google Scholar CrossRef Search ADS   Silva Marlene N., Vieira Paulo N., Coutinho Sílvia R., Minderico Cláudia S., Matos Margarida G., Sardinha Luís B., Teixeira Pedro J. ( 2010), “ Using Self-Determination Theory to Promote Physical Activity and Weight Control: A Randomized Controlled Trial in Women,” Journal of Behavioral Medicine , 33 2, 110– 22. Google Scholar CrossRef Search ADS   Snyder C. R. ( 1992), “ Product Scarcity by Need for Uniqueness Interaction: A Consumer Catch-22 Carousel?” Basic and Applied Social Psychology , 13 1, 9– 24. Google Scholar CrossRef Search ADS   Snyder C. R., Fromkin Howard L. ( 1980), Uniqueness, the Human Pursuit of Difference , New York: Springer US. Spiller Stephen A., Fitzsimons Gavan J., Lynch John G.Jr., McClelland Gary ( 2013), “ Spotlights, Floodlights, and the Magic Number Zero: Simple Effects Tests in Moderated Regression,” Journal of Marketing Research , 50 2, 277– 88. Google Scholar CrossRef Search ADS   Swaminathan Vanitha, Stilley Karen M., Ahluwalia Rohini ( 2009), “ When Brand Personality Matters: The Moderating Role of Attachment Styles,” Journal of Consumer Research , 35 6, 985– 1002. Google Scholar CrossRef Search ADS   Thomson Matthew ( 2006), “ Human Brands: Investigating Antecedents to Consumers’ Strong Attachments to Celebrities,” Journal of Marketing , 70 3, 104– 19. Google Scholar CrossRef Search ADS   Tian Kelly Tepper, Bearden William O., Hunter Gary L. ( 2001), “ Consumers’ Need for Uniqueness: Scale Development and Validation,” Journal of Consumer Research , 28 1, 50– 66. Google Scholar CrossRef Search ADS   Vignoles Vivian L. ( 2009), “The Motive for Distinctiveness: Identity Construction in a Cultural Context,” in Oxford Handbook for Positive Psychology , ed. Snyder C. R., Lopez Shane, New York: Oxford University Press, 489– 99. Vignoles Vivian L. ( 2011), “Identity Motives,” in Handbook of Identity Theory and Research , ed. Schwartz Seth J., New York: Springer Science+Business Media, 403– 32. Google Scholar CrossRef Search ADS   Vignoles Vivian L., Chryssochoou Xenia, Breakwell Glynis M. ( 2002), “ Sources of Distinctiveness: Position, Difference and Separateness in the Identities of Anglican Parish Priests,” European Journal of Social Psychology , 32 6, 761– 80. Google Scholar CrossRef Search ADS   Vignoles Vivian L., Regalia Camillo, Manzi Claudia, Golledge Jen, Scabini Eugenia ( 2006), “ Beyond Self-Esteem: Influence of Multiple Motives on Identity Construction,” Journal of Personality and Social Psychology , 90 2, 308– 33. Google Scholar CrossRef Search ADS   Wan Echo Wen, Xu Jing, Ding Ying ( 2014), “ To Be or Not to Be Unique? The Effect of Social Exclusion on Consumer Choice,” Journal of Consumer Research , 40 6, 1109– 22. Google Scholar CrossRef Search ADS   Waytz Adam, Gray Kurt, Epley Nicholas, Wegner Daniel M. ( 2010a), “ Causes and Consequences of Mind Perception,” Trends in Cognitive Sciences , 14 8, 383– 8. Google Scholar CrossRef Search ADS   Waytz Adam, Heafner Joy, Epley Nicholas ( 2014), “ The Mind in the Machine: Anthropomorphism Increases Trust in an Autonomous Vehicle,” Journal of Experimental Social Psychology , 52 (May), 113– 7. Google Scholar CrossRef Search ADS   Waytz Adam, Morewedge Carey K., Epley Nicholas, Monteleone George, Gao Jia-Hong, Cacioppo John T. ( 2010b), “ Making Sense by Making Sentient: Effectance Motivation Increases Anthropomorphism,” Journal of Personality and Social Psychology , 99 3, 410– 35. Google Scholar CrossRef Search ADS   White Katherine, Argo Jennifer J. ( 2011), “ When Imitation Doesn’t Flatter: The Role of Consumer Distinctiveness in Responses to Mimicry,” Journal of Consumer Research , 38 4, 667– 80. Google Scholar CrossRef Search ADS   White Katherine, Dahl Darren W. ( 2006), “ To Be or Not Be? The Influence of Dissociative Reference Groups on Consumer Preferences,” Journal of Consumer Psychology , 16 4, 404– 14. Google Scholar CrossRef Search ADS   White Katherine, Dahl Darren W. ( 2007), “ Are All Out-Groups Created Equal? Consumer Identity and Dissociative Influence,” Journal of Consumer Research , 34 4, 525– 36. Google Scholar CrossRef Search ADS   Williams Geoffrey G., Gagné Marylène, Ryan Richard M., Deci Edward L. ( 2002), “ Facilitating Autonomous Motivation for Smoking Cessation,” Health Psychology , 21 1, 40– 50. Google Scholar CrossRef Search ADS   Xu Jing, Shen Hao, Wyer Robert S.Jr. ( 2012), “ Does the Distance between Us Matter? Influences of Physical Proximity to Others on Consumer Choice,” Journal of Consumer Psychology , 22 3, 418– 23. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Consumer Research Oxford University Press

Brands as Rivals: Consumer Pursuit of Distinctiveness and the Role of Brand Anthropomorphism

Loading next page...
 
/lp/ou_press/brands-as-rivals-consumer-pursuit-of-distinctiveness-and-the-role-of-56LwaG43wJ
Publisher
University of Chicago Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0093-5301
eISSN
1537-5277
D.O.I.
10.1093/jcr/ucy035
Publisher site
See Article on Publisher Site

Abstract

Abstract Although past research has shown that anthropomorphism enhances consumers’ attraction to a brand when social-connectedness or effectance motives are active, the current research demonstrates that anthropomorphizing a brand becomes a detrimental marketing strategy when consumers’ distinctiveness motives are salient. Four studies show that anthropomorphizing a brand positioned to be distinctive diminishes consumers’ sense of agency in identity expression. As a result, when distinctiveness goals are salient, consumers are less likely to evaluate anthropomorphized (vs. nonanthropomorphized) brands favorably and are less likely to choose them to express distinctiveness. This negative effect of brand anthropomorphism, however, is contingent on the brand’s positioning strategy—brand-as-supporter (supporting consumers’ desires to be different) versus brand-as-agent (communicating unique brand features instead of focusing on consumers’ needs) versus brand-as-controller (limiting consumers’ freedom in expressing distinctiveness). Our results demonstrate that an anthropomorphized brand-as-supporter enhances consumers’ sense of agency in identity expression, compared to both an anthropomorphized brand-as-agent and an anthropomorphized brand-as-controller. In turn, enhancing or thwarting consumers’ sense of agency in expressing their differences from others drives the differential impact of anthropomorphizing a brand positioned to be distinctive. brand anthropomorphism, distinctiveness motive, agency in identity expression Prior research demonstrates that people often seek to differentiate themselves from others (Snyder and Fromkin 1980). The need to express one’s distinctiveness is reflected in people’s unique product choices (Snyder 1992), intentions to recustomize the product (White and Argo 2011), and preferences for unique shopping venues (Burns and Warren 1995). In other words, distinctiveness motives are predictive of people’s consumption behaviors because products are seen as their self-extensions and as signals of their identity (Snyder and Fromkin 1980). However, missing from prior work is an understanding of the effects of the innate features or positioning strategies of brands, such as imbuing brands with humanlike features that are used by managers to help consumers signal their distinctiveness. Recent industry trends suggest that in attempts to find unique positioning strategies, many managers turn to the tactic of anthropomorphizing their brands (i.e., imbuing brands with humanlike features; Aggarwal and McGill 2012). For example, the Swatch brand introduced a collection of watches that activate various human schemas ranging from a Blue Wild Face to the hot-pink Lady Spy. Similarly, the Hammacher Schlemmer Institute retails a variety of unique products that are often designed to project humanlike features (e.g., an infrared supine sauna with a giant face). One possible reason for the popularity of this approach is that it might increase consumer connection to these unique products (Ahn, Kim, and Aggarwal 2014). In fact, people who lack social connectedness prefer anthropomorphized (vs. nonanthropomorphized) brands to help compensate for that gap (Chen, Wan, and Levy 2017). However, when purchasing unique brands, consumers pursue the goal of being distinct from others. The source of this distinctiveness motive is the desire to be different through signaling one’s unique identity (Vignoles 2009). When brands become important for identity signaling, factors that inhibit consumers’ need for self-expression should negatively affect brand evaluations. If anthropomorphized brands are perceived as being mindful (Puzakova and Kwak 2017), can anthropomorphizing a brand designed to uniquely express one’s identity become a detrimental brand strategy? With this research, we attempt to answer this question and make three important contributions. Our key contribution is to show that under salient distinctiveness motives, brand anthropomorphism has a detrimental effect on evaluations of distinctive brands. As our theoretical foundation, we rely on prior work showing that anthropomorphized brands are viewed as intentional agents that are able to control others’ actions (Epley and Waytz 2009). Consequently, anthropomorphized distinctive brands retain their own identities and, being seen as agentic, are perceived as reducing consumers’ sense of agency in self-expression, thereby leading to less favorable brand evaluations. Thus, our contribution lies in demonstrating the novel process through which brand anthropomorphism affects consumer preference under distinctiveness motives (i.e., reducing consumers’ agency in identity expression). This is in contrast to prior work showing positive effects of anthropomorphism when consumers’ belongingness (Chen et al. 2017) or effectance (Waytz et al. 2010b) goals are salient. Next, we establish important boundary conditions of the negative effect of brand anthropomorphism: the effect will occur only when a brand is positioned to be distinctive and the product usage context is public. This is because a distinctiveness motive enhances consumer preference for brands that are able to satisfy this goal, and because the signaling value of distinctive brands is higher in a public context (Tian, Bearden, and Hunter 2001). Finally, we contribute to prior work by identifying a theoretically relevant moderator (i.e., positioning a brand as supporting vs. controlling consumers’ sense of agency in self-expression). Because people develop strong ties to others who enhance their sense of agency (Thomson 2006), we show that positioning an anthropomorphized distinctive brand as a supporter to consumers increases their perceived sense of agency in identity expression and brand evaluations, compared to the situation when an anthropomorphized brand is positioned either as a controller or as an independent agent (an entity focused on its own features vs. consumers’ needs). The findings reveal an interesting nuance by indicating that distinctiveness motives enhance consumers’ sensitivity to cues that signal a potential threat to their sense of agency, and positioning a distinctive brand as an agent can make consumers view the brand as a less controllable entity, and hence, as less desirable. Overall, unlike prior work showing that anthropomorphism has positive effects under salient belongingness or effectance goals, our work demonstrates that anthropomorphism has negative effects on preferences for distinctive brands under distinctiveness motives. Four studies provide converging evidence in support of our theory; implications for managers and future research are discussed. CONSUMERS’ DISTINCTIVENESS MOTIVES AND PRODUCT CONSUMPTION Distinctiveness Motivation and Identity Expression Prior work defines distinctiveness motivation as the desire to differentiate oneself from others in appearance, qualities, traits, and abilities (Vignoles et al. 2006; White and Argo 2011). According to the uniqueness theory (Snyder and Fromkin 1980), people seek to be unique because they learn over time that they are different from others, and this learned self-image drives an active search for ways to differentiate from others. Prior work also shows that possessions are extensions of the self (Belk 1988) and hence can be used for self-expression (White and Argo 2011). Thus, people express their distinctiveness through consumption of products that are scarce, unique, customized, or odd (Snyder 1992). For example, Tian et al. (2001) find that people with a high need for uniqueness prefer products that are less popular and publicly visible. This finding highlights the crucial role of publicly visible products in satisfying people’s distinctiveness motives. Most pertinent to our work is research delineating how people motivated by distinctiveness diverge from others in their choices as they communicate their identity to others (Berger and Heath 2007). Importantly, past work emphasizes that the distinctiveness motive is paramount to identity construction (Vignoles 2009). Hence, people tend to make unique choices in product domains that have strong signaling value (e.g., car models; Berger and Heath 2007). In this regard, White and Argo (2011) show that when being mimicked by similar others, consumers high in distinctiveness motive respond with choices that differentiate them (e.g., recustomize the product). Similarly, consumers who are high in distinctiveness buy unique products to obtain a favorable evaluation from others (Tian et al. 2001). Overall, we adopt the view that consumers motivated by distinctiveness make choices to self-express and to ensure that others understand who they are. Distinctiveness Motivation and Agency in Identity Expression We further build on prior research showing that identity expression is motivated by another fundamental drive: desire for agency (Baumeister 1998). In general, agency refers to an individual’s need to feel that one’s actions are self-endorsed, as opposed to controlled and pressured by others (Deci and Ryan 2000). Prior research shows that when consumers express their identity, a certain type of marketing message (e.g., directly defining self-expression) can suggest that forces external to the self are determining one’s behavior. In other words, such messages diminish the extent to which identity-expressing choices are perceived to be determined by the self, thus limiting consumer perceptions of agency and reducing evaluations of products being promoted with these messages (Bhattacharjee, Berger, and Menon 2014). Similarly, Mochon, Norton, and Ariely (2012) show that a feeling of control over one’s environment is the driving force behind consumers’ self-expressive tasks (e.g., assembling furniture). Overall, this line of work emphasizes the importance of one’s agency in identity expression. Here, we merge two streams of research: the first shows that the distinctiveness motive is driven by the desire to communicate one’s difference to others, and the second emphasizes the importance of one’s own agency in identity expression. Because the latter is a crucial force behind the motive to be distinct, we argue that a brand positioning strategy that reduces consumers’ sense of agency in identity expression should decrease consumers’ brand evaluations. BRAND ANTHROPOMORPHISM Brand anthropomorphism is the process of imbuing brands with humanlike features that enhance consumers’ attributions of mind to the brands (Puzakova and Kwak 2017). Marketers encourage consumers to see products in humanlike forms, either by creating humanlike characters (e.g., the California Raisins) or through visual representations on product packages (e.g., Capri-Sun). Brands are also given names that signal social roles and familial relationships (e.g., Uncle Ben’s). These tactics are employed to enhance consumer–brand connections and to entice consumers to choose these brands over others (Fournier 1998). Prior work shows that anthropomorphism influences product evaluations (Aggarwal and McGill 2007), emotional responses (Delbaere, McQuarrie, and Phillips 2011), and intentions to replace a product (Chandler and Schwarz 2010). However, simply imbuing products with humanlike traits does not always improve evaluations. The attributes of the consumer, the nature of the context, and brand features impact the valence of the anthropomorphism influence (Maeng and Aggarwal 2018; Puzakova and Kwak 2017). Importantly, past research shows that salient consumer motivations influence the direction of anthropomorphism effects (Kim and McGill 2011). In this regard, prior work identifies two key consumer motivations—social connectedness and effectance—that lead to positive downstream consequences of anthropomorphism (Epley, Waytz, and Cacioppo 2007). For example, people who desire to connect with others tend to prefer anthropomorphized (vs. nonanthropomorphized) products (Chen et al. 2017). In turn, effectance motivation that is driven by the desire to explain and predict an agent’s action also enhances a tendency to humanize (Waytz et al. 2010b). Here, we theorize that the workings of brand anthropomorphism under high distinctiveness motives go beyond the consequences of this phenomenon under social-connectedness or effectance goals. One could argue that a distinctiveness motive is the opposite of a social-connectedness goal in that it leads to a lowered desire for social interactions, which could potentially explain the predicted negative reactions to anthropomorphized brands. However, prior research points out that a distinctiveness goal should be seen as a desire to establish a difference from others rather than to increase social separateness (Tian et al. 2001). For example, Vignoles (2009) distinguishes between the two motives by illustrating a distinctiveness motive as the difference from others through the presentation of unique geometrical figures (e.g., six squares vs. one circle), and the social-separateness goal via the spatial distancing of one circle from the cluster of other circles. To that end, Baumeister and Sommer (1997) posit that if people were to pursue social separateness by being distinctive, it would be most beneficial to cultivate bad unique traits; however, people tend to differentiate themselves on choices that positively distinguish them from others. Marketing scholars also argue that the desire to be different drives the choice of unique products, whereas the motive for social separateness leads to product choices based on internal tastes, which may or may not be unique (Tian et al. 2001). Thus, past work concludes that social connectedness and distinctiveness motives have different underlying forces. Although both distinctiveness and effectance motives are driven by a desire to control, we argue that the fundamental source of this desire varies. The primary drive of the distinctiveness motive is to have agency (or control) over one’s ability to signal identity, whereas the major force of effectance motivation is to understand and predict an agent that is unpredictable in order to more effectively interact with it (Epley et al. 2007). For example, Waytz et al. (2010b) show that the effectance motivation is triggered only when an object behaves randomly or unexpectedly, and giving it a human intent makes the random behavior seem more understandable. In fact, other work shows that effectance motivation leads to a greater preference for anthropomorphized entities (Chen, Sengupta, and Adaval 2013). In contrast, we predict that consumers’ evaluations of an anthropomorphized (vs. nonanthropomorphized) distinctive brand decrease under distinctiveness motives because a brand that is anthropomorphized is seen as having humanlike agency, which then reduces consumers’ own sense of agency. Overall, adopting the view that the distinctiveness motive is driven by the desire for difference rather than social separateness or efficacy, we elaborate on the role of brand anthropomorphism next. BRAND ANTHROPOMORPHISM UNDER DISTINCTIVENESS MOTIVES As documented by prior work (Tian et al. 2001; White and Argo 2011), people motivated by distinctiveness are more attracted to products that are positioned to be distinctive (i.e., products with unique features or that are advertised to be different) because these products satisfy consumers’ differentiation drives compared to nondistinctively positioned products. However, as discussed earlier, consumers with high distinctiveness motives are also concerned with maintaining a sense of agency in self-expression (Bhattacharjee et al. 2014). Here, we propose that humanizing a distinctive brand decreases consumers’ perceived agency in identity expression, thus counteracting the increased preference for a distinctive brand under high distinctiveness goals. This contention is borne out by several streams of research. First, research on self-representation suggests a hierarchy of self-allied objects (McClelland 1951). Studies involving the task of sorting items on the continuum of self to not-self show that possessions are viewed as closer to the self than other people (Prelinger 1959). Second, the reason why products are more self-defining than other humans is that people exert greater control over external objects than over other people (Belk 1988). Pertinent to this, nonhuman entities lack “free will” (Epley and Waytz 2009) and could be controlled and subsumed as “self” by consumers, thus giving consumers more agency in self-expression. In contrast, because other people retain their own identities, this “meaning” or identity of other people cannot be fully controlled by an individual. Since anthropomorphized entities are also viewed as mindful (Epley and Waytz 2009), they are perceived as less likely to be controlled by others and thus are evaluated less positively. It is important to note that we expect this negative effect to occur only when a brand is distinctive and when a distinctiveness goal is salient. This is because a nondistinctive brand is unable to satisfy consumers’ differentiation goals (Tian et al. 2001), and because consumers low in distinctiveness motive are generally less sensitive to cues that reduce their sense of agency in self-expression. This last premise is consistent with prior work showing that when no specific goals are salient, consumers’ preference for anthropomorphized versus nonanthropomorphized brands is no different (Waytz et al. 2010a). For example, Chen et al. (2017) find no effect of product anthropomorphism in the low-social-connectedness condition, whereas Hur, Koo, and Hoffman (2015) show that anthropomorphizing a tempting product does not reduce consumers’ experiences of conflict when a product is goal-irrelevant. H1:When the distinctiveness motive is salient, consumers will develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) distinctive brand. In contrast, we expect no differences between an anthropomorphized and a nonanthropomorphized distinctive brand when the distinctiveness motive is not salient. H2:When the distinctiveness motive is salient, the negative impact of an anthropomorphized (vs. nonanthropomorphized) distinctive brand on brand attitudes will be driven by consumers’ perception of reduced agency in identity expression. Next, given that the motive to communicate one’s distinctiveness is stronger when a product is public (White and Dahl 2006), we argue that the product usage context will moderate the core negative effect. Prior work shows that consumers motivated to be different are less willing to generate positive word of mouth for a public product because a product that becomes popular will lose its exclusivity (Cheema and Kaikati 2010). Dissociative group influence on product choice is also heightened in a public context (White and Dahl 2007). When consumers strive to signal their image through brands, the signaling role of brand personality is restricted to public contexts (Swaminathan, Stilley, and Ahluwalia 2009). Overall, because public products facilitate self-expression (Tian et al. 2001), we expect that the negative effect of brand anthropomorphism under a high distinctiveness motive will emerge only for products used publicly (vs. privately). H3: In the public usage context, consumers will develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) distinctive brand under high distinctiveness motives, whereas this effect will be attenuated in the private usage context. In contrast, we expect no differences between an anthropomorphized and a nonanthropomorphized distinctive brand when the distinctiveness motive is not salient, either in public or private usage contexts. CAN AN ANTHROPOMORPHIZED BRAND BE AGENCY-SUPPORTIVE? Prior work demonstrates that people’s sense of agency depends on contexts and agency-supportive or agency-controlling social roles (Williams et al. 2002). In particular, agency-supportive social contexts emphasize opportunities for free choice and encouragement of personal initiative (Gagne, Ryan, and Bargmann 2003). For example, agency-supportive physicians emphasize freedom of choice rather than forcing a change (Williams et al. 2002). In turn, this supportive style enhances one’s own sense of agency in a goal pursuit and increases satisfaction with the process (Williams et al. 2002). In contrast, agency-controlling contexts where someone else is presumably in control makes people perceive that their actions are a consequence of imposition by others (Deci and Ryan 2008). In turn, these contexts result in a lower sense of agency and decreased satisfaction with the goal-pursuit process (Williams et al. 2002). Here, we argue that positioning a distinctive brand as being agency-supportive may reverse the effect predicted in hypothesis 1. Prior work shows that people are attracted to others who support their sense of agency (La Guardia et al. 2000). For example, the extent to which a human brand (e.g., a celebrity) provides consumers with a sense of agency determines their intensity of brand attachment (Thomson 2006). Thus, positioning an anthropomorphized distinctive brand as a supporter is unlikely to impinge upon consumers’ sense of agency in self-expression. As a result, consumers are likely to evaluate an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-supporter more favorably. Next, we propose that a brand could also be positioned as limiting consumers’ perceived agency in self-expression (i.e., brand-as-controller). Because people develop more distant relationships with others who thwart their agency (La Guardia et al. 2000), we expect that an anthropomorphized (vs. nonanthropomorphized) brand-as-controller should lead to less favorable consumer attitudes toward the brand. Similarly, we predict that consumers are likely to evaluate an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-agent (i.e., a brand positioned as an independent entity focusing on its own features without emphasizing consumer needs) less favorably. Anthropomorphizing a brand increases consumer perception of it as having its own unique identity and as being capable of controlling others’ actions (Epley and Waytz 2009). Because consumers high in distinctiveness are sensitive to cues that thwart their agency in self-expression, they are likely to be sensitive to cues indicating that the brand cannot be fully controlled. Thus, a distinctive anthropomorphized brand-as-agent would signal to consumers that they cannot use it to communicate who they are to others, much like they are unable to use others’ identities to express their own. Finally, we expect that these effects of brand anthropomorphism will occur only under high distinctiveness motives. Consumers with low distinctiveness motives are less concerned with having agency in self-expression and are unlikely to be influenced by social entities (including anthropomorphized brands) that enhance or reduce their sense of perceived agency. H4: When distinctiveness motives are salient, consumers will have more favorable evaluations of an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-supporter, whereas they will exhibit less favorable evaluations of an anthropomorphized (vs. nonanthropomorphized) distinctive brand-as-agent and brand-as-controller. In contrast, when distinctiveness motives are low, we expect no differences between anthropomorphized and nonanthropomorphized distinctive brands positioned as a supporter, a controller, or an agent. Four studies test hypotheses 1–4. Study 1 provides evidence for hypothesis 1 and examines the moderating role of consumers’ need for agency. Study 2 shows that hypothesis 1 holds only under high distinctiveness motives and only for a distinctive brand; it also demonstrates the mediating role of agency in identity expression and rules out alternative explanations (i.e., a brand’s ability to provide social distance, perceived personal space violation). Study 3 establishes the moderating role of a product usage context (private vs. public) and rules out the role of perceived efficacy. Finally, using a moderation approach, study 4 shows that positioning an anthropomorphized distinctive brand as supporter (vs. controller or agent) enhances consumers’ sense of agency, thus reversing the effect predicted in hypothesis 1. STUDY 1: THE MODERATING ROLE OF AGENCY The main goal of study 1 is to test hypothesis 1. To activate distinctiveness motive, study 1 employs a subtle manipulation—namely, the visual presentation of either unique or homogenous geometrical shapes. Prior research shows that exposure to unique stimuli that have no a priori meaning, such as arrays of geometrical shapes, enhances the appeal of unique products (Maimaran and Wheeler 2008). Accordingly, the exposure to unique visual arrays is expected to activate a distinctiveness motive and reduce consumer attraction to anthropomorphized (vs. nonanthropomorphized) brands. Next, hypothesis 2 predicts the mediating role of agency in self-expression. Thus, the second goal of study 1 is to show the underlying mechanism by examining individual differences in the need for agency (Clee and Wicklund 1980). We expect that the negative effect of anthropomorphizing a distinctive brand will occur only among individuals motivated to be different who are high in the need for agency and not for those low in need for agency. Participants, Design, and Procedure Three hundred eighty US participants (average age = 36 years old, 43% female) were recruited from Amazon Mechanical Turk (MTurk) and paid $1.00. To enhance the quality of data (Hauser and Schwarz 2016), only MTurk workers with high reputation (i.e., 98% Human Intelligence Task [HIT] approval rate, 5,000 approved HITs) were eligible to complete this task. Across all studies, we employed three data exclusion criteria; that is, we omitted data of those participants who 1) were not diligent in reading and following our instructions, and hence failed the attention check (Oppenheimer, Meyvis, and Davidenko 2009), 2) were extreme outliers (i.e., 2.5 deviations from the mean; Meyvis and van Osselaer 2017), and 3) correctly guessed the true research hypothesis being examined (Sawyer 1975). We believe that it is prudent to use these criteria, as failure to account for these factors has been shown to increase noise and reduce experimental power. For study 1, to identify nondiligent respondents, we used a single-item self-reported measure of attentiveness. None of the MTurk workers selected the lowest value on the attention scale, there were also no outliers, and none of the participants correctly guessed the true research question being examined; thus, we retained the full sample of 380 participants. We employed a 2 (distinctiveness motivation: uniqueness vs. homogeneity visual arrays) × 2 (anthropomorphized vs. nonanthropomorphized brand) between-subjects factorial design; we also included a scale to measure individual differences in need for agency. After exposure to a disclosure form, participants learned that they would take part in two ostensibly unrelated studies. The first was introduced as a “visual perception” study. Respondents were told that the researchers were interested in the effect of figure arrangement on visual perception and accuracy. Consistent with Maimaran and Wheeler (2008), respondents viewed five pictures and indicated the number of squares and circles they saw in each. Each image consisted of different geometrical shapes and conveyed either uniqueness (e.g., OOOOΔOO) or homogeneity (e.g., OOOOOOO). A separate pretest (n = 82) showed that priming respondents with the unique visual (vs. homogeneity) arrays increased their desire for unique and unusual products (4.45 vs. 3.68; F(1, 80) = 6.74, p = .011, η2 = .08), measured on four items (α = .92) used in prior work (Ruvio, Shoham, and Brencic 2008). After the priming task, participants were invited to take part in a “product evaluation” study. They were asked to imagine that they were looking for a new pair of sunglasses and viewed one of two versions of an advertisement (i.e., showing an anthropomorphized or a nonanthropomorphized brand) for a fictitious brand of sunglasses, Safiro. Following prior work (Aggarwal and McGill 2007), we manipulated brand anthropomorphism using verbal (i.e., first- vs. third-person) communication (figure 1). The brand was positioned as emphasizing its distinctive features. The results of the pretest confirmed the effectiveness of brand anthropomorphism and brand distinctiveness manipulations (see web appendix W1 for details). After viewing the ad, participants rated their brand attitudes, responded to the need-for-agency scale, and answered demographic questions. Finally, they reported how much attention they paid to the survey (1 = “not too much attention,” 7 = “paid a lot of attention”) and responded to the demand probe. FIGURE 1 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM IN STUDY 1 FIGURE 1 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM IN STUDY 1 Measures We measured brand attitude with four items (e.g., 1 = “unfavorable, unpleasant, bad, dislike,” 7 = “favorable, pleasant, good, like”; α = .98). Following prior work (Bhattacharjee et al. 2014), participants’ need for agency was operationalized as reactance to compliance and resistance to influence from others and measured with the six-item reactance scale, adopted from Hong and Faedda (1996). A factor analysis with a varimax rotation revealed a one-factor structure, with factor loadings of all items being higher than .64. Thus, we combined all items in one scale (α = .87). The details of all measures are presented in web appendix W2. Results We performed a 2 (distinctiveness motivation: uniqueness vs. homogeneity visual arrays) × 2 (anthropomorphized vs. nonanthropomorphized brand) ANOVA with brand attitude as the dependent variable. There were no significant main effects of either brand anthropomorphism or distinctiveness motivation (ps > .19). However, consistent with hypothesis 1, we found a significant interaction effect between distinctiveness motivation and brand anthropomorphism (F(1, 376) = 5.33, p = .022, ηp2 = .01). We followed up with an analysis of planned contrasts. To elaborate, when distinctiveness motivation was high, consumers developed less favorable brand attitudes when the brand was anthropomorphized (M = 3.67, SD = 1.60) than when it was nonanthropomorphized (M = 4.15, SD = 1.63; F(1, 376) = 3.27, p = .071, ηp2 = .01), but this effect was only marginally significant. In contrast, there was no significant effect of brand anthropomorphism in the homogeneity condition (M = 3.85, SD = 1.78 vs. M = 3.51, SD = 1.74; F(1, 376) = 2.07, p = .151). Furthermore, before examining the role of consumers’ need for agency, we ran a 2 (distinctiveness motivation: uniqueness vs. homogeneity visual arrays) × 2 (anthropomorphized vs. nonanthropomorphized brand) ANOVA with need for agency as the dependent variable to ensure that our manipulation of distinctiveness motive and brand anthropomorphism did not impact the need for agency. The results revealed no main or interaction effects of manipulated variables on need for agency (ps > .60). Next, to test for the moderating effect of consumers’ need for agency, we conducted a moderation analysis (model 3; Hayes 2013) with distinctiveness motivation, brand anthropomorphism, consumers’ need for agency, and two- and three-way interactions among them as the independent variables, and brand attitude as the dependent variable. The results revealed a significant two-way interaction between brand anthropomorphism and distinctiveness motivation (b = –.20, t(372) = – 2.28, p = .023). This effect was qualified by a marginally significant three-way interaction among brand anthropomorphism, distinctiveness motivation, and consumers’ need for agency (b = –.16, t(372) = – 1.84, p = .066). There were no other significant effects (ps > .11). We further performed a floodlight analysis in the high distinctiveness condition using the Johnson-Neyman technique (Spiller et al. 2013). The results revealed a significant negative effect of brand anthropomorphism on brand attitudes among participants with values of need for agency above 3.62 (BJN = –.25, SE = .13, p = .05), with this effect being nonsignificant when need for agency was below 3.62. Discussion Overall, study 1 shows the negative effect of brand anthropomorphism on attitudes toward a brand positioned to be distinctive when distinctiveness motivation is salient. In addition, study 1 provides preliminary evidence of the underlying process by highlighting the moderating role of consumers’ need for agency; the results show that under high distinctiveness motivation, consumers who are high on need for agency develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) brand positioned to be distinctive, whereas no such effect emerges for consumers low in need for agency. STUDY 2: THE ROLE OF BRAND DISTINCTIVENESS Study 2 builds on study 1 in a number of ways. First, study 2 investigates two important boundary conditions: 1) whether the effect is specific only to consumers with high distinctiveness concerns, and 2) whether the effect holds only for brands that are positioned distinctively (vs. nondistinctively). Because a high distinctiveness motive attracts consumers to products that satisfy this goal (Tian et al. 2001), and because consumers are unlikely to be concerned with agency in identity expression under low distinctiveness motives, we expect the hypothesized negative effect of brand anthropomorphism to emerge only when a brand is positioned to be distinctive and only for consumers with a high desire to differentiate themselves. Second, study 2 measures consumers’ agency in identity expression to more directly examine the role of this factor. The third goal of study 2 is to rule out alternative explanations. Prior work shows that humanizing an entity enhances its ability to provide a sense of social connectedness (Epley et al. 2008b). Although our conceptualization of distinctiveness led to our predictions regarding the mediating role of agency in identity expression, it is important to empirically rule out an alternative explanation that an anthropomorphized (vs. nonanthropomorphized) distinctive brand provides a reduced sense of social separateness, thereby lowering brand evaluations of consumers high in distinctiveness motives. Furthermore, prior research by Xu, Shen, and Wyer (2012) finds that physical proximity to other people in a crowded context increases the motivation to seek uniqueness. Thus, as another alternative explanation, it is possible that having a distinctiveness motive may increase consumers’ goal to avoid violation of personal space by increasing their desire to stay away from other people and anthropomorphized entities. Study 2 directly tests these two potential alternative explanations. Participants, Design, and Procedure Four hundred twenty workers who met a 98% HIT approval rate and 5,000 approved HITs as the qualification criteria were recruited from Amazon MTurk. They provided their consent by agreeing to a disclosure and received a $1.00 financial compensation for their participation. We applied the same three data exclusion criteria as in study 1. However, to identify inattentive respondents we used an instructional manipulation check (IMC) (“Instead of entering your MTurk ID number, please enter number six”; Oppenheimer et al. 2009). Prior work identifies that a self-reported measure of attentiveness (as was used in study 1) is less reliable than an IMC because participants may not want to admit their negligence (Oppenheimer et al. 2009). Overall, we omitted the data of those participants (n = 9) who failed the IMC included at the end of the study and the data of extreme outliers (n = 9), leaving a sample of 402 participants (average age = 34 years old, 36% female). Similar to study 1, none of the participants were able to accurately guess the research hypothesis being examined. We employed a 2 (distinctiveness motivation: uniqueness visual arrays vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (brand positioning: distinctive vs. nondistinctive) between-subjects design. The procedure and manipulation of distinctiveness motive were similar to that of study 1. In the control condition, respondents immediately proceeded to the product-evaluation study. All participants were presented with one of four versions of an advertisement (figure 2) for a fictitious brand of watch, Tesmer. After viewing the ad, participants indicated their brand attitudes, responded to questions that measured potential mediating variables, and completed a manipulation check on distinctiveness motivation. The presentation order of mediating variables was randomized to avoid a carryover effect from one construct to the other. Neither the effect of presentation order nor its interaction with any other variables was significant (ps > .11). At the end of the study, participants responded to an IMC question, filled out demographic variables, and responded to the demand probe question. FIGURE 2 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND DISTINCTIVENESS IN STUDY 2 FIGURE 2 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND DISTINCTIVENESS IN STUDY 2 Anthropomorphism and Brand Distinctiveness Manipulation Following prior research (Aggarwal and McGill 2007), we manipulated brand anthropomorphism using first-person (vs. third-person) communication (figure 2). We carefully selected the stimulus for the watch product image since we intended its appearance to be ambiguous with respect to anthropomorphism, such that it could resemble a human face when accompanied by anthropomorphized ad copy but not necessarily be seen as a human face if not prompted by other cues in the ad. We manipulated brand distinctiveness by modifying the ad copy. In the distinctive brand condition, the ad copy was oriented toward distinguishing the brand from other watches. In the nondistinctive brand condition, the message emphasized typical product features and suitability for everyday use (figure 2). The results of the pretest confirmed the effectiveness of brand anthropomorphism and brand distinctiveness manipulations (web appendix W1). Measures We measured brand attitude (α = .96) and desire for unique products (α = .95) with the same items used in study 1. Following prior work (Bellezza, Gino, and Keinan 2014; Bhattacharjee et al. 2014), we assessed participants’ sense of agency in identity expression with two items (e.g., “Owning a Tesmer watch would allow me to freely express who I am”; 1 = “strongly disagree,” 7 = “strongly agree”; r = .89, p < .001; web appendix W2). The brand’s ability to provide social distance was measured with two items (e.g., “Owning a Tesmer watch would allow me to be distant from others”; 1 = “strongly disagree,” 7 = “strongly agree”; r = .80, p < .001). We constructed the questions by adopting items from previous work that delineates social separateness (Vignoles, Chryssochoou, and Breakwell 2002) and modifying them for our context. Finally, the violation of personal space was measured with three items on a Likert-type scale (e.g., “I would feel my personal space is violated”; α = .94), adopted from Harrell, Hutt, and Anderson (1980). The construct correlation matrix is presented in web appendix W3. Results Manipulation Check. A manipulation check on the distinctiveness motive—a 2 (unique visual array priming vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (distinctive vs. nondistinctive brand) ANOVA with the desire for unique products as the dependent variable—showed that participants viewing the uniqueness visual array (vs. control) reported greater desire for unique products (3.76 vs. 3.38; F(1, 392) = 5.05, p = .025, ηp2 = .01), with the main or interaction effects of other variables being nonsignificant (ps > .13). Hypothesis Tests. We performed a 2 (distinctiveness motivation: unique visual arrays vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (distinctive vs. nondistinctive brand) ANOVA with brand attitude as the dependent variable. There was a significant effect of brand anthropomorphism (4.90 vs. 5.18; F(1, 394) = 4.51, p = .034, ηp2 = .01) and a significant interaction between brand distinctiveness and brand anthropomorphism (F(1, 394) = 4.92, p = .027, ηp2 = .01). In support of hypothesis 1, we also found that when the brand was positioned to be distinctive, there was a significant interaction between brand anthropomorphism and distinctiveness motivation (F(1, 394) = 4.28, p = .039, ηp2 = .01). No other main or interaction effects were significant (ps > .14). We followed up with an analysis of planned contrasts. First, confirming our prediction, respondents with high distinctiveness motive developed more favorable attitudes toward a nonanthropomorphized distinctive (vs. nondistinctive) brand (5.62 vs. 5.04; F(1, 394) = 4.79, p = .029, ηp2 = .01; table 1). Because anthropomorphizing a brand counteracts the increased preference for a distinctive brand under high distinctiveness goals, the planned contrast tests comparing an anthropomorphized and a nonanthropomorphized brand in the high distinctiveness condition revealed that anthropomorphizing a distinctive brand decreased respondents’ brand attitudes (4.67 vs. 5.62; F(1, 394) = 14.82, p < .001, ηp2 = .04). Finally, because a nondistinctive brand is unable to satisfy consumers’ differentiation goals, the planned contrasts in the condition of high distinctiveness motive and when a brand was nondistinctive showed no significant effects of anthropomorphism (5.06 vs. 5.04; F(1, 394) = .005, p = .94). We also found no effect of brand anthropomorphism on brand attitudes in the low distinctiveness condition (distinctive brand: 4.85 vs. 5.04; F(1, 394) = .63, p = .48; nondistinctive brand: 5.03 vs. 5.03; F < .001, p = .98). TABLE 1 CELL MEANS IN STUDY 2   Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)    Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a significant difference of the cell mean from (b). TABLE 1 CELL MEANS IN STUDY 2   Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)    Distinctiveness motive condition   Control condition     Distinctively positioned brand   Nondistinctively positioned brand   Distinctively positioned brand   Nondistinctively positioned brand     NAB (n = 50)  AB (n = 63)  NAB (n = 46)  AB (n = 53)  NAB (n = 41)  AB (n = 52)  NAB (n = 40)  AB (n = 57)  Brand attitude  5.62 (1.02)b  4.67 (1.33)*  5.04 (1.32)a  5.06 (1.54)  5.04 (1.04)a  4.85 (1.38)  5.03 (1.34)  5.03 (1.31)  Agency in identity expression  4.39 (1.56)  3.68 (1.66)*  3.89 (1.48)  4.23 (1.90)  4.04 (1.26)  3.66 (1.56)  4.01 (1.77)  3.85 (1.82)  A brand’s ability to provide social distance  2.83 (1.52)  2.67 (1.27)  2.82 (1.48)  2.60 (1.61)  2.68 (1.35)  2.51 (1.46)  2.53 (1.51)  2.61 (1.64)  Perceived personal space violation  1.54 (.88)  2.12 (1.47)*  1.86 (1.25)  2.19 (1.68)  2.06 (1.41)  1.95 (1.47)  1.88 (1.41)  2.32 (1.57)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a significant difference of the cell mean from (b). Process Analysis. First, the results of a 2 (distinctiveness motivation: unique visual arrays vs. control) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (distinctive vs. nondistinctive brand) ANOVA with agency in identity expression as the dependent variable confirmed that anthropomorphizing a distinctive brand reduced consumers’ perceived agency in identity expression in the high distinctiveness condition (3.68 vs. 4.39; F(1, 394) = 5.26, p = .022, ηp2 = .01). This effect was nonsignificant in all other conditions (p > .27; table 1). Next, we conducted a moderated mediation analysis using a bootstrapping approach (model 12; Hayes 2013) with distinctiveness motivation, brand anthropomorphism, brand distinctiveness, and all two- and three-way interactions among them as the independent variables, brand attitude as the dependent variable, and agency in identity expression as the mediating variable. Results revealed that when distinctiveness motive was high and the brand was distinctive, there was a significant effect of brand anthropomorphism on brand attitude through consumers’ agency in identity expression (a point estimate of the effect = –.36; 95% CI [–.68, –.07]). There were no significant effects in other conditions (web appendix W4). Overall, the results confirmed hypothesis 2: the negative effect of anthropomorphism on brand attitude is driven by the perception of an anthropomorphized distinctive brand reducing consumers’ sense of agency in identity expression. Ruling Out Alternative Explanations. To further examine the underlying process, we used a bootstrapping approach (model 12; Hayes 2013) to test the regression model with brand anthropomorphism, brand distinctiveness, distinctiveness motivation, and all two-way and three-way interactions among them as the independent variables; consumers’ agency in identity expression, a brand’s ability to provide social distance, and perceived personal space violation as three mediators; and brand attitude as the dependent variable. Consistent with our expectations, we found that in the high distinctiveness condition and when a brand was positioned distinctively, consumers’ agency in identity expression mediated the effect of brand anthropomorphism on brand attitudes (a point estimate of the effect = –.37; 95% CI [–.70, –.05]). In the same condition, the indirect paths for the other two mediators were nonsignificant (the brand’s ability to provide social distance: a point estimate of the effect = .009; 95% CI [–.01, .08]; perceived personal space violation: a point estimate of the effect = .03; 95% CI [–.01, .11]). All other paths for all three mediators in other conditions were nonsignificant (details are presented in web appendix W4). Next, although the results of the parallel mediation test reveal a mediating role of agency in self-expression, it is possible that brand anthropomorphism reduces consumers’ perceptions of distance, thus simultaneously increasing perceived personal space violation and, as a result, leading to a diminished sense of agency. To ensure that agency in identity expression mediates over and above the other two process variables, in the condition of high distinctiveness motive and when a brand is positioned distinctively, we conducted a serial mediation analysis (i.e., a brand’s ability to provide social distance → perceived personal space violation → agency in identity expression). The only significant path from brand anthropomorphism to brand attitude emerged for agency in identity expression (a point estimate of the effect = –.27; 95% CI [–.59, –.01]; web appendix W4). Discussion Study 2 replicates the findings of study 1 with a different product (i.e., a watch) and establishes two important boundary conditions: the negative impact of brand anthropomorphism emerges only for brands that are likely to signal consumers’ distinctiveness and when consumers’ distinctiveness concerns are high. Furthermore, study 2 shows that consumers’ perceptions of lowered agency in identity expression drive the negative impact of brand anthropomorphism when distinctiveness motives are salient. The nonsignificant indirect paths through both a brand’s ability to provide social distance and perceived personal space violation rule out alternative process explanations and further enhance our confidence in the role of agency in identity expression. STUDY 3: THE ROLE OF PRIVATE VERSUS PUBLIC USAGE CONTEXT The major goal of study 3 is to examine the premise that the negative effect of brand anthropomorphism will hold only in a public (vs. private) product usage context because public products are visible to others and are important for self-expression (Tian et al. 2001), as laid out in hypothesis 3. The second goal of study 3 is to empirically rule out the alternative explanation that the distinctiveness motive leads to effects similar to those triggered by effectance motivation. However, as noted earlier, we propose that distinctiveness motive results in a greater need for agency in identity expression, while effectance leads to a greater need to make sense of an agent’s unpredictability. If distinctiveness motivation were really a subset of effectance motivation, then it should increase consumers’ desire to understand and predict the performance of the brand (Epley et al. 2007). If this is the case, then one possible reason why anthropomorphizing a distinctive brand reduces brand attitudes when distinctiveness goals are salient is that anthropomorphizing an entity that is already unique and different (e.g., a distinctive brand) can potentially reduce a consumer’s ability to comprehend it, thereby challenging consumers’ effectance motivation. Hence, study 3 aims to empirically rule out the potential mediating role of consumers’ perceived efficacy. In study 3 we manipulate the distinctiveness motive utilizing an alternative procedure. We asked participants to elaborate on the importance of distinctiveness (Cheema and Kaikati 2010). Finally, study 3 examines the impact of brand anthropomorphism on an additional dependent variable that is closer to an actual purchasing behavior—namely, brand consideration. Participants, Design, and Procedure Study 3 used a 2 (distinctiveness motivation: high vs. low) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 2 (product usage context: public vs. private) between-subjects design. Three hundred forty-two undergraduate business students at a private US East Coast university participated in study 3 in exchange for extra credit. Following the procedure of studies 1 and 2, we omitted the data of extreme outliers (n = 4) and of one respondent who reported an awareness of the study purpose. There was no attention check in this study. Overall, our final sample consisted of 337 participants (57% female). Upon arrival at a laboratory room with the maximum capacity of 27 students, participants were provided with the consent form and were told that they would take part in two unrelated studies. The first was introduced as examining the use of language in different situations, whereas the second study involved brand evaluations. Consistent with Cheema and Kaikati (2010), participants in the high distinctiveness condition elaborated on the importance of distinctiveness (being different from others). In the low distinctiveness (control) condition, respondents elaborated on the importance of a neutral activity (eating three times a day). A separate pretest (n = 168) with participants from the same population as in the main study confirmed that elaborating on distinctiveness (vs. on the importance of eating three times a day) increased participants’ desire for unique and unusual products (4.80 vs. 4.30; F(1, 166) = 9.57, p = .002, ηp2 = .06), measured with the same four items (α = .83) as used in studies 1 and 2. After the elaboration exercise, approximately half of the participants were asked to imagine shopping for an electric outdoor bike (a public usage context), and the rest were asked to imagine shopping for an in-home exercise electric bike (a private usage context) called Zidley. Next, they viewed one of four versions of an advertisement where we manipulated brand anthropomorphism as well as product features that made the product more appropriate for public or private usage. Brand anthropomorphism was manipulated via a combination of visual and verbal elements (Aggarwal and McGill 2007; figure 3). The results of the pretest showed that our manipulations were successful (web appendix W1). After viewing the ads, participants provided their ratings on brand attitude and brand consideration and responded to questions measuring agency in identity expression and perceived efficacy. The survey ended with demographic questions and a demand probe. FIGURE 3 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND PRODUCT USAGE CONTEXT IN STUDY 3 FIGURE 3 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND PRODUCT USAGE CONTEXT IN STUDY 3 Measures Brand attitude (α = .93) and agency in identity expression (r = .84, p < .001) were measured with the same items as in study 2. As a measure of brand consideration (Puzakova, Kwak, and Rocereto 2013), respondents provided their agreement with two statements (e.g., “I can imagine buying a Zidley bike”; 1 = “strongly disagree,” 7 = “strongly agree”; r = .81, p < .001; web appendix W2). Perceived efficacy was measured with two items adopted from prior work (Waytz et al. 2010b) and modified for our context (e.g., “Please indicate the extent to which you feel you can understand the Zidley brand”; 1 = “not at all,” 7 = “very much”; r = .33, p < .001). The construct correlation matrix is presented in web appendix W3. Results Hypothesis Tests. We ran two ANOVA tests (one with brand attitude and the other with brand consideration as the dependent variables) that included brand anthropomorphism, usage context, and distinctiveness motivation as the independent variables. The results revealed a significant interaction effect between distinctiveness motive and brand anthropomorphism for brand attitude (F(1, 329) = 3.92, p = .049, ηp2 = .01) and a marginally significant effect for brand consideration (F(1, 329) = 2.97, p = .086, ηp2 = .01). In addition, the findings showed a significant interaction effect between brand distinctiveness and brand anthropomorphism on brand attitude when the product usage context was public (F(1, 329) = 4.55, p = .034, ηp2 = .01), with no interaction effect when the context was private (F(1, 329) = .54, p = .462). Finally, the results revealed that when the distinctiveness motive was high, there was a marginally significant interaction between brand anthropomorphism and product usage context for brand attitude (F(1, 329) = 3.49, p = .063, ηp2 = .01); however, this effect became nonsignificant in the control condition (F(1, 329) = .22, p = .64). No other main or interaction effects were significant (ps > .11). Further planned contrasts showed that in the high distinctiveness condition and when the product usage context was public, participants had less favorable brand attitudes (4.42 vs. 4.91; F(1, 329) = 4.20, p = .041, ηp2 = .01) and reported lower brand consideration (3.40 vs. 3.95; F(1, 329) = 3.44, p = .065, ηp2 = .01) when the brand was anthropomorphized (vs. nonanthropomorphized). Consistent with a theoretical prediction that a high distinctiveness motive enhances attraction to brands positioned to be distinctive, we found that respondents exhibited more favorable attitudes toward a nonanthropomorphized brand in the distinctiveness (vs. control) condition only when the usage context was public (4.91 vs. 4.41; F(1, 329) = 4.49, p = .091, ηp2 = .01). This effect was nonsignificant for brand consideration; however, the mean difference was in the predicted direction (3.95 vs. 3.50; F(1, 329) = 1.51, p = .22). Interestingly, a marginally significant difference emerged in brand attitudes between the anthropomorphized and nonanthropomorphized brand in the control condition and when the product usage context was private (4.65 vs. 4.14; F(1, 329) = 2.89, p = .09, ηp2 = .009). We address this finding in the discussion section. Finally, we found no significant differences between the means across all other conditions for both dependent variables (ps > .19; table 2). TABLE 2 CELL MEANS IN STUDY 3   Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)    Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (**)indicates a marginally significant difference (p < .10) between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a marginally significant difference of the cell mean from (b); (a1)indicates a significant difference of the cell mean from (b1); (c)indicates a significant difference of the cell mean from (b). TABLE 2 CELL MEANS IN STUDY 3   Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)    Distinctiveness motive condition   Control condition     Public product   Private product   Public product   Private product     NAB (n = 53)  AB (n = 57)  NAB (n = 36)  AB (n = 41)  NAB (n = 28)  AB (n = 53)  NAB (n = 36)  AB (n = 33)  Brand attitude  4.91 (.96)b  4.42 (1.30)*  4.31 (1.64)c  4.51 (1.21)  4.41 (1.14)a  4.73 (1.22)  4.14 (1.32)  4.65 (1.19)**  Brand consideration  3.95 (1.49)  3.40 (1.67)**  3.81 (1.72)  3.59 (1.49)  3.50 (1.16)  3.63 (1.59)  3.39 (1.49)  3.70 (1.84)  Agency in identity expression  3.42 (1.27)b1  2.87 (1.42)**  3.17 (1.62)  2.73 (1.49)  2.71 (1.27)a1  2.79 (1.30)  2.44 (1.40)  2.96 (1.31)  Perceived efficacy  4.26 (.99)b  4.11 (1.27)  4.22 (1.00)  4.11 (1.24)  3.77 (1.13)a  3.98 (1.01)  3.74 (1.48)  4.12 (.97)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; *indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition; (**)indicates a marginally significant difference (p < .10) between an anthropomorphized and a nonanthropomorphized brand condition; (a)indicates a marginally significant difference of the cell mean from (b); (a1)indicates a significant difference of the cell mean from (b1); (c)indicates a significant difference of the cell mean from (b). Process Analysis. First, consistent with our expectations, we found that when the usage context was public and the distinctiveness motive was high, brand anthropomorphism attenuated participants’ agency in identity expression compared to a nonanthropomorphized brand condition (2.87 vs. 3.42; F(1, 329) = 4.28, p = .039, ηp2 = .01), whereas we observed no significant effects of anthropomorphism on perceived agency in other conditions (ps > .12). Next, we conducted two moderated mediation analyses (one for brand attitude and one for brand consideration as the dependent variables) using a bootstrapping approach (model 12; Hayes 2013) with brand anthropomorphism as the independent variable, product usage context and distinctiveness motivation as two moderators, and agency in identity expression as the mediating variable. The results revealed that when the distinctiveness motive was high and the product was public, there was a significant effect of brand anthropomorphism on brand attitude (a point estimate of the effect = –.21; 95% CI [–.43, –.01]) and on brand consideration (a point estimate of the effect = –.29; 95% CI [–.57, –.04]) through perceived agency in identity expression. There were no other significant indirect paths through agency in identity expression (web appendix W4). Ruling Out the Role of Perceived Efficacy. First, the results of a confirmatory factor analysis provided evidence for discriminant validity between agency in identity expression and perceived efficacy: their squared correlation of .14 was lower than both the average variance extracted for agency in identity expression (.84) and the average variance extracted for perceived efficacy (.34). Second, to further examine the underlying process of the effect of brand anthropomorphism, we used a bootstrapping approach (model 12; Hayes 2013) to test the regression model with brand anthropomorphism, product usage context, distinctiveness motive, and all two-way and three-way interactions among them as the independent variables; consumers’ agency in identity expression and perceived efficacy as two mediators; and brand attitude as the dependent variable. Consistent with our expectations, we found that in the high distinctiveness condition and in the public usage context, consumers’ agency in identity expression mediated the effect of brand anthropomorphism on brand attitudes (a point estimate of the effect = –.18; 95% CI [–.39, –.02]). In the same condition, the indirect path for perceived efficacy was nonsignificant (a point estimate of the effect = –.04; 95% CI [–.17, .08]). All other paths for the two mediators in other conditions were nonsignificant (web appendix W4). The outcomes for brand consideration were similar (web appendix W4). Discussion The results of study 3 are consistent with our theoretical prediction that when the distinctiveness motivation is activated, consumers evaluate an anthropomorphized (vs. nonanthropomorphized) brand positioned to be distinctive less positively; however, the negative effect of brand anthropomorphism occurs only for public (vs. private) usage contexts. These results are in accord with prior work showing that a signaling of uniqueness through choice of distinctive products occurs when products are public (Tian et al. 2001). The findings of study 3 further rule out the role of perceived efficacy in driving the negative effect of anthropomorphizing a distinctive brand under high distinctiveness goals. These outcomes are not surprising given that effectance motivation arises primarily in response to entities that are unexpected (Waytz et al. 2010b), and anthropomorphizing a distinctive brand does not decrease consumers’ perceived efficacy. Interestingly, the findings of study 3 also show that consumers develop more favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) brand in a private usage context in the control condition. Our speculation is that elaborating on the importance of eating three times a day (which was intended to be a control) might have inadvertently activated participants’ health goals, and an in-home exercise bike was perceived as related to this goal. Because attributions of mental capacities to technological agents enhance trust that they will perform their intended functions competently (Waytz, Heafner, and Epley 2014), it is possible that participants viewed an electric exercise bike as being able to assist them with their goal more effectively. In study 4, we directly test the possibility that when a distinctive brand is positioned as supporting consumer agency in achieving distinctiveness, the negative effect of brand anthropomorphism under high distinctiveness concerns reverses. STUDY 4: BRAND AS SUPPORTER, AGENT, OR CONTROLLER Agency-supportive social roles enhance consumers’ agency in achieving a goal (Williams et al. 2002). Hence, we expect the negative effect of brand anthropomorphism on consumer evaluations under high distinctiveness motivation to be reversed when the anthropomorphized brand is positioned as supporting consumers’ agency in identity expression, thereby providing further evidence for the underlying role of consumers’ agency in self-expression. Consequently, in study 4 we examine the moderating effect of anthropomorphizing a brand-as-supporter (vs. -controller and -agent) on consumer responses. Participants, Design, and Procedure Participants were recruited on Amazon MTurk and received a $1 monetary compensation. The HIT was available to workers with high reputation (98% of HIT approval rate and 5,000 approved HITs). To identify inattentive workers, we included a prescreen criterion based on an IMC (i.e., “Please indicate number ‘six’ in the text box below instead of ‘Today’s date’”; Oppenheimer et al. 2009), similar to that used in study 2. Five hundred forty-five MTurk workers passed the IMC question and were allowed to participate in the study. As in study 3, we omitted the data of extreme outliers (n = 5), thus leaving a sample of 540 participants (average age = 35; 39% female). None of the participants could accurately guess the true nature of the research hypothesis being examined. We employed a 2 (distinctiveness motivation: high vs. low) × 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) between-subjects factorial design. First, we manipulated high versus low distinctiveness motivation in a manner similar to that of study 3. In the high distinctiveness condition, participants were asked to elaborate on the importance of distinctiveness (being different from others). In the low distinctiveness (control) condition, respondents wrote about an innocuous topic (US geography; Berger and Shiv 2011). Next, participants were instructed to imagine that they were considering buying a new pair of sneakers and that, while searching for the shoes, they were exposed to an ad for the Geono (hypothetical) brand. Subsequently, they viewed one of six versions of the advertisement (three brand roles × two brand anthropomorphism conditions). Brand anthropomorphism was manipulated similar to that of previous studies (figure 4). We manipulated brand roles by modifying the ad copy, relying on prior research about autonomy-supportive and autonomy-controlling social roles. Prior work shows that central features of agency support are encouraging individuals to make their own choices, promoting a sense of ownership over one’s behaviors, emphasizing self-direction, and providing goal-related advice intended as information rather than direct pressure (Gagne et al. 2003; Williams et al. 2002). Building on this prior work, in the brand-as-supporter condition, we positioned the brand as encouraging consumers to stay true to their own unique style and emphasizing consumers’ own choices and behaviors, while providing options and supporting these behaviors (figure 4). In contrast to agency-supportive styles, in a controlling social role individuals generally direct others to behave in a certain way, fail to elicit others’ perspectives, provide no options, and use assertive language such as “must” and “need to” (Sheldon and Filak 2008; Silva et al. 2010). Relying on this work, in the brand-as-controller condition, we positioned the brand as overtly controlling consumers’ behaviors and choices in expressing their distinctiveness (figure 4). Finally, in the brand-as-agent condition, the ad copy emphasized the brand’s features and its distinctiveness, without focusing on the consumer. The results of the pretest revealed the effectiveness of our manipulations of brand anthropomorphism and the three brand roles (web appendix W1). After viewing the ad, participants reported their brand attitudes and agency in identity expression as the key dependent variables, and distinctiveness motivation as the manipulation-check measure. To ensure that our manipulation of brand role did not inadvertently influence perceived effectiveness of the brand or the impact of the brand in fulfilling distinctiveness goals, we measured perceived brand effectiveness. The study ended with the demographic information section and the demand probe. FIGURE 4 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND ROLES IN STUDY 4 FIGURE 4 View largeDownload slide MANIPULATIONS OF BRAND ANTHROPOMORPHISM AND BRAND ROLES IN STUDY 4 Measures Brand attitude (α = .97), agency in identity expression (r = .88, p < .001), and manipulation check on consumer distinctiveness motivation (α = .94) were measured with the same items as in studies 1–3. As a measure of brand effectiveness, we asked participants to indicate how certain they felt that the Geono brand would make them distinct from others (1 = “not at all certain,” 7 = “very certain”). Results Distinctiveness Motivation Manipulation Check and Brand Effectiveness. The results of a 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) × 2 (distinctiveness motivation: high vs. low) ANOVA revealed that consumers had greater desire for unique and unusual products in the high (vs. low) distinctiveness condition (4.24 vs. 3.80; F(1, 524) = 9.47, p = .002, ηp2 = .02), with all other effects being nonsignificant (ps > .27). Next, the results of a 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) × 2 (distinctiveness motivation: high vs. low) ANOVA test with brand effectiveness as the dependent variable showed no significant main or interaction effects (ps > .11). Importantly, the contrast tests showed no significant differences in perceived brand effectiveness among brand roles (Msupporter = 4.46 vs. Mcontroller = 4.32 vs. Magent = 4.50; ps > .31). Hypothesis Tests. A 2 (anthropomorphized vs. nonanthropomorphized brand) × 3 (brand role: supporter vs. agent vs. controller) × 2 (distinctiveness motivation: high vs. low) ANOVA with brand attitude as the dependent variable showed a significant effect of distinctiveness motivation (Mhigh = 4.42 vs. Mlow = 3.64; F(1, 528) = 26.10, p < .001, ηp2 = .05) and a two-way interaction between brand anthropomorphism and brand role (F(1, 528) = 5.29, p = .005, ηp2 = .02), with all other effects being nonsignificant (ps > .16). Further contrast analysis showed that participants had more favorable attitudes toward a nonanthropomorphized brand when their distinctiveness motives were high (vs. low) (brand-as-controller: 4.72 vs. 3.56; F(1, 528) = 9.18, p = .003, ηp2 = .02; brand-as-agent: 4.52 vs. 3.67; F(1, 528) = 5.65, p = .018, ηp2 = .01; brand-as-supporter: 4.19 vs. 3.49; F(1, 528) = 3.45, p = .064, ηp2 = .006). Next, within the high distinctiveness condition, results revealed that participants developed less favorable brand attitudes toward an anthropomorphized (vs. nonanthropomorphized) brand in the brand-as-agent condition (3.73 vs. 4.52; F(1, 528) = 4.68, p = .033, ηp2 = .01). In contrast, in the brand-as-supporter condition, consumers had more favorable brand attitudes (5.18 vs. 4.19; F(1, 528) = 6.83, p = .009, ηp2 = .01) when the brand was anthropomorphized (vs. nonanthropomorphized). Furthermore, although the mean difference was in the expected direction in the brand-as-controller condition, the effect of brand anthropomorphism was nonsignificant (4.16 vs. 4.72; F(1, 528) = 1.98, p = .16). However, the results also showed that participants reported more favorable attitudes toward an anthropomorphized brand-as-supporter compared to both an anthropomorphized brand-as-agent and a brand-as-controller (Msupporter = 5.18 vs. Mcontroller = 4.16 vs. Magent = 3.73; ps < .01), whereas there were no significant differences among brand roles in the nonanthropomorphized brand condition (ps > .16). Finally, the contrast tests within the low distinctiveness condition showed no significant effects of brand anthropomorphism on brand attitudes (ps > .25). Overall, hypothesis 4 was partially supported (table 3). TABLE 3 CELL MEANS IN STUDY 4   Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)    Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; (*) indicates a significant difference between anthropomorphized and nonanthropomorphized brand conditions; (**) indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition at p = .071; pairs of cell means (b1–b2), (c1–c2), (d1–d2), (h1–h2), (i1–i2), (g1–g2), and (L1–L2) are significantly different at p < .05; a pair of cell means (a1–a2) is marginally significantly different at p = .064; (b1) is significantly different from both (f) and (g) at p < .05; (i1) is significantly different from both (k1) and (m1) at p < .05. TABLE 3 CELL MEANS IN STUDY 4   Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)    Distinctiveness motive condition   Control condition     Brand-as-supporter   Brand-as-agent   Brand-as-controller   Brand-as-supporter   Brand-as-agent   Brand-as-controller     NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  NAB  AB  (n = 46)  (n = 40)  (n = 49)  (n = 44)  (n = 38)  (n = 41)  (n = 44)  (n = 47)  (n = 49)  (n = 45)  (n = 48)  (n = 49)  Brand attitude  4.19a1  5.18b1*  4.52c1  3.73f*  4.72d1  4.16g  3.49a2  3.92b2  3.67c2  3.73  3.56d2  3.45  (1.86)  (1.49)  (1.92)  (1.78)  (1.51)  (1.81)  (1.66)  (1.68)  (1.73)  (1.99)  (1.74)  (1.85)  Agency in identity expression  4.13h1  4.88i1*  4.45g1  3.80k1**  4.40L1  3.77m1  3.22h2  3.65i2  3.61g2  3.38k2  3.08L2  3.32 m2  (1.79)  (1.48)  (1.87)  (1.62)  (1.75)  (1.64)  (1.73)  (1.71)  (1.65)  (2.04)  (1.65)  (1.86)  NOTE.—AB = anthropomorphized brand; NAB = nonanthropomorphized brand; standard deviations are in parentheses; (*) indicates a significant difference between anthropomorphized and nonanthropomorphized brand conditions; (**) indicates a significant difference between an anthropomorphized and a nonanthropomorphized brand condition at p = .071; pairs of cell means (b1–b2), (c1–c2), (d1–d2), (h1–h2), (i1–i2), (g1–g2), and (L1–L2) are significantly different at p < .05; a pair of cell means (a1–a2) is marginally significantly different at p = .064; (b1) is significantly different from both (f) and (g) at p < .05; (i1) is significantly different from both (k1) and (m1) at p < .05. Process Analysis. We employed a moderated mediation analysis with a multicategory moderator variable (brand role) using a sequential coding for three brand-role conditions (model 12; Hayes 2013). Results showed that in the high distinctiveness condition, there was a significant negative indirect effect of brand anthropomorphism on brand attitude through agency in identity expression when the brand was positioned as an agent (a point estimate of the effect = –.53; 95% CI [–1.08, –.01]) or as a controller (a point estimate of the effect = –.62; 95% CI [–1.24, –.06]). In contrast, consumers perceived a greater agency in self-expression when a brand-as-supporter was anthropomorphized (vs. nonanthropomorphized), which in turn led to a more positive effect of brand anthropomorphism on brand attitudes (a point estimate of the effect = .55; 95% CI [.01, 1.08]). There were no significant indirect paths in other conditions (web appendix W4). Further analysis within brand anthropomorphism revealed that in the high distinctiveness condition, positioning an anthropomorphized brand as supporter (vs. agent) increased perceived agency in identity expression, which led to more positive brand attitudes (a point estimate of the effect = .83; 95% CI [.31, 1.34]). Similarly, greater agency in identity expression explained more favorable attitudes toward an anthropomorphized brand-as-supporter compared to a controller (a point estimate of the effect = .41; 95% CI [.15, 1.69]). Discussion Results of study 4 partially support hypothesis 4. When the brand is positioned as an agent, in keeping with hypothesis 4 consumers with salient distinctiveness motives develop less favorable attitudes toward an anthropomorphized (vs. nonanthropomorphized) distinctive brand, and view it as reducing their agency in identity expression. In contrast, brand-as-supporter increases the extent to which consumers view the brand as enhancing their agency in identity expression, thus leading to more favorable evaluations of an anthropomorphized (vs. nonanthropomorphized) brand. In the brand-as-controller condition, we found only a directionally lower attitude toward an anthropomorphized (vs. nonanthropomorphized) brand. However, our results are consistent with our theorizing that consumers perceive an anthropomorphized brand-as-supporter more favorably than both an anthropomorphized brand-as-agent and a brand-as-controller. These outcomes dovetail well with prior work showing that people develop stronger attachments to others who satisfy (vs. thwart) their agency needs (Thomson 2006) and contribute to this work by establishing the key role of positioning an anthropomorphized distinctive brand-as-supporter under consumers’ heightened concerns to be different. Overall, study 4 provides evidence for the mediating mechanism of agency in identity expression using a moderation approach (through manipulating different brand roles). GENERAL DISCUSSION This research provides converging evidence for our theory that anthropomorphism has a detrimental effect on evaluations of a distinctive brand under high distinctiveness motives. This finding contributes to prior work establishing the role of salient consumer goals in the direction of anthropomorphism effects (Kim and McGill 2011; Waytz et al. 2010b). In combination, the results of studies 1–4 show convergent evidence that brand anthropomorphism negatively affects perceived sense of agency in identity expression, through the moderating roles of consumers’ need for agency (study 1) and positioning the brand as agency-supporting versus agency-controlling (study 4), and through direct measurements of perceived agency in identity expression (studies 2–4). Study 2 also rules out an alternative explanation related to the potential impact of an anthropomorphized brand’s ability to provide social connectedness and a perceived violation of personal space when a distinctiveness motive is salient, whereas study 3 further rules out the role of perceived efficacy. In addition, studies 2 and 3 shed light on boundary conditions by establishing that the core negative effect of anthropomorphizing a brand holds only when consumers’ distinctiveness goals are heightened, only when a brand is positioned to be distinctive, and only when a product is publicly consumed. Importantly, our findings provide novel insights into consumer reactions to positioning a brand as supporting or reducing consumers’ sense of agency in self-expression. That is, study 4 demonstrates that when consumers’ distinctiveness concerns are high, an anthropomorphized brand-as-supporter enhances consumers’ sense of agency in identity expression, and thus brand evaluations, compared to both an anthropomorphized brand-as-agent and an anthropomorphized brand-as-controller. Theoretical and Managerial Implications Our work contributes to the literature on the role of products and brands in expressing the self (Belk 1988). Prior research shows that people infer the identities of others from both their possessions (Escalas and Bettman 2005) and the friendships that they develop (Cialdini and De Nicholas 1989). However, when practitioners imbue brands with humanlike qualities, consumers perceive these brands in a manner similar to how they perceive other humans (Kwak, Puzakova, and Rocereto 2015). As a result, our work provides crucial insights regarding which entities (i.e., anthropomorphized vs. nonanthropomorphized) consumers choose to self-express when seeking to be different from others. That is, by bringing together diverse research streams on signaling power of brands (Ahuvia 2005) and identity expression as the major driver of distinctiveness (Vignoles 2011), and by highlighting the key role of agency in self-expression (Bhattacharjee et al. 2014), the current work is the first to shed light on the extent to which consumers would prefer to use a nonanthropomorphized versus an anthropomorphized distinctive brand to signal their uniqueness. Our results are consistent with prior work showing that the intention to signal one’s identity through distinctive products is attenuated in private conditions (Berger and Heath 2007), and we add to this work by revealing that the negative effect of anthropomorphizing a distinctive brand dissipates when a product is used privately. Our research also contributes to anthropomorphism theory by unveiling counterintuitive outcomes of the influence of distinctiveness goals on the valence of the anthropomorphism effects. The current work gives insights into the unique process of brand anthropomorphism through reduced consumer agency in identity expression, as opposed to being the instrument for social connectedness or effectance motivation, as shown in prior work (Epley et al. 2008b). This understanding is crucial from a theoretical standpoint because knowledge about the role of unique goals in consumer reactions to anthropomorphized entities leads to significant insights about the downstream effects of anthropomorphism. This research further contributes to prior work highlighting consumers’ and marketers’ actions that can decrease the uniqueness of a product. For example, Cheema and Kaikati (2010) show that positive word of mouth about a unique brand decreases its perceived uniqueness. Our work adds novel insights by showing that anthropomorphizing a distinctive brand decreases its attractiveness under distinctiveness concerns. This conclusion has an important managerial implication because the decreased ability of a brand to confer distinctiveness on the consumer could result in lower firm profits. Our findings thus suggest that the managers of uniquely positioned brands or otherwise unique products should exercise caution when employing a brand anthropomorphism strategy. While brand anthropomorphism could be successful with certain consumer groups (e.g., with the assimilation goal), it might backfire with other consumers who seek distinctiveness. For example, if a woman buys upscale “Wink Face” shoes from the Alice and Olivia brand with the goal to assimilate with others at a sophisticated social function, humanlike designs of shoes may be appealing. In contrast, if a man buys a watch to signal his distinctiveness, the anthropomorphized look of the Blue Wild Face Swatch will likely be less attractive due to a decreased sense of agency in identity expression. Extensions and Directions for Future Research Several interesting research extensions are possible. The optimal distinctiveness theory states that the forces of distinctiveness and assimilation are interconnected (Brewer 1991). Thus, people tend to choose brands that simultaneously signal difference from one group and similarity to another group (Chan, Berger, and Van Boven 2012). Our work examined the goal of difference from others; however, the effects of brand anthropomorphism under the simultaneous goals of being distinct and similar to others could be a fruitful area for future research. Interestingly, given that distinctiveness in the form of difference from others has negative connotations in collectivistic cultures (Vignoles et al. 2002), our findings raise the question of whether our hypothesized effects would hold across cultures. In addition, prior work shows that some cultures (e.g., Italy, Australia) define human nature more in terms of warm traits and emotions (vs. mental states; Bain et al. 2012). Hence, it is possible that in these cultures, the negative effect of humanizing a brand on agency in identity expression may not occur. Furthermore, prior research shows that consumers strategically choose unique products to cope with social exclusion (Wan, Xu, and Ding 2014). Will social exclusion then lead to lower evaluations of anthropomorphized (vs. nonanthropomorphized) unique brands because consumers are motivated to be different? If so, such an effect could present a boundary condition for prior research showing that consumers are attracted to anthropomorphized brands when they feel excluded (Epley et al. 2008a). Future studies might explore in greater detail the specific circumstances when social exclusion may change evaluations of anthropomorphized brands. In conclusion, this research establishes a systematic impact that consumers’ differentiation motives have on the evaluations of anthropomorphized (vs. nonanthropomorphized) distinctive brands. Results suggest that anthropomorphizing unique brands hampers consumers’ perceived sense of agency in identity expression. When consumers desire uniqueness, they dislike humanness. DATA COLLECTION INFORMATION Data for studies 1, 2, and 4 were collected on Amazon Mechanical Turk by the first author. The data collection periods are as follows: study 1 in summer 2015, study 2 in fall 2016, and study 4 in spring 2017. The first author collected data for study 3 at Lehigh University in fall 2017. The first author conducted primary analyses, and both authors reviewed and discussed the results. The authors thank the editor, the associate editor, the three anonymous reviewers, and the trainee reviewer for their excellent guidance throughout the review process. Financial support from the College of Business and Economics, Lehigh University, and the Social Sciences and Humanities Research Council of Canada is gratefully acknowledged. Supplementary materials are included in the web appendix accompanying the online version of this article. References Aggarwal Pankaj, McGill Ann L. ( 2007), “ Is That Car Smiling at Me? Schema Congruity as a Basis for Evaluating Anthropomorphized Products,” Journal of Consumer Research , 34 4, 468– 79. Google Scholar CrossRef Search ADS   Aggarwal Pankaj, McGill Ann L. ( 2012), “ When Brands Seem Human, Do Humans Act Like Brands? Automatic Behavioral Priming Effects of Brand Anthropomorphism,” Journal of Consumer Research , 39 2, 307– 23. Google Scholar CrossRef Search ADS   Ahn Hee-Kyung, Kim Hae Joo, Aggarwal Pankaj ( 2014), “ Helping Fellow Beings: Anthropomorphized Social Causes and the Role of Anticipatory Guilt,” Psychological Science , 25 1, 224– 9. Google Scholar CrossRef Search ADS   Ahuvia Aaron C. ( 2005), “ Beyond the Extended Self: Loved Objects and Consumers’ Identity Narratives,” Journal of Consumer Research , 32 1, 171– 84. Google Scholar CrossRef Search ADS   Bain Paul, Vaes Jeroen, Kashima Yoshihisa, Haslam Nick, Guan Yanjun ( 2012), “ Folk Conceptions of Humanness: Beliefs about Distinctive and Core Human Characteristics in Australia, Italy, and China,” Journal of Cross-Cultural Psychology , 43 1, 53– 8. Google Scholar CrossRef Search ADS   Baumeister Roy F. ( 1998), “The Self,” in The Handbook of Social Psychology , ed. Gilbert Daniel T., Fiske Susan T., Lindzey Gardner, New York: McGraw-Hill, 680– 740. Baumeister Roy F., Sommer Kristin L. ( 1997), “ What Do Men Want? Gender Differences and Two Spheres of Belongingness: Comment on Cross and Madson,” Psychological Bulletin , 122 1, 38– 44. Google Scholar CrossRef Search ADS   Belk Russell W. ( 1988), “ Possessions and the Extended Self,” Journal of Consumer Research , 15 2, 139– 68. Google Scholar CrossRef Search ADS   Bellezza Silvia, Gino Francesca, Keinan Anat ( 2014), “ The Red Sneakers Effect: Inferring Status and Competence from Signals of Nonconformity,” Journal of Consumer Research , 41 1, 35– 54. Google Scholar CrossRef Search ADS   Berger Jonah, Heath Chip ( 2007), “ Where Consumers Diverge from Others: Identity Signaling and Product Domains,” Journal of Consumer Research , 34 2, 121– 34. Google Scholar CrossRef Search ADS   Berger Jonah, Shiv Baba ( 2011), “ Food, Sex, and the Hunger for Distinction,” Journal of Consumer Psychology , 21 4, 464– 72. Google Scholar CrossRef Search ADS   Bhattacharjee Amit, Berger Jonah, Menon Geeta ( 2014), “ When Identity Marketing Backfires: Consumer Agency in Identity Expression,” Journal of Consumer Research , 41 2, 294– 309. Google Scholar CrossRef Search ADS   Brewer Marilynn B. ( 1991), “ The Social Self: On Being the Same and Different at the Same Time,” Personality and Social Psychology Bulletin , 17 5, 475– 82. Google Scholar CrossRef Search ADS   Burns David J., Warren Homer B. ( 1995), “ Need for Uniqueness: Shopping Mall Preference and Choice Activity,” International Journal of Retail & Distribution Management , 23 12, 4– 12. Google Scholar CrossRef Search ADS   Chan Cindy, Berger Jonah, Van Boven Leaf ( 2012), “ Identifiable but Not Identical: Combining Social Identity and Uniqueness Motives in Choice,” Journal of Consumer Research , 39 3, 561– 73. Google Scholar CrossRef Search ADS   Chandler Jesse, Schwarz Norbert ( 2010), “ Use Does Not Wear Ragged the Fabric of Friendship: Thinking of Objects as Alive Makes People Less Willing to Replace Them,” Journal of Consumer Psychology , 20 2, 138– 45. Google Scholar CrossRef Search ADS   Cheema Amar, Kaikati Andrew M. ( 2010), “ The Effect of Need for Uniqueness on Word of Mouth,” Journal of Marketing Research , 47 3, 553– 63. Google Scholar CrossRef Search ADS   Chen Fangyuan, Sengupta Jaideep, Adaval Rashmi ( 2013), “The Effects of Anthropomorphization on Brand Personality Perceptions: A Motivational Account,” in NA—Advances in Consumer Research , Vol. 41, ed. Botti Simona, Labroo Aparna, Duluth, MN: Association for Consumer Research, 27– 8. Chen Rocky Peng, Wan Echo Wen, Levy Eric ( 2017), “ The Effect of Social Exclusion on Consumer Preference for Anthropomorphized Brands,” Journal of Consumer Psychology , 27 1, 23– 34. Google Scholar CrossRef Search ADS   Cialdini Robert B., De Nicholas Maralou E. ( 1989), “ Self-Presentation by Association,” Journal of Personality and Social Psychology , 57 4, 626– 31. Google Scholar CrossRef Search ADS   Clee Mona A., Wicklund Robert A. ( 1980), “ Consumer Behavior and Psychological Reactance,” Journal of Consumer Research , 6 4, 389– 405. Google Scholar CrossRef Search ADS   Deci Edward L., Ryan Richard M. ( 2000), “ The ‘What’ and ‘Why’ of Goal Pursuits: Human Needs and the Self-Determination of Behavior,” Psychological Inquiry , 11 4, 227– 68. Google Scholar CrossRef Search ADS   Deci Edward L., Ryan Richard M. ( 2008), “ Facilitating Optimal Motivation and Psychological Well-Being across Life’s Domains,” Canadian Psychology , 49 1, 14– 23. Google Scholar CrossRef Search ADS   Delbaere Marjorie, McQuarrie Edward F., Phillips Barbara J. ( 2011), “ Personification in Advertising: Using a Visual Metaphor to Trigger Anthropomorphism,” Journal of Advertising , 40 1, 121– 30. Google Scholar CrossRef Search ADS   Epley Nicholas, Akalis Scott, Waytz Adam, Cacioppo John T. ( 2008a), “ Creating Social Connection through Inferential Reproduction: Loneliness and Perceived Agency in Gadgets, Gods, and Greyhounds,” Psychological Science , 19 2, 114– 20. Google Scholar CrossRef Search ADS   Epley Nicholas, Waytz Adam ( 2009), “Mind Perception,” in The Handbook of Social Psychology , 5th edition, ed. Fiske Susan T., Gilbert Daniel T., Lindzey Gardner, New York: Wiley, 498– 541. Epley Nicholas, Waytz Adam, Akalis Scott, Cacioppo John T. ( 2008b), “ When We Need a Human: Motivational Determinants of Anthropomorphism,” Social Cognition , 26 2, 143– 55. Google Scholar CrossRef Search ADS   Epley Nicholas, Waytz Adam, Cacioppo John T. ( 2007), “ On Seeing Human: A Three-Factor Theory of Anthropomorphism,” Psychological Review , 114 4, 864– 86. Google Scholar CrossRef Search ADS   Escalas Jennifer Edson, Bettman James R. ( 2005), “ Self-Construal, Reference Groups, and Brand Meaning,” Journal of Consumer Research , 32 3, 378– 89. Google Scholar CrossRef Search ADS   Fournier Susan ( 1998), “ Consumers and Their Brands: Developing Relationship Theory in Consumer Research,” Journal of Consumer Research , 24 4, 343– 73. Google Scholar CrossRef Search ADS   Gagné Marylène, Ryan Richard M., Bargmann Kelly ( 2003), “ Autonomy Support and Need Satisfaction in the Motivation and Well-Being of Gymnasts,” Journal of Applied Sport Psychology , 15 4, 372– 90. Google Scholar CrossRef Search ADS   Harrell Gilbert D., Hutt Michael D., Anderson James C. ( 1980), “ Path Analysis of Buyer Behavior under Conditions of Crowding,” Journal of Marketing Research , 17 1, 45– 51. Google Scholar CrossRef Search ADS   Hauser David J., Schwarz Norbert ( 2016), “ Attentive Turkers: MTurk Participants Perform Better on Online Attention Checks than Do Subject Pool Participants,” Behavior Research Methods , 48 1, 400– 7. Google Scholar CrossRef Search ADS   Hayes Andrew F. ( 2013), Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach , New York: Guilford. Hong Sung-Mook, Faedda Salvatora ( 1996), “ Refinement of the Hong Psychological Reactance Scale,” Educational and Psychological Measurement , 56 1, 173– 82. Google Scholar CrossRef Search ADS   Hur Julia D., Koo Minjung, Hoffman Wilhelm ( 2015), “ When Temptations Come Alive: How Anthropomorphism Undermines Self-Control,” Journal of Consumer Research , 42 2, 340– 58. Kim Hyeongmin (Christian), Kramer Thomas ( 2015), “ Do Materialists Prefer the ‘Brand-as-Servant’? The Interactive Effect of Anthropomorphized Brand Roles and Materialism on Consumer Responses,” Journal of Consumer Research , 42 2, 284– 99. Google Scholar CrossRef Search ADS   Kim Sara, McGill Ann L. ( 2011), “ Gaming with Mr. Slot or Gaming the Slot Machine? Power, Anthropomorphism, and Risk Perception,” Journal of Consumer Research , 38 1, 94– 107. Google Scholar CrossRef Search ADS   Kwak Hyokjin, Puzakova Marina, Rocereto Joseph F. ( 2015), “ Better Not Smile at the Price: The Differential Role of Brand Anthropomorphization on Perceived Price Fairness,” Journal of Marketing , 79 4, 56– 76. Google Scholar CrossRef Search ADS   La Guardia Jennifer G., Ryan Richard M., Couchman Charles E., Deci Edward L. ( 2000), “ Within-Person Variation in Security of Attachment: A Self-Determination Theory Perspective on Attachment, Need-Fulfillment and Well-Being,” Personality and Social Psychology Bulletin , 79 3, 367– 84. Google Scholar CrossRef Search ADS   Lynn Michael, Harris Judy ( 1997), “ The Desire for Unique Consumer Products: A New Individual Differences Scale,” Psychology and Marketing , 14 6, 601– 16. Google Scholar CrossRef Search ADS   Maeng Ahreum, Aggarwal Pankaj ( 2018), “ Facing Dominance: Anthropomorphism and the Effect of Product Face Ratio on Consumer Preference,” Journal of Consumer Research , 44 5, 1104– 22. Google Scholar CrossRef Search ADS   Maimaran Michal, Wheeler S. Christian ( 2008), “ Circles, Squares, and Choice: The Effect of Shape Arrays on Uniqueness and Variety Seeking,” Journal of Marketing Research , 45 6, 731– 40. Google Scholar CrossRef Search ADS   McClelland David ( 1951), Personality , New York: Holt, Rinehart, & Winston. Google Scholar CrossRef Search ADS   Meyvis Tom, Van Osselaer Stijn M. J. ( 2017), “ Increasing the Power of Your Study by Increasing the Effect Size,” Journal of Consumer Research , 44 5, 1157– 73. Mochon Daniel, Norton Michael I., Ariely Dan ( 2012), “ Bolstering and Restoring Feelings of Competence via the IKEA Effect,” International Journal of Research in Marketing , 29 4, 363– 9. Google Scholar CrossRef Search ADS   Oppenheimer Daniel M., Meyvis Tom, Davidenko Nicolas ( 2009), “ Instructional Manipulation Checks: Detecting Satisficing to Increase Statistical Power,” Journal of Experimental Social Psychology , 45 4, 867– 72. Google Scholar CrossRef Search ADS   Parish Steven M. ( 1991), “ The Sacred Mind: Newar Cultural Representations of Mental Life and the Production of Moral Consciousness,” Ethos: Journal of the Society for Psychological Anthropology , 19 3, 313– 51. Prelinger Ernst ( 1959), “ Extension and Structure of the Self,” Journal of Psychology , 47 1, 13– 23. Google Scholar CrossRef Search ADS   Puzakova Marina, Kwak Hyokjin ( 2017), “ Should Anthropomorphized Brands Engage Customers? The Impact of Social Crowding on Brand Preferences,” Journal of Marketing , 81 6, 99– 115. Google Scholar CrossRef Search ADS   Puzakova Marina, Kwak Hyokjin, Rocereto Joseph F. ( 2013), “ When Humanizing Brands Goes Wrong: The Detrimental Effect of Brand Anthropomorphization amid Product Wrongdoings,” Journal of Marketing , 77 3, 81– 100. Google Scholar CrossRef Search ADS   Ruvio Ayalla, Shoham Aviv, Makovec Brenčič Maja ( 2008), “ Consumers’ Need for Uniqueness: Short-Form Scale Development and Cross-Cultural Validation,” International Marketing Review , 25 1, 33– 53. Google Scholar CrossRef Search ADS   Sawyer Alan G. ( 1975), “ Demand Artifacts in Laboratory Experiments in Consumer Research,” Journal of Consumer Research , 1 4, 20– 30. Google Scholar CrossRef Search ADS   Sheldon Kennon M., Filak Vincent ( 2008), “ Manipulating Autonomy, Competence, and Relatedness Support in a Game-Learning Context: New Evidence That All Three Needs Matter,” British Journal of Social Psychology , 47 2, 267– 83. Google Scholar CrossRef Search ADS   Silva Marlene N., Vieira Paulo N., Coutinho Sílvia R., Minderico Cláudia S., Matos Margarida G., Sardinha Luís B., Teixeira Pedro J. ( 2010), “ Using Self-Determination Theory to Promote Physical Activity and Weight Control: A Randomized Controlled Trial in Women,” Journal of Behavioral Medicine , 33 2, 110– 22. Google Scholar CrossRef Search ADS   Snyder C. R. ( 1992), “ Product Scarcity by Need for Uniqueness Interaction: A Consumer Catch-22 Carousel?” Basic and Applied Social Psychology , 13 1, 9– 24. Google Scholar CrossRef Search ADS   Snyder C. R., Fromkin Howard L. ( 1980), Uniqueness, the Human Pursuit of Difference , New York: Springer US. Spiller Stephen A., Fitzsimons Gavan J., Lynch John G.Jr., McClelland Gary ( 2013), “ Spotlights, Floodlights, and the Magic Number Zero: Simple Effects Tests in Moderated Regression,” Journal of Marketing Research , 50 2, 277– 88. Google Scholar CrossRef Search ADS   Swaminathan Vanitha, Stilley Karen M., Ahluwalia Rohini ( 2009), “ When Brand Personality Matters: The Moderating Role of Attachment Styles,” Journal of Consumer Research , 35 6, 985– 1002. Google Scholar CrossRef Search ADS   Thomson Matthew ( 2006), “ Human Brands: Investigating Antecedents to Consumers’ Strong Attachments to Celebrities,” Journal of Marketing , 70 3, 104– 19. Google Scholar CrossRef Search ADS   Tian Kelly Tepper, Bearden William O., Hunter Gary L. ( 2001), “ Consumers’ Need for Uniqueness: Scale Development and Validation,” Journal of Consumer Research , 28 1, 50– 66. Google Scholar CrossRef Search ADS   Vignoles Vivian L. ( 2009), “The Motive for Distinctiveness: Identity Construction in a Cultural Context,” in Oxford Handbook for Positive Psychology , ed. Snyder C. R., Lopez Shane, New York: Oxford University Press, 489– 99. Vignoles Vivian L. ( 2011), “Identity Motives,” in Handbook of Identity Theory and Research , ed. Schwartz Seth J., New York: Springer Science+Business Media, 403– 32. Google Scholar CrossRef Search ADS   Vignoles Vivian L., Chryssochoou Xenia, Breakwell Glynis M. ( 2002), “ Sources of Distinctiveness: Position, Difference and Separateness in the Identities of Anglican Parish Priests,” European Journal of Social Psychology , 32 6, 761– 80. Google Scholar CrossRef Search ADS   Vignoles Vivian L., Regalia Camillo, Manzi Claudia, Golledge Jen, Scabini Eugenia ( 2006), “ Beyond Self-Esteem: Influence of Multiple Motives on Identity Construction,” Journal of Personality and Social Psychology , 90 2, 308– 33. Google Scholar CrossRef Search ADS   Wan Echo Wen, Xu Jing, Ding Ying ( 2014), “ To Be or Not to Be Unique? The Effect of Social Exclusion on Consumer Choice,” Journal of Consumer Research , 40 6, 1109– 22. Google Scholar CrossRef Search ADS   Waytz Adam, Gray Kurt, Epley Nicholas, Wegner Daniel M. ( 2010a), “ Causes and Consequences of Mind Perception,” Trends in Cognitive Sciences , 14 8, 383– 8. Google Scholar CrossRef Search ADS   Waytz Adam, Heafner Joy, Epley Nicholas ( 2014), “ The Mind in the Machine: Anthropomorphism Increases Trust in an Autonomous Vehicle,” Journal of Experimental Social Psychology , 52 (May), 113– 7. Google Scholar CrossRef Search ADS   Waytz Adam, Morewedge Carey K., Epley Nicholas, Monteleone George, Gao Jia-Hong, Cacioppo John T. ( 2010b), “ Making Sense by Making Sentient: Effectance Motivation Increases Anthropomorphism,” Journal of Personality and Social Psychology , 99 3, 410– 35. Google Scholar CrossRef Search ADS   White Katherine, Argo Jennifer J. ( 2011), “ When Imitation Doesn’t Flatter: The Role of Consumer Distinctiveness in Responses to Mimicry,” Journal of Consumer Research , 38 4, 667– 80. Google Scholar CrossRef Search ADS   White Katherine, Dahl Darren W. ( 2006), “ To Be or Not Be? The Influence of Dissociative Reference Groups on Consumer Preferences,” Journal of Consumer Psychology , 16 4, 404– 14. Google Scholar CrossRef Search ADS   White Katherine, Dahl Darren W. ( 2007), “ Are All Out-Groups Created Equal? Consumer Identity and Dissociative Influence,” Journal of Consumer Research , 34 4, 525– 36. Google Scholar CrossRef Search ADS   Williams Geoffrey G., Gagné Marylène, Ryan Richard M., Deci Edward L. ( 2002), “ Facilitating Autonomous Motivation for Smoking Cessation,” Health Psychology , 21 1, 40– 50. Google Scholar CrossRef Search ADS   Xu Jing, Shen Hao, Wyer Robert S.Jr. ( 2012), “ Does the Distance between Us Matter? Influences of Physical Proximity to Others on Consumer Choice,” Journal of Consumer Psychology , 22 3, 418– 23. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Consumer ResearchOxford University Press

Published: Apr 25, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off