In a series of four experiments, the authors examine the implications of one consumer’s possession being mimicked by another consumer. The results demonstrate that when distinctiveness concerns are heightened, greater dissociation responses (i.e., possession disposal intentions, recustomization behaviors, and exchange behaviors) arise in response to being mimicked by a similar as opposed to dissimilar other. These effects are driven by threats to distinctiveness. Finally, these effects are mitigated when the imitated possession is nonsymbolic in nature and when a low degree of effort is exerted to initially obtain the possession. Implications for marketers and consumers are discussed.
Imitation is the sincerest form of flattery. (Colton 1820, no. 183)
The need to see oneself as unique is a potent and continuous force in our society. (Snyder and Fromkin 1980)
The first quotation above suggests that imitation by another individual is a positive occurrence that is best construed as a genuine compliment. Consistent with this notion, research demonstrates that mimicry leads to a variety of favorable responses on the part of the mimicked individual, including increased liking for and affiliation toward the mimicker (e.g., Chartrand, Maddux, and Lakin 2005; Lakin and Chartrand 2003). However, the second quotation suggests that people have a fundamental need to see the self as unique and distinct from others. As such, it seems reasonable to expect that, under certain conditions, imitation may lead to negative consequences. We propose that one such instance may arise when a consumer’s distinctiveness becomes compromised due to another person’s selecting the same product the consumer already owns. We explore the importance of consumer distinctiveness in the context of possession imitation by proposing that when distinctiveness concerns are heightened (i.e., when an independent self-construal is primed or need for uniqueness is high) and one’s distinctiveness is compromised (i.e., in the form of having one’s possession mimicked by a similar, rather than dissimilar, other), dissociation responses (i.e., possession disposal intentions, recustomization behaviors, and exchange behaviors) will be more likely to occur.
The current research makes several contributions to the marketing and psychology literatures. First, we extend work that has identified many positive outcomes of mimicry by demonstrating that mimicry can also have negative downstream consequences. To do so, we go beyond past research that has focused on the implications of mimicry that the imitated individual is unaware of (e.g., Chartrand and Bargh 1999; Chartrand et al. 2005; Lakin et al. 2003; van Baaren et al. 2003, 2004) and instead examine the consequences that emerge when consumers are aware that their possessions have been imitated. Second, our results contribute to recent work finding that consumers are often motivated to avoid and abandon products that are associated with dissimilar, dissociative others because of the costs of being misrecognized (Berger and Heath 2007, 2008) or being associated with an undesired reference group (White and Dahl 2006, 2007). While this past work demonstrates that consumers often diverge in product preferences from dissimilar others, we show that, at the individual level (i.e., one consumer mimicking another consumer’s possession), similar others can drive dissociation responses.
Third, while optimal distinctiveness theory (Brewer 1991) proposes that individuals attain a state of ideal differentiation from others by being similar to in-group members and dissimilar from out-group members, we empirically demonstrate conditions under which being too similar to in-group referents compromises one’s sense of distinctiveness. This, in turn, motivates individuals high in distinctiveness concerns to differentiate the self from highly similar, in-group members. Fourth, while past work has suggested that high levels of similarity may lead to negative emotional reactions (Snyder and Fromkin 1980), we show that increased similarity can lead to dissociation responses (i.e., possession disposal intentions, recustomization behaviors, and exchange behaviors) that enable them to reassert their distinctiveness on the imitated dimension. Importantly, we find that mimicry leads to negative reactions only when similarity is high and distinctiveness concerns are heightened (i.e., an independent self-construal or a high need for uniqueness). Fifth, we examine the moderating roles of both product type (i.e., the degree to which the imitated product is symbolic in nature) and acquisition effort (i.e., the degree to which obtaining the product requires effort). In doing so, we highlight important boundary conditions for our findings and show that, when the product is nonsymbolic in nature or when little effort was expended to acquire the product, consumers for whom distinctiveness is particularly relevant do not exhibit negative reactions to being mimicked by a similar other. Our final contribution lies in the broader focus of the research. Although work has examined consumer responses to marketers consciously imitating others’ product offerings (Ofek and Turut 2008; Warlop and Alba 2004; Wilcox, Kim, and Sen 2009), to our knowledge, our research is the first to examine consumer reactions to being copied by another individual in their product choices.
The Incidence and Consequences of Behavioral Mimicry
Behavioral mimicry is a pervasive phenomenon (see Chartrand and Dalton 2009 for a review). Research reveals that individuals often automatically imitate an interaction partner’s nonverbal (e.g., gestures, postures, and facial expressions) and verbal (e.g., speech patterns, syntax, and accents) behaviors (e.g., Chartrand and Bargh 1999; Chartrand et al. 2005; Dimberg, Thunberg, and Elmehed 2000). This work has shown that behavioral mimicry largely leads to positive outcomes. For example, social encounters characterized by a high degree of posture and mannerism sharing tend to reflect a high degree of liking, rapport, and empathy between interaction partners (Bernieri 1988; Chartrand and Bargh 1999; Chartrand et al. 2005; La France 1979; Lakin and Chartrand 2003; Tanner et al. 2008). Such nonconscious behavioral mimicry is typically referred to as the “chameleon effect,” based on the notion that mimickers are generally unaware of their tendency to match an interaction partner’s behavior (e.g., Chartrand and Bargh 1999; Cheng and Chartrand 2003; Lakin and Chartrand 2003; Tanner et al. 2008). However, conscious imitation can also occur (Bandura 1977).
Research that has examined the consequences of intentional mimicry (i.e., where the mimicker is aware of the imitation, but the mimicked individual is not) finds that it leads the mimicked individual to engage in positive prosocial behaviors such as picking up dropped pens, making a monetary donation to charity (van Baaren et al. 2004), and tipping a server in a restaurant (van Baaren et al. 2003). Consumer researchers have further found that individuals are more positive about their evaluations and consume more of a product presented by a mimicker, especially when the mimicker is perceived to be highly invested in the success of the product (Tanner et al. 2008). Finally, Bailenson and Yee (2005) find that mimicry on the part of a digital avatar results in greater agreement with the avatar’s persuasive message.
Although past research has consistently shown positive downstream effects of behavioral mimicry, there may also be instances under which mimicry can have negative outcomes (e.g., Dalton, Chartrand, and Finkel 2010). To date, research demonstrating positive mimicry effects has largely focused on mimicry that is unknown to the imitated individual (e.g., Chartrand and Bargh 1999; Chartrand et al. 2005; Lakin et al. 2003; Tanner et al. 2008; van Baaren et al. 2003, 2004). One condition under which mimicry might have negative implications is when the imitated individual is cognizant that mimicry has occurred. We propose that when a consumer is aware that his/her possession has been imitated, and this mimicry compromises the copied individual’s degree of distinctiveness, negative consequences will arise. That is, consumers may sometimes engage in dissociation responses (i.e., possession disposal intentions, recustomization behaviors, and exchange behaviors) that enable them to reassert distinctiveness on the mimicked dimension.
Consumer Distinctiveness and Reactions to Being Mimicked
According to optimal distinctiveness theory (Brewer 1991) people have two opposing fundamental needs that they strive to retain in balance—assimilation and similarity with others (i.e., a need to belong) and differentiation and distinctiveness from others (i.e., a need to be unique). One way this optimal balance can be achieved is through one’s group memberships; the need for assimilation can be satisfied through identifying with the in-group and the need for distinctiveness can be fulfilled by distinguishing the in-group from out-groups (Brewer 1991, 2003). Research has shown that both assimilation and differentiation motives can emerge. While consumers are sometimes motivated to behave similarly to those around them (Bearden, Netemeyer, and Teel 1989; Burnkrant and Cousineau 1975), they will often demonstrate preferences that distinguish themselves from others (Ariely and Levav 2000) or behave in ways that allow them to simultaneously balance their needs for assimilation and differentiation (Chan, Berger, and Van Boven 2011). One further implication of optimal distinctiveness theory is that if this balance is usurped (i.e., people’s sense of either distinctiveness or belongingness is violated in some way) people will exhibit responses that allow for the restoration of an optimal balance between the two motivations (e.g., Brewer 1993; Brewer, Manzi, and Shaw 1993; Pickett and Brewer 2001).
In the present research we are interested in the implications of an imbalance arising when distinctiveness is compromised by a similar (vs. dissimilar) consumer explicitly imitating another consumer’s possession. As possessions are seen as an extension of the self (Belk 1988), we anticipate that consumers for whom distinctiveness concerns are particularly self-relevant will attempt to restore balance in response to having their distinctiveness compromised by having a possession mimicked by a similar other. They may achieve this by engaging in dissociation responses (i.e., possession disposal intentions, recustomization behaviors, and exchange behaviors) because such responses allow them to reassert their distinctiveness on the dimension that has been mimicked. In sum, our framework suggests that being mimicked by a similar (vs. dissimilar) other compromises consumer distinctiveness, but that this only leads to negative consequences when consumers have distinctiveness concerns heightened (i.e., when an independent self-construal is primed or need for uniqueness is high).
Similarity of the Mimicker
The first factor we consider in our framework is the similarity between the consumer and the mimicker. This factor is particularly relevant to explore given that imitation is most likely to occur when a high degree of similarity exists (Bernieri, Reznick, and Rosenthal 1988; Gueguen and Martin 2009). We predict that when too much shared similarity exists between the consumer and the mimicker this will compromise the consumer’s sense of distinctiveness. This expectation is supported by work that finds that people strive to avoid appearing too similar to others (e.g., Vignoles, Chryssochoou, and Breakwell 2000; Vignoles et al. 2006) and that too much similarity can have negative affective consequences (Lynn and Snyder 2002; Snyder and Fromkin 1980). We extend this past work in two ways. First, we show that higher levels of similarity can lead to dissociation responses (i.e., possession disposal intentions, recustomization behaviors, and exchange behaviors). Second, as we discuss next, we identify an important boundary condition for this effect—distinctiveness concerns.
We propose that although similarity of the mimicker will lead individuals to reassert their distinctiveness, this tendency will be particularly pronounced among those who have heightened distinctiveness concerns (i.e., the degree to which being a unique and autonomous individual is important to the self). This is because being copied by a similar other is particularly threatening to those for whom being distinct from others is important. In the present research we operationalize distinctiveness concerns in two ways.
First, we propose that primed self-construal reflects the activation of distinctiveness concerns. Research has highlighted that two distinct construals, independent and interdependent, exist with regard to the way in which the self is viewed in relation to others (e.g., Singelis 1994). People with an independent self-construal see the self as autonomous and unique, whereas those with more interdependent self-construals see the self as relational and interconnected with others (e.g., Brewer and Gardner 1996; Markus and Kitayama 1991; Triandis 1989). Although people tend to differ on which of these self-construals is more chronically accessible (Singelis 1994; Triandis 1989), situational cues can influence which construal is activated at a given time (Brewer and Gardner 1996; Gardner, Gabriel, and Lee 1999; Trafimow, Triandis and Goto 1991; van Baaren et al. 2003; White, Lehman, and Cohen 2006). Research suggests that independents (vs. interdependents) view the self as more unique (Burns and Brady 1992; Yamaguchi, Kuhlman, and Sugimori 1995), prefer options that represent uniqueness (Markus and Kim 1999), have more negative reactions to brands associated with out-groups (an effect attributed to a greater desire for self-differentiation; Escalas and Bettman 2005), and define the self more in terms of unique identifying traits (e.g., Brewer and Gardner 1996; Oyserman and Lee 2008; Trafimow et al. 1991). Thus, extant research shows that independents, rather than interdependents, are more concerned with being unique and distinct from others.
Second, we examine measured individual differences in need for uniqueness as an additional proxy for distinctiveness concerns. Although most people appear to be motivated to see themselves as unique to some degree (e.g., Snyder and Fromkin 1980), the extent to which the motivation exists varies at the individual level (e.g., Lynn and Harris 1997; Snyder and Fromkin 1980; Tian, Bearden, and Hunter 2001). Specifically, those high in need for uniqueness tend to be more likely to desire unique, counternormative, or unconventional options (Bloch 1995; Lynn and Harris 1997; Simonson and Nowlis 2000; Snyder 1992; Tian et al. 2001; Tian and McKenzie 2001). In addition, compared to those low in need for uniqueness, those high on this dimension are somewhat more likely to avoid a product that the majority of others have chosen (Berger and Heath 2007) and will avoid engaging in positive word of mouth regarding a publicly consumed product that they own out of concern that others will also purchase it (Cheema and Kaikati 2010). Thus, those high in need for uniqueness are particularly concerned with maintaining their distinctiveness.
In sum, our framework proposes that when consumer distinctiveness is jeopardized (i.e., a possession is mimicked by a similar rather than dissimilar other) and distinctiveness concerns are heightened (i.e., independent self-construal is primed or need for uniqueness is high), consumers will distance themselves from the mimicked possession to differentiate themselves from the mimicker. This is expected to arise because when distinctiveness concerns are activated and a similar other mimics one’s possession, threat to distinctiveness should be highest. Such distinctiveness threat should then, in turn, influence consumer tendencies to dissociate within the mimicked domain. More formally,
H1a Consumers high in distinctiveness concerns will report greater possession dissociation responses when their possession is mimicked by a similar rather than a dissimilar other.
H1b For consumers low in distinctiveness concerns, differences in dissociation responses as a function of the nature of the mimicker will be eliminated.
H2 Perceptions of distinctiveness threat will mediate the above effects.
The notion that similarity can drive divergence is counterintuitive in light of previous research that finds that dissimilarity drives product abandonment and avoidance at a collective level (Berger and Heath 2007, 2008; White and Dahl 2006, 2007). That is, this past work suggests that consumers will disfavor products that are associated with particular out-groups. The current work departs from this past research by examining explicit mimicry that occurs at the individual level (e.g., one consumer is obviously copied by another consumer). The current research forwards the novel prediction that, under certain conditions, similarity, rather than dissimilarity will drive consumer dissociation behaviors. In particular, we argue that explicit mimicry at the individual level is highly relevant to the self, and as a result, when it is done by a similar other it is particularly threatening to one’s sense of distinctiveness.
The Current Research
We test our hypotheses regarding dissociation behaviors in response to possession mimicry across four studies. In study 1, we demonstrate that those high in distinctiveness concerns are most likely to report possession disposal intentions when they are copied by a similar rather than dissimilar other. In study 2, we replicate this finding using an alternative manipulation of similarity and show that distinctiveness threat underlies the findings. In studies 3 and 4 we investigate conditions under which those high in distinctiveness concerns will not exhibit negative reactions to being mimicked by a similar other. Specifically, we examine the moderating role of product symbolism and show that our effects emerge for recustomization behaviors only when the possession is symbolic in nature (study 3). Finally, in study 4, we examine another behavior—possession exchange—and demonstrate a boundary condition such that our effects do not arise when minimal effort is required to attain the product.
In study 1 we vary distinctiveness concerns by manipulating the activation of an independent versus interdependent construal of self. Following the framework outlined above, those who have an independent self-construal primed should be more motivated to view the self as a unique individual (Brewer and Gardner 1996; Oyserman and Lee 2008) and, as a result, be more inclined to dispose of a possession that is mimicked by a similar as opposed to a dissimilar other.
Sixty undergraduates (32 males, 28 females) participated in a 2 (self-construal: independent vs. interdependent) × 2 (nature of the mimicker: similar vs. dissimilar) between-subjects design in exchange for a $10 honorarium. Participants took part in small groups of 4–10 people.
Upon arrival, participants were informed that they would be completing multiple tasks during the study session. The first task they were presented with was the manipulation for self-construal. Following past research (e.g., Brewer and Gardner 1996; Trafimow et al. 1991; White et al. 2006), participants were told that the study was interested in obtaining a measure of verbal competence and to examine this they were going to read a short story involving a trip to a nearby city about which they would later be asked questions. They were further told that to determine whether people are able to comprehend the story when they are distracted, they would be asked to circle pronouns appearing in the text of the short story. The texts in the two conditions differed only in the extent to which different pronouns were used: in the independent condition, individual-level pronouns such as I and me were used (e.g., So I ventured downtown and into the new Victoria Square Mall. I stopped for lunch at a fast-food chain in the food court. After that I stopped for a look at the historic Parliament Buildings, where the Government conducts its affairs. I got some wonderful photographs in front of the fountain, using the buildings as a backdrop), whereas in the interdependent condition collective-level pronouns such as we and us were used (e.g., So we ventured downtown and into the new Victoria Square Mall. We stopped for lunch at a fast-food chain in the food court. After that we stopped for a look at the historic Parliament Buildings, where the Government conducts its affairs. We got some wonderful photographs in front of the fountain, using the buildings as a backdrop). In order to provide evidence that our manipulation of self-construal activates distinctiveness concerns, a pretest was conducted. Participants (n = 36) completed the self-construal manipulation (described in detail below) and then answered the following items on 7-point scales: “To what degree do you feel like a unique individual?” “To what degree do you feel distinct from others?” and “To what degree do you feel different from others?” (distinctiveness activation index α=.90). An independent samples t-test revealed that those primed with independence reported greater feelings of uniqueness (M = 5.77) than did those primed with interdependence (M = 4.87; t(35) = 2.59, p <..02).
Participants were then asked to read and imagine a scenario that described a social interaction that involved a discussion about a brand of perfume/cologne that the participant owned. More specifically, participants read that, while they were in the photocopy room at work, a coworker commented that s/he really liked the participant’s perfume/cologne and then indicated that although s/he knew the participant wore that specific brand, s/he recently purchased the same one. To achieve the nature of the mimicker manipulation, in the similar condition, the scenario described the coworker as being the participant’s friend while in the dissimilar condition the coworker was someone whom the participant did not like. After completing the scenario they completed a short survey to indicate their disposal intentions regarding the mimicked possession. In particular, participants were asked to what extent they would “dispose of,” “throw out,” “pack away,” and “give away” the copied perfume/cologne on 7-point item scales (1 = not at all likely, 7 = very likely). These four items were averaged together to create a disposal intentions index (α =.92). Participants then completed three items that assessed the nature of the mimicker manipulation, indicating the extent to which they felt “similar” to, “dissimilar (reverse scored)” from, and “different (reverse scored)” from the other person in the scenario (1 = not at all, 7 = very; similarity index, α =.82). Finally, participants indicated their gender and completed an open-ended suspicion probe. Analysis revealed that gender did not predict or interact with any of the other independent variables to predict variance in the dependent variable. This was true in both of the studies that included both males and females in the sample (i.e., studies 1 and 2). Examination of the suspicion probe revealed that participants were not cognizant of the experimental hypotheses in any of the studies.
To assess our manipulation of the nature of the mimicker a 2 (self-construal: independent vs. interdependent) × 2 (nature of the mimicker: similar vs. dissimilar) analysis of variance (ANOVA) was conducted including the similarity index as the dependent variable. Results indicated a significant main effect for mimicker (F(1, 56) = 16.81, p <..001), revealing that participants perceived more similarity to the other person in the similar condition (M = 4.29) than in the dissimilar condition (M = 3.23).
Test of Hypotheses
A 2 (self-construal: independent vs. interdependent) × 2 (nature of the mimicker: similar vs. dissimilar) ANOVA with the disposal intentions index as the dependent variable revealed significant main effects for both self-construal (F(1, 56) = 6.74, p <..05) and mimicker (F(1, 56) = 13.82, p <..001; see fig. 1). Importantly, these findings were qualified by a two-way interaction (F(1, 56) = 7.18, p <..05). Simple effects tests revealed that, consistent with expectations, participants primed with an independent self-construal were more likely to dispose of the perfume/cologne when they had been mimicked by a similar (M = 3.92) as compared to a dissimilar (M = 1.54; t(56) = 4.45, p <..001) other. In contrast, participants primed with an interdependent construal did not differ in their possession disposal intentions as a function of the mimicker’s similarity (Msimilar = 1.95, Mdissimilar = 1.57; t(56) =.75, NS). Finally, when the mimicker was similar, those primed with an independent self-construal were more likely to indicate a willingness to dispose of the possession than those primed with an interdependent self-construal (t(56) = 3.79, p <. 01).
The results reveal that among those primed with an independent self-construal, greater disposal intentions were reported when possessions was copied by a similar as compared to a dissimilar other. Those primed with interdependence, however, did not exhibit differential disposal intentions regardless of who mimicked their possession.
In a conceptual replication of study 1, a follow-up study employed a similar design and used the same manipulation of nature of the mimicker, but used measured need for uniqueness (Tian et al. 2001) to index distinctiveness concerns. In addition, we assessed consumer evaluations as the dependent variable (e.g., “How likely would your preference for the perfume/cologne decrease?” and “How likely do you think the importance of the perfume/cologne to you would decrease?” (r =.84, p <..001). Results revealed a significant interaction (β =.392; t(110) = 2.37, p <..03). Simple slopes analysis shows that while those low in need for uniqueness did not report differences in evaluations as a function of the nature of the mimicker (t(110) = 1.11, NS), those high in need for uniqueness exhibited more negative preferences when the mimicker was similar as opposed to dissimilar (t(110) = 2.28, p <..03). Taken together then, the results of study 1 and our follow-up study support the notion that having one’s distinctiveness compromised (being mimicked by a similar other) only leads to greater disposal intentions and negative product evaluations when the consumer is also particularly concerned with being distinct from others.
In study 2, we sought to extend study 1’s findings in two ways. First, we wished to further elucidate the mechanism underlying the effects by examining the mediational role of distinctiveness threat. Recall that our framework proposes that consumers will exhibit dissociative responses to products when similarity is high and distinctiveness concerns are relevant because these are the conditions under which threats to distinctiveness are highest. Second, we wished to rule out some potential alternative explanations for the effects of our mimicker manipulation. Although our manipulation check in study 1 showed that the mimicker in the similar condition was viewed as being significantly more similar than the mimicker in the dissimilar condition, the manipulation may have affected the perceived closeness of the other person as well. Thus, in study 2 we use an alternative manipulation of similarity that has been shown to lead people to draw more similarities versus differences when thinking of another person (e.g., Markman and Gentner 1996; Mussweiler 2001). We predict that priming a similar versus dissimilar mind-set will have effects analogous to the nature of the mimicker manipulation in study 1. Consumers primed with independence will report greater intentions to dispose of a possession when the possession is mimicked and a similar versus dissimilar mind-set is activated. When consumers are primed with interdependence, differences in possession disposal intentions as a function of a similar versus dissimilar mind-set will not emerge.
Eighty undergraduates (46 males, 34 females) participated in a 2 (self-construal: independent vs. interdependent) × 2 (mind-set: similar vs. dissimilar) between-subjects design in exchange for a $10 honorarium.
Upon arrival, participants were informed that they would be completing multiple tasks during the study session. They were then given the manipulation to induce either a similar or dissimilar mind-set. Following a procedure outlined by Markman and Gentner (1996; see also Mussweiler 2001) participants were informed that the first study was a pretest for later studies that the experimenter planned to complete on event memory. They were then given a priming task that consisted of drawings of two household scenes. In the first scene a female was shown leaning over a table while holding a cup of coffee. In addition, the scene included a Christmas tree with gifts below it and a fireplace in the background. In the second scene a male was shown standing in front of a table while reaching for a bowl that was located in the middle of the table. The scene also included a bottle and a few glasses located on the table and a fireplace in the background. In the similar condition, participants were asked to examine the two scenes and list the similarities that appeared between them. In the dissimilar condition they were asked to list the differences that appeared between the two drawings. In both conditions, participants were instructed to be sure to inspect the two scenes as thoroughly as possible and list as many features as they could find. Once participants had completed the mind-set manipulation, the experimenter explained the second study they were to complete, which was the same manipulation for self-construal as described in study 1. Participants were then presented with a scenario that was the same as the one described in study 1 except that specific information about the nature of the mimicker was not provided.
After completing the scenario, participants indicated in a short survey the extent to which their distinctiveness was threatened on 7-point scales (1 = not at all, 7 = very much) including “To what degree did having someone select the same product threaten your sense of distinctiveness?” “To what degree did having someone select the same product as you make you feel less unique?” “To what degree did having someone select the same product threaten your sense of being a unique individual?” “To what degree did having someone select the same product threaten your individuality?” (distinctiveness threat index, α =.90). Participants then reported their disposal tendencies using the items indicated in study 1 (disposal intentions index, α =.90). In addition, our similarity manipulation was assessed using the same three items reported in study 1 (similarity index, α =.89). Finally, two 7-point scale items (1 = not at all, 7 = very) were included to assess whether the similarity manipulation influenced the extent to which participants felt they shared a relationship with the mimicker (i.e., “close” and “connected”; relationship index r =.43).
To assess our similarity manipulation a 2 (self-construal: independent vs. interdependent) × 2 (mind-set: similar vs. dissimilar) ANOVA was conducted including the similarity index as the dependent variable. Results revealed only a significant main effect for similarity (F(1, 76) = 12.74, p <..001). Participants reported feeling more similar to the mimicker in the similar mind-set (M = 4.01) as compared to the dissimilar mind-set (M = 2.81). A 2 (self-construal) × 2 (mind-set) ANOVA with the relationship index included as the dependent variable did not produce any significant effects (p’s >.20). Thus, the similarity manipulation influences feelings of similarity toward the mimicker, but not feelings of having a close relationship with the mimicker.
Test of Hypotheses
A 2 (self-construal: independent vs. interdependent) × 2 (mind-set: similar vs. dissimilar) ANOVA with the disposal intentions index as the dependent variable revealed significant main effects for both self-construal (F(1, 76) = 6.28, p <..05) and mind-set (F(1, 76) = 3.85, p =.05). Importantly, these findings were qualified by the anticipated two-way interaction (F(1, 76) = 4.75, p <..05; see fig. 2). Simple effects tests revealed that, consistent with our theorizing, participants primed with an independent self-construal were more likely to dispose of the perfume/cologne when they had a similar (M = 2.98) as compared to a dissimilar (M = 1.92; t(76) = 2.97, p <..01) mind-set. In contrast, participants primed with an interdependent self-construal did not differ in their possession disposal intentions as a function of mind-set (Msimilar = 1.78, Mdissimilar = 1.83; t(76) =.152, NS). Finally, when a similar mind-set was primed, those with an independent self-construal were more likely to indicate a willingness to dispose of the possession than those with an interdependent self-construal (t(76) = 3.23, p <..01).
Mediational Role of Distinctiveness Threat
Mediation analysis was used to examine whether distinctiveness threat underlies the effects. The bootstrapping methodology was used based on the recommendation by Shrout and Bolger (2002) and Preacher and Hayes (2008). The results show that when distinctiveness threat is examined as the mediating factor, the 95% BCa (bias corrected and accelerated) bootstrap CI of.040 to.345 was obtained. Since zero was not included in the lower and upper bounds of this confidence interval, this was an indication that the distinctiveness threat had a significant indirect effect in the relationship between our interaction term and disposal tendencies.
Utilizing a contextual manipulation for similarity to the mimicker, we once again demonstrate that those primed with independence are more inclined to dispose of a copied product when the mimicker is perceived as being similar rather than dissimilar (due to a manipulated mind-set). On the other hand, participants primed with an interdependent self-construal did not differ in their possession disposal intentions as a function of similarity. Consistent with our conceptualization, distinctiveness threat mediates the effects.
Study 3 builds on the previous studies in a number of ways. First, we employ an alternative operationalization of distinctiveness concerns to increase confidence that it is concerns about being unique and distinct that drive the effects. While the self-construal manipulation does influence thoughts about uniqueness (i.e., as demonstrated in our pretest in study 1), it is possible that the manipulation may have affected other dimensions as well (e.g., Aaker and Lee 2001). We thus include a previously validated individual difference measure that more specifically taps into one’s need for uniqueness (Tian et al. 2001). Second, we manipulate similarity utilizing an alternative procedure (Tesser and Campbell 1980) that directly manipulates the similarity of the mimicker while holding that person’s valence constant. Third, we broaden the scope of our investigation by examining recustomization behaviors toward actual product bundles in a more involving context. The use of this methodology enables us to examine whether the individual must be explicitly told they have been copied (as in the previous studies) or if the effects also occur when consumers simply notice they have been mimicked. This method also allows us to examine whether consumers will alter a self-customized bundle of products that initially is composed of preferred options, to a recustomized bundle that includes less desirable, yet differentiated, items. Fourth, given that no gender differences emerged in the previous studies, we restrict our investigation to females. This allows us to examine the recustomization of product bundles that are personally relevant.
In addition, we investigate the moderating role of product type. While certain products are symbolic in nature, other products are less so (e.g., Berger and Heath 2007). Because symbolic products communicate information about the self to others, we propose that being mimicked on a more symbolic product compromises one’s sense of being a unique and distinct individual. Consistent with this reasoning, we predict that when the imitated product is symbolic in nature those high (vs. low) in distinctiveness concerns will be more likely to report dissociation behaviors related to a possession that has been copied by a similar rather than dissimilar other. However, when the imitated possession is a nonsymbolic product no differences in dissociation behaviors will emerge. Thus:
H3a When the imitated product is symbolic, those high in distinctiveness concerns will indicate greater dissociation responses when it is mimicked by a similar rather than a dissimilar other. Among those low in distinctiveness concerns, differences in dissociation responses as a function of the nature of the mimicker will be mitigated.
H3b When the imitated product is nonsymbolic, the above differences will be eliminated.
Seventy-six female undergraduates took part in a 2 (information about mimicker: similar vs. dissimilar) × need for uniqueness between-subjects design in exchange for course credit. We included two dependent variables—whether participants chose to recustomize their bundle on a symbolic product dimension and whether participants chose to recustomize their bundle on a nonsymbolic dimension.
In order to select products that were high and low in symbolism, a pretest was conducted. On an a priori basis, we expected that lip gloss might be perceived as more symbolic, whereas loofah sponges might be perceived as nonsymbolic. Fourteen participants evaluated both lip gloss and loofah sponges and evaluated each product on the following scales: “To what extent does _____ express who you are?” “To what extent does _____ communicate something about the person who uses it?” and “How much _____ symbolize what kind of person uses it?” (on 7-point scales, lip gloss α =.92, loofah α =.90). A paired samples t-test indicated that the lip gloss was perceived to be significantly more symbolic (M = 5.00) than the loofah sponge (M = 2.45; t(13) = 5.15, p <..001).
Prior to arrival, participants received an e-mail in which they were asked to provide some personal information (i.e., month of birth, major, and favorite television show) and e-mail their responses back to the researcher. Responses to these questions were later used to manipulate information about the mimicker (i.e., similar vs. dissimilar). Upon arrival at the experimental session, participants, who were run individually, were greeted by the experimenter and another “participant” who was already present. Unbeknownst to the participant, the other “participant” was really a confederate. To manipulate information about the mimicker, we adapted a procedure utilized by Tesser and Campbell (1980). The experimenter told both participants that although the session was normally run one person at a time, because of time constraints they had been scheduled in pairs. Further, they were told that, to eliminate any potential bias, each participant was paired with someone who was very “similar to” or “different from” them. Not only were they born in the same (different) month, but they were also in the same (different) major, and their favorite television show was the same (different). “So all in all, you’re really (not) very similar to one another (in any way), so there shouldn’t be any problem in running you two together.”
The participants were then told that the purpose of the study was to investigate product customization and to explore this they would be asked to create a bundle consisting of personal care products. They were asked to put together a gift basket containing personal care products that they would like to take home with them. There were two product categories available—one that was symbolic (i.e., lip gloss) and one that was nonsymbolic (i.e., loofah sponges). Participants then selected one lip gloss (from three options) and one loofah sponge (from three options). After the participant made her two product choices and customized her basket, the confederate made her two selections which consisted of the exact same two options the participant had chosen.
Participants then completed a short survey. After responding to a couple of filler questions related to the cover story, participants were asked if they would like to change any of the products they had initially selected. If a participant indicated an interest in recustomizing her bundle the research assistant presented her with an option that was pretested as being less desirable than the original options. Our first dependent variable was whether or not participants recustomized their bundle (by trading in their preferred option for one that was less desirable) on the symbolic product dimension. Our second dependent variable was whether or not the bundle was recustomized on the nonsymbolic product dimension. Participants then completed the need-for-uniqueness scale (Tian et al. 2001). Examples of items in the scale include, “I collect unusual products as a way of telling people I am different” and “I have sometimes purchased unusual products or brands as a way to create a more distinctive personal image” (α =.95). The need-for-uniqueness scale assesses enduring individual differences (rather than situational variation), and we note that the manipulation of information about the mimicker did not significantly influence ratings of need for uniqueness (p >.20). They were then asked to complete the similarity index used in study 1 (α =.77).
To test the success of the information about the mimicker manipulation, linear regression analysis was conducted that included information about the mimicker, the continuous mean-centered need for uniqueness index, and their interaction term as the independent variables and the similarity index as the dependent variable. Results revealed a main effect only for information about the mimicker (β =.424, t(72) = 3.97, p <..001). Thus, our manipulation was effective.
Tests of Hypotheses
Binary logistic regression was conducted with information about the mimicker, the continuous mean-centered need-for-uniqueness scale, and their interaction term as the independent variables and whether or not participants recustomized their product bundle on the symbolic dimension as the dependent variable. This analysis revealed the predicted significant interaction (β = 1.355, Wald = 5.75, p <..02). The main effects for similarity (β =.597, Wald = 2.85, p =.092) and need for uniqueness (β = 1.06, Wald = 3.52, p =.061) did not reach significance. To facilitate an examination of the interaction for the symbolic dimension, a median split was performed on the need-for-uniqueness index (e.g., White and Argo 2009). On the symbolic dimension, those high in need for uniqueness were more likely to recustomize their product bundle when the mimicker was similar (58%) versus dissimilar (11%; χ2(1) = 8.88, p <..01; see fig. 3). When need for uniqueness was low, individuals did not differ in the tendency to recustomize their product bundle in response to a similar (15%) versus dissimilar (12%) mimicker (χ2(1) =.08, p >.70). Similar analysis was conducted using the nonsymbolic dimension as the dependent variable. As anticipated, this analysis did not reveal a significant interaction (β = −.539, Wald =.784, p >.30).
Using a more involving methodology, study 3 shows that consumers who notice that they are mimicked are more likely to recustomize the symbolic element of a product bundle when they are high in need for uniqueness and the mimicker is similar (vs. dissimilar). Consistent with our framework, we do not find that high need for uniqueness individuals recustomize their bundle on nonsymbolic elements when mimicked by similar others. Also, consistent with theorizing, those low in need for uniqueness did not recustomize on symbolic or nonsymbolic elements, regardless of the information about the mimicker.
An additional follow-up study also investigated the role of product type by manipulating it as a between-subjects variable. This study primed self-construal and manipulated the nature of the mimicker as in study 1 and then varied whether the product copied was symbolic (i.e., perfume) or nonsymbolic (e.g., an iron). Results of a 2 (nature of the mimicker: similar vs. dissimilar) × 2 (self-construal: independent vs. interdependent) × 2 (product type: symbolic vs. nonsymbolic) ANOVA with disposal intentions (as measured in study 1) revealed a significant three-way interaction (F(1, 219) = 5.05, p <..05). When the product was symbolic, a significant interaction between mimicker and self-construal emerged (F(1, 111) = 10.19, p <..01). Those primed with independence reported higher disposal intentions when the possession was copied by a similar (M = 2.57) as opposed to dissimilar (M = 1.68) other (t(111) = 2.67, p <..01). Among those primed with interdependence, no differences in disposal intentions emerged when the mimicker was dissimilar (M = 1.46) versus similar (M = 2.06; t(111) = 1.70, p >.05). When the product was nonsymbolic in nature, the interaction between mimicker and self-construal was nonsignificant (F(1, 108) =.011, NS). These results thus conceptually replicate the findings reported in study 3 and suggest that the effects are most pronounced when the product is more symbolic in nature.
Taken together, our first three studies provide evidence for our proposed conceptual framework. When distinctiveness concerns are heightened (i.e., independent self-construal has been primed or the individual is high in need for uniqueness) and distinctiveness is compromised (i.e., consumers’ possessions are mimicked by a similar rather than dissimilar other) consumers exhibit increased product dissociation responses on the mimicked dimension (studies 1–3), particularly when the product is symbolic in nature (study 3). Further, our findings appear to be driven by threats to consumer distinctiveness (study 2).
Again using an involving context where the opportunity for consumers to notice they are being mimicked exists, study 4 has two primary objectives. First, we introduce another type of dissociation response—possession exchange. In particular, we explore whether or not participants will exchange their original product choice for an alternative, less-preferred option. Second, we explore conditions under which those who have their distinctiveness concerns activated (i.e., those primed with independence) do not demonstrate the tendency to dispose of a symbolic product in response to being imitated by a similar other. To achieve this, we manipulate the degree to which the individual views their product acquisition as effortful.
Past research shows that the higher the degree of effort expended on a task, the more important and favorable the task and its outcomes are viewed (Beatty and Smith 1987; Cardozo 1965). It may be that exerting a lot of effort in acquiring a symbolic product makes that product more important to the self. As such, being copied on a product that was acquired following a high degree of effort should be particularly threatening to those high in distinctiveness concerns. In contrast, when minimal effort is expended to acquire the product, product imitation should not be threatening. Thus, we predict that, holding mimicker constant as similar, when an individual high in distinctiveness concerns (independent self-construal is primed) is copied and a low degree of effort was involved in acquiring the product the effects will be eliminated.
H4a Consumers high in distinctiveness concerns will be less likely to exhibit dissociation responses regarding an imitated possession when the effort to acquire the possession is low rather than high.
H4b Among consumers low in distinctiveness concerns, differences in dissociation responses will not emerge as a function of effort.
Seventy-four female undergraduates participated in a study that used a 2 (self-construal: independent vs. interdependent) × 2 (effort to obtain the product: low vs. high) between-subjects experimental design in exchange for $10. Mimicker was held constant as a similar other.
Participants took part in the study in pairs. However, only one individual in each pair was a real participant. Unbeknownst to the real participant, the other individual taking part in the study was a confederate. The confederate was also a female student, and participants were led to believe that she was also an undergraduate student from the same academic program. We assumed that this would lead participants to perceive some degree of similarity with the participant. Participants were informed that they would be taking part in a series of unrelated studies. First, they completed the self-construal manipulation described in study 1. They then completed a computer task that manipulated the amount of effort they needed to engage in to acquire the target product—their choice of a pair of sandals. Regardless of the effort condition, after clicking the start button on a computer, participants were presented with instructions about the task they would be required to complete in order to earn a pair of sandals. Three keys on the keyboard (B, N, M) were assigned to three positions on the screen (“left,” “middle,” “right,” respectively). During each “trial” of the task, three randomly chosen letters were presented on the computer screen and participants were instructed to indicate by pressing one of three keys, which of the letters appears first in the alphabet. To achieve the effort manipulation, in the low effort condition, participants completed three trials of the computer task whereas in the high effort condition they completed 50 trials. A pretest confirmed that the two tasks indeed differed in terms of the perceived effort that was required.
After completing the effort manipulation, both the participant and the confederate were notified that they had each earned a pair of sandals. Both the participant and the confederate were then taken to view a display containing six different options of sandals in assorted colors and styles. The different styles of sandals pretested as being equal in attractiveness. The participant was then asked to make a product selection by pointing out which pair she preferred. After the participant indicated her choice, the confederate also indicated to the experimenter that she would like the same pair of sandals. While the experimenter retrieved the shoes in the appropriate sizes, the participant and the confederate both returned to their respective work spaces (located out of view from each other) to begin completing a survey. The first series of items contained a number of filler questions to provide the experimenter with sufficient time to distribute the sandals. The survey then indicated that participants had an opportunity to trade the pair of sandals they initially selected for another alternative. The instructions indicated that if they would like to see the alternative they should raise their hand and the experimenter would bring the substitute pair over for them to inspect. In the instances when participants raised their hand, the experimenter took over a plain pair of white sandals that had pretested as being significantly less attractive and desirable than the sandals in the original choice set. Once participants saw the alternative they were asked if they would like to switch. The percentage of participants who ultimately traded down from their preferred alternative to the less desirable option was the key dependent variable.
To ensure that participants did indeed perceive the confederate to be similar, a one-sample t-test was conducted that compared the reported similarity mean (M = 4.18) to a baseline of 3.5 (midpoint on the similarity index; t(67) = 5.12, p <.01).
Test of Hypothesis
Examination of the frequencies with which participants traded their selection when mimicked by a similar other revealed results consistent with our predictions, given that the interaction between self-construal and effort was significant (β =.517, Wald = 3.70, p =.05; see fig. 4). Those primed with an independent self-construal traded their preferred option for a less desirable alternative more often when the task was effortful (56%) as compared to when the effort to attain the product was low (20%, χ2 = 5.06, p <..05). In contrast, those primed with an interdependent self-construal were equally unlikely to trade their original option regardless of the amount of effort that was exerted to obtain the product (effort low = 28%, effort high = 20%, χ2 =.317, p >.5).
Study 4 utilized an involving laboratory methodology to examine whether being mimicked by a similar other can lead consumers to exchange a possession for a less-preferred alternative. Those primed with independence were most likely to trade their original product selection for a less-desirable alternative when they exerted high (versus low) effort to obtain the product. Among those primed with interdependence, no differences in possession exchange occurred as a function of task effort.
This study not only identifies a condition under which consumers high in distinctiveness concerns are less motivated to dispose of a mimicked possession but also shows a counterintuitive result. Research on cognitive dissonance and effort justification often finds that when a high rather than low degree of effort is expended on a task, the task and its outcomes are often preferred (e.g., Aronson and Mills 1959; Beatty and Smith 1987; Cardozo 1965; Festinger 1957; Kivets and Simonson 2002). Furthermore, research shows that when people put more effort into obtaining a tangible good this makes them value it more (Lowenstein and Issacharoff 1994). The current work, then, builds on the cognitive dissonance literature by showing when these effort justification effects can be reversed—increased effort can sometimes lead consumers to value the product less. According to our conceptualization, this arises when one’s distinctiveness concerns are heightened and this distinctiveness is compromised by being copied by a similar other. Under these conditions, when the consumer puts a great deal of effort into obtaining a symbolic product, being copied by a similar other can be particularly threatening, leading those high in distinctiveness concerns to exchange the product.
The current research demonstrates the important role of consumer distinctiveness in determining reactions to having one’s possession mimicked by another individual. We find that the most negative reactions to being explicitly copied by another consumer emerge when distinctiveness concerns are high (i.e., independent self-construal is activated or need for uniqueness is high) and distinctiveness is jeopardized in some way (i.e., similarity is heightened when mimicry occurs). In particular, we find that consumers with high distinctiveness concerns report stronger disposal intentions when they were copied by a similar rather than a dissimilar other (study 1), thoughts of similarity to the other are activated based on the context (study 2), or similarity is manipulated through information about the mimicker (study 3). These effects only emerge in symbolic (but not nonsymbolic) product domains (study 3) and appear to be driven by perceptions of distinctiveness threat (study 2). Finally, in study 4 we demonstrate that when distinctiveness concerns are high, consumers are more likely to exchange a preferred possession for another, less-desirable option when they are copied by a similar other and a high (versus low) degree of effort is required to obtain the product. This tendency does not emerge for those for whom distinctiveness concerns are low.
The current research adds to a growing body of literature on mimicry that largely suggests that being imitated by another individual has positive consequences (Bernieri 1988; Chartrand and Bargh 1999; Chartrand et al. 2005; La France 1979; Lakin and Chartrand 2003; Tanner et al. 2008). It does so by highlighting a novel negative consequence of being imitated by another individual. Consumer research shows that imitation of others can occur when individuals incidentally view the consumption behaviors of others (Ferarro, Bettman, and Chartrand 2009) as well as when they consume in the presence of others (Ramanathan and McGill 2007), which in turn leads to congruent product preferences and consumption behaviors on the part of the observer. The current research extends this past work by showing when negative consequences of mimicry (i.e., increases in dissociation responses) can occur.
The current research also extends work showing that consumers often avoid products imbued with negative meanings because they are associated with negatively viewed out-groups (Berger and Heath 2007, 2008; Escalas and Bettman 2005; White and Dahl 2006, 2007). We demonstrate that when distinctiveness concerns are paramount, similar in-group members can drive consumers to engage in dissociation responses related to a copied product. While individuals often fulfill their affiliation and assimilation needs through connection with similar in-group members (Brewer 1991), the current work shows that consumers can also balance their needs for differentiation and distinctiveness through distancing their product selections from those of similar others.
This research also contributes to our understanding of an understudied topic in the consumer research literature—dissociation responses regarding already owned products (i.e., possessions). Although dissociation responses of possessions (e.g., product disposal) are an important facet of consumer behavior (Solomon 2009), they have not been extensively examined. A small body of research explores factors that relate to product disposal when consumers have ended their use of a possession (Belk, Sherry, and Wallendorf 1988; Lastovicka and Fernandez 2005). The current research extends this past work by examining factors that increase tendencies to dispose of, recustomize, or exchange a previously preferred possession before the product’s utility has expired. Further, we demonstrate that factors external to the actual product itself can lead to dissociation responses. Ironically, when distinctiveness concerns are high and one is copied by a similar other, the mimicked individual will recustomize or exchange a preferred possession for a less desirable option. Thus, dissociation responses that arise due to mimicry have less to do with the qualities of the product itself and more to do with the consumer’s desire for self-differentiation.
From a practical standpoint, marketers may want to consider what happens when consumers can no longer assert their distinctiveness through a possession. For example, if the consumer’s possession is copied or if a product becomes too easy to obtain or commonplace it may be abandoned or viewed less positively (Berger and Heath 2007, 2008; Granovetter and Soong 1986). The results of study 3 suggest that marketers should provide consumers with ways to customize their possessions over time. As a result, if distinctiveness becomes compromised, consumers could update their product’s uniqueness through customized alterations rather than simply disposing of or exchanging it. Examples of this include changing the style of the face plate on a cellular phone to altering the color of a watch wristband, to changing one’s options on an automobile.
The current work points to a number of opportunities for future research. One interesting question is whether our effect is limited to being copied on a product choice per se or whether more general forms of mimicry would lead to similar effects. For example, would being mimicked by a similar other on some behavior, such as nonverbal gestures, lead those high in distinctiveness concerns to assert their individuality by engaging in behaviors that make them appear unique? Also, do consumers need to be consciously aware that their possession is being mimicked or can these effects arise at an unconscious level? Future research could examine such possibilities.
Another interesting question is whether the effects of being mimicked by a similar other persist over time. That is, would consumers exchange a possession that was copied if they only had the chance to do so at a later point in time? In addition, it would be interesting to examine other consequences of being mimicked by similar others. For example, would consumers prefer products that are more scarce or difficult to obtain after being mimicked (e.g., Snyder 1992)? Would this occur for products only related to the mimicked dimension or for any product that conveys distinctiveness? Further, the consumption context might have implications for dissociation responses. For example, distinctiveness might be increasingly compromised when the consumption is in a more public setting and diminished under conditions that are more private in nature (e.g., White and Dahl 2006), which might moderate distinctiveness-seeking responses. Contrary to the notion that mimicry has ubiquitously positive effects, the current research shows that negative responses to mimicry can sometimes arise. We hope this work is a first step toward providing a more nuanced view of mimicry in the domain of consumption.