Uncertainty Evokes Consumers’ Preference for Brands Incongruent with their Global

Abstract

This research demonstrates that under states of certainty, consumers with a relatively stronger global (local) identity prefer global (local) brands, whereas under states of uncertainty, consumers with a relatively stronger global (local) identity prefer local (global) brands. This effect occurs because uncertainty (certainty) activates a divergent (convergent) thinking style, which results in a preference for options that are more distant from (closer to) the identity to which consumers associate more strongly. The effect holds both when individuals’ global–local citizenship identity is measured and when it is manipulated. The research further establishes an important boundary condition for the effect. The effect holds in the citizenship identity context because people normally associate themselves with both local and global citizenship identities, and situational or dispositional factors only influence the degree to which they associate with each identity. The effect does not surface when individuals construe their local–global citizenship identities as interfering, meaning they conceive that holding one identity conflicts with holding the other.

Keywords

globalization branding identity uncertainty creativity divergent thinking

Uncertainty is an integral part of consumers’ lives. Whether the source of uncertainty is distant (e.g., the possible impact of Brexit or of an upcoming presidential election) or proximal (e.g., “will I get this job?”), prior research has shown that incidental uncertainty exerts consequential influence on consumers’ consumption decisions in a variety of domains (e.g., Faraji-Rad and Pham 2017). Building on prior literature, in this research we investigate how such incidental uncertainty may also influence consumers’ brand choices in today’s globalized world. Globalization has dramatically shaped the lives of modern consumers (Arnett 2002; Robertson 1995). As a result of globalization, consumers are faced with choices between global brands (i.e., multicountry brands marketed across multiple markets) and local brands (i.e., brands marketed in a more restricted locality) (Alden, Steenkamp, and Batra 2006; Strizhakova and Coulter 2015). Significantly, they make these choices with reference to the identities they have formed in reaction to globalization. Most consumers possess two independent and noninterfering identities: a global identity and a local identity (Arnett 2002; Strizhakova, Coulter, and Price 2011), and the relative strength of these identities has the potential to shape their preferences for global versus local brands.

A common understanding from the brand preference literature in general (e.g., Aaker 2000) and the global/local identity research more specifically (Strizhakova, Coulter, and Price 2011; Zhang and Khare 2009) is that consumers generally prefer brands that are congruent with their identities. However, we show that this preference for identity-consistent brands does not hold under states of uncertainty (vs. certainty). Instead, we propose that uncertainty (certainty) activates a divergent (convergent) thinking style (Guilford 1967, 1968), which results in a preference for brands that are incongruent (congruent) with one’s local/global identity. Thus, under states of certainty, consumers with more salient global (local) identities prefer global (local) brands. However, under states of uncertainty, consumers with more salient global (local) identities prefer local (global) brands. We further show that this reversal in brand preferences holds in the local/global identity context because individuals generally view global and local identities as noninterfering identities. This reversal effect does not surface when individuals construe them as interfering (Settles 2004; Strizhakova, Coulter, and Price 2011).

In the following sections, we describe the theoretical background of our research. We then present the results of five studies that provide evidence for the proposed effect and its process explanation.

Theoretical Background

Global–Local Identity and Brand Choices

For marketers, an important consequence of globalization is the diffusion of global brands around the world. As a result of this diffusion, consumers increasingly face choices between global and local brands (Alden, Steenkamp, and Batra 2006). Such tussles between global and local brands exist in many industries, from ride sharing (e.g., Uber vs. Didi in China) to food and beverages (e.g., KFC vs. Jollibee in the Philippines). In addition, firms often face a strategic decision regarding whether to position themselves as more global or local, regardless of how objectively global or local their operations are. A clear example of such duality is HSBC’s “the world’s local bank” slogan.

A psychological result of globalization is that people increasingly perceive themselves to be interconnected with the rest of the world (Robertson 1995), a perception that leads to the development of a global citizen identity. A poll conducted by the British Broadcasting Corporation involving over 20,000 people in 18 countries showed that people around the world are increasingly identifying themselves as global citizens (Grimley 2016). At the same time, beyond such identification of oneself as a global citizen, every individual is also a member of a local community and holds a local citizen identity (Arnett 2002; Strizhakova and Coulter 2019).

Although consumers often face an “either/or” choice between local and global brands (Strizhakova and Coulter 2015), theoretical and empirical work has shown that an individual’s association with global and local identities is not a matter of “either/or” but rather “both/and” (Alden, Steenkamp, and Batra 2006; Arnett 2002; Strizhakova and Coulter 2019; Strizhakova, Coulter, and Price 2011, 2012; Tu, Khare, and Zhang 2012). As Arnett (2002, p. 777, emphasis in original) originally observed two decades ago, “…as a consequence of globalization, most people now develop a bicultural identity, in which one part of their identity is rooted in their local culture while another part stems from an awareness of their relation to the global culture.” Since then, a large body of literature (e.g., Alden, Steenkamp, and Batra 2006; Strizhakova, Coulter, and Price 2011, 2012) has empirically established that consumers simultaneously possess both global and local identities and, depending on various factors, consumers may feel more connected to their global identity, their local identity, both identities, or neither identity (Strizhakova and Coulter 2015, 2019). Because global and local identities coexist, global and local identities are not inherently “interfering” in nature; in other words, association with one identity does not require one to dissociate from the other (Settles 2004). Consumers’ global identity allows them to connect with people from diverse populations, especially through media technology, while they still use their local identity in their daily interactions with family, friends, and community members (Arnett 2002, p. 777). In summary, although consumers normally associate with both local and global identities, depending on individual or situational factors, one identity may be more salient and therefore more influential in guiding behavior.

Such a conceptualization of citizenship identity is consistent with broader research on social identities and culture, which shows that individuals may possess multiple identities simultaneously (Brewer 1991; Hong et al. 2000). For example, an individual may be a professor, a father, and a community leader, and may hold multiple cultural identities (e.g., independent vs. interdependent self). The relative strength of each identity compared with the others at any given moment has an impact on the individual’s behavior (Aaker and Lee 2001; Brewer 1991; Stayman and Deshpande 1989).

Given the importance of global and local identities for modern consumers, it is therefore natural that they would evaluate global and local brands in relation to these identities. An accessible identity provides a set of internal scripts or “guides” that define how an individual should behave at any given time (Oyserman 2009). Consumers’ preferences for brands and products are shaped by the brand’s degree of congruence versus incongruence with these guides (e.g., Aaker 2000). Previous research on citizenship identities has shown that consumers with stronger local (global) identities prefer local (global) brands (Strizhakova, Coulter, and Price 2011; Zhang and Khare 2009), an effect that is consistent with similar congruity effects in many other contexts (e.g., Adaval 2001; Han and Shavitt 1994; Higgins 2000; Liberman and Trope 1998; Sirgy 1982).

However, are consumers always motivated to follow the guide dictated by their identity and prefer identity-congruent brands, or would they sometimes prefer brands that are incongruent with their identity? Previous research both within the context of global–local identity (Zhang and Khare 2009) and other contexts (Yang et al. 2011) has shown that under some circumstances, consumers may prefer identity-incongruent brands. Zhang and Khare (2009, Study 2), for example, demonstrated that consumers who were primed with a motivation to differentiate from (integrate with) their social group were more likely to prefer an identity-incongruent (identity-congruent) brand. We argue that the level of incidental uncertainty consumers experience is also an important situational variable that influences whether they prefer identity-(in)consistent brands.

Uncertainty and Thinking Styles

Various economic, social, and political occurrences—such as those many contemporary consumers experience—may trigger uncertainty. Uncertainty may be triggered as a result of positive or negative events in consumers’ daily lives as well (Faraji-Rad and Pham 2017). Although the economics and decision making literatures often conceptualize uncertainty as a description of the objective state of knowledge regarding a decision (Edwards 1954; Kahneman, Slovic, and Tversky 1982; Knight 1921), uncertainty may also be defined as an attribute of an individual’s state (Bar-Anan, Wilson, and Gilbert 2009). When conceptualized as the state of an individual, uncertainty refers to the psychological condition of the person making the decision (i.e., a feeling state of uncertainty) (Kagan 1972; Van den Bos 2009). A psychological state of uncertainty (vs. certainty) may be caused by a person’s surrounding environment. Experimentally, a state of uncertainty (vs. certainty) may also be induced through different priming manipulations (Bar-Anan, Wilson, and Gilbert 2009; Clore and Parrott 1994; Faraji-Rad and Pham 2017; Faraji-Rad, Samuelsen, and Warlop 2015; Hogg 2007; Lee and Qiu 2009; Tiedens and Linton 2001).

Regardless of how it is induced, a psychological state of uncertainty has important motivational consequences in terms of how people process information and make decisions (Kagan 1972). Namely, uncertainty is a signal to the decision maker that the currently available pool of information is not sufficient (Bar-Anan, Wilson, and Gilbert 2009). This signal induces psychological discomfort and thus motivates the individual to resolve the uncertainty (Kagan 1972; Laurin, Kay, and Moscovitch 2008; Van den Bos 2009). Uncertainty encourages individuals to move away from routine and comfortable (and often less cognitively demanding) decision processes and to instead adopt decision processes that would help expand the pool of knowledge available to them. For example, uncertainty has been shown to cause curiosity (Bar-Anan, Wilson, and Gilbert 2009; Lee and Qiu 2009), which has been theorized as a motivational “drive state” caused by “a gap in knowledge and understanding” (Loewenstein 1994). A drive state, such as hunger, is theorized as an all-encompassing excitatory state, which drives individuals’ behaviors until balance is reached (Kagan 1972; Seward 1956). Thus, the causal link between uncertainty and curiosity is consistent with the notion that uncertainty causes a general motivation to learn more about the environment and to solve problems. Other research shows that, compared with certainty, uncertainty causes a shift in consumers’ information processing styles. Uncertainty has been shown to increase the systematic processing of information, which helps an individual uncover more detailed information about their environment (Tiedens and Linton 2001; Vaughn and Weary 2003). Uncertainty also increases reliance on affect-based decision making, which results in individuals having greater access to the vast amount of information they have previously encoded about their environment (Faraji-Rad and Pham 2017).

In summary, these observations suggest that a state of uncertainty causes a shift in consumers’ decision-making and information-processing approach, motivating them to adopt decision processes that help them expand the pool of knowledge available to them and solve problems. Consistent with these observations, in this research we propose that uncertainty also influences another decision-making dimension, which has downstream consequences in shaping consumers’ preference for identity-(in)consistent brands. Namely, we propose that uncertainty (vs. certainty) encourages a divergent (vs. convergent) thinking style.

To elaborate, a large body of research conducted within the past five decades distinguishes between divergent and convergent styles of thinking (Guilford 1967, 1968; Mednick 1962). These thinking styles describe how individuals search for information within the associative network and make decisions (Yang et al. 2011). From an information-processing perspective, divergent (convergent) thinking involves conducting a broader (narrower) information search within the associative network. From a decision-making perspective, divergent (vs. convergent) thinking involves forgoing familiar and conventional approaches one would ordinarily employ in decision making and favoring unconventional approaches that one would ordinarily not consider (Burroughs and Mick 2004; Cropley 2006; Moreau and Dahl 2005; Yang et al. 2011). Building on our argument that uncertainty motivates individuals to adopt decision processes that help them problem solve and expand the pool of knowledge available to them, we propose that uncertainty (certainty) activates a divergent (convergent) thinking style.

Uncertainty, Identity, and Preferences for Identity-Inconsistent Brands

Our proposition that uncertainty (certainty) activates a divergent (convergent) thinking style further suggests that uncertainty (certainty) leads to a preference for identity-incongruent (congruent) brands. As explained previously, a prominent identity provides a set of guides on how an individual should behave. Consumers adopting a convergent thinking style gravitate toward conventional decision-making approaches and are likely to follow the guides set by their identity (“I am a global person, and my identity suggests that I should choose the global brand, so I will go with my identity and choose the global brand”). However, according to previous research, a divergent thinking style not only motivates individuals to forgo conventional decision approaches but it also motivates them to pursue unconventional and novel approaches (Friedman and Förster 2002; Yang et al. 2011). Friedman and Förster (2002), for example, associated divergent thinking not only with an avoidance of more conservative decision strategies but also a pursuit of riskier strategies (also see Fredrickson 2004). Previous research has also maintained that stimuli incompatible with one’s own internal states (e.g., one’s own emotional states, identity) are considered more unconventional and novel. Aaker and Williams (1998), for example, showed that emotional appeals in advertisements that are incompatible with an individual’s emotional focus are considered more novel and unconventional. Thus, divergent thinkers’ motivation to pursue unconventional approaches leads them to seek identity-inconsistent brands rather than identity-consistent ones. Yang et al. (2011) tested the relationship between divergent thinking and preference for inconsistency by exploring consumers’ preference for advertisements that were congruent versus incongruent with their mental construal (i.e., the level of abstractness at which objects are construed in the mind; Liberman and Trope 1998). The authors observed that participants in a less creative mindset (akin to convergent thinking style) preferred advertisements that were congruent with their level of construal. However, those in a more creative mindset (akin to divergent thinking style) not only had a reduced preference for the congruent advertisement but also demonstrated a higher preference for the incongruent advertisement. In other words, divergent thinkers’ motivation to try new approaches in decision making, and to think “out of the box,” not only led them to prefer congruent advertisements less but also created a greater preference for advertisements that were incongruent with their mental construal. We argue that a similar process applies to our context of global and local identities and brand choices: A divergent thinking style should cause consumers not only to break away from the decision guides provided by their identity but also to choose options that are incompatible with these guides (“I am a global person, but I want to think differently today, so I will go against my identity and choose the local brand”).

Given our earlier proposition that uncertainty (certainty) leads to a divergent (convergent) thinking style, we therefore now propose that uncertainty (certainty) leads to a preference for identity-inconsistent brands. Specifically, we propose that under states of certainty, consumers with a relatively stronger global (local) identity would exhibit a greater preference for global (local) brands, whereas under states of uncertainty, consumers with a relatively stronger global (local) identity would exhibit a stronger preference for local (global) brands. We test the proposition regarding the effect of uncertainty on preference for identity-(in)congruent brands in all our studies. Furthermore, we provide support for our proposition that a change in thinking style under uncertainty (vs. certainty) mediates the effect by both measuring the mediator (Study 3) and manipulating it (Study 4).

The Moderating Role of Identity Type

The arguments discussed previously propose that uncertainty (vs. certainty) evokes a preference for brands that are inconsistent (vs. consistent) with consumers’ local/global identities. However, we must ask whether this effect holds across all forms of identity. Relatedly, how can our propositions be reconciled with prior findings in the literature showing that, with regard to religious identities, people sometimes demonstrate stronger convictions after being primed with uncertainty (e.g., Hogg 2007)? This research has shown that uncertainty causes people to express a stronger belief in God (Kay, Moscovitch, and Laurin 2010; Laurin, Kay, and Moscovitch 2008), evaluate religious ingroups more favorably and religious outgroups more unfavorably (Greenberg et al. 1991), and express their moral identity more emphatically (McGregor et al. 2001; Rosenblatt et al. 1989). Prior research has argued that uncertainty is self-threatening, and the self-threat experienced after primed uncertainty motivates people to reaffirm their religious convictions by dissociating from the outgroup and associating with the ingroup (Hogg 2007; Hogg, Adelman, and Blagg 2010; Van den Bos 2009). One may reasonably ask: If uncertainty causes people to exhibit stronger religious convictions, why does it not also cause people to affirm their citizenship identity by choosing an identity-consistent brand?

As reviewed earlier, an important attribute of global and local identities is that they are often construed as coexisting and noninterfering identities, rather than interfering identities. According to previous identity research, when two identities are viewed as interfering, people usually see a conflict between the roles, values, and cultures surrounding them, and association with one identity is perceived to require some dissociation from the other (e.g., Sellers et al. 2003; Settles 2004). In contrast, when two identities are seen to be noninterfering, people perceive less conflict between them (Aryee 1992; Bagger, Li, and Gutek 2008; Kopelman, Greenhaus, and Connolly 1983; Kossek and Ozeki 1998; Netemeyer, Boles, and McMurrian 1996; Sellers et al. 1997; Sellers et al. 2003; Settles 2004). We suggest that this important attribute of citizenship identity, which may not be present in identities such as religious identity (e.g., being a Christian vs. not being a Christian), leads to our proposed effect of uncertainty on individuals’ preference for brands inconsistent with their citizenship identity.

When two alternative identities are seen as interfering, associating with one identity will require some dissociation from the other. In such a context, making a choice inconsistent with an important identity (e.g., purchasing a product that is inconsistent with being a Christian) would not only work against affirming that identity (e.g., being a Christian) but would also signal the rejection of that important identity. In such a situation, the consumer would perceive the identity-inconsistent choice as unacceptable and would exclude it from their initial consideration set. Thus, even when consumers have divergent thinking motivations, an identity-inconsistent choice is not an option they would consider in this context. In contrast, when the two identities are noninterfering, choosing a product that is inconsistent with an important identity would not signal a rejection of that identity. As such, from a self-affirmation standpoint, choosing an identity-consistent versus an identity-inconsistent option is nondiagnostic. Thus, the consumer would deem both identity-consistent and identity-inconsistent options as acceptable and include them in their consideration set. In this context, whether the consumer chooses an identity-consistent option or an identity-inconsistent option will depend on that consumer’s thinking style.

To summarize, we predict that uncertainty evokes a preference for identity-inconsistent brands in the context of citizenship identity because global and local citizenship identities are normally seen as noninterfering. However, if individuals were to perceive their global and local citizenship identities as interfering, the effect would reverse, leading to uncertainty increasing individuals’ preference for the identity-consistent brand. We test this prediction in Study 5.

Empirical Approach

In the following sections, we report the results of five studies that test our propositions. In our studies, to induce uncertainty versus certainty independently of the judgments to be made and independently of the information provided for these judgments, we use the priming method. We also measure and manipulate the relative salience of local versus global identities, rather than focusing on each identity individually. Although the independence of global and local identities has been established in the literature, it is also widely accepted that consumers view the choice between local and global brands as an either/or choice¹ (Alden, Steenkamp, and Batra 2006; Steenkamp and De Jong 2010; Strizhakova and Coulter 2015). As such, in our research, we study the choice of local and global brands (or preference between them) instead of individuals’ evaluation of each brand. Given the independence of local and global identities, examining the effect of an identity on this choice separately would omit important information about the strength of the other identity (e.g., a person may associate highly with one identity but even higher with the other) that would be needed for us to interpret whether the choice is identity-consistent or identity-inconsistent.

Study 1

The aim of Study 1 was to test whether the relative strength of consumers’ predisposed global versus local identity interacts with uncertainty (vs. certainty) to predict the preference for local versus global brands. We expected that consumers experiencing certainty would prefer a brand that was congruent with their identity, whereas consumers experiencing uncertainty would prefer a brand that was incongruent with their identity.

Method

Participants and design

A total of 383 students from a large U.S. university participated in the study for course credit. The study adopted a 2 (prime: uncertainty vs. certainty) × 2 (brand type: global vs. local) between-subjects design. Consistent with prior research (e.g., Alden, Steenkamp, and Batra 2006; Tu, Khare, and Zhang 2012), we measured the relative strength of participants’ global (vs. local) identities. The dependent variable was participants’ evaluation of the brand.

Procedure and stimuli

The study ostensibly consisted of several unrelated parts. The first part used an essay-writing task to prime either certainty or uncertainty. Depending on the condition, participants read a brief paragraph about certainty or uncertainty and were then asked to describe what they felt most certain or uncertain about in their lives at the moment (see Web Appendix A). Then, as a manipulation check, the participants indicated the extent to which they felt uncertain and the extent to which life was unpredictable for them using two seven-point items. We calculated a composite uncertainty score by averaging these two items (r = .67).

After completing the first part of the study, participants moved on to the second part, which was presented as a market research survey for a pair of new headphones. All participants were shown a review of headphones described as portable, having a sleek design, and providing high sound quality. However, depending on the brand type condition, the headphones were described as being a global or a local brand (see Web Appendix A).

After reading the review, participants were asked to evaluate the headphones on a four-item scale ranging from 1 (“negative,” “unfavorable,” “not a good buy,” and “not likeable”) to 7 (“positive,” “favorable,” “a good buy,” and “very likeable”). An overall evaluation score was created from the mean of these four items (α = .93). Finally, to assess the relative strength of participants’ global and local identities, we adopted the scales from Tu et al. (2012), which consist of four items to measure global identity (α = .66) and four items to measure local identity (α = .72) (see Web appendix A). We subtracted the mean of the local identity items from the global identity items to obtain the relative strength of global (vs. local) identity.

Results and Discussion

Preliminary analysis

Compared with participants in the certainty prime condition (M = 4.97, SE = .092), those in the uncertainty prime condition reported feeling more uncertain (M = 5.30, SE = .094; F(1, 379) = 5.97, p = .015). The manipulation did not affect participants’ responses to the identity scale (F < 1). Therefore, the priming manipulation was successful.

Brand preference

We conducted a regression analysis with brand attitude as the dependent variable and identity, uncertainty (−1 = “certain,” and 1 = “uncertain”), brand type (−1 = “local,” and 1 = “global”), and all interactions as the independent variables. Analysis showed a significant identity × brand type interaction (β = .101, SE = .05; t(374) = 2.15, p = .03) and a significant three-way interaction (β = −.30, SE = .05; t(374) = −6.19, p < .001). All other effects were nonsignificant (p values > .197). Further analysis to investigate this three-way interaction showed that among participants primed with certainty, the identity × brand type interaction was significant (F(1, 374) = 19.65, p < .001) (see Figure 1). Among these participants, a stronger local identity (at −1SD) led to a more favorable attitude toward the local brand (M = 4.92) than the global brand (M = 4.10; β = −.41, SE = .12; t(374) = −3.39, p = .001), whereas a stronger global identity (at +1SD) led to a more favorable attitude toward the global brand (M = 5.14) than the local brand (M = 3.89; t(374) = 4.17, p < .001). This pattern of results demonstrates a preference for the identity-consistent brand among participants primed with certainty. The identity × brand type interaction was also significant among participants primed with uncertainty (F(1, 374) = 18.60; p < .001). However, among these participants, the interaction pattern was reversed. As predicted, in this case, a stronger local identity (−1SD) led to a more favorable attitude toward the global brand (M = 4.90) than the local brand (M = 4.41; β = .25, SE = .13; t(374) = 1.96, p = .05); whereas a stronger global identity (+1SD) led to a more favorable attitude toward the local brand (M = 4.97) than the global brand (M = 4.45; β = −.26, SE = .12; t(374) = −2.10, p = .037). This pattern of results demonstrates a preference for identity-inconsistent brands among participants primed with uncertainty.²

Figure 1.

Brand evaluation as a function of uncertainty prime, brand type, and relative strength of global versus local citizenship identity (Study 1).

Study 2

Study 2 built on Study 1 in several ways. First, whereas Study 1 measured the relative strength of global versus local identities, Study 2 aimed to provide stronger evidence of causality by manipulating the salience of participants’ global versus local identities. Second, whereas in Study 1 we measured a relative preference for global versus local brands between subjects, in Study 2, we asked participants to make a choice. Third, we also provided participants the option to defer their choice in Study 2. We did this because in actual consumption, consumers do have an option to defer choice, and providing participants with a no-choice option mirrors actual market behavior more closely and is a more stringent test of our hypothesis.

Method

Participants and design

We randomly assigned 805 participants recruited from Prolific (UK) (M_age = 36.13 years, 66.3% female) to the conditions of a 2 (prime: uncertainty vs. certainty) × 2 (identity: global vs. local) between-subjects study.

Procedure and stimuli

We conducted the study under the guise of three unrelated parts. The first part primed global versus local identity using a scrambled sentence task adapted from Zhang and Khare (2009), modifying it to reduce potential demand effects. We asked participants in each condition to unscramble 16 scrambled sentences. Six of these sentences were non-identity-related filler sentences (e.g., “is waiter friendly the very”) that were identical across conditions and were added to conceal the prime’s intent. The other ten sentences primed global or local identity, depending on the condition (e.g., “events know I local” in the local-identity condition vs. “events know I global” in the global-identity condition; see Web Appendix B for the full set of sentences). The order of all sentences was randomized. As reported in detail in Web Appendix B, a pretest among a separate group of 115 participants demonstrated that the prime influenced the strength of participants’ global and local identities, as intended.

After completing the identity-priming task, participants moved to the next ostensibly unrelated part in which they completed either the certainty or uncertainty manipulation task described in Study 1. They then proceeded to a purported third part, presented as a market research survey. Unlike Study 1, in which participants only read the information on one brand (either local or global), in this study, we provided participants information on both a local and a global brand. Specifically, we showed participants a table of information describing the features of two sets of headphones in the form of a side-by-side, comparative review (see Web Appendix B). We pretested the descriptions of the headphones before the study (see Web Appendix B). Participants were told one of these headphones, presented as “Brand G,” was a global brand, whereas the other, presented as “Brand L,” was a local brand. Participants were told that the real brand names were disguised for confidentiality. Presentation of the brand information was counterbalanced (i.e., for the same headphone description, the product was labeled as Brand L for half of the participants and Brand G for the other half). This ensured that differences in participants’ responses would not be attributed to differences in the product descriptions. After reading the review, we asked participants to indicate whether they would choose the local brand (“I would choose Brand L”), the global brand (“I would choose Brand G”), or whether they would prefer to defer choice (“I wouldn’t make a choice at this moment”) if they had to purchase a set of headphones. Finally, as a check of the uncertainty manipulation, we measured participants’ degree of uncertainty using two items identical to the ones used in Study 1 (r = .68).

Results and Discussion

Manipulation checks

Participants in the uncertainty-prime condition felt more uncertain (M = 5.40, SE = .06) than those in the certainty-prime condition (M = 4.76, SE = .06; F(1, 801) = 53.77, p < .001). All other effects were nonsignificant (Fs < 1).

Choice

A χ² analysis showed that, overall, brand choices differed across conditions (χ² (2) = 36.52; p < .001). Further analysis showed that brand choices among participants primed with certainty were significantly different across the identity conditions (χ² (2) = 21.98; p < .001). Among participants primed with certainty, those in the local-identity condition were more likely to choose the local brand (46%) than the global brand (19.3%; Z = 5.51, p < .001), with 34.8% of participants deferring choice. Participants in the global-identity condition were more likely to choose the global brand (37.7%) than the local brand (27.5%; Z = 2.36, p = .018), with 34.7% of participants deferring choice. These results provide evidence that individuals primed with certainty prefer identity-congruent options.

Brand choices among participants primed with uncertainty were also significantly different across the identity conditions (χ² (2) = 14.30; p < .001). However, as hypothesized, those in the local-identity condition were more likely to choose the global brand (39.6%) than the local brand (29.8%; Z = 2.18, p = .029), with 30.7% of participants deferring their choice. Participants in the global-identity condition were more likely to choose the local brand (45.6%) than the global brand (23.9%; Z = 4.32, p < .001), with 30.6% of participants deferring their choice. These results provide evidence that individuals primed with uncertainty prefer identity-incongruent options. We did not observe a significant difference in choice deferral across conditions (all p-values > .21).³

Study 3

The previous studies showed that uncertainty leads to a preference for brands that are incongruent with consumers’ global–local citizenship identity. We argued that this effect occurs because uncertainty (certainty) activates a divergent (convergent) thinking style, which leads to a preference for identity-inconsistent (consistent) brands. Study 3 aims to provide evidence of this process explanation through a test of mediation. As reviewed earlier, divergent thinking involves employing unfamiliar (i.e., “out-of-the-box”) approaches to solve decision problems. We therefore measured participants’ self-reported motivations to “think out of the box” and to “problem solve” as a proxy for divergent thinking. We predicted that these motivations would mediate the effect of uncertainty on brand preference.

Method

Participants and design

We randomly assigned 240 participants from a large Singapore university (M_age = 21.27 years, 55% female) to either the uncertainty-prime or the certainty-prime condition. The relative strength of participants’ global (vs. local) identity was measured.

Procedure and stimuli

The study ostensibly consisted of several unrelated parts. The first part used an essay-writing task identical to the one used in the earlier studies to prime uncertainty versus certainty. The second part was presented as a market research survey on high-definition televisions (HDTVs). We presented participants with information about two HDTV sets. The information described the specifications of the TVs (i.e., sound and video quality) and their user ratings in a side-by-side, comparative review format (see Web Appendix C). We told participants that the real brand names were disguised for confidentiality reasons and that one of the TVs was a global brand (Brand G) and the other was a local brand (Brand L). As reported in detail in Web Appendix C, a pretest among 60 participants from the same population had shown that when the product features were presented without any (global vs. local) brand information, perceptions and evaluations of the two products were essentially the same. We then asked participants to indicate their relative evaluation of the two TVs on a four-item bipolar scale: which product they liked more, which one they thought was a better choice, which one they felt more favorable toward, and which one they felt more positive toward (1 = “Brand L,” and 7 = “Brand G”). These items were averaged for a composite relative brand evaluation score (α = .95), with higher values indicating a preference for the global brand.

In addition, in this study, two items measured the extent to which participants were motivated to “think out of the box” and to “problem solve” at that moment. Given that divergent thinking is often conceptualized as “out-of-the-box” thinking to problem solve (e.g., Baer 1993), we used these two items to capture our construct of interest. We computed a composite divergent thinking score from the average of these two items (r = .52). Participants’ relative strength of global (vs. local) identity was then measured using items identical to the ones used in Study 1. Lastly, as a manipulation check of uncertainty, we also measured participants’ degree of uncertainty in a manner identical to Study 1 (r = .82).

Results and Discussion

Manipulation check

Compared with participants in the certainty-prime condition (M = 4.16, SE = .13), participants in the uncertainty-prime condition reported feeling more uncertain (M = 5.21, SE = .13; F(1, 238) = 33.48, p < .001). Thus, the priming manipulation of uncertainty was successful.

Brand preference

We conducted a regression analysis with participants’ relative preference for the global brand as the dependent variable and uncertainty (certain = −1, uncertain = 1), identity, and the interaction of the two as independent variables. The main effects of uncertainty (β = −.14, SE = .09; t(236) = −1.57, p = .12) and identity (β = .05, SE = .09; t(236) = .51, p = .61) were not significant. However, importantly, we observed a significant interaction effect (β = −.36, SE = .09; t(236) = −4.17, p < .001). As shown in the figure in Web Appendix C, among participants primed with certainty, a relatively stronger global identity led to a stronger preference for the global brand (M_+SD = 4.31 vs. M_−SD = 3.51; β = .41, SE = .13; t(236) = 3.16, p = .002). In contrast, among participants primed with uncertainty, a relatively stronger global identity led to a stronger preference for the local brand (M_+SD = 3.58 vs. M_−SD = 4.21; β = −.32, SE = .12; t(236) = −2.73, p = .007). These results demonstrate that priming uncertainty (certainty) leads to a preference for identity-inconsistent (identity-consistent) brands, which is a conceptual replication of the results of the previous studies.

Divergent thinking mediation

According to our conceptualization, the identity × uncertainty moderation observed in this study should be mediated by divergent thinking. To test this, we conducted a mediated moderation analysis using Model 14 in the Hayes Process Model, with 5,000 bootstrap iterations. This moderated mediation model produced a regression estimate of the effect of the uncertainty prime on divergent thinking, which was significant such that being primed with uncertainty (vs. certainty) was associated with higher levels of divergent thinking (M_uncertain = 4.03 vs. M_certain = 3.65; β = .19, SE = .08; t(238) = 2.23, p = .027). This result supports our proposition that uncertainty leads to divergent thinking.

Furthermore, the model produced a regression estimate of the identity × divergent-thinking interaction (while controlling for uncertainty, divergent thinking, and identity), which was significant (β = −.18, SE = .06; t(235) = −2.88, p = .004). As expected, among participants who indicated higher divergent thinking motivations (+1SD), stronger global identity was associated with lower preference for the global brand (β = −.29, SE = .14; t(235) = −2.11, p = .036), indicating an identity-inconsistent preference. However, among participants who indicated lower divergent thinking motivations (−1SD), stronger global identity was associated with higher preference for the global brand (β = .19, SE = .11; t(235) = 1.75, p = .081), indicating an identity-consistent preference. These results show that what indeed drives consumers to prefer an identity-inconsistent (consistent) brand is their divergent (convergent) thinking styles. The moderated mediation index was also significant (index = −.034, SE = .02, C.I. [−.088, −.002]), indicating a significant overall moderated mediation.

Study 4

The aim of Study 4 was to provide further support for our process explanation that the effect of uncertainty (certainty) on preference for identity-inconsistent (consistent) brands is because uncertainty (certainty) evokes a divergent (convergent) thinking style. Whereas Study 3 provided support for the thinking style–based process explanation by measuring the mediator, Study 4’s aim was to test the process using a moderation-of-process paradigm (Spencer, Zanna, and Fong 2005). To this end, in this study, in addition to priming participants with uncertainty versus certainty, we also primed them with different thinking styles (divergent thinking vs. convergent thinking vs. control condition). We assessed how the interaction of these two factors influenced the effect of citizenship identity on preference for global versus local brands. Among participants who were not primed with any particular thinking style (i.e., the control condition), we expected to replicate the uncertainty × identity interaction observed in the earlier studies. However, consistent with a moderation-of-process paradigm, we reasoned that if differences in brand preference between the uncertainty and certainty conditions were triggered by differences in thinking style, manipulation of the thinking style itself should override the effects caused by the uncertainty versus certainty prime and produce main effects consistent with the effect that would be expected from the primed thinking style. In other words, regardless of the uncertainty versus certainty prime, among participants primed with a convergent thinking style, we expected a preference for the identity-consistent brand, whereas among those primed with a divergent thinking style, we expected a preference for the identity-inconsistent brand.

Method

Participants and design

We randomly assigned 481 students from a large university in Singapore (M_age = 21.39 years, 60.6% female) to one of the conditions of a 2 (prime: certainty vs. uncertainty) × 3 (thinking style: divergent vs. convergent vs. control) between-subjects study. As in Studies 1 and 3, global and local identity were measured.

Procedure and stimuli

The study ostensibly consisted of several unrelated parts. The first part primed certainty versus uncertainty using a task identical to the ones used in the earlier studies. Participants then moved to the second part, which involved completing a task that has been used widely in previous creativity research to manipulate divergent versus convergent thinking (Friedman and Förster 2001). In this task, participants in the convergent thinking condition were asked to generate and write down as many common uses of a brick as possible, whereas those in the divergent thinking condition were asked to generate as many uncommon uses of a brick as possible. By asking participants to focus on generating uncommon (common) uses for a brick, the task primed participants to conduct a broad (narrow) search within the associative network and to think divergently (convergently). Participants in the study’s control condition skipped this task.

We pretested the effectiveness and discriminant validity of the thinking styles and uncertainty manipulations on a separate group of 297 participants. As reported in Web Appendix D, the pretest demonstrates that (1) our manipulations of uncertainty and thinking style shifted participants’ level of uncertainty and thinking style as intended, (2) the thinking style manipulation did not influence participants’ level of uncertainty, and (3) among participants in the control condition (i.e., when no thinking style was primed), the uncertainty prime led to divergent thinking whereas the certainty prime led to convergent thinking—a pattern of results that provides additional support for our proposition that uncertainty leads to divergent thinking.

After completing the uncertainty and thinking style primes, participants then proceeded to the next part of the study, which, under the guise of a market research survey for headphones, asked them to read a side-by-side, comparative review about two sets of headphones as described in Study 2. The actual brand names were not provided, purportedly to keep the brand name confidential. However, we presented one of the headphones as a global brand and the other as a local brand. To ensure that the information presented about the two brands (i.e., which attributes were assigned to which brand) as well as the order of presentation of the global versus local brands (i.e., which brand was presented on the left vs. the right side of the screen) did not affect our results, we counterbalanced which set of headphones was presented as local and which was presented as global. Neither factor affected our results. After reading the information about the headphones, participants reported their relative preference for them using four items identical to the ones used in Study 3.

Then, as a manipulation check of uncertainty, we asked participants to indicate the extent to which they felt life was uncertain. Finally, we measured the relative strength of global versus local identities using items identical to the ones used in Studies 1 and 3.

Results and Discussion

Preliminary analysis

Participants primed with uncertainty reported feeling more uncertain (M = 4.65, SE = .10) than those primed with certainty (M = 3.52, SE = .10; F(1, 475) = 63.55, p < .001). Thus, our uncertainty manipulation was effective. Importantly, the main effect of thinking style (F(1, 475) = 1.93, p = .146) and the interaction term were not significant (F < 1). Given that we administered the uncertainty manipulation check after participants evaluated the headphones, together with the results of the pretest, these results support our expectation that the uncertainty versus certainty manipulation was not affected by the thinking style manipulation task. Furthermore, the priming manipulations did not influence participants’ rating on the identity scale (p-values > .60).

Brand preference

We ran a full factorial regression model with identity, uncertainty (coded as a binary variable: −1 = certain and 1 = uncertain), and two dummy variables to represent the thinking style conditions (D1: 0 = convergent, 1 = other conditions; D2: 1 = divergent, 0 = other conditions) as the predictor variables and relative brand preference as the dependent variable. Central to testing our predictions were the uncertainty × identity × D1 (β = −4.89, SE = .76, t(469) = −6.39, p < .001) and uncertainty × identity × D2 (β = 3.46, SE = .77, t(469) = 4.48, p < .001) interactions, which were both significant.

To test our predictions, we examined the two-way interaction within each thinking style condition (see Figure 2). In the control condition (i.e., when no specific thinking style was induced), the uncertainty × identity interaction was significant (F(1, 469) = 53.68, p < .001). Mirroring the findings of earlier studies, when certainty was primed, a stronger global identity led to a greater preference for the global brand (M_−SD = 3.06 vs. M_+SD = 4.84; β = 1.83, SE = .42, t(469) = 4.31, p < .001), whereas when uncertainty was primed, a stronger global identity led to a greater preference for the local brand (M_−SD = 5.60 vs. M_+SD = 3.11; β = −2.56, SE = .42, t(469) = −6.06, p < .001).

Figure 2.

Relative preference for global brand as a function of uncertainty prime, identity, and thinking style prime (Study 4).

Contrary to what we observed in the control condition, in the convergent thinking condition, both when certainty was primed (M_−SD = 3.63 vs. M_+SD = 4.66; β = 1.06, SE = .32, t(469) = 3.34, p < .001) and when uncertainty was primed (M_−SD = 3.43 vs. M_+SD = 4.90; β = 1.51, SE = .334, t(469) = 4.46, p < .001), a stronger global identity led to a greater preference for the global brand (uncertainty × identity interaction: F(1, 469) = .95, p = .330). That is, when convergent thinking was primed, participants preferred the identity-consistent brand, regardless of their level of uncertainty.

In the divergent thinking condition, the opposite pattern emerged. Here, both when certainty was primed (M_−SD = 4.71 vs. M_+SD = 4.02; β = −.70, SE = .27, t(469) = −2.58, p < .01) and when uncertainty was primed (M_−SD = 5.18 vs. M_+SD = 3.59; β = −1.63, SE = .41, t(469) = −3.99, p < .001), a stronger global identity led to a lower preference for the global brand (uncertainty × identity interaction: F(1, 469) = 3.53, p = .061). In other words, when divergent thinking was primed, participants preferred the identity-inconsistent brand, regardless of their level of uncertainty.

In summary, the preferences of participants who were primed with divergent (convergent) thinking were parallel to the preferences of participants who did not receive a thinking style prime but were primed with uncertainty (certainty). These findings provide further support for the process explanation that the effect of uncertainty (certainty) on preference for identity-inconsistent (consistent) brands is driven by the divergent (convergent) thinking style it activates.

Study 5

Does uncertainty lead to a preference for identity-consistent brands for all identity types, and how do we reconcile our findings with the results of earlier research showing that, in the religious identity context, uncertainty leads to stronger religious convictions? As reviewed earlier, an important attribute of citizenship identity is that people normally associate themselves with both local and global identities (despite their differences in the relative strength of these identities), and they see these identities as noninterfering. We have argued that our proposed effect of uncertainty on preference for citizenship identity–inconsistent brands is contingent on this attribute of citizenship identity, which may not be present in identities such as religious identity (e.g., being a Christian vs. not being a Christian).

The aim of Study 5 was to provide support for our proposition that identity interference moderates the effect of uncertainty on preference for identity-inconsistent brands. To this end, in this study, in addition to priming participants with uncertainty versus certainty, we manipulated their beliefs about whether global and local identities were interfering or not. Previous research has demonstrated that, over and above identification with different identities, people may form beliefs about the structure/function of such identities (Hong et al. 2009). As conceptualized earlier, we predicted that uncertainty would lead to a preference for identity-inconsistent brands only when global and local identities were perceived as noninterfering.

Method

Participants and design

We randomly assigned a total of 911 participants from the Prolific (UK) online panel (M_age = 36.55 years, 51.3% female) to one of the conditions of a 2 (prime: certainty vs. uncertainty) × 2 (identity type: interfering vs. noninterfering) between-subjects design, in which citizenship identity was additionally measured as a dichotomous variable.

Procedure and stimuli

The study consisted of several ostensibly unrelated parts. The first part was presented as a “citizenship identity survey” and was used to measure participants’ global–local citizenship identity and to manipulate beliefs about the identity being interfering. In this survey, participants first read, “Some people see themselves mostly as global citizens of the world, identify with the broader global population, and feel a sense of interconnectedness with the global world. However, there are other people who see themselves mostly as local citizens of their own communities, identify mainly with the local population they were raised in, and feel a sense of interconnectedness with their local community. We would like to know how you see yourself. Do you see yourself as mostly a global citizen of the world or a local citizen of your own community?” Participants then chose from two options to indicate whether they saw themselves as a “global citizen of the world” or a “local citizen of [their] own community.”⁴ This dichotomous item was used as the measure of citizenship identity in our analysis.

After indicating their identity, we asked participants to read a short paragraph that ostensibly described some recent research on citizenship identity (see Web Appendix E). In reality, we used this paragraph to manipulate beliefs about whether citizenship identities are interfering or noninterfering. We told participants they would be asked a few questions about the paragraph after reading it. Those in the interfering identities condition read, “When a person identifies with a local identity, it typically means that they do not feel a need to connect with the wider global world. Similarly, when a person identifies with the global world, they typically do not feel that it is necessary to connect to their local culture…if you are a local person, that identity is probably ingrained in your psyche. Similarly, if you are a global person, that identity is very much entrenched in your personality. Each identity is unique and well defined, with clear and rigid boundaries.” In contrast, those in the noninterfering identities condition read that “when a person identifies with a local identity, it typically does not mean that they have lost connection with the wider global world. Similarly, when a person identifies with the global world, it typically does not mean that they have lost connection to their local culture…some people are more global, and some people are more local. However, these identity designations are ‘relatively speaking.’ Local and global identities are somewhat fluid, and everyone has a little bit of each identity in them; albeit, depending on the individual’s personality, each person identifies with one identity more strongly.” A pretest among 199 participants showed that this identity type manipulation influenced participants’ beliefs about the compatibility (interference) of global and local identities as intended, without actually influencing the strength of their identification with either identity per se (see Web Appendix E for details), suggesting the independence of the identity-type and identity-interference constructs.

After completing the “citizenship identity survey,” participants moved on to the next survey, which primed uncertainty versus certainty in a manner identical to the previous studies. Participants then proceeded to a subsequent survey, which asked them to read some information about two sets of headphones (one presented as a global brand and the other as a local brand) and to evaluate them. The procedure and the stimuli used were similar to the ones in Study 4. Finally, as a check for demand effects, we asked participants if they could guess the hypothesis being tested in the study.

Results and Discussion

Preliminary analysis

None of the participants were able to correctly guess the hypothesis being tested in the study. A total of 475 participants (52.1%) self-identified with a global identity.

Brand preferences

An ANOVA on brand preferences showed a main effect of identity such that, overall, participants with a global identity (M = 4.03, SD = 1.15) were more likely to prefer the global brand (relative to the local brand) than participants with a local identity (M = 3.75, SD = 1.27; F(1, 903) = 12.54, p < .001, η_p ² = .014). This effect was qualified by an identity × interference interaction (F(1, 903) = 8.20, p = .004, η_p ² = .009) and, central to testing our predictions, an identity × interference × uncertainty interaction (F(1, 903) = 13.86, p < .001, η_p ² = .015). All other effects were nonsignificant (p > .23; see Figure 3).

Figure 3.

Relative preference for global brand as a function of uncertainty prime, identity, and identity type (Study 5).

Follow-up analysis revealed a significant identity × uncertainty interaction among participants who were primed to believe the identities were noninterfering (F(1, 903) = 11.83, p < .001, η_p ² = .013). Here, among participants who were primed with uncertainty, relative preference for the global brand was lower if they had a global identity (M = 3.71, SD = 1.05) than if they had a local identity (M = 4.04, SD = 1.24; F(1, 903) = 4.33, p = .037, η_p ² = .013). In comparison, among participants who were primed with certainty, relative preference for the global brand was higher if they had a global identity (M = 4.19, SD = 1.18) than if they had a local identity (M = 3.75, SD = 1.21; F(1, 903) = 7.78, p = .005, η_p ² = .008). These results replicate the results of our previous studies and show that when consumers consider global and local identities to be noninterfering, uncertainty (certainty) leads to a preference for identity-inconsistent (consistent) brands.

Among participants who were primed to believe global and local identities are interfering, we observed a marginally significant identity × uncertainty interaction (F(1, 903) = 3.28, p = .07, η_p ² = .004) as well. However, here, both participants who were primed with certainty (M_{global_identity} = 4.03, SD = 1.26 vs. M_{local_identity} = 3.73, SD = 1.27; F(1, 903) = 3.89, p = .048, η_p ² = .004) and those who were primed with uncertainty (M_{global_identity} = 4.20, SD = 1.02 vs. M_{local_identity} = 3.49, SD = 1.32; F(1, 903) = 19.96, p < .001; η_p ² = .021) demonstrated a relative preference for the identity-consistent brand. Thus, the interaction pattern indicates a stronger preference for the identity-consistent brand among participants primed with uncertainty, which is consistent with what one would predict given the previous religious identity research.

General Discussion

Summary of Findings

Uncertainty and globalization are both unavoidable parts of our lives. How does uncertainty in our environment interact with the rise in globalization to affect our brand choices? We report the results of five studies demonstrating that under states of certainty, consumers are more likely to prefer a global (local) brand if they have a relatively stronger global (local) identity, whereas under uncertainty, consumers are more likely to prefer a global (local) brand if they have a relatively stronger local (global) identity. In other words, in the citizenship-identity context, we observe that uncertainty (certainty) leads to a preference for identity-inconsistent (consistent) brands. We obtained these effects across various product contexts, with both measured (Studies 1 and 3–5) and manipulated identity (Study 2), and even when we gave participants an option to defer their choice (Study 2).

We further show that a preference for identity-incongruent (congruent) brands under uncertainty (certainty) is a consequence of uncertainty (certainty) activating a divergent (convergent) thinking style. Study 3 showed that participants’ self-reported divergent thinking style following uncertainty (vs. certainty) mediated the effect of uncertainty on preference for identity-inconsistent (vs. consistent) brands. Study 4 provided additional evidence of the process through moderation by showing that when divergent or convergent thinking styles were primed, the impact of uncertainty (vs. certainty) on the preference for identity-incongruent brands was eliminated. Testing the boundaries of the effect, Study 5 further showed that the preference for identity-inconsistent brands under uncertainty did not hold in contexts in which global and local identities were believed to be interfering (i.e., in conflict) with one another.

Alternative Explanations

Although our studies have provided strong process evidence for the hypothesized effect, a number of potential alternative explanations for these effects must be discussed. First, given that uncertainty is often considered a feeling state, could priming of uncertainty have lowered participants’ mood, leading to the observed results? Although uncertainty has been considered a feeling state in some previous research (e.g., Bar-Anan, Wilson, and Gilbert 2009; Clore and Parrott 1994; Faraji-Rad and Pham 2017; Faraji-Rad, Samuelsen, and Warlop 2015), uncertainty and mood are conceptually distinct (Tiedens and Linton 2001). Feelings have multiple dimensions beyond mere positive versus negative valence (see Lerner, Han, and Keltner 2007 for a review). Still, even if one assumes that the uncertainty primes in our studies lowered consumers’ mood, it is unclear how a more negative mood would evoke a preference for identity-inconsistent brands. In reality, prior research would support the opposite effect—that is, a more positive (and not negative) mood would favor more flexible and liberal approaches to decision making (e.g., Fredrickson 2004). Thus, a mood explanation would not account for our results.

Second, could our findings have been driven by a misattribution of the feeling of uncertainty to either/both brands, causing participants to “question” either/both brands? A misattribution explanation fails to explain our pattern of results either when one assumes uncertainty caused people to question just one of the brands, or when one assumes uncertainty caused people to question both brands. If one assumes uncertainty caused participants to question both brands (i.e., uncertainty was attributed to both options), then we should have observed a reduction in preference for both the identity-consistent and identity-inconsistent brand (a main effect on preferences), not an interaction effect. If one assumes uncertainty caused participants to question only one of the options (i.e., uncertainty was attributed to either the identity-inconsistent brand or the identity-consistent brand only), then we should have observed only a reduction in the strength of the preference for the option to which uncertainty had been attributed and not an increase in preference for the other option. In our studies, however, consistent with our thinking style–based explanation, we observed that uncertainty increased preference for identity-inconsistent brands.

Theoretical Contributions

The results of our studies contribute to three streams of research. First, they contribute to the literature on identity and the self. As reviewed earlier, whereas research is rife with evidence that consumers prefer brands congruent with their identity (e.g., Sirgy 1982; Zhang and Khare 2009), our research adds to the few recently emerging studies showing that the preference for congruity does not always prevail—people sometimes prefer brands that are incongruent with their dominant identities. Furthermore, by demonstrating the role of identity interference on the effect of uncertainty on preference for identity-(in)consistent brands, our findings show that research on identity may benefit from exploring not only the effects caused by the content of consumers’ identities but also consumers’ beliefs and lay theories about the structure and relationship between those identities, such as whether the identities interfere. To our knowledge, although the interference of identities has been explored in social psychological research (e.g., Settles 2004), research on consumer psychology has not yet investigated this important construct.

Second, the results also contribute to the literatures on globalization and branding (Alden, Steenkamp, and Batra 1999, 2006; Batra et al. 2000; Steenkamp, Batra, and Alden 2003; Strizhakova, Coulter, and Price 2008). Because globalization has made global brands more available in various markets around the world, thereby increasing competition for local brands, more research is needed on the processes underlying how consumers make these brand choices and how markets and policy makers can influence these choices. While prior research has shown consumers make these choices in an identity-congruent manner (Zhang and Khare 2009), our research contributes by showing that the preference for identity-congruent brands does not hold under all circumstances. Specifically, we identify an important and managerially relevant factor—incidental uncertainty—that moderates the effect. Our findings thus provide a more nuanced understanding of how a combination of uncertainty and identity salience may influence consumers’ choices between global and local brands, and how citizenship identity effects are different from effects expected from other types of identities (e.g., religious identity). We also contribute to research on globalization more generally. Although previous research on globalization has theorized that global and local identities coexist and are noninterfering (e.g., Arnett 2002), the implications of this coexistence have not yet been studied, and our results show that these implications have important downstream consequences.

Third, our results contribute to the research on the decision-making consequences of incidental uncertainty, which has not received much attention in the previous literature, especially in consumer research (except for a handful of studies such as Faraji-Rad and Pham 2017 and Lee and Qiu 2009). Our research is the first to provide evidence that uncertainty (vs. certainty) leads to a divergent thinking style, adding to other prior findings that uncertainty favors decision processes geared toward expanding the pool of knowledge available to decision makers (Tiedens and Linton 2001). Although prior research has shown that a threatening situation—such as being physically confined or lacking resources—may increase a desire to pursue variety (Levav and Zhu 2009) and may make people more creative (Moreau and Dahl 2005; also see Mehta and Zhu 2015), it has not examined how incidental uncertainty may lead to a divergent or convergent thinking style.

Managerial and Substantive Implications

Our research presents some interesting insights to marketers. Uncertainty is an inescapable part of human life. At the time this article was being written, such inescapability was being highlighted by the COVID-19 pandemic, making our findings especially relevant to any marketer operating in such an environment. In many countries and societies, a sense of uncertainty is further coupled with an intensified debate over globalization. The past few decades have clearly been ones in which global commerce, culture, and identity seemed to be in a smooth, unstoppable upswing. Now, however, debate over the issue of globalization in countries such as the United States, the UK, France, Turkey, China, Australia, and many others has been increasing. Such debates will only strengthen the question and salience of global versus local identities in consumers’ minds. Marketers of both global and local brands thus now have to navigate far more turbulent waters, and our findings should make marketers reconsider seemingly obvious strategies (such as emphasizing a brand’s global meaning to cosmopolitan consumers and local origins to more ethnocentric or patriotic consumers). Our findings suggest that today’s heightened uncertainty may actually make the opposite strategies more fruitful.

The implications of our findings are not limited to the domain of global and local identities. Although we examined the effect in the context of global and local identities, our findings will also allow marketers to better understand how identities in general—interfering or noninterfering—may influence consumer choices. As mentioned earlier, every individual carries with them multiple identities. Beyond global and local identities, we also define ourselves in terms of our gender, occupation, family situation, and other attributes. The extent to which target consumers see identities (e.g., female and successful professional) as interfering (or noninterfering) may influence how they respond to a marketing pitch that aligns with one of these identities under conditions of certainty and uncertainty.

Future Research

In this research, we tested our proposed effect in multiple product categories (e.g., headphones, HDTV). However, a more systematic study of whether the product category would moderate our findings is a fruitful avenue for further research. An interesting question to ask is whether the effect is stronger or weaker in product categories in which consumer preferences are generally more “localized,” highly influenced by a consumers’ upbringing and cultural habits, or based primarily on emotional factors (e.g., for foods; Özsomer 2012). For example, would the emotional comfort received from the purchase of some product categories interact with our effects to shape brand preferences?

In addition, in our lab studies, we manipulated uncertainty using priming tasks. However, the uncertainties consumers experience in real life vary both in terms of the degree to which lack of information exists (i.e., the degree of uncertainty), as well as how consequential the uncertain outcome is to the individual (i.e., outcome importance). Future research would benefit from exploring whether either of these factors influence the effects observed in this article. Relatedly, research would benefit from a more systematic investigation of the level of uncertainty needed for the emergence of the effects observed in this research. For example, could uncertainty have a nonlinear effect on divergent thinking such that it increases divergent thinking at moderate levels but not at high levels?

Future research could also probe more deeply into the relationships between our dependent variable (preference for identity-consistent vs. identity-inconsistent options) and other variables such as a preference for nondefault, novel, and unfamiliar options for which divergent thinking may have direct implications. Although it needs to be empirically tested, we speculate that a divergent thinking style itself is likely to increase consumers’ preference for more novel, unfamiliar, or nondefault options more broadly. However, uncertainty may produce other motivations in parallel with divergent thinking that may compete with divergent thinking motivation. It is possible, therefore, that similarly to the way in which identity interference moderates the effect of uncertainty on preference for identity-(in)consistent brands in our context, other moderators specific to those dependent variables may influence the effect of uncertainty on those variables (see Min and Schwarz 2017).

Supplemental Material

Supplemental Material, sj-pdf-1-mrj-10.1177_0022243720972956 - Uncertainty Evokes Consumers’ Preference for Brands Incongruent with their Global–Local Citizenship Identity

Supplemental Material, sj-pdf-1-mrj-10.1177_0022243720972956 for Uncertainty Evokes Consumers’ Preference for Brands Incongruent with their Global–Local Citizenship Identity by Sharon Ng, Ali Faraji-Rad and Rajeev Batra in Journal of Marketing Research

Footnotes

Associate Editor

Rui (Juliet) Zhu

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by the Ministry of Education, Singapore, under its MOE AcRF Tier 1 (RG5/11).

Notes

Online supplement:

References

Aaker

Jennifer L.

(2000), “Accessibility or Diagnosticity? Disentangling the Influence of Culture on Persuasion Processes and Attitudes,” Journal of Consumer Research, 26 (4), 340–57.

Aaker

Jennifer L.

Lee

Angela Y.

(2001), “I Seek Pleasures and We Avoid Pains: The Role of Self-Regulatory Goals in Information Processing and Persuasion,” Journal of Consumer Research, 28 (1), 33–49.

Aaker

Jennifer L.

Williams

Patti

(1998), “Empathy Versus Pride: The Influence of Emotional Appeals Across Cultures,” Journal of Consumer Research, 25 (3), 241–61.

Adaval

Rashmi

(2001), “Sometimes It Just Feels Right: The Differential Weighting of Affect-Consistent and Affect-Inconsistent Product Information,” Journal of Consumer Research, 28 (1), 1–17.

Alden

Dana L.

Steenkamp

Jan-Benedict E. M.

Batra

Rajeev

(1999), “Brand Positioning Through Advertising in Asia, North America, and Europe: The Role of Global Consumer Culture,” Journal of Marketing, 63 (1), 75–87.

Alden

Dana L.

Steenkamp

Jan-Benedict E. M.

Batra

Rajeev

(2006), “Consumer Attitudes Toward Marketplace Globalization: Structure, Antecedents and Consequences,” International Journal of Research in Marketing, 23 (3), 227–39.

Arnett

Jeffrey J.

(2002), “The Psychology of Globalization,” American Psychologist, 57 (10), 774–83.

Aryee

Samuel

(1992), “Antecedents and Outcomes of Work–Family Conflict Among Married Professional Women: Evidence from Singapore,” Human Relations, 45 (8), 813–37.

Baer

John

(1993), Creativity and Divergent Thinking: A Task-Specific Approach. New York: Psychology Press.

10.

Bagger

Jessica

Andrew

Gutek

Barbara A.

(2008), “How Much Do You Value Your Family and Does It Matter? The Joint Effects of Family Identity Salience, Family Interference with Work, and Gender,” Human Relations, 61 (2), 187–211.

11.

Bar-Anan

Yoav

Wilson

Timothy D.

Gilbert

Daniel T.

(2009), “The Feeling of Uncertainty Intensifies Affective Reactions,” Emotion, 9 (1), 123–27.

12.

Batra

Rajeev

Ramaswamy

Venkatram

Alden

Dana L.

Steenkamp

Jan-Benedict E. M.

Ramachander

(2000), “Effects of Brand Local and Nonlocal Origin on Consumer Attitudes in Developing Countries,” Journal of Consumer Psychology, 9 (2), 83–95.

13.

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.

14.

Burroughs

James E.

Mick

David G.

(2004), “Exploring Antecedents and Consequences of Consumer Creativity in a Problem-Solving Context,” Journal of Consumer Research, 31 (2), 402–11.

15.

Clarkson

Joshua J.

Tormala

Zakary L.

DeSensi

Victoria L.

Christian Wheeler

(2009), “Does Attitude Certainty Beget Self-Certainty?” Journal of Experimental Social Psychology, 45 (2), 436–39.

16.

Clore

Gerald L.

Parrott

W. Gerrod

(1994), “Cognitive Feelings and Metacognitive Judgments,” European Journal of Social Psychology, 24 (1), 101–15.

17.

Cropley

Arthur

(2006), “In Praise of Convergent Thinking,” Creativity Research Journal, 18 (3), 391–404.

18.

Edwards

Ward

(1954), “The Theory of Decision Making,” Psychological Bulletin, 51 (4), 380–417.

19.

Ernst Kossek

, Ellen

Ozeki

Cynthia

(1998), “Work–Family Conflict, Policies, and the Job–Life Satisfaction Relationship: A Review and Directions for Organizational Behavior–Human Resources Research,” Journal of Applied Psychology, 83 (2), 139–49.

20.

Faraji-Rad

Ali

Pham

Michel Tuan

(2017), “Uncertainty Increases the Reliance on Affect in Decisions,” Journal of Consumer Research, 44 (1), 1–21.

21.

Faraji-Rad

Ali

Samuelsen

Bendik Meling

Warlop

Luk

(2015), “On the Persuasiveness of Similar Others: The Role of Mentalizing and the Feeling of Certainty,” Journal of Consumer Research, 42 (3), 458–71.

22.

Fredrickson

Barbara L.

(2004), “The Broaden-and-Build Theory of Positive Emotions,” Philosophical Transactions of the Royal Society B: Biological Sciences, 359 (1449), 1367–77.

23.

Friedman

Ronald S.

Förster

Jens

(2001), “The Effects of Promotion and Prevention Cues on Creativity,” Journal of Personality and Social Psychology, 81 (6), 1001–13.

24.

Friedman

Ronald S.

Förster

Jens

(2002), “The Influence of Approach and Avoidance Motor Actions on Creative Cognition,” Journal of Experimental Social Psychology, 38 (1), 41–55.

25.

Greenberg

Jeff

Pyszczynski

Tom

Solomon

Sheldon

Rosenblatt

Abram

Veeder

Mitchell

Kirkland

Shari

, et al. (1991), “Evidence for Terror Management Theory II: The Effects of Mortality Salience on Reactions to Those Who Threaten or Bolster the Cultural Worldview,” Journal of Personality and Social Psychology, 58 (2), 308–18.

26.

Grimley

Naomi

(2016), “Identity 2016: ‘Global Citizenship’ Rising, Poll Suggests,” BBC News (April 28), http://www.bbc.com/news/world-36139904.

27.

Yangjie

Botti

Simona

Faro

David

(2013), “Turning the Page: The Impact of Choice Closure on Satisfaction,” Journal of Consumer Research, 40 (2), 268–83.

28.

Guilford

Joy P.

(1967), The Nature of Human Intelligence. New York: McGraw-Hill.

29.

Guilford

Joy P.

(1968), Intelligence, Creativity, and Their Educational Implications. San Diego, CA: Knapp.

30.

Han

Sang Pil

Shavitt

Sharon

(1994), “Persuasion and Culture: Advertising Appeals in Individualistic and Collectivistic Societies,” Journal of Experimental Social Psychology, 30 (4), 326–50.

31.

Higgins

E. Tory

(2000), “Making a Good Decision: Value from Fit,” American Psychologist, 55 (11), 1217–30.

32.

Hogg

Michael A.

(2007), “Uncertainty–Identity Theory,” Advances in Experimental Social Psychology, 39, 69–126.

33.

Hogg

Michael A.

Adelman

Janice R.

Blagg

Robert D.

(2010), “Religion in the Face of Uncertainty: An Uncertainty–Identity Theory Account of Religiousness,” Personality and Social Psychology Review, 14 (1), 72–83.

34.

Hong

Ying-Yi

Morris

Michael W.

Chiu

Chi-Yue

Benet-Martinez

Veronica

(2000), “Multicultural Minds: A Dynamic Constructivist Approach to Culture and Cognition,” American Psychologist, 55 (7), 709–20.

35.

Kagan

Jerome

(1972), “Motives and Development,” Journal of Personality and Social Psychology, 22 (1), 51–66.

36.

Kahneman

Daniel

Slovic

Paul

Tversky

Amos

(1982), Judgment Under Uncertainty: Heuristics and Biases. New York: Cambridge University Press.

37.

Kay

Aaron C.

Moscovitch

David A.

Laurin

Kristin

(2010), “Randomness, Attributions of Arousal, and Belief in God,” Psychological Science, 21 (2), 216–18.

38.

Knight

Frank H.

(1921), Risk, Uncertainty, and Profit. Boston: Houghton Mifflin.

39.

Kopelman

Richard E.

Greenhaus

Jeffrey H.

Connolly

Thomas F.

(1983), “A Model of Work, Family, and Inter-Role Conflict: A Construct Validation Study,” Organizational Behavior and Human Performance, 32 (2), 198–215.

40.

Laurin

Kristin

Kay

Aaron C.

Moscovitch

David A.

(2008), “On the Belief in God: Towards an Understanding of the Emotional Substrates of Compensatory Control,” Journal of Experimental Social Psychology, 44 (6), 1559–62.

41.

Lee

Yih Hwai

Qiu

Cheng

(2009), “When Uncertainty Brings Pleasure: The Role of Prospect Imageability and Mental Imagery,” Journal of Consumer Research, 36 (4), 624–33.

42.

Leotti

Lauren A.

Iyengar

Sheena S.

Ochsner

Kevin N.

(2010), “Born to Choose: The Origins and Value of the Need for Control,” Trends in Cognitive Sciences, 14 (10), 457–63.

43.

Lerner

Jennifer S.

Han

Seunghee

Keltner

Dacher

(2007), “Feelings and Consumer Decision Making: Extending the Appraisal-Tendency Framework,” Journal of Consumer Psychology, 17 (3), 181–87.

44.

Levav

Jonathan

Zhu

Rui

(2009), “Seeking Freedom Through Variety,” Journal of Consumer Research, 36 (4), 600–610.

45.

Liberman

Nira

Trope

Yaacov

(1998), “The Role of Feasibility and Desirability Considerations in Near and Distant Future Decisions: A Test of Temporal Construal Theory,” Journal of Personality and Social Psychology, 75 (1), 5–18.

46.

Loewenstein

George

(1994), “The Psychology of Curiosity: A Review and Reinterpretation,” Psychological Bulletin, 116 (1), 75–98.

47.

McGregor

Ian

Zanna

Mark P.

Holmes

John G.

Spencer

Steven J.

(2001), “Compensatory Conviction in the Face of Personal Uncertainty: Going to Extremes and Being Oneself,” Journal of Personality and Social Psychology, 80 (3), 472–88.

48.

Mednick

Sarnoff

(1962), “The Associative Basis of the Creative Process,” Psychological Review, 69 (3), 220–32.

49.

Mehta

Ravi

Zhu

Meng

(2015), “Creating When You Have Less: The Impact of Resource Scarcity on Product Use Creativity,” Journal of Consumer Research, 42 (5), 767–82.

50.

Min

Bora

Schwarz

Nobert

(2017), “Novelty as Risk and Opportunity: Opposite Effects of Low Personal Control and Perceived Unpredictability on Novelty Seeking,” Advances in Consumer Research, 45, 152–57.

51.

Moreau

C. Page

Dahl

Darren W.

(2005), “Designing the Solution: The Impact of Constraints on Consumers’ Creativity,” Journal of Consumer Research, 32 (1), 13–22.

52.

Netemeyer

Richard G.

Boles

James S.

McMurrian

Robert

(1996), “Development and Validation of Work–Family Conflict and Family–Work Conflict Scales,” Journal of Applied Psychology, 81 (4), 400–410.

53.

Oyserman

Daphna

(2009), “Identity-Based Motivation and Consumer Behavior,” Journal of Consumer Psychology, 19 (3), 276–79.

54.

Özsomer

Aysegul

(2012), “The Interplay Between Global and Local Brands: A Closer Look at Perceived Brand Globalness and Local Iconness,” Journal of International Marketing, 20 (2), 72–95.

55.

Robertson

Roland

(1995), Globalization: Social Theory and Global Culture. London: SAGE.

56.

Rosenblatt

Abram

Greenberg

Jeff

Solomon

Sheldon

Pyszczynski

Tom

Lyon

Deborah

(1989), “Evidence for Terror Management Theory: I. The Effects of Mortality Salience on Reactions to Those Who Violate or Uphold Cultural Values,” Journal of Personality and Social Psychology, 57 (4), 681–90.

57.

Sellers

Robert M.

Caldwell

Cleopatra H.

Schmeelk-Cone

Karen H.

Zimmerman

Marc A.

(2003), “Racial Identity, Racial Discrimination, Perceived Stress, and Psychological Distress Among African American Young Adults,” Journal of Health and Social Behavior, 44 (3) 302–17.

58.

Sellers

Robert M.

Rowley

Stephanie A. J.

Chavous

Tabbye M.

Nicole Shelton

Smith

Mia A.

(1997), “Multidimensional Inventory of Black Identity: A Preliminary Investigation of Reliability and Construct Validity,” Journal of Personality and Social Psychology, 73 (4), 805–15.

59.

Settles

Isis H.

(2004), “When Multiple Identities Interfere: The Role of Identity Centrality,” Personality and Social Psychology Bulletin, 30 (4), 487–500.

60.

Seward

John P.

(1956), “Drive, Incentive, and Reinforcement,” Psychological Review, 63 (3), 195–203.

61.

Sirgy

M. Joseph

(1982), “Self-Concept in Consumer Behavior: A Critical Review,” Journal of Consumer Research, 9 (3), 287–300.

62.

Spencer

Steven. J.

Zanna

Mark P.

Fong

Geoffrey. T.

(2005), “Establishing a Causal Chain: Why Experiments Are Often More Effective Than Mediational Analyses in Examining Psychological Processes,” Journal of Personality and Social Psychology, 89 (6), 845–51.

63.

Stayman

Douglas M.

Deshpande

Rohit

(1989), “Situational Ethnicity and Consumer Behavior,” Journal of Consumer Research, 16 (3), 361–71.

64.

Steenkamp

Jan-Benedict E. M.

Jong

Martin G. de

(2010), “A Global Investigation into the Constellation of Consumer Attitudes Toward Global and Local Products,” Journal of Marketing, 74 (6), 18–40.

65.

Steenkamp

Jan-Benedict E. M.

Batra

Rajeev

Alden

Dana L.

(2003), “How Perceived Brand Globalness Creates Brand Value,” Journal of International Business Studies, 34 (1), 53–65.

66.

Strizhakova

Yulia

Robin

A. Coulter

(2015), “Drivers of Local Relative to Global Brand Purchases: A Contingency Approach,” Journal of International Marketing, 23 (1), 1–22.

67.

Strizhakova

Yulia

Robin

A. Coulter

(2019), “Consumer Cultural Identity: Local and Global Cultural Identities and Measurement Implications,” International Marketing Review, 36 (5), 610–27.

68.

Strizhakova

Yulia

Coulter

Robin A.

Price

Linda L.

(2008), “Branded Products as a Passport to Global Citizenship: Perspectives from Developed and Developing Countries,” Journal of International Marketing, 16 (4), 57–85.

69.

Strizhakova

Yulia

Coulter

Robin A.

Price

Linda. L.

(2011), “Branding in a Global Marketplace: The Mediating Effects of Quality and Self-identity Brand Signals,” International Journal of Research in Marketing, 28 (4), 342–51.

70.

Strizhakova

Yulia

A. Coulter

Robin

Price

Linda L.

(2012), “The Young Adult Cohort in Emerging Markets: Assessing Their Glocal Cultural Identity in a Global Marketplace,” International Journal of Research in Marketing, 29 (1), 43–54.

71.

Tiedens

Larissa Z.

Linton

Susan

(2001), “Judgment Under Emotional Certainty and Uncertainty: The Effects of Specific Emotions on Information Processing,” Journal of Personality and Social Psychology, 81 (6), 973–88.

72.

Lingjiang

Khare

Adwait

Zhang

Yinlong

(2012), “A Short 8-item Scale for Measuring Consumers’ Local–Global Identity,” International Journal of Research in Marketing, 29 (1), 35–42.

73.

Van den Bos

Kees

(2009), “Making Sense of Life: The Existential Self Trying to Deal with Personal Uncertainty,” Psychological Inquiry, 20 (4), 197–217.

74.

Vaughn

Leigh Ann

Weary

Gifford

(2003), “Causal Uncertainty and Correction of Judgments,” Journal of Experimental Social Psychology, 39 (5), 516–24.

75.

Yang

Xiaojing

Ringberg

Torsten

Mao

Huifang

Peracchio

Laura A.

(2011), “The Construal (In)compatibility Effect: The Moderating Role of a Creative Mind-Set,” Journal of Consumer Research, 38 (4), 681–96.

76.

Zhang

Yinlong

Khare

Adwait

(2009), “The Impact of Accessible Identities on the Evaluation of Global Versus Local Products,” Journal of Consumer Research, 36 (3), 524–37.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.38 MB