The Cost of Being “True to Yourself” for Mixed Selves: Frame Switching Leads to Perceived Inauthenticity and Downstream Social Consequences for Biculturals

Abstract

A growing population of biculturals—who identify with at least two cultures—often frame switch, adapting their behavior to their shifting cultural contexts. We demonstrate that frame switching biculturals are perceived as inauthentic by majority Americans and consequently seen as less likable, trustworthy, warm, and competent compared to biculturals who do not frame switch or a neutral control (Studies 1–3, N = 763). In Study 2, describing the bicultural’s behavior as authentic despite its inconsistency partly alleviated the negative effects of frame switching. In our preregistered Study 3, majority American women were less romantically interested in and less willing to date a bicultural who frame switched in his dating profiles (mediated by inauthenticity). The way biculturals negotiate their cultures can have social costs and create a barrier to intercultural relations.

Keywords

frame switching authenticity bicultural multicultural intercultural relations intergroup relations

Do I contradict myself? Very well then I contradict myself, (I am large, I contain multitudes.)

—Walt Whitman, “Song of Myself”

As diversity increases in many nations, including the United States (Horowitz, 2019), so too has the population of biculturals—people who belong to at least two cultural groups. Biculturals themselves are also diverse and can include immigrants and their progeny, biracials, and people who are immersed in multiple cultures. Being bicultural can be challenging—not only must biculturals negotiate different cultural norms but they also face misunderstandings and discrimination from others. Mainstream Americans may be suspicious of biculturals’ dual cultural identification (Kunst et al., 2018) and assume biculturals are confused about their identity and are untrustworthy (Albuja et al., 2018). Yet in addition to biases against biculturals based on who they are, another source of bias may come from what they do. We posit that biculturals’ behavior as they negotiate their cultures can have powerful effects on the way others perceive them (West et al., 2017, 2018).

Here, we focus on the bicultural phenomenon of frame switching or adapting oneself in response to the immediate cultural context (Hong & Khei, 2014). This process can occur consciously or unconsciously (Doucerain et al., 2013; Mok & Morris, 2013) and involves shifting between culturally normative styles of cognition, emotion, and behavior (e.g., Perunovic et al., 2007). Frame switching enables biculturals to gain acceptance and maintain relationships within each of their cultural groups, fostering their well-being (LaFromboise et al., 1993). Although frame switching has a clear function for biculturals, its potential consequences are not well-understood. Does frame switching come with social costs for biculturals, even as they strive to be true to themselves?

Inconsistency Signals Inauthenticity

Western cultures emphasize the individual as an autonomous agent, ideally uninfluenced by external forces (Nisbett et al., 2001). Behaviors ought to reflect one’s singular, true self and not change across situations (Cross et al., 2003). People who behave inconsistently are seen as inauthentic (Kashima et al., 2004), and authenticity is upheld as a virtue (Kernis & Goldman, 2006). This is problematic for biculturals because frame switching requires changing the way they behave according to the cultural context, and this inconsistency might undermine their perceived authenticity and have downstream social costs.

Social Costs of Inauthenticity

One reason biculturals frame switch is to gain acceptance by being mindful of each cultures’ norms and values. Paradoxically, frame switching may undercut biculturals’ acceptance in Western society because switching between cultural frames violates the dominant culture’s expectation of behavioral consistency (English & Chen, 2011). The social consequences of frame switching may be far from trivial as inauthenticity comes with many costs.

At a person-perception level, frame switching may damage general impressions of biculturals as a fallout of being seen as inauthentic. Extant research with majority Americans shows that perceived authenticity strongly relates to impressions of likeability and trustworthiness (Krumhuber et al., 2007). Further, authenticity is related to perceptions of warmth and competence (West et al., 2018), which are considered universal dimensions in impression formation (Cuddy et al., 2008). Thus, we hypothesize that a bicultural’s frame switching will undermine their perceived authenticity, and subsequently, their likeability, trustworthiness, warmth, and competence.

Frame switching may have additional, powerful consequences for biculturals, particularly in romantic relationships. In Western societies, feeling and being perceived as authentic is fundamental to forming and maintaining romantic relationships (Josephs et al., 2019), and perceived inauthenticity can diminish relationship satisfaction, commitment, and support (Lopez & Rice, 2006; Wickham, 2013). Our final study examines the consequences of frame switching on biculturals’ online dating prospects, an impactful real-world context in which concerns about authenticity are heightened (Toma et al., 2008).

In the current research, we test our key prediction that frame switching undermines a bicultural’s perceived authenticity, subsequently damaging general impressions and their romantic relationship prospects in America. All studies’ materials, data, syntax, and the preregistration for Study 3 are available on Open Science Framework (osf.io/4397c/); pretests, power analyses, additional and meta-analyzed results are also in the Online Supplementary Materials (OSMs).

Study 1

We hypothesize that participants will see a bicultural as less authentic if he frame switches than if he does not and that this reduction in perceived authenticity will have downstream consequences such that the bicultural will be seen as less likeable, trustworthy, warm, and competent.

Method

Participants

Majority Americans (N = 150) participated online via Prolific. Power analyses (α = .05) based on the effect size of frame switching (vs. no switching) on authenticity obtained in a pilot study (d = 2.04; West et al., 2018) indicated 99.9% power with N = 150. To be eligible, participants had to be White, U.S. citizens, born and residing in the United States, English as first language, and had parents born in the United States, Canada, or Western Europe excluding Southern Europe (Lalonde et al., 2013; n = 9 excluded). We excluded participants who failed more than one of four attention checks (recall the bicultural’s name and cultures, n = 8) or indicated that they did not complete the study honestly and attentively (self-report item, n = 0). Final sample N = 133 (57 females, M _age = 34.38, SD _age = 13.46).

Procedure

Following informed consent, participants were randomly assigned to read one of three vignettes: (1) Switching (n = 44), the bicultural’s behavior differed depending on which cultural group he is with; (2) No Switching (n = 46), the bicultural’s behavior was the same regardless of which cultural group he was with; or (3) Neutral (n = 43), only background information and none on how a bicultural behaved with his cultural groups. After reading the vignette and answering attention checks, participants reported their impressions of the bicultural’s authenticity and provided their impressions of their likeability, trustworthiness, warmth, and competence. Finally, participants completed demographics before debriefing.

Materials

Bicultural vignettes

Participants read vignettes featuring Miguel Wong, a U.S.-born Mexican Chinese bicultural American (West et al., 2018). We selected two minority cultures as the focus of switching to avoid any confounding effects of in-group/out-group biases (e.g., concerns about disloyalty). Both cultures represented minority out-groups for participants, which isolates the effects of frame switching from group biases that may occur if the bicultural was switching between his majority Americans and a minority culture. Vignettes began with the same description of Miguel as an American graduate student who identifies equally with his father’s Chinese culture and his mother’s Mexican culture. The next part of the vignette differed by condition.

The Switching condition described, “Miguel behaves differently depending on which cultural group he is with, so his behavior is more typically Chinese when he is with Chinese people, and more typically Mexican when he is with Mexicans” and then provided examples of how his behavior changes with each culture.

The No Switching condition described, “Miguel doesn’t tend to behave any differently depending on which cultural group he is with, so his behavior is largely the same regardless of whether he is with Chinese people or Mexicans” and provided examples of how he behaves with each culture.

The Neutral condition vignette did not provide any additional information.

Pretesting ensured that the descriptions of Miguel’s specific behaviors did not differ in desirability by condition.

Authenticity

We adapted a four-item measure of subjective authenticity (English & Chen, 2011; α = .94) to assess a target’s perceived authenticity rather than one’s own authenticity, for example, “Miguel is being himself with others” (1 = strongly disagree to 7 = strongly agree).

General impressions

Likeability

Participants responded to nine items gauging how likeable they found the bicultural (Cila & Lalonde, 2019; α = .88), for example, “Miguel seems like a really nice guy” (1 = strongly disagree to 7 = strongly agree).

Trustworthiness

A single item asked, “Overall, I think Miguel is a trustworthy person” (1 = strongly disagree to 7 = strongly agree).

Warmth and competence

Participants also rated two fundamental trait dimensions: warmth (six items; α = .87) and competence (seven items; α = .84; Cuddy et al., 2007) on 5-point scales (1 = not at all to 5 = extremely).

Results

See Table 1 for descriptive statistics.

Table 1.

Descriptive Statistics for Study 1.

Outcome Variable	Switching M [95% CI]	No Switching M [95% CI]	Neutral M [95% CI]
Authenticity	4.30 [4.06, 4.54]	6.40 [6.17, 6.64]	5.87 [5.62, 6.11]
Likeability	5.23 [4.98, 5.47]	5.58 [5.34, 5.81]	5.71 [5.46, 5.95]
Trustworthiness	5.02 [4.70, 5.35]	5.76 [5.44, 6.08]	5.49 [5.16, 5.82]
Warmth	3.81 [3.65, 3.98]	4.12 [4.00, 4.31]	4.10 [3.94, 4.27]
Competence	3.71 [3.55, 3.86]	4.10 [3.95, 4.26]	4.01 [3.85, 4.17]

A one-way analysis of variance (ANOVA) revealed that authenticity ratings differed significantly across conditions, F(2, 130) = 82.11, p < .001, $η_{p}^{2}$ = .56. Consistent with our primary hypothesis, participants saw Miguel as less authentic when he frame switched compared to when he actively did not frame switch, t(130) = 12.38, p < .001, d = 2.17, and to when no information was given about his behavior, t(130) = 9.08, p < .001, d = 1.59.

One-way ANOVAs on likeability, trustworthiness, warmth, and competence showed significant effects of condition, Fs(2, 130) > 4.18, ps < .02, $η_{p}^{2}$ s > .06 (Table 2—total effects). Across all measures, participants in the Switching condition formed less favorable impressions of Miguel compared to those in the No Switching condition, likeable t(130) = 2.07, p = .04, d = 0.36; trustworthy t(130) = 3.20, p = .002, d = 0.56; warm t(129) = 3.00, p = .003, d = 0.53; competent t(129) = 3.56, p = .001, d = 0.63, and compared to those in the Neutral condition, likeable t(130) = 2.79, p = .006, d = 0.49; trustworthy t(130) = 1.98, p = .05, d = 0.35; warm t(129) = 2.51, p = .01, d = 0.44; competent t(129) = 2.64, p = .009, d = 0.47.

Table 2.

Total and Indirect Effects of Frame Switching (vs. No Switching and vs. Control) for Study 1.

	Total Effectof Condition			Switch vs. No Switchvia Authenticity		Switch vs. Neutralvia Authenticity
Outcome Variable	F	p	$η_{p}^{2}$	b (SE)	95% CI	b (SE)	95% CI
Likeability	4.18	.02	.06	−1.14 (.18) −1.37 (.19)	[−1.51, −0.79] [−1.77, −1.02]	−0.85 (.16) −1.02 (.17)	[−1.17, −0.56] [−1.36, −0.72]
Trust	5.20	.006	.07	−1.52 (.22) −1.34 (.16)	[−1.94, −1.10] [−1.66, −1.04]	−1.13 (.19) −1.00 (.15)	[−1.53, −0.77] [−1.31, −0.72]
Warmth	5.18	.007	.07	−0.71 (.12) −1.27 (.20)	[−0.95, −0.49] [−1.67, −0.90]	−0.54 (.10) −0.97 (.17)	[−0.75, −0.36] [−1.33, −0.66]
Competence	6.84	.002	.10	−0.45 (.12) −0.83 (.20)	[−0.69, −0.23] [−1.22, −0.42]	−0.35 (.09) −0.63 (.16)	[−0.54, −0.17] [−0.96, −0.32]

Note. For total effects, df ₁ = 2, df ₂ = 130. For indirect effects, nonitalicized coefficients refer to the unstandardized indirect effects, and italicized coefficients below refer to the partially standardized indirect effects. The 95% bias-corrected bootstrap confidence intervals (10,000 samples) that do not contain zero indicate a statistically significant effect. All indirect effects above are significant.

To test whether frame switching negatively affected general impressions by reducing authenticity, we conducted mediation analyses using PROCESS (Version 3) following procedures for multicategorical independent variables (Hayes & Preacher, 2014). Conditions were coded into two orthogonal contrasts: Switching versus No Switching and Switching versus Neutral. Supporting our prediction, confidence intervals for all indirect effects were below zero (Table 2—indirect effects), demonstrating that frame switching significantly decreased evaluations on all traits by diminishing Miguel’s perceived authenticity.

Study 2

Study 1 demonstrated that majority Americans saw a frame switching bicultural as less authentic compared to when he actively did not frame switch and when no information was given about his behavior. This loss of perceived authenticity consequently damaged general impressions of the frame switching bicultural. However, statistical mediation in cross-sectional designs is limited to only testing a correlation between the mediator and outcome (Spencer et al., 2005); thus, our next study manipulates the mediator to establish a causal chain between frame switching to authenticity to general impressions. If the consequences of frame switching are truly due to perceived inauthenticity, then assuring participants that a frame switching bicultural is still being authentic should mitigate the harsher impressions found in Study 1. We predicted that majority Americans would form more favorable impressions of a frame switching bicultural when told that he is behaving authentically with each culture compared to when his authenticity is not affirmed.

Method

Participants

Majority Americans (N = 435) participated online via Prolific. Eligibility and exclusion criteria were consistent with Study 1; final sample N = 390. Power analyses based on an initial study (see OSM) indicated that N = 390 provided 80% power (α = .05) to detect the smallest observed effect—authentic switching versus switching on competence, d = 0.29.

Procedure

Overall, the design and procedure followed Study 1. The major difference was adding a new Authentic Switching condition (n = 129) that was based on the previous Switching condition vignette but included an additional paragraph affirming the bicultural’s authenticity. This study also included the same Switching (n = 132) and No Switching (n = 129) conditions from Study 1, allowing us to test whether the previous effects replicated along with the current hypothesis. Thus, there were three randomly assigned conditions: Switching, Authentic Switching, and No Switching. After reading one of the vignettes, participants rated the bicultural’s likeability, trustworthiness, warmth, and competence. They also rated the bicultural’s authenticity as a manipulation check before completing demographics and debriefing.

Materials

Bicultural vignettes

The Switching and No Switching vignettes were identical to those in Study 1. The new Authentic Switching vignette provided the same content as the Switching vignette, followed by information affirming the bicultural’s authenticity:

Miguel is not trying to pretend or misrepresent himself when he is with either cultural group, and he has no intention to deceive or manipulate others through his behaviour. Rather, Miguel’s behavior with each cultural group reflects different sides of himself that are both equally a part of who he truly is.

Pretesting these vignettes confirmed the effectiveness of the manipulation—Miguel was deemed more authentic in the Authentic Switching (vs. Switching) condition.

General impressions

Measures of likeability (α = .91), warmth (α = .89), and competence (α = .86) were the same as in Study 1. To improve our assessment of trustworthiness beyond a single item, we adapted a three-item measure (Fletcher et al., 2000; α = .93). All response scales ranged from 1 (strongly disagree) to 7 (strongly agree).

Authenticity

The authenticity measure from Study 1 provided a manipulation check; results ensured that the authenticity manipulation in the Authentic Switching condition was successful.

Results

See Table 3 for descriptive statistics.

Table 3.

Descriptive Statistics for Study 2.

Outcome Variable	Authentic Switching M [95% CI]	Switching M [95% CI]	No Switching M [95% CI]
Likeability	5.56 [5.40, 5.72]	5.31 [5.16, 5.47]	5.67 [5.51, 5.82]
Trustworthiness	5.12 [4.95, 5.30]	5.00 [4.82, 5.17]	5.49 [5.32, 5.67]
Warmth	4.03 [3.92, 4.13]	3.83 [3.72, 3.93]	4.09 [3.99, 4.20]
Competence	3.89 [3.79, 3.98]	3.82 [3.72, 3.92]	4.09 [4.00, 4.19]

One-way ANOVAs revealed significant differences between the three conditions on all four general impressions: likability, F(2, 385) = 5.28, p = .005, $η_{p}^{2}$ = .03; trustworthiness, F(2, 387) = 8.42, p < .001, $η_{p}^{2}$ = .04; warmth, F(2, 385) = 6.70, p = .001, $η_{p}^{2}$ = .03; and competence, F(2, 385) = 8.37, p < .001, $η_{p}^{2}$ = .04. Negative effects of the Switching (vs. No Switching) condition also replicated on all impressions, ts(385–387) > 3.17, ps < .002, ds > 0.32. Assuring participants of Miguel’s authenticity when frame switching (i.e., Authentic Switching vs. Switching) partially mitigated the negative consequences of frame switching, Miguel was judged less harshly in terms of likeability, t(385) = 2.19, p = .03, d = 0.22, and warmth, t(385) = 2.65, p = .008, d = 0.27, but not trustworthiness, t(387) = 1.00, p = .32, d = 0.10, or competence, t(385) = 0.98, p = .33, d = 0.10. Further, affirming Miguel’s authenticity when frame switching partially nullified the benefits of actively not frame switching (i.e., Authentic Switching vs. No Switching), as his perceived likeability and warmth did not differ significantly between these two conditions: likeability, t(385) = 0.97, p = .34, d = 0.10; warmth, t(385) = 0.85, p = .40, d = 0.09. However, actively not frame switching still produced advantages over authentically frame switching (i.e., No Switching vs. Authentic Switching) for Miguel’s perceived trustworthiness, t(387) = 2.94, p = .004, d = 0.30, and competence, t(385) = 2.95, p = .003, d = 0.30. Thus, affirming the bicultural’s authenticity countered some, but not all, of the costs from frame switching as well as the benefits from actively not frame switching.

Study 3

Next, we raise the stakes on the social consequences by examining how frame switching negatively impacts biculturals’ romantic relationship prospects. We also address two limitations of the prior studies. First, Studies 1–2 used vignettes explicitly describing the bicultural’s frame switching and so may have had high demand characteristics—participants may have felt expected to react negatively to the bicultural’s inconsistency. Although we would argue that the demand characteristics are likely outweighed by the social desirability of not appearing prejudiced (McConahay et al., 1981), we improve our manipulation in Study 3 to be less explicit by using online dating profiles that display frame switching (or not) in more discrete ways. Second, reading third-person vignettes may not reflect how people naturally learn about others. Study 3 simulates a more realistic situation: seeing a bicultural’s frame switching in action in dating profiles that one could find easily online. We predict that (1) frame switching (vs. No Switching vs. Neutral) will negatively affect majority Americans’ perceptions of a bicultural’s authenticity and (2) majority Americans will form less favorable general and dating-relevant impressions of a frame switching bicultural and these effects will be mediated by authenticity. Our preregistration is available here: (osf.io/8yp7x).

Method

Participants

Heterosexual, mainstream American women (N = 292) participated online via Prolific or MTurk. Power analyses approximated that N = 300 provided 94% power (α = .05) to detect the effect of Switching versus No Switching on authenticity (d = 0.50) observed in a pretest.

As preregistered, we excluded participants who did not meet eligibility criteria: majority American, heterosexual women (age 18–40) not currently in a relationship (n = 49 excluded). We excluded participants who indicated that they did not complete the study honestly and attentively (self-report item; n = 1) or did not provide post-debrief consent (n = 2). Attention check items were also included, and all participants passed. Final sample N = 240.

Procedure

Participants were led to believe that they would see five single, American men’s profiles from one or more dating websites. In reality, all participants only saw dating profiles ostensibly created by Miguel Wong from Studies 1–2. Participants were randomly assigned to one of three conditions: (1) Switching (n = 81): Miguel had profiles on two cultural-niche dating websites, each highlight different aspects of himself depending on the cultural context of each site; (2) No Switching (n = 79): Miguel had nearly identical profiles on the same two cultural-niche sites and did not emphasize either culture over the other; (3) Neutral (n = 80): Miguel had one profile on a general (not cultural-niche) dating site and did not emphasize either culture, thereby establishing his bicultural background without demonstrating his (in)consistency. The No Switching condition presented the same content in each profile with slight variations in how statements were worded to isolate the effects of actively not frame switching from a more mundane form of consistency (i.e., exactly duplicating content).

Participants opened website links to pdfs of Miguel’s profile(s) and were instructed to review them carefully. After freely perusing the profiles, we directed participants’ focus to key aspects with attention checks about the profile photo and content. Participants then rated Miguel on authenticity, general impressions from prior studies, and new dating-relevant impressions. Further, we assessed hypothetical dating intentions toward Miguel. Finally, participants were informed there were no other profiles currently available to rate and provided demographics before debriefing.

Materials

Bicultural dating profiles

All participants saw either one (Neutral) or two dating profiles (Switching or No Switching). All profiles contained the same basic information about Miguel’s demographics, lifestyle, and cultural background. His profile photos (Figure 1) and subtle aspects of the profile content varied between conditions.

Figure 1.

Study 3 profile photos (left to right): (1) Mexican profile photo in the Switching condition, (2) Chinese profile photo in the Switching condition, (3) profile photo in the No Switching and Neutral condition. Note. For full profiles, see Online Supplementary Material or Open Science Framework page.

In the Switching condition, Miguel had profiles on two real cultural-niche dating websites: MexicanCupid.com and ChinaLoveCupid.com. His MexicanCupid profile photo showed him wearing a shirt with a Calavera (Day-of-the-Dead skull), and the profile content emphasized his interest in more Mexican-associated foods, hobbies/sports, and travel. In contrast, his ChinaLoveCupid.com profile photo showed him wearing a shirt with a Chinese dragon, and the profile content emphasized his interest in more Chinese-associated foods, hobbies/sports, and travel. Importantly, nothing stated in either profile was mutually exclusive—for instance, saying he visited Mexico City in one profile does not contradict the trip to Beijing described in his other profile.

In the No Switching condition, Miguel also had two profiles on the same two cultural-niche websites. In both his MexicanCupid and ChinaLoveCupid profile photos, he was wearing a blank shirt, and the content described his interest in international foods, exercise and sports in general, and a trip to Sydney. Again, the intention here was to demonstrate Miguel’s active nonswitching with culturally neutral content.

In the Neutral condition, Miguel had just one profile on the fabricated, culturally neutral LoveCupid.com, which we created by covering elements of the ChinaLoveCupid.com layout. His photo showed him wearing the same blank shirt, and the profile content was the same as the No Switching condition.

Pretests ensured that participants noticed Miguel’s frame switching between profiles in the Switching condition and made the intended cultural associations (e.g., recognized highlighting of Mexican/Chinese culture) and did not see Miguel as more or less American in the Switching versus No Switching profiles.

Authenticity

Measured the same as previous (α = .93). Two additional exploratory mediators, deceptiveness and manipulativeness, assessed malicious forms of inauthenticity.

Dating-relevant impressions

Impressions of Miguel as a potential dating partner were assessed using a four-item measure of Interpersonal Attraction and Intentions to Meet (Alves, 2018; α = .94), for example, “How much would you like to meet Miguel?” (1 = not at all to 9 = extremely). We also created two new items to assess how attractive (physically and more broadly) participants found Miguel to be (α = .84) and another two items to assess how interested participants were in Miguel as a dating partner (α = .95), for example, “Miguel seems like someone I would be open to dating” (1 = strongly disagree to 7 = strongly agree). Participants also reported how likely they would be to recommend Miguel as a dating partner to a friend using an existing dating endorsement item (1 = strongly disagree to 5 = strongly agree; Rycyna et al., 2009).

Dating intentions

Next, participants indicated how likely they would be to engage in three dating behaviors with Miguel (α = .91). Imagining they had come across Miguel’s profile(s) outside of this study, participants reported their willingness to (1) send Miguel a message, (2) respond to a message from Miguel, and (3) go on a date with Miguel (1 = strongly disagree to 7 = strongly agree).

General impressions

Participants also evaluated Miguel’s likeability, trustworthiness, warmth, and competence using four single-item measures (1 = strongly disagree to 7 = strongly agree), for example, “Overall, I think Miguel is a likeable person.”

Results

See Table 4 for descriptive statistics.

Table 4.

Descriptive Statistics for Study 3.

Outcome Variable	Switching M [95% CI]	No Switching M [95% CI]	Neutral M [95% CI]
Authenticity	4.07 [3.84, 4.30]	5.69 [5.45, 5.92]	5.49 [5.26, 5.73]
Interpersonal attraction and intentions to meet	4.58 [4.19, 4.97]	5.94 [5.54, 6.33]	5.62 [5.23, 6.02]
Attractiveness	3.80 [3.52, 4.09]	4.79 [4.50, 5.08]	4.39 [4.11, 4.68]
Dating interest	3.33 [2.98, 3.68]	4.35 [4.00, 4.71]	4.01 [3.66, 4.37]
Dating endorsements	2.80 [2.58, 3.03]	3.82 [3.60, 4.05]	3.71 [3.49, 3.94]
Dating intentions	3.12 [2.77, 3.48]	4.08 [3.72, 4.44]	3.86 [3.51, 4.22]
Likeability	4.91 [4.70, 5.12]	5.89 [5.68, 6.09]	5.73 [5.53, 5.94]
Trustworthiness	4.14 [3.89, 4.39]	5.59 [5.34, 5.84]	5.34 [5.09, 5.58]
Warmth	4.71 [4.46, 4.96]	5.58 [5.32, 5.83]	5.21 [4.96, 5.46]
Competence	5.14 [4.92, 5.36]	5.87 [5.65, 6.10]	5.70 [5.48, 5.92]

Effects on authenticity

One-way ANOVAs revealed significant differences between conditions on ratings of authenticity, F(2, 237) = 56.21, p < .001, $η_{p}^{2}$ = .32. Participants who witnessed Miguel’s frame switching saw him as less authentic compared to both control conditions (No Switching and Neutral): Switching versus No Switching, t(237) = 9.68, p < .001, d = 1.26, Switching versus Neutral, t(237) = 8.55, p < .001, d = 1.11. Miguel was not seen as any more or less authentic when he actively did not frame switch (No Switching) compared to when no information about his behavior was given (Neutral), t(237) = 1.15, p = .25, d = 0.15. Thus, frame switching had strong negative effects on authenticity, the proposed mediator.¹

Consequences for dating-relevant impressions

ANOVA results indicated significant differences between conditions on each of the dating-relevant impressions, Fs(2, 237) < 8.48, ps < .001, $η_{p}^{2}$ s > .07 (Table 5—total effects). When Miguel frame switched instead of actively not switching or when only one noncultural-niche profile was presented, majority American women formed less favorable dating-relevant impressions. Miguel’s frame switching reduced participants’ Interpersonal Attraction and Intentions to Meet, ts(237) > 3.70, ps < .001, ds > 0.48; their attraction to him physically and more broadly, ts(237) > 2.90, ps < .004, ds > 0.38; their interest in him as a dating partner, ts(237) > 2.70, ps < .007, ds > 0.35; and their endorsement of him as a dating partner, ts(237) > 5.61, ps < .001, ds > 0.07. To test the role of authenticity as mediating these negative effects, simple mediation models were constructed in line with the analyses described in Study 1. Supporting our hypothesis, confidence intervals for all of the indirect effects were below zero, showing that frame switching significantly diminished majority Americans’ dating-relevant impressions (vs. No Switching and vs. Neutral) because they saw Miguel as less authentic (Table 5—indirect effects). These results show that frame switching in a dating context can make majority Americans feel that a bicultural is being less authentic and, in turn, a less appealing potential romantic partner.

Table 5.

Total and Indirect Effects of Frame Switching (vs. No Switching and vs. Control) via Perceived Authenticity for Study 3.

	Total Effectof Condition			Switch vs. No Switchvia Authenticity		Switch vs. Neutralvia Authenticity
Outcome Variable	F	p	$η_{p}^{2}$	b (SE)	95% CI	b (SE)	95% CI
Interpersonal attraction and intentions to meet	12.66	<.001	.10	−1.47 (.24) −0.79 (.12)	[−1.98, −1.02] [−1.03, −0.57]	−1.30 (.23) −0.69 (.11)	[−1.79, −0.90] [−0.92, −0.69]
Attractiveness	11.85	<.001	.09	−0.76 (.17) −0.56 (.12)	[−1.11, −0.45] [−0.80, −0.35]	−0.67 (.15) −0.50 (.11)	[−0.99, −0.39] [−0.72, −0.30]
Dating interest	8.48	<.001	.07	−0.88 (.20) −0.53 (.12)	[−1.32, −0.52] [−0.80, −0.32]	−0.77 (.18) −0.47 (.11)	[−1.16, −0.45] [−0.69, −0.28]
Dating endorsements	23.84	<.001	.17	−0.60 (.14) −0.53 (.12)	[−0.88, −0.35] [−0.77, −0.32]	−0.52 (.13) −0.47 (.11)	[−0.78, −0.30] [−0.69, −0.27]
Dating intentions	7.77	<.001	.06	−0.83 (.20) −0.50 (.12)	[−1.25, −0.46] [−0.75, −0.28]	−0.73 (.18) −0.44 (.11)	[−1.13, −0.40] [−0.67, −0.25]
Likeability	24.71	<.001	.17	−0.91 (.16) −0.89 (.12)	[−1.26, −0.63] [−1.14, −0.67]	−0.80 (.15) −0.78 (.11)	[−1.11, −0.53] [−1.02, −0.57]
Trustworthiness	37.90	<.001	.24	−1.19 (.18) −0.93 (.11)	[−1.55, −0.85] [−1.16, −0.71]	−1.04 (.18) −0.82 (.11)	[−1.41, −0.71] [−1.05, −0.59]
Warmth	11.41	<.001	.09	−0.98 (.18) −0.82 (.13)	[−1.35, −0.66] [−1.08, −0.59]	−0.86 (.17) −0.72 (.12)	[−1.20, −0.56] [−0.98, −0.50]
Competence	11.84	<.001	.09	−0.82 (.16) −0.78 (.12)	[−1.16, −0.53] [−1.02, −0.57]	−0.72 (.15) −0.69 (.11)	[−1.03, −0.46] [−0.92, −0.69]

Note. For total effects, df ₁ = 2, df ₂ = 237. For indirect effects, nonitalicized coefficients refer to the unstandardized indirect effects, and italicized coefficients below refer to the partially standardized effects. The 95% bias-corrected bootstrap confidence intervals (10,000 samples) that do not contain zero indicate a statistically significant effect. All indirect effects above are significant.

Consequences for dating intentions

The strength of participants’ intentions to communicate with and date Miguel significantly varied between conditions, F(2, 237) = 7.77, p = .001, $η_{p}^{2}$ = .06 (Table 5—total effects). Participants felt that they would be less likely to send or respond to a message or go on a date with Miguel when he frame switched compared to when he actively did not frame switch and compared to neutral control, ts(237) > 2.91, ps < .004, ds > 0.38. Further, simple mediation results revealed that frame switching reduced participants’ dating intentions (vs. No Switching and vs. Neutral) because Miguel’s frame switching undermined his perceived authenticity (Table 5—indirect effects). Thus, majority Americans were not only less impressed with Miguel as a potential partner when he frame switched but felt they would also be less likely to actually engage with him romantically if they had found these dating profiles on their own in the real world.

Consequences for general impressions

Finally, the results show a significant effect of condition on each of the general impressions, Fs(2, 237) > 11.41, ps < .001, $η_{p}^{2}$ s > .09 (Table 5—total effects). Specifically, frame switching cost Miguel in terms of his likeability, trustworthiness, warmth, and competence compared to when he did not frame switch and compared to neutral control, ts(234–237) > 2.77, ps < .006, ds > 0.36. Mediation analyses confirmed that these consequences of frame switching on general impressions, compared to No Switching and to Neutral, are all mediated by a loss of Miguel’s perceived authenticity when he frame switches (Table 5—indirect effects). These findings directly replicate the second pretest’s results and conceptually replicate each of the earlier studies’ results, adding strong evidence that majority Americans dislike frame switching because they infer that inauthenticity drives the bicultural’s behavior.

General Discussion

Biculturals frame switch as a way to navigate their complex cultural worlds. Across four experiments, however, we demonstrate that frame switching is perceived as inauthentic and, in turn, has social costs. In all studies, the hit to authenticity led to worse impressions of a bicultural’s likeability, trustworthiness, warmth, and competence. Affirming the bicultural’s authenticity in Study 2 partly mitigated frame switching’s negative effects on likeability and warmth, but not trustworthiness or competence. Study 3 targeted the impactful arena of romantic relationships, demonstrating that frame switching in dating profiles diminishes biculturals’ perceived authenticity and reduces their chances of dating success with majority Americans. These results illustrate how frame switching creates a paradox for biculturals living in Western cultures: It allows them to fit in with their cultural groups, but it can backfire when behaving inconsistently violates perceivers’ expectations and values. That is, frame switching biculturals can incur powerful social penalties to impression formation and romantic relationships.

These findings illuminate a novel barrier to intercultural relations in Western society. Previous research has shown that majority Americans are suspicious of biculturals by default because of their dual identities (Kunst et al., 2018) that are assumed to confuse biculturals about who they truly are (Skinner et al., 2019). While these biases may be at play, our results showed that the negative effect on authenticity and its downstream consequences held when frame switching was compared to a neutral control condition in which participants only knew about the bicultural’s dual cultural identities but did not know about his behavior with his cultural groups. This implies that majority Americans’ reactions were driven by the way the bicultural behaved beyond any biases they may hold against his particular cultures or against his dually identified bicultural status.

Recent studies have uncovered that “passing” behavior, whereby a biracial presents as only one racial identity based on the context, also evokes negative reactions from majority Americans (Albuja et al., 2018). In our studies, we were able to isolate a different source of bias against biculturals—switching between their multiple identities—providing some of the first evidence that biculturals’ overt behavior across cultural contexts affects the way they are seen by others. Cumulatively, the previous and current work unveil the quagmire that biculturals face in Western society—they are punished by majority members not only when they deny one of their identities but also when they present both identities and adapt themselves to their cultural contexts by frame switching. This raises the question: Is there any socially accepted way to be “true to yourself” for mixed selves?

Limitations and Future Research

These studies have some limitations. We only created one bicultural target used across the studies and so we have not examined how target gender or how other minority cultures might change reactions to frame switching. Because Study 3 participants were heterosexual women, we do not know how men or non-heterosexual people would react to prospective bicultural partners’ frame switching. We anticipate that the shared Western understanding of authenticity and its incompatibility with frame switching would be strong enough to influence most majority Americans’ reactions to biculturals, but future research is needed to uncover potential moderators of frame switching’s negative effects. Additionally, the control conditions in these studies depicted a particular form of “not frame switching,” whereby the bicultural’s behavior was intended to be not directly linked to either culture, rather than aligned with one culture over the other (e.g., always more Chinese, as in assimilation) or uniquely mixed together (i.e., hybridizing; West et al., 2017). Future studies should pit frame switching against these and other cultural negotiation strategies for a more complex understanding of how biculturals’ behavior is perceived.

Notably, Study 2 failed to explicitly affirm the biculturals’ authenticity to mitigate the damage of frame switching on trust and competence, even though Studies 1 and 3 establish authenticity as a statistical mediator. It is possible that our manipulation did not cover aspects of authenticity more relevant to trust and competence, or that other mediators may factor more heavily for these two outcomes. Alternatively, affirming authenticity may have weaker benefits for a frame switching bicultural because Americans may not hold an authentic mixed self in as high regard as they would an authentic singular self that personifies their understanding of authenticity. Of these two downstream consequences, implications for trustworthiness are particularly impactful because trust is regarded as fundamental to harmonious relationships (Rempel et al., 1985). This fits well with the results of Study 3, which examined a romantic relationship domain and also suggests that frame switching may lead to particularly harsh penalties in contexts where trust is important. Future research may investigate the fallout of frame switching for bicultural politicians, job applicants, and those already in intercultural romantic relationships. In contrast, Study 2 successfully restored impressions of likeability and warmth by affirming authenticity—results with implications for ameliorating intercultural relations. At least for these traits, our results demonstrate that Americans can form favorable impressions of a bicultural despite their frame switching. Due to the limits of cross-sectional mediation, these data are not ideally suited to comparing alternate models (e.g., parallel or sequential mediation between perceived authenticity and other trait impressions). Future longitudinal studies should examine how impressions may change and develop over the course of multiple interactions with a frame switching bicultural to more comprehensively test the role of perceived authenticity over time.

Conclusion

A growing population of biculturals endeavor to be true to their mixed selves. However, the strategies biculturals use to successfully navigate their multiple cultures can have social costs. As many nations become increasingly diverse, it is more important than ever to identify and break down these barriers to intercultural relations so that all people can thrive while being true to themselves.

Supplemental Material

West,_Muise,_and_Sasaki_Consequences_of_Frame_Switching_R2_Online_Supplement - The Cost of Being “True to Yourself” for Mixed Selves: Frame Switching Leads to Perceived Inauthenticity and Downstream Social Consequences for Biculturals

West,_Muise,_and_Sasaki_Consequences_of_Frame_Switching_R2_Online_Supplement for The Cost of Being “True to Yourself” for Mixed Selves: Frame Switching Leads to Perceived Inauthenticity and Downstream Social Consequences for Biculturals by Alexandria L. West, Amy Muise and Joni Y. Sasaki in Social Psychological and Personality Science

Footnotes

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Alexandria L. West

Amy Muise

Supplemental Material

The supplemental material is available in the online version of the article.

Note

References

Albuja

A. F.

Sanchez

D. T.

Gaither

S. E.

(2018). Fluid racial presentation: Perceptions of contextual “passing” among biracial people. Journal of Experimental Social Psychology, 77, 132–142.

Alves

(2018). Sharing rare attitudes attracts. Personality and Social Psychology Bulletin, 44(8), 1270–1283.

Cila

Lalonde

R. N.

(2019). What’s in a name? Motivations for baby-naming in multicultural contexts. In Al-Hoorie

MacIntyre

P. D.

(Eds.), Contemporary Language Motivation Research: 60 Years Since Gardner and Lambert (1959) (pp. 130–152). Multilingual Matters.

Cross

S. E.

Gore

J. S.

Morris

M. L.

(2003). The relational-interdependent self-construal, self-concept consistency, and well-being. Journal of Personality and Social Psychology, 85(5), 933–944. https://doi.org/10.1037/0022-3514.85.5.933

Cuddy

A. J. C.

Fiske

S. T.

Glick

(2007). The BIAS map: Behaviors from intergroup affect and stereotypes. Journal of Personality and Social Psychology, 92(4), 631–648. https://doi.org/10.1037/0022-3514.92.4.631

Cuddy

A. J. C.

Fiske

S. T.

Glick

(2008). Warmth and competence as universal dimensions of social perception: The stereotype content model and the BIAS map. Advances in Experimental Social Psychology, 40, 61–149. https://doi.org/10.1016/s0065-2601(07)00002-0

Doucerain

Dere

Ryder

A. G.

(2013). Travels in hyper-diversity: Multiculturalism and the contextual assessment of acculturation. International Journal of Intercultural Relations, 37(6), 686–699. https://doi.org/10.1016/j.ijintrel.2013.09.007

English

Chen

(2011). Self-concept consistency and culture: The differential impact of two forms of consistency. Personality and Social Psychology Bulletin, 37(6), 838–849. https://doi.org/10.1177/0146167211400621

Fletcher

G. J.

Simpson

J. A.

Thomas

(2000). The measurement of perceived relationship quality components: A confirmatory factor analytic approach. Personality and Social Psychology Bulletin, 26(3), 340–354.

10.

Hayes

A. F.

Preacher

K. J.

(2014). Statistical mediation analysis with a multicategorical independent variable. British Journal of Mathematical and Statistical Psychology, 67, 451–770. https://doi.org/10.1111/bmsp.12028

11.

Hong

Y.-y.

Khei

(2014). Dynamic multiculturalism: The interplay of socio-cognitive, neural, and genetic mechanisms. In Benet-Martínez

Hong

(Eds.), The Oxford handbook of multicultural identity (pp. 11–34). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199796694.013.026

12.

Horowitz

J. M.

(2019). Americans see advantages and challenges in country’s growing racial and ethnic diversity. https://www.pewsocialtrends.org/2019/05/08/americans-see-advantages-and-challenges-in-countrys-growing-racial-and-ethnic-diversity/

13.

Josephs

Warach

Goldin

K. L.

Jonason

P. K.

Gorman

B. S.

Masroor

Lebron

(2019). Be yourself: Authenticity as a long-term mating strategy. Personality and Individual Differences, 143, 118–127.

14.

Kashima

Farsides

Kim

Strack

Werth

Yuki

(2004). Culture and context-sensitive self: The amount and meaning of context-sensitivity of phenomenal self differ across cultures. Self and Identity, 3(2), 125–141.

15.

Kernis

M. H.

Goldman

B. M.

(2006). A multicomponent conceptualization of authenticity: Theory and research. Advances in Experimental Social Psychology, 38, 283–357. https://doi.org/10.1016/S0065-2601(06)38006-9

16.

Krumhuber

Manstead

A. S.

Cosker

Marshall

Rosin

P. L.

Kappas

(2007). Facial dynamics as indicators of trustworthiness and cooperative behavior. Emotion, 7(4), 730–735. https://doi.org/10.1037/1528-3542.7.4.730

17.

Kunst

J. R.

Thomsen

Dovidio

J. F.

(2018). Divided loyalties: Perceptions of disloyalty underpin bias toward dually-identified minority-group members. Journal of Personality and Social Psychology. https://doi.org/10.1037/pspi0000168

18.

LaFromboise

Coleman

H. L. K.

Gerton

(1993). Psychological impact of biculturalism: Evidence and theory. Psychological Bulletin, 114, 395–412. https://doi.org/10.1037/0033-2909.114.3.395

19.

Lalonde

R. N.

Cila

Lou

Giguère

(2013). Delineating groups for cultural comparisons in a multicultural setting: Not all Westerners should be put into the same melting pot. Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement, 45(4), 296–304. https://doi.org/10.1037/a0034257

20.

Lopez

F. G.

Rice

K. G.

(2006). Preliminary development and validation of a measure of relationship authenticity. Journal of Counseling Psychology, 53(3), 362–371. https://doi.org/10.1037/0022-0167.53.3.362

21.

McConahay

J. B.

Hardee

B. B.

Batts

(1981). Has racism declined in America? It depends on who is asking and what is asked. Journal of Conflict Resolution, 25(4), 563–579.

22.

Mok

Morris

M. W.

(2013). Bicultural self-defense in consumer contexts: Self-protection motives are the basis for contrast versus assimilation to cultural cues. Journal of Consumer Psychology, 23(2), 175–188.

23.

Nisbett

R. E.

Peng

Choi

Norenzayan

(2001). Culture and systems of thought: Holistic versus analytic cognition. Psychological Review, 108(2), 291–310. https://doi.org/10.1037/0033-295X.108.2.291

24.

Perunovic

W. Q. E.

Heller

Rafaeli

(2007). Within-person changes in the structure of emotion: The role of cultural identification and language. Psychological Science, 18, 607–613. https://doi.org/10.1111/j.1467-9280.2007.01947.x

25.

Rempel

J. K.

Holmes

J. G.

Zanna

M. P.

(1985). Trust in close relationships. Journal of Personality and Social Psychology, 49(1), 95.

26.

Rycyna

C. C.

Champion

C. D.

Kelly

A. E.

(2009). First impressions after various types of deception: Less favorable following expectancy violation. Basic and Applied Social Psychology, 31(1), 40–48.

27.

Skinner

A. L.

Perry

S. P.

Gaither

(2019). Not quite monoracial: Biracial stereotypes explored. Personality and Social Psychology Bulletin. https://doi.org/10.1177/0146167219858344

28.

Spencer

S. J.

Zanna

M. P.

Fong

G. 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.

29.

Toma

C. L.

Hancock

J. T.

Ellison

N. B.

(2008). Separating fact from fiction: An examination of deceptive self-presentation in online dating profiles. Personality and Social Psychology Bulletin, 34(8), 1023–1036.

30.

West

A. L.

Zhang

Yampolsky

Sasaki

(2017). More than the sum of its parts: A transformative theory of biculturalism. Journal of Cross-Cultural Psychology, 48 (7), 963–990. https://doi.org/10.1177/0022022117709533

31.

West

A. L.

Zhang

Yampolsky

M. A.

Sasaki

(2018). The potential cost of cultural fit: Frame switching undermines perceptions of authenticity. Frontiers in Psychology, 9, 2622.

32.

Wickham

R. E.

(2013). Perceived authenticity in romantic partners. Journal of Experimental Social Psychology, 49(5), 878–887. https://doi.org/10.1016/j.jesp.2013.04.001

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

0.41 MB