Discerning Cultural Identification From a Thinly Sliced Behavioral Sample

Abstract

This research examined whether individual differences in cultural identification can be discerned at zero acquaintance. This issue was examined in Hong Kong, where the idiosyncrasy of cultural identification is a salient social-psychological issue. The participants were able to perceive accurately the targets’ identification with Western culture from a video clip and from a still image. Findings also indicated that a stereotype of Western cultural identity (i.e., extraversion and particular hairstyle) facilitated these perceptions. Specifically, (a) the participants with a stronger stereotype were more accurate in perceiving Western cultural identification, (b) the targets who were experimentally manipulated to appear extraverted were rated as more strongly identifying with Western culture, and (c) the participants relatively unfamiliar with these stereotypes did not correctly perceive Western cultural identification. Implications of these findings on research on multiculturalism are also discussed.

Keywords

thin slicing cultural identification multiculturalism person perception diversity

Individuals identifying with a particular cultural tradition have their self-concept referenced to the beliefs, values, and customs of that culture (Hong, Wan, No, & Chiu, 2007). A pattern of a particular cultural tradition, in turn, is endorsed and expressed more prominently among individuals who have a stronger identification with that culture.

To the extent that cultural identification anchors the self-system and implicates a wide range of psychological processes, a perception of cultural identification may facilitate social understanding. As such, cultural identification may be actively discerned in social interactions. Among a group of friends, a perception of cultural identification would be relatively straightforward through intimate knowledge of one another. One intriguing question is whether people with no knowledge of one another are able to discern others’ cultural identification. If the answer is yes, then how much information is needed? What is the process through which such an inference is made? What are the implications of such a perception? Addressing these questions enables an integration and extension of prior research on person perception and cultural identification.

The current research also aims to contribute a new perspective to the contemporary debate between multiculturalism and assimilationism (Plaut, 2010). As detailed below, the question of whether cultural identity is perceived at zero acquaintance and whether such a perception influences subsequent interpersonal interactions is one of the key issues in this debate.

Perceiving Culture From a Thinly Sliced Behavioral Sample

Research has found that a great deal of information about a person is revealed through a short observation of social behaviors (Ambady & Rosenthal, 1992). For example, strangers’ perceived ratings of a target’s personality corresponded well with the target’s self-ratings (Albright, Kenny, & Malloy, 1988); socioeconomic status (SES) could be accurately perceived from a brief silent video clip of a target engaging in a get-acquainted conversation with another person (Kraus & Keltner, 2009); and sexual orientation could be accurately judged from a brief observation of a dynamic nonverbal behavior (Ambady, Hallahan, & Conner, 1999; Rule, Ambady, Adams, & Macrae, 2008). These findings indicate that individuals quickly and accurately draw an inference pertaining to a wide range of characteristics from cues extracted from social behaviors.

Cross-cultural research on this topic suggests that a similar inference process (i.e., associating a specific perceptual cue with a particular attribute) is often available across cultures (e.g., Albright et al., 1997; Cunningham, Roberts, Barbee, Druen, & Wu, 1995; Rule et al., 2010). However, research also suggests cross-cultural variability in these associations (e.g., Cunningham et al., 1995). These patterns have been extensively studied in the area of emotion recognition—whereas evidence suggests that individuals across cultures recognize a set of basic emotions expressed by others from different cultural backgrounds (Ekman, Sorenson, & Friesen, 1969), a meta-analysis revealed that emotion recognition was generally more accurate when judging the facial expression of people from their own culture (Elfenbein & Ambady, 2002).

An attempt to integrate cross-cultural regularities and variations has resulted in the theory of “nonverbal accents”: universal regularity in nonverbal behaviors localized differently across cultures, akin to linguistic dialects of a basic language (Elfenbein & Ambady, 2002). In a social interaction, nonverbal accents convey cues about cultural membership, similar to an inference made upon a detection of linguistic dialect. In one study, researchers compiled a series of photographs of Japanese and Japanese American targets, which were identified as having equivalent facial expressions based on Facial Action Coding System (FAS) coding, and asked a group of American and Canadian participants to judge the nationality of each target. The participants in this study were able to judge accurately the nationality of the targets (Marsh, Elfenbein, & Ambady, 2003). Presumably, the participants inferred the targets’ cultural membership by stereotyping observed nonverbal cues.

Marsh, Elfenbein, and Ambady (2007) further elucidated the role of stereotype in cultural membership judgment. In this study, American participants evaluated a series of photographs of American and Australian targets. The participants were able to judge accurately the targets’ nationality, but the accuracy was moderated by the extent to which the targets were perceived as more “stereotypical” of that culture. For example, Australian targets who appeared likable and nondominant were more likely to be judged as Australians. In sum, research suggests that from minimum information about a stranger, individuals can accurately recognize the person’s cultural membership. The key to this inference is a detection of nonverbal behaviors that match an extant stereotype about a particular cultural group.

A similar process may enable an inference of cultural identification. That is, to the extent that nonverbal accents of a particular culture are more prominently displayed among individuals with strong cultural identity, perceivers may use an extant stereotype about that culture to discern those who strongly (vs. weakly) identify with that culture. The current research examined this possibility.

Cultural Identification Inference in a Multicultural Society

Individuals belonging to the same cultural group differ in the strength of their cultural identification, which in turn implicates different expression of beliefs and values associated with that culture (Jetten, Postmes, & McAuliffe, 2002). In a culturally diverse society, idiosyncrasy of cultural identity is amplified because individuals also differ on how they identify with cultural traditions other than their own—they may identify with none, some, or all other cultures (Benet-Martínez, Leu, Lee, & Morris, 2002; Ryder, Alden, & Paulhus, 2000). Hence, a culturally diverse society is characterized by a wide variation of cultural identity.

Cultural diversity also implicates potential conflicts among beliefs, values, and practices associated with different cultural traditions. How best these conflicts could be managed has been discussed (Plaut, 2010). Two approaches in particular have received much attention. One is multiculturalism, which advocates recognition and accommodation of cultural differences on an equal basis. The other is assimilationism, which advocates disregarding group differences in favor of treating each person on his or her own right. Research examining implications of these approaches has converged in suggesting the advantage of multiculturalism over assimilationism in fostering positive intergroup relations (e.g., Verkuyten, 2005).

Assimilationism has also been critiqued for its viability especially in interracial relations based on research indicating a rapid perception of race and an inference drawn from such a perception: These findings suggest that being blind to others’ race is improbable (Apfelbaum, Norton, & Sommers, in press). Is cultural identification similar to race in that it is also perceived even at zero acquaintance and influence subsequent interpersonal interaction? Or is cultural identification discernible only with intimate contact? The findings that affirm the former possibility would suggest the difficulty of disregarding cultural identity in social interactions.

Current Research

This research was conducted in Hong Kong, where Western and Chinese cultural traditions coexist. We focused on the perceptions of Western cultural identification. Our objective was to examine (a) whether participants in Hong Kong can perceive fellow Hong Kongers’ Western cultural identification and (b) to the extent that such a perception is possible, its implications and underlying mechanism.

It should be noted that the same set of questions could have been asked about the perception of Chinese cultural identification. As individuals’ identifications with multiple cultural traditions are independent (Ryder et al., 2000), it is likely that such an examination would reveal distinct underlying processes (i.e., cues) involved in the perceptions of Chinese cultural identification. In other words, we do not anticipate that Hong Kongers infer others’ Chinese cultural identification from a perception of their Western cultural identification. Examining these issues, however, was beyond the scope of the current research.

A series of studies examined the two objectives outlined above. The first set of studies examined whether Hong Kongers perceive different levels of Western cultural identification in fellow Hong Kongers. In these studies, the amount of information conveyed in the stimulus was varied to determine the minimum condition for such a perception. We also examined interpersonal implications of being perceived as identifying with Western culture.

The second set of studies examined the hypothesized mechanism, that a perception of Western cultural identity is facilitated by its stereotype. We focused on the role of perceived extraversion and hairstyle based on the findings from Studies 1 and 2 as well as prior research implicating these factors in a stereotype of Western culture among Hong Kongers (Chen & Bond, 2010) and in judgments of social identity (Rule et al., 2008), respectively.

Role of these stereotypes in facilitating cultural identification judgment was examined in multiple ways. To examine the role of perceived extraversion, Study 2 tested whether judges who strongly associate extraversion with Western cultural identity were more accurate in perceiving Western cultural identification. In addition, Study 3 examined whether the targets manipulated to appear extraverted were rated as more strongly identifying with Western culture.

To examine the role of targets’ hairstyle, Study 2a tested whether perceivers fail to accurately perceive Western cultural identification when the targets’ hairstyle information was removed. In addition, Study 4 recruited two groups of foreigners (Western exchange students in Hong Kong and an American community sample). To the extent that these individuals are relatively unfamiliar with types of hairstyle that signal Western cultural identification among Hong Kongers, they may fail to perceive Hong Kongers’ Western cultural identification.

Study 1a

Method

Phase 1: Creating stimuli

More than 300 students from a large university in Hong Kong completed the Vancouver Index of Acculturation (VIA), which separately assesses identification with two cultural traditions, typically host and heritage cultures (Ryder et al., 2000). Here, we measured only an identification with Western culture on a 9-point scale (M = 5.98, SD = 1.21).

We selected 4 males and 6 females who indicated the highest level of Western cultural identification (M = 8.19, SD = 0.31) and 4 males and 5 females who indicated the lowest level of Western cultural identification (M = 3.74, SD = 0.19). Of the 10 participants who strongly identified with Western culture, 2 of them had previously lived in Western societies (1 month and 4 months). None of the participants in the weak identification group had such an experience.

Next, these 19 participants were invited to the laboratory to take part in an interview that had a standardized protocol, answering a series of questions about daily events (e.g., “Do you have any stress? If so, what do you do to manage it?” and “What are things you hate or dislike?”). The interview was videotaped with the camera directed at the target, who sat on a chair against a plain white background. The experimenter was not captured in the video. From each interview, which lasted about 3 min, a 60-s video clip was created by extracting the 60th through the 120th second of the interview.

After the interview, the targets indicated their liking and influence of Western culture to check whether the two groups reliably differed in their Western cultural identification (liking: “How much do you like Western culture?” and influence: “How much influence does Western culture have on you [your thoughts, emotions, and behaviors]?”), with response options for both items ranging from 1 (not at all) to 6 (extremely). The strong identification group indicated a significantly greater liking of Western culture (M = 5.20, SD = 0.42) relative to the weak identification group (M = 3.89, SD = 0.78), t(17) = 4.62, p < .001, d = 2.18. Similarly, the strong identification group indicated that they were more influenced by Western culture (M = 5.00, SD = 0.67) relative to the weak identification group (M = 3.33, SD = 0.87), t(17) = 4.73, p < .001, d = 2.17.

To gain insight into the cues associated with a perception of cultural identification, in a pilot study we asked groups of participants to engage in a get-acquainted conversation and freely write about their impression of others regarding Western cultural identification. Analyses of the responses revealed that four aspects of extraversion (sociable, talkative, expressive, and confident; Lucas, Diener, Grob, Suh, & Shao, 2000) were frequently associated with the perception of Western culture identification.

Confirming this observation, among the 19 targets, extraversion (assessed by the above four items, α = .89) was positively correlated with Western cultural identification, r = .51, p = .03. That is, we found that the stereotype associating Western cultural identification and extraversion (Chen & Bond, 2010) was based on the actual covariation between these two variables.

Phase 2: Perceiving cultural identification

Forty-eight students from the same university (22 males and 26 females) arrived in groups of 8 to 15 and were individually seated in front of a workstation, where they viewed the video clips on a monitor with headphones worn for audio. Nineteen video clips were randomly assembled into one sequence for half of the participants, and the sequence was reversed for the remaining half. The participants were told that after each video, their impression of the target would be assessed.

After each video clip, the participants rated the target person in the video on perceived identification with Western culture, which was assessed by two items: “How much do you think this person likes Western culture?” and “How much influence does Western culture have on this person (his or her thoughts, emotions, and behaviors)?” with response options ranging from 1 (not at all) to 6 (extremely). Ratings for these two items were correlated across the participants in all the studies (average within-participant r between the two items was greater than .77 across the studies).

In addition, the participants rated the targets for liking and perceived extraversion. Liking was assessed by two items: “How much do you like this person?” and “How much would you like to know more about this person?” with response options ranging from 1 (not at all) to 6 (extremely). The analysis was conducted on the average of the two responses (average within-participant r = .81). Perceived extraversion was assessed by the extent to which the four items mentioned earlier, presented with several other traits serving as a filler, were applicable to each target, with response options ranging from 1 (not at all) to 6 (entirely true). In all the studies reported below, the four ratings were highly intercorrelated across the perceivers (average within-participant α >.79).

Finally, the participants indicated whether they personally knew the targets. In 2.6% of the cases (or 24 of 912 combinations of 48 judges across 19 targets), the participants indicated knowing a target prior to the study, and the data for these cases were discarded.¹

Results

To test the hypothesis that perceivers differentiate targets’ Western cultural identification, each perceiver’s ratings were averaged separately for strong and weak identification targets, and a pairwise t test compared averaged ratings.²

A pairwise t test revealed that perceivers indeed differentiated targets with strong (M = 3.43, SD = 0.52) and weak Western cultural identification (M = 3.22, SD = 0.54), t(47) = 3.96, p < .001, d = 0.57 (Table 1).

Table 1.

Perceived Cultural Identification and Its Correlation With Perceived Extraversion (Studies 1 and 2)

Study	Strong identification targets	Weak identification targets	Effect size (d)	Average within-participant correlation with perceived extraversion
Study 1a (video)	3.43 (.52)	3.22 (.54)	0.57	.48 (.22)
Study 1b (video: no audio)	3.49 (.61)	3.14 (.63)	0.75	.59 (.22)
Study 2a (photo: whole body)	3.30 (.70)	2.91 (.65)	0.93	.60 (.21)
Study 2a (photo: face)	3.03 (.76)	2.75 (.72)	0.68	.56 (.20)
Study 2a (photo: facial features)	2.99 (.64)	3.03 (.63)	−0.11	.52 (.24)
Study 2b (photo: hair style)	3.27 (.60)	3.15 (.64)	0.31	.51 (.28)

Note: Numbers inside parentheses are standard deviations.

Analysis of a within-participant correlation indicated that perceived Western cultural identification was associated with greater liking (M = 0.47, SD = 0.25, t = 11.55, p < .001) and perceived extraversion (M = 0.48, SD = 0.22, t = 13.33, p < .001).³

In Study 1b, audio information was removed from the video to examine whether cultural identification is perceived from nonverbal behaviors alone.

Study 1b

Method

Forty-two students from the same university (18 males and 24 females) arrived in groups of 10 to 15, were individually seated in front of a workstation, and viewed the same video clips in one of the two randomized orders. Audio was muted. After each video clip, the participants answered the same set of questions asked in Study 1a. In 3.6% of the cases, the participants indicated knowing a target prior to the study, and the data for these cases were discarded.

Results

The participants differentiated targets with strong (M = 3.49, SD = 0.61) and weak Western cultural identification (M = 3.14, SD = 0.63), t(41) = 4.88, p < .001, d = 0.75. Perceived Western cultural identification was correlated with perceived extraversion (M = 0.59, SD = 0.22, t = 14.05, p < .001) and liking (M = 0.53, SD = 0.27, t = 10.86, p < .001). Cultural identification was perceived from nonverbal behaviors alone.

Discussion for Study 1

We found initial evidence that cultural identification can be accurately perceived from a brief observation of social behaviors. Moreover, this perception was associated with the perceived extraversion and liking.

Study 2 was designed to determine nonverbal cues that enable a cultural identification judgment. Specifically, we extracted still images from the videos and prepared three different versions: showing the whole body, faces only (removing anything below neck), and facial features only (removing any traces of hairstyle). Based on prior work indicating that controllable aspects of appearance such as clothing and hairstyle convey cues regarding one’s identity (Rule et al., 2008), Study 2 focused on the role of these cues in enabling cultural identification inference.

Study 2 also tested the role of stereotype in facilitating cultural identification inference. Specifically, we examined whether perceived extraversion facilitates perception of Western cultural identification. This hypothesis was tested in two ways. First, we examined whether perceivers who strongly associate extraversion with Western culture were more accurate in differentiating targets’ Western cultural identification relative to perceivers with a weaker stereotype. Second, we examined whether perceivers who themselves identify with Western culture are more accurate in perceiving targets’ Western cultural identification. To the extent that individuals who strongly identify with a particular culture are better versed with that culture (Hong et al., 2007), their judgment of others’ cultural identification are expected to be well calibrated.

Study 2a

Method

Phase 1: Creating stimuli

From the videos used in Study 1, several still images for each of the 19 targets were extracted. The images showed a target sitting on a chair against a plain white background. Attempts were made to extract images that showed little sign of emotion. We focused on neutral expressions to examine perceptions of cultural identification that are not tied to any particular emotion expressions. The raters who were blind to the targets’ cultural identification (n = 11) were asked whether any emotion was expressed in each image (yes or no). For each target, a picture that was rated as most neutral was selected. On average, the selected images were rated as not showing any emotion by 78% of the raters, and this proportion did not differ as a function of the targets’ cultural identification (p = .84).

Then, we created two additional versions of the images (Figure 1). In one version, the images showed only the targets’ faces (removing anything below the neck), and in another version, the images showed only facial features (removing any traces of hairstyle).

Figure 1.

Sample images

Phase 2: Perceiving cultural identification

One hundred and forty-seven students from the same university (70 males and 77 females) were randomly assigned to three image conditions (i.e., whole body, face, or facial features). In each condition, the images were presented in a random order.

After viewing each picture, the judges rated the targets on perceived extraversion and Western cultural identification. We added one item on perceived cultural identification. Specifically, in addition to the perceived liking and perceived influence of Western culture questions, the judges rated the targets directly on his or her perceived identification with Western culture (“How much do you think this person identifies with Western culture?”), with response options ranging from 1 (not at all) to 6 (extremely). Three items were highly intercorrelated across the participants (average within-participant α = .88), and the analysis was conducted on the composite score.

Finally, participants completed the VIA (Ryder et al., 2000) assessing their identification with Western culture.

Results and Discussion

Perceived Western cultural identification

The ratings on perceived cultural identification were analyzed by an ANOVA in which a within-subject factor of the targets’ Western cultural identification (high or low) was crossed with a between-subject factor of image conditions (whole body, face, or facial features). The main effect of the targets’ cultural identification was significant, F(1, 144) = 42.16, p < .001: The perceivers differentiated targets who strongly identified with Western culture (M = 3.10, SD = 0.71) from targets who did not identify with Western culture (M = 2.89, SD = 0.67), d = 0.49. However, this effect was qualified by a significant interaction with image conditions, F(2, 144) = 15.47, p < .001. Although the perceivers accurately differentiated the targets’ cultural identification when the images showed the whole body, t(47) = 6.45, p < .001, d = 0.93, or face, t(50) = 4.87, p < .001, d = 0.68, this was not the case when only facial features were shown, t(47) = −0.78, p = .44, d = −0.11. The main effect of image condition was not significant, F(2, 144) = 1.16, p = .26.

Does perceived extraversion moderate judgment of Western cultural identification?

Next, we examined whether individual difference in the stereotype associating extraversion and Western cultural identification (M = 0.56, SD = 0.21, t = 24.62, p < .001) moderates the perception of Western cultural identification. In the hierarchical linear modeling (HLM) analysis, the targets’ cultural identification was entered as a Level 1 factor and the perceivers’ stereotype as a Level 2 factor in predicting perceptions of Western cultural identification. The analysis excluded the facial features condition as the participants failed to differentiate the targets’ cultural identification in that condition. This analysis revealed a significant interaction between targets’ cultural identification and the perceivers’ stereotype in predicting perceived Western cultural identification (b = .79, p < .001).⁴ Perceivers who had a stronger stereotype were more accurate in differentiating the targets’ cultural identification (b = .51, p < .001) relative to perceivers with a weaker stereotype (b = .19, p < .001). Supporting the hypothesis, a stereotype of Western culture facilitated perceptions of Western cultural identification.

Does judge’s cultural identity moderate judgment of Western cultural identification?

We also examined whether perceivers’ cultural identification moderated their perception of Western cultural identification. In the HLM analysis, the targets’ cultural identification was entered as a Level 1 factor and the perceivers’ cultural identification as a Level 2 factor in predicting perceptions of Western cultural identification. The analysis, which excluded the facial features condition, revealed a marginally significant cross-level interaction (b = .09, p = .09), indicating that perceivers who strongly identified with Western culture were more accurate in differentiating the targets’ cultural identification (b = .43, p < .001) relative to perceivers who weakly identified with Western culture (b = .28, p < .001). Confirming the prediction, judges who strongly identified with Western culture were better calibrated in perceiving Western cultural identification.

Interestingly, when hairstyle was deleted from the images, the targets’ Western cultural identification was no longer perceived. Previously, Rule and colleagues (2008) found that controllable aspects of appearance, such as hairstyle, convey strong cues about male sexuality. Similarly, current findings suggest that hairstyles convey strong cues regarding Western cultural identification among Hong Kongers. An additional test of this claim would be to present the targets’ hairstyles on their own, dissociated from facial features, and examine whether Western cultural identification could be perceived. Study 2b examined this possibility.

Study 2b

Method

From the images used in Study 2a, we created images that showed only the targets’ hairstyles (Figure 1). Sixty-five students from the same university (32 males and 33 females) saw these images presented in a randomized order. The participants were asked to form an impression of the targets based only on their hairstyle. Targets were rated on the same set of questions as in Study 2a.

Results

The participants differentiated targets with strong (M = 3.27, SD = 0.60) and weak (M = 3.15, SD = 0.64) Western cultural identification, t(65) = 2.54, p = .014, d = 0.31,⁵ and perceived Western cultural identification was correlated with perceived extraversion (M = 0.51, SD = 0.28, t = 12.78, p < .001). These findings suggest the role of hairstyle in conveying cues regarding Western cultural identification among Hong Kongers.

Discussion for Study 2

In sum, Study 2 found that the perception of cultural identification is possible, even from still images of targets posing with neutral facial expressions. We also found three moderators: Western cultural identity was perceived more accurately when perceivers strongly associated it with extraversion, strongly identified with Western culture (thought this effect was marginal), and when the targets’ hairstyle information was available.

In Studies 1 and 2, the judges constantly associated perceived Western cultural identification with extraversion, and individual difference in this association moderated the cultural identification inference. These findings support the hypothesis that perceived extraversion facilitates judgments of Western cultural identification. To test this process directly, Study 3 manipulated the targets’ levels of confidence, which is an ascendance aspect of extraversion (Lucas et al., 2000), and examined whether the targets induced to appear confident were seen as more strongly identifying with Western culture. With random assignment, the role of hairstyle was controlled in this study. That is, this study examined the role of perceived extraversion independently of the cues conveyed by hairstyle.

Study 3

Method

Phase 1: Creating stimuli

Sixteen participants from the same university were randomly assigned to either a confident (three males and four females) or a nonconfident (four males and five females) condition. The participants were instructed to prepare a 5-min speech about a time in which they felt very confident or not at all confident. The speech was videotaped in the same manner as in Study 1. After the speech, the participants responded to questions about their liking and influence of Western culture. The two groups (confident vs. nonconfident) did not differ on these measures: liking (p = .95) and influence (p = .26).

From the videos, a research assistant who was blind to the targets’ condition extracted one still image per target showing the whole body and a neutral facial expression. Then, a group of 11 raters evaluated whether any emotion was expressed in each image. On average, the selected images were rated as not showing any emotion by 63% of the raters, and this proportion did not differ between the conditions (p = .97).

Phase 2: Perceiving cultural identification

Fifty-six students from the same university (18 males and 38 females) participated in an on-line survey in which 16 images were randomly ordered. For each image, the participants rated the target on perceived identification with Western culture, using the same set of items as in Study 1. In 3.2% of the cases, the participants indicated knowing a target prior to the study, and the data for these cases were discarded.

Results and Discussion

A significantly greater identification with Western culture was perceived among the targets in the confident condition (M = 3.05, SD = 0.79) relative to those in the nonconfident condition (M = 2.92, SD = 0.87), t(55) = 2.53, p = .014, d = 0.34. Manipulating the targets’ confidence produced different perceptions of Western cultural identification.

This finding supports the role of perceived extraversion in signaling Western cultural identification. Importantly, this finding was obtained independently of the cues conveyed in hairstyle. In sum, findings from Studies 2 and 3 suggest that hairstyle and extraversion represent independent channels of information for perceiving Western cultural identification

Study 4 Overview

Findings thus far indicate that the perceivers inferred Western cultural identification from two cues stereotyped with Western culture among Hong Kongers: one pertaining to hairstyle and another pertaining to cues of extraversion. This rationale also suggests that individuals unfamiliar with such a stereotype may fail to perceive Hong Kongers’ identification with Western culture.

One finding consistent with this rationale is the pattern in Study 2a that the judges’ cultural identification moderated their inference, to the extent that weaker Western cultural identification indicated weaker knowledge of Western culture. However, this rationale is indirect and the pattern was marginal. Study 4 examined this issue directly. We invited two groups of individuals who are relatively unfamiliar with Hong Kong, exchange students from Western countries (Study 4a) and an American community sample (Study 4b). To the extent that these individuals, particularly the latter group, lack exposures to Hong Kongers to learn about cues that signal Western cultural identification, these individuals may fail to perceive Hong Kongers’ Western cultural identification.

Study 4a

Method

Thirty-five exchange students (18 males and 17 females) of non-Asian ethnicities from various Western nations (15 Americans, 4 French, 3 Australians, etc.), studying at the same university as in previous studies, were approached at various locations on campus and they agreed to participate in the study for HK$20. On average, the participants had lived in Hong Kong for about 6 months (M = 5.79, SD = 5.33, range = 2-23 months).

In a paper-and-pencil questionnaire, the participants were presented with the still images used in Study 2, which were printed in color and showed the target’s whole body. Each target was rated on his or her perceived identification with Western culture and perceived extraversion.

Images were sequenced in two randomized orders, and the two versions were randomly assigned to the participants. In 0.3% of the cases, the participants indicated knowing a target prior to the study, and the data for these cases were discarded.

Results

The participants differentiated the targets’ Western cultural identification—strong identification targets: M = 3.59, SD = 0.71; weak identification targets: M = 3.23, SD = 0.65; t(34) = 5.38, p < .001, d = 0.91 (Table 2). In addition, participants associated perceived Western cultural identification with perceived extraversion (M = 0.56, SD = 0.23, t = 11.49, p < .001). Although this finding was unexpected, the exchange students might have had sufficient exposures to Hong Kong (even before coming to Hong Kong) to learn about the cues of Western cultural identity among Hong Kongers. For this reason, Study 4b recruited participants less exposed to Hong Kong.

Table 2.

Perceived Cultural Identification and Its Correlation With Perceived Extraversion (Studies 4)

Study	Strong identification targets	Weak identification targets	Effect size (d)	Average within-participant correlation with perceived extraversion
Study 4a (exchange students)	3.59 (.71)	3.23 (.65)	.91	.56 (.23)
Study 4b (American community sample)	3.87 (.76)	3.81 (.78)	.20	.54 (.29)

Note: Numbers inside parentheses are standard deviations.

Study 4b

Method

Forty-eight Americans (18 males and 30 females; age: M = 35.44, SD = 11.41; 50% completed university education) were recruited via MTurks. The same materials used in Study 4a were presented online.

Results

The participants failed to differentiate the targets’ Western cultural identification (strong identification target: M = 3.87, SD = 0.76; weak identification target: M = 3.81, SD = 0.78); t(47) = 1.38, p = .17, d = 0.20 (Table 2). Nevertheless, the participants associated perceived Western cultural identification with perceived extraversion (M = 0.54, SD = 0.29, t = 11.29, p < .001). The finding supported the hypothesis in that perceivers unfamiliar with Hong Kong (thus unfamiliar with cues of Western cultural identity among Hong Kongers) were unable to discriminate the targets’ Western cultural identification.

Discussion for Study 4

In both samples, perceived extraversion was associated with perceived Western cultural identification. To the extent that extraversion is one self-stereotyped attribute among many Westerners (Terracciano et al., 2005), the participants might have applied this knowledge to infer the targets’ Western cultural identification. Nevertheless, between the two samples, only the exchange students accurately differentiated Hong Kongers’ Western cultural identification.

Regarding the cause of the divergence, one possibility is that the American participants, but not the exchange students, inaccurately perceived the targets’ extraversion, which in turn might have led to inaccurate perceptions of Western cultural identification. However, evidence does not support this account. The ratings of perceived extraversion were correlated with the targets’ self-report extraversion similarly between the two samples, Study 4a: r(19) = .42, p = .08; Study 4b: r(19) = .48, p = .04. This replicates past research in showing cross-cultural agreement in extraversion judgment (Albright et al., 1997).

Another possibility is that the divergence reflects different familiarity with hairstyle associated with Western culture in Hong Kong. That is, the finding seems to suggest that the exchange students were familiar with the role of hairstyle in signaling Western cultural identity among Hong Kongers and used this knowledge to guide their judgment. In contrast, the failure of the American participants seems attributable to the lack of this knowledge.

General Discussion

Studies reported here found that cultural identification is perceived at a zero acquaintance situation. We also found the role of an extant stereotype in enabling this perception. This was shown in a few different ways: (a) the perceivers’ stereotype moderating their cultural identification judgment (Study 2a), (b) manipulation of a stereotyped trait in the targets resulting in different perceptions of cultural identity (Study 3), and (c) a failure to correctly perceive cultural identification when stereotyped cues were unavailable—when hairstyle information was removed from the picture (Study 2a) and when the perceivers were unfamiliar with the role of this cue in signaling Western culture identity (Study 4b). Prior research found the role of stereotype in facilitating perceptions of nationality (Marsh et al., 2007). The current research found that the similar process is at play in perceiving cultural identification.

We also found implications of cultural identification inference in that targets perceived as identifying with Western culture were better liked (Study 1). This finding is consistent with a theory of thin slicing that personal attributes are recognized from thin slices to the extent that they enable an affective evaluation regarding the pleasantness or safety in interacting with that person (Ambady & Rosenthal, 1992). This suggests that cultural identification is a cue in social interaction that implicates a quick approach-avoidance judgment. A perception of Western cultural identification may also implicate attributes beyond liking. For example, one may infer individualistic values from such a perception. Such an inference, in turn, may trigger individualistic mind-set as an appropriate schema for subsequent interaction with that person. Future research should examine these possibilities.

Contrary to the past findings that neutral expressions convey relatively weak information regarding the targets’ cultural membership (Marsh et al., 2003, 2007), we found that even neutral face convey rich signals for Western cultural identification. Although this inference is limited by our coding which was done by novice raters and might have overlooked some subtle cues of emotions, it is nevertheless intriguing that an image showing a neutral face conveyed as much information about a target’s Western cultural identification as a 60-s video did—The average effect size for Study1 did not differ from the average effect size for Study 2a (excluding the facial features condition), Q = .92, p = .34. One possibility is that cues of Western cultural identification—“Western hairstyle” as well as extraversion—are easily detectable, even in neutral expressions (Naumann, Vazire, Rentfrow, & Gosling, 2009). In turn, this consideration suggests that neutral expressions may not be sufficient in judging cultural identification stereotyped with less visible traits (e.g., conscientiousness). Future research should examine this issue.

One unanticipated finding was the positive correlation between the targets’ self-reported extraversion and their self-reported Western cultural identification. In cross-cultural research, a relationship between personality and cultural identity has been attributed to a shift in personality as individuals acculturate to a new culture (e.g., McCrae, Yik, Trapnell, Bond, & Paulhus, 1998). However, an alternative account based on self-selection—that individuals are pulled toward a cultural tradition with which a particular personality profile fits—is also plausible in culturally diverse societies. Future research examining these possibilities would be important in understanding the nature of the relationship between personality and cultural identification.

Finally, the current research makes contribution to the contemporary debate between multiculturalism and assimilationism (Plaut, 2010). Studies reported here found that cultural identification is perceived even from a minimal amount of information, and such a perception implicates subsequent interpersonal interactions. These findings suggests that cultural identification plays a role in lay theories of social behaviors, similar to other attributes that are discerned at zero acquaintances like personality traits, SES, sexual orientation, and so forth. Based on these findings, the emphasis on disregarding cultural background advocated in assimilationism seems divorced from the tendency to perceive and extract information from others’ cultural identification even at a zero acquaintance situation. In sum, our findings contribute new evidence in assessing the viability of assimilationism as an approach to cultural diversity.

Footnotes

Acknowledgements

We thank Emma Buchtel and Sylvia Chen for comment.

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 work was supported by the Direct Allocation Grant from the Chinese University of Hong Kong (Grant Number 2021102).

Notes

References

Albright

Kenny

D. A.

Malloy

T. E.

(1988). Consensus in personality judgments at zero acquaintance. Journal of Personality and Social Psychology, 55, 387-395.

Albright

Malloy

T. E.

Dong

Kenny

D. A.

Fang

Winquist

(1997). Cross-cultural consensus in personality judgments. Journal of Personality and Social Psychology, 72, 558-569.

Ambady

Hallahan

Conner

(1999). Accuracy of judgments of sexual orientation from thin slices of behavior. Journal of Personality and Social Psychology, 77, 538-547.

Ambady

Rosenthal

(1992). Thin slices of expressive behavior as predictors of interpersonal consequences: A meta-analysis. Psychological Bulletin, 111, 256-274.

Apfelbaum

E. P.

Norton

M. I.

Sommers

S. R.

(2012). Racial colorblindness: Emergence, practice, and implications. Current Directions in Psychological Science, 21, 205-209.

Benet-Martínez

Leu

Lee

Morris

M. W.

(2002). Negotiating biculturalism. Journal of Cross-Cultural Psychology, 33, 492-516.

Chen

S. X.

Bond

M. H.

(2010). Two languages, two personalities? Examining language effects on the expression of personality in a bilingual context. Personality and Social Psychology Bulletin, 36, 1514-1528.

Cunningham

M. R.

Roberts

A. R.

Barbee

A. P.

Druen

P. B.

C. H.

(1995). “Their ideas of beauty are, on the whole, the same as ours”: Consistency and variability in the cross-cultural perception of female physical attractiveness. Journal of Personality and Social Psychology, 68, 261-279.

Ekman

Sorenson

E. R.

Friesen

W. V.

(1969). Pan-cultural elements in facial displays of emotion. Science, 164, 86-88.

10.

Elfenbein

H. A.

Ambady

(2002). On the universality and cultural specificity of emotion recognition: A meta-analysis. Psychological Bulletin, 128, 203-235.

11.

Hong

Wan

Chiu

(2007). Multicultural identities. In Kitayama

Cohen

(Eds.), Handbook of cultural psychology (pp. 323-345). New York, NY: Guilford.

12.

Jetten

Postmes

McAuliffe

B. J.

(2002). “We’re all individuals”: Group norms of individualism and collectivism, levels of identification and identity threat. European Journal of Social Psychology, 32, 189-207.

13.

Kraus

M. W.

Keltner

(2009). Signs of socioeconomic status. Psychological Science, 20, 99-106.

14.

Lucas

R. E.

Diener

Grob

Suh

E. M.

Shao

(2000). Cross-cultural evidence for the fundamental features of extraversion. Journal of Personality and Social Psychology, 79, 452-468.

15.

Marsh

A. A.

Elfenbein

H. A.

Ambady

(2003). Nonverbal “accents.” Psychological Science, 14, 373-376.

16.

Marsh

A. A.

Elfenbein

H. A.

Ambady

(2007). Separated by a common language. Journal of Cross-Cultural Psychology, 38, 284-301.

17.

McCrae

R. R.

Yik

M. S. M.

Trapnell

P. D.

Bond

M. H.

Paulhus

D. L.

(1998). Interpreting personality profiles across cultures: Bilingual, acculturation, and peer rating studies of Chinese undergraduates. Journal of Personality and Social Psychology, 74, 1041-1055.

18.

Naumann

L. P.

Vazire

Rentfrow

P. J.

Gosling

S. D.

(2009). Personality judgments based on physical appearance. Personality and Social Psychology Bulletin, 35, 1661-1671.

19.

Plaut

V. C.

(2010). Diversity science: Why and how difference makes a difference. Psychological Inquiry, 21, 77-99.

20.

Preacher

Curran

Bauer

(2006). Computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational and Behavioral Statistics, 31, 437.

21.

Rule

N. O.

Ambady

Adams

R. B.

Jr. Macrae

C. N.

(2008). Accuracy and awareness in the perception and categorization of male sexual orientation. Journal of Personality and Social Psychology, 95, 1019-1028.

22.

Rule

N. O.

Ambady

Adams

R. B.

Jr. Ozono

Nakashima

Yoshikawa

Watabe

(2010). Polling the face: Prediction and consensus across cultures. Journal of Personality and Social Psychology, 98, 1-15.

23.

Ryder

A. G.

Alden

L. E.

Paulhus

D. L.

(2000). Is acculturation unidimensional or bidimensional? A head-to-head comparison in the prediction of personality, self-identity, and adjustment. Journal of Personality and Social Psychology, 79, 49-65.

24.

Terracciano

Abdel-Khalek

Adam

Adamovova

Ahn

McCrae

R. R.

(2005). National character does not reflect mean personality trait levels in 49 cultures. Science, 310, 96-100.

25.

Verkuyten

(2005). Ethnic group identification, and group evaluations among minority and majority groups: Testing the multiculturalism hypothesis. Journal of Personality and Social Psychology, 88, 121-138.