Investigating direct and indirect globalization-based acculturation

Cultural globalization affects most people around the world in contemporary, modern societies. The resulting intercultural contact have been examined using the theory of globalization-based acculturation. However, little is known about possible differences and similarities in processes underlying the effects of direct (e.g., through contact with immigrants) and indirect (e.g., engagement with cultural elements through media) forms of new cultural exposure. Drawing on the contact hypothesis, social identity theory, and symbolic threat theory, we examined whether perceived intercultural threat and local and global identities would explain whether both forms of contact result in multicultural acquisition or in ethnic protection. In Study 1 (N = 402), indirect, but not direct, intercultural contact was positively associated with multicultural acquisition; and both types of intercultural contact were negatively linked with ethnic protection. Global identity significantly mediated the association of both direct and indirect intercultural contact with both multicultural acquisition and ethnic protection, whereas perceived cultural threat only significantly mediated the associations of direct intercultural contact with multicultural acquisition and ethnic protection. In Study 2 (N = 424), higher levels of ethnic protection, and lower levels of multicultural acquisition, emerged in the experimental group primed with indirect, versus direct, intercultural contact. Furthermore, intercultural threat was negatively, and global identity positively, associated with multicultural acquisition, while intercultural threat was positively, and global identity negatively, associated with ethnic protection. Results are discussed in relation to similarities and differences across direct and indirect intercultural contact, providing a nuanced understanding of contemporary intercultural contact and globalization-based acculturation among majority populations.


Introduction
Cultural globalization has potentiated intercultural experiences in people's everyday lives within most contemporary societies. Consequently, the psychological study of globalization has advanced as a central field of research, emphasizing the importance and ramifications of intercultural interactions (Hong & Cheon, 2017;Ozer, 2019). Within the acculturation-based psychological approach to intercultural contact (Berry, 2011), new theoretical developments have examined the ways in which individuals shape, react to, and adapt to globalized new cultural influences (Chen, Benet-Martínez, & Bond, 2008;Ferguson & Bornstein, 2012;Kunst, Lefringhausen, Sam, Berry, & Dovidio, resubmitted). These theoretical developments advance classical acculturation theory and research not only to examine cultural transitions based on international relocation, but also to investigate the acculturation processes experienced by people within their home country.
Hence, this new line of research examines acculturation processes among the majority population in addition to that of minority groups (Haugen & Kunst, 2017;Kunst, Lefringhausen, Skaar, & Obaidi, 2021;Lefringhausen & Marshall, 2016). Such acculturative phenomena include both (1) direct intercultural contact and (2) indirect intercultural contact (Ferguson & Bornstein, 2012;Ozer, 2019). In the present article, we define direct intercultural contact as experiences of face-to-face interaction with people from foreign cultures (e.g., international immigrants). Such contact greatly fluctuates across time and situations in terms of frequency, valence (e.g., positive and negative), and in terms of how voluntary/involuntary the experiences are. Indirect intercultural contact refers to experiences of engaging with foreign cultural elements through media and Internet. Such experiences are primarily experienced as voluntary and positive (i.e., listening to foreign music or exploring the Pokémon universe). From these different forms of intercultural contact, a central question emerges: do direct and indirect globalization-based acculturation represent aspects of the same phenomenon; and what are the similar and different processes associated with direct versus indirect globalization-based acculturation?
Globalization relates to several dimensions such as political, economic, technological, and cultural areas of development through increasingly dense networks of interconnection and interdependence across geographical distance (Tomlinson, 2007). A country's degree of globalization can be quantified using the Konjunkturforschungsstelle (KOF) globalization index (Dreher, Gaston, & Martens, 2008). This index uses markers such as trade in cultural goods and number of McDonald's and IKEA stores. According to the index, Denmark-a small Scandinavian country-represents one of the most globalized societies in the world, as reflected in a cultural globalization score of 90.89 in 2017. For comparison, the world average KOF cultural globalization score was 54.53 in 2017 (Gygli, Haelg, Potrafke, & Sturm, 2019).
Besides new cultural influences affecting individuals indirectly through media, entertainment, and consumerism that often originate from foreign countries (e.g., United States, European countries, or Japan), Denmark has experienced a relatively recent increase in immigration. In 2018, a report concluded that immigrants represented 10 % of the Danish population and, of these, 58 % originated from non-Western countries. Among the European immigrants, Poles, Germans, and Romanians represented the three largest groups, whereas immigrants from Syria, Turkey, and Iraq constituted the largest groups of non-Western immigrants (Danmarks Statistik, 2018). As a consequence of this and of previous waves of migration, Denmark has become an ethnically diverse society, in which a growing number of Danes are directly exposed to new cultural traditions and values, often on a daily basis. Hence, both direct and indirect globalization-based acculturation processes take place in contemporary Danish society.
Acculturation psychology has provided a comprehensive empirical investigation of how immigrants adapt to their new cultural environment (Sam & Berry, 2016) and how such adaptation is influenced by the context of reception (Horenczyk et al., 2013). However, little is known about how majority populations (i.e., non-immigrant nationals in the destination society) are influenced by new cultural influences within their home country (Haugen & Kunst, 2017). To facilitate a more nuanced understanding of the globalization-based acculturation process experienced by the majority populations, we experimentally and cross-sectionally investigated similarities and differences in direct and indirect intercultural processes within the context of Denmark. That is, we examined whether globalized influences (e.g., indirect through media) prompt openness whereas others (e.g., direct through immigration) promote a greater degree of defensiveness. Accordingly, we investigated how reactions to direct and indirect globalization-based acculturation differed with regard to previous experiences of intercultural contact, global/local identity configuration, and perceived cultural threat. Consequently, our study examines how a majority population in a Western country adapts to cultural globalization, expanding the focus of the classical acculturation literature and nuancing our understanding of globalized intercultural contact.

Globalization-based acculturation
Psychological research on globalized cultural interaction, multiculturalism, and cultural mixing has flourished with the acknowledgement that not only immigrants, but also host nationals, negotiate multiple cultural elements in their everyday lives (Haugen & Kunst, 2017;Hong & Cheon, 2017;Lefringhausen & Marshall, 2016). Within acculturation psychology, the classical definition of acculturation, which refers to "those phenomena which result when groups of individuals having different cultures come into continuous first-hand contact, with subsequent changes in the original culture patterns of either or both groups" (Redfield, Linton, & Herskovitz, 1936, p. 149), has been expanded. To capture acculturation processes initiated by globalization, Ferguson (2013) defined acculturation as "what happens when groups or individuals of different cultures come into contact-whether continuous or intermittent, firsthand or indirect-with subsequent changes in the original culture patterns of one or more parties" (p. 249). This expansion of the definition of acculturation includes indirect and intermittent cultural interaction (e.g., exposure to social media) that occurs through globalization. As a result of globalization-based acculturation, individuals are thought to selectively integrate cultural elements from their local culture and an often media-transferred global culture into their self-identity without relocating internationally (Chen, Lam et al., 2016;Chen, Leung et al., 2016). With increasing connectivity across geographical distance, in addition to global economic, political, and cultural interaction, a more or less coherent global cultural stream is emerging. Consequently, in globalized societies people can develop a global identification that includes awareness of practices, values, and lifestyles reflecting the global culture. This global identity augments one's local identity, which is rooted in local circumstances and traditions (Arnett, 2002).
The theoretical understanding of globalization-based acculturation is relatively new (Chen et al., 2008), and little is known about differences in the ways in which such intercultural contact is experienced and about the reactions it produces. Indeed, it is essential to compare direct (interacting with immigrants and tourists) versus indirect (media exposure) forms of globalized intercultural contact. A primary difference between these two forms of intercultural contact involves the extent to which social-psychological reactions (such as multicultural endorsement versus ethnic protection) are activated. Furthermore, such reactions are influenced by prior intercultural contact. We introduce these concepts in the following sections.

Reactions to globalization-based acculturation
Through globalization-driven intercultural contact and interaction, people develop new forms of cultural hybridity, or polycultural configurations, with multiple and combined cultural affiliations (Morris, Chiu, & Liu, 2015). Such integrative reactions to novel cultural exposure reflect a proactive response to new cultural learning that involves acquiring new cultural elements. This form of reaction has been termed multicultural acquisition (Chen, Lam et al., 2016;Chen, Leung et al., 2016). However, other exclusionary reactions to new cultural influences are shaped by simultaneous exposure to dissimilar cultures, perceived cultural incompatibility, as well as perceived cultural contamination and low individual need for cognition (e.g. engaging in and enjoying activities that require thinking; Torelli, Chiu, Tam, Au, & Keh, 2011). That is, new cultural influences experienced as intrusive cultural mixing or as threatening to one's ethnic culture are associated with exclusionary reactions to cultural globalization, which are further related to strong local cultural identification and to low scores on personality traits such as openness (Chen, Lam et al., 2016;Chen, Leung et al., 2016;Shi, Shi, Luo, & Cai, 2016).
Such exclusionary and defensive reactions to globalization have been termed ethnic protection and are characterized by an enhanced affirmation of one's national culture during exposure to other cultures (Chen, Lam et al., 2016;Chen, Leung et al., 2016). Hence, whereas the general integrative reaction to globalization and cultural mixing has been operationalized as multicultural acquisition through borrowing and adapting foreign ideas and knowledge, the exclusionary reaction has been operationalized as ethnic protection, reflecting the perception of other cultural streams as incompatible and impossible to integrate (Chen, Lam et al., 2016;Chen, Leung et al., 2016;Morris et al., 2015). These respectively proactive and defensive responses to cultural globalization can be thought to relate to individual differences in global orientation, and both responses include affective, behavioral, and cognitive components. Multicultural acquisition and ethnic protection encompass (1) learning and using new languages, obtaining new cultural knowledge and experiences and learning new cultural customs, norms, and traditions or (2) sticking to one's cultural norms and practices despite the presence of other cultural influences, holding essentialist beliefs about cultural groups, and objecting to intercultural contact (Chen, Lam et al., 2016;Chen, Leung et al., 2016). As such, these orientations map onto the two dimensions of majority-group acculturation, namely adoption of other cultures and maintenance of one's heritage culture, respectively (Haugen & Kunst, 2017;Lefringhausen & Marshall, 2016). However, national identification is not necessarily defensive as conceptualized in ethnic protection (e.g., see Lefringhausen, Ferenczi, Marshall, & Kunst, 2021). Multicultural acquisition is positively associated with openness to experience, tolerance, and psychological adaptation, whereas ethnic protection is negatively associated with these variables (Chen, Lam et al., 2016;Chen, Leung et al., 2016). Notably, global identity has been associated with a positive attitude toward intercultural experiences, whereas local identification has often been associated with perceived cultural contamination or threat (Reese, Rosenmann, & Cameron, 2019).

Intercultural contact
The contact hypothesis (Allport, 1954;Pettigrew & Tropp, 2006) states that people's attitudes toward outgroups (i.e., groups to which they do not belong) are shaped by previous intergroup contact. More specifically, contact with outgroup members can reduce stereotypes and prejudice, thereby promoting tolerance and acceptance (Weber & Crocker, 1983). Although this hypothesis is strongly supported and strengthened by certain conditions (e.g., equal status, cooperative norms, common goals, and institutional support), the frequency and valence of intergroup contact appear especially pivotal, as negative contact can increase stereotypes and compromise multicultural engagement (Meleady & Forder, 2019). Accordingly, we might expect that globalization-based intercultural contact would exert overall positive effects. Indeed, previous research has found that prior intercultural contact with foreigners is associated with positive and/or negative reactions to new cultural influences, where such reactions can be categorized in terms of global (multicultural versus ethnic-protectionist) orientation (Chen, Lam et al., 2016;Chen, Leung et al., 2016).
Research on globalization-based acculturation has found that one's number of close friends from ethnic outgroups is positively associated with multicultural acquisition and negatively linked with ethnic protection (Chen, Lam et al., 2016;Chen, Leung et al., 2016). Furthermore, multicultural acquisition has been associated with frequency of intercultural contact, reflecting greater motivation for interacting with people from other cultures as well as positive experiences with such intercultural contact (Chen, Lam et al., 2016;Chen, Leung et al., 2016). Accordingly, research on majority group members' acculturation process suggests that maintaining one's majority culture and not adopting any immigrant cultural elements (the separation acculturation strategy) is associated with perceiving more ethnic identity threat and discrimination from immigrants (Haugen & Kunst, 2017). Moreover, majority individuals endorsing the separation acculturation strategy experience other cultures as more of a threat, and less as an enrichment, as compared to individuals endorsing the integration acculturation . Indeed, research has found that intercultural contact may decrease tendencies toward foreigner exclusion and that this association might be mediated by other factors such as cultural threat and social identifications (Escandell & Ceobanu, 2009;Schmid, Hewstone, Küpper, Zick, & Tausch, 2014). However, research is still needed to examine these processes vis-à-vis direct and indirect globalization-based acculturation in order to facilitate greater understanding of the complexity characterizing cultural globalization and its influences.

Direct and indirect globalization-based acculturation
Immigration within one's local context represents one direct mechanism that increases globalized intercultural contact and exposes people to other cultural influences within their own country. Within acculturation psychology, the phenomenon of acculturation has been conceptualized as a mutual process taking place among the minority and majority populations, acknowledging the importance of both the ideologies and policies of the dominant society and the acculturation attitudes of its individual members (Berry, 2011). Most research so far has investigated how minority group members acculturate, whereas the study of majority members has mostly been limited to the acculturation expectations they hold toward immigrants (Bourhis, Moïse, Perreault, & Senécal, 1997). Hence, despite the conceptualization of acculturation as a mutual process of accommodation, there has been a general lack of research on acculturative processes among majority groups (Haugen & Kunst, 2017;Horenczyk et al., 2013;Lefringhausen & Marshall, 2016). This is where the psychological acculturation approach to globalization aims to contribute by examining how majority members are reacting and adapting to new cultural influences (i.e., majority-group acculturation). Specifically, this approach focuses on the changes resulting from direct forms of globalization-based intercultural contactthat is, contact with immigrants, tourists, and others (Chen, Lam et al., 2016;Chen, Leung et al., 2016;Ozer, 2019).
In addition to direct forms of intercultural contact, globalization-based acculturation also includes indirect forms through media, which has increased due to recent developments in information technology (Chen, Lam et al., 2016;Chen, Leung et al., 2016). The primary channels of cultural influence during globalization include music, television, movies, and news, potentiated by dissemination through the internet and through social media. Through these channels of remote acculturation, people can integrate cultural elements from dissimilar cultural contexts across the world without leaving their home country (Ferguson, Tran, Mendez, & van de Vijver, 2017). As a case in point, Ferguson and Adams (2016) examined cultural orientation toward six cultural streams among South African emerging adults. Tapping into both the orientation toward direct acculturation in the multicultural South African context as well as indirect intercultural contact (or "remote acculturation") through European American and African American cultural streams, they identified four different acculturation clusters. These clusters were characterized by (1) a traditional South African orientation and three variations of combining South African cultural orientation with (2) multiculturalism, (3) European American culture, or (4) African American culture. These findings reflect the complexity of globalization-based acculturation involving several interacting cultural streams. Indirect intercultural contact has been conceptualized as a process characterized by agency and selectivity, and such cultural interactions canin the same way as direct intercultural contactbe associated with significant ramifications regarding identity development, well-being, and intergroup relations (Ozer, 2019).

Cultural threat and social identifications during globalization
The associations of intercultural contact with exclusionary versus integrative reactions to new cultural influences could be shaped by several intergroup processes. One central factor vis-à-vis attitudes toward outgroup members and prejudice during increasing intercultural and intergroup connectivity is the possibility of new cultural influences being perceived as a cultural threat (Riek, Mania, & Gaertner, 2006;Stephan & Stephan, 2013). For example, symbolic threat -reflecting the experience that one's groups beliefs and values may be at risk -has been found to predict prejudice toward Muslim immigrants (Obaidi, Kunst, Kteily, Thomsen, & Sidanius, 2018). Furthermore, symbolic threat has been found to be negatively associated with both (a) contact with Muslims and (b) endorsement of multiculturalism (González, Verkuyten, Weesie, & Poppe, 2008). Such perceived symbolic threats appear to prompt outgroup hostility across sociocultural contexts (Obaidi et al., 2018). The association between intergroup contact and perceived threat is complex as these two factors change across time and in regard to the situation. Accordingly, the valence and context of intercultural contact becomes pivotal in terms of whether it encourages future contact or is perceived as a threat (Abrams & Eller, 2016). Generally, intergroup anxiety and perceived cultural threat have emerged as mediating factors in the relationship between contact with immigrants and attitudes toward immigrants (Esses, Dovidio, Jackson, & Armstrong, 2001;Stephan & Stephan, 2013).
Social and cultural identifications represent another relevant factor that may underlie the associations between intercultural contact and attitudes toward other cultural groups. Drawing from social identity theory, identifications with one's ingroup, or with outgroups, may exert different effects on attitudes toward others and be differentially associated with possible perceived threats (Phalet, Baysu, & Verkuyten, 2010;Tajfel & Turner, 1979). In one study (Müller, 2009), watching multicultural television was associated with lower levels of perceived ethnic threat. This association was mediated by decreased identifications with one's ingroup and increased identifications with outgroups. Even without direct intercultural contact, merely imagining intergroup contact has been associated with increased identification with outgroup members and with more positive attitudes toward outgroups (Igartua, Wojcieszak, & Kim, 2019). Although perceived cultural threat and social identification have been found to represent important factors that may account for the association between intergroup contact and attitudes toward outgroups (Tausch, Tam, Hewstone, Kenworthy, & Cairns, 2007), the actual experience and knowledge gained from direct intercultural contact could decrease perceived cultural threats. In turn, such decreased cultural threats could lead to inclusionary reactions to globalization through a global identification that encompasses the diversity of information, values, and practices reflected in the global cultural stream. Indeed, intergroup bias may be reduced if people conceive of themselves as part of the same group and identify with a global culture (Gaertner & Dovidio, 2000) The association of intercultural contact with perceived threat and identifications greatly depends on both the frequency and the valence of the contact experiences (Abrams & Eller, 2016). Accordingly, direct contact that is experienced as negative and involuntary could increase the salience of local identifications, resulting in ethnic protection as a defensive and exclusionary attitude toward globalization. On the contrary, positive contact experiences could lead to more inclusive identifications that combine both local and global orientations. That is, national or local cultural identification does not necessarily transform into ethnic protection . Rather, multicultural orientation could reflect the integration of both local and global identity if one does not perceive cultural threats. Furthermore, selective and intentional indirect cultural contact could decrease perceived cultural threat and enhance one's global identification and consequently multicultural acquisition (i.e., a broadly inclusionary attitude toward cultural globalization; Ozer, 2019). Although both direct and indirect intercultural contact could enhance multicultural acquisition and decrease ethnic protection, the indirect form of contact (as more voluntary and proactive, e.g., one can choose which media to consume) could be more strongly associated with openness to new cultural influences as compared to the direct form of cultural contact, which in some cases could spur perceived intercultural threat.

The current studies
The current studies were guided by two goals comprising four hypotheses based on the theoretical argument and literature review presented in the introduction. These hypotheses address the associations of direct and indirect intercultural contact with multicultural acquisition and ethnic protection through both global and local identity as well as perceived cultural threat.
Our first goal of experimentally distinguishing reactions to direct and indirect globalization-based acculturation was pursued by testing the following two hypotheses: priming direct intercultural contact reflecting both voluntary and involuntary experiences would result in more defensive reactions (i.e., higher ethnic protectionism, more intercultural threat and local identification) than would priming indirect contact (H1). Additionally, reactions to indirect intercultural priming reflecting primarily voluntary and positive experiences would result in greater levels of multicultural acquisition through global identification as compared to the direct form of intercultural contact (H2).
Our second goal involved testing models that assume that the associations of intercultural contact with multicultural acquisition and ethnic protection would be mediated by perceived cultural threat and by identifications with the local and global contexts. Here, we examined two hypotheses. In accordance with the contact hypothesis (Allport, 1954), we hypothesized that the frequency of both direct and indirect intercultural contact would generally reflect positive experiences of contactand would consequently be positively associated with global identity and, in turn, multicultural acquisition (H3). Additionally, we expected that both previous direct and indirect intercultural contact would mitigate perceived intercultural threat and endorsement of local identity, thereby resulting in a lower ethnic protectionist orientation (H4). That is, although experimentally primed reactions to intercultural contact would prompt openness with regard to indirect contact and defensiveness in relation to direct contact, we expected that the frequency of previous intercultural contact (both direct and indirect) would be associated with greater multicultural acquisition and lesser ethnic protection through lower levels of intercultural threat and local identity and through greater endorsement of global identity.

Method Participants
Using G*Power (Faul, Erdfelder, Buchner, & Lang, 2009), we conducted a power analysis for a t-test to detect a difference between two independent means with at least small to medium-sized effects (d = .40) and 80 % power (Cohen, 1992). This power analysis produced a minimum required sample size of N = 200 to detect significant effects (alpha level of .05). Furthermore, we recruited additional participants to achieve adequate power for SEM analysis (Kline, 2015).
Participants were 402 Danish university students who were all Danish citizens with both parents born in Denmark. Regarding gender, 84.1 % were female, 15.2 % were male, and 0.7 % did not wish to state their gender. The mean participant age was 24.03 (SD = 4.62). When participants were asked to characterize their family's income, 4.5 % reported below average, 8.7 % reported just below average, 37.6 % reported average, 31.8 % indicated just above average, and 17.4 % reported above average.

Procedure
Participants were recruited through student mailing lists and relevant student Facebook groups within Denmark. Participants were assigned to receive one of two questionnaires completely at random. Both questionnaires consisted of the same scales described below. The only difference was that before answering the last scale-the Global Orientations Scale-one group (n = 193, M age = 24.2, 83.4 % female) was primed with a prompt asking them to reflect on the influences from immigration in Denmark and to fill out the questionnaire based on these reflections. The other group (n = 209, M age = 23.9, 84.7 % female) was asked to reflect on how they experience the influence from foreign cultures through media and the Internet and to answer the Global Orientations Scale with these reflections in mind. 1 Consequently, one group reported reactions to direct intercultural contact, and the other group reported reactions to indirect intercultural contact.

Measures
The following measures were presented in their original English versions in the following order (see Table 1 for Cronbach's alphas).
The Intercultural Contact Scale was developed for this study to tap into both direct contact with immigrants and indirect contact with media representing new cultural influences. The measure consists of eight items answered using a 5-point Likert scale ranging from never to often. Sample items included: "I have contact with immigrants in my everyday life" (direct) and "I enjoy entertainment from other countries" (indirect). The scale assesses frequency of both quantitative and qualitative aspects of intercultural contact. The factor structure of the items was examined through exploratory factor analysis. Principal axis factoring was used, employing promax rotation (Kappa = 4), which indicated a two-factor solution that fit the data with four items pertaining to direct and four items to indirect intercultural contact. The value of the Kaiser-Meyer-Olkin test was .77, and all loadings ranged from .53 to .89, except for one item yielding poor loadings (<.40) and high cross loadings. Consequently, this item was not included in further analyses. The two-factor solution supported the conception of direct and indirect intercultural contact as independent dimensions. 2 The local-global Identity Scale (Tu, Khare, & Zhang, 2012) is a validated scale for assessing local and global identifications. The scale consists of eight items that are answered using a 7-point Likert scale ranging from strongly disagree to strongly agree. Sample items included, "I identify that I am a global citizen" and "My heart mostly belongs to my local community".
The Intercultural Threat Scale was developed for this study to tap into a generic experience of threat toward the Danish national culture. The scale consisted of four items pertaining to culture in general as well as identity, values, and practices (suggested acculturation domains by Schwartz, Unger, Zamboanga, & Szapocznik, 2010). Sample items included, "Danish cultural identity is currently threatened by new cultural influences". All items were answered using a 7-point Likert scale ranging from strongly disagree to strongly agree. The items were examined through exploratory factor analysis. Principal axis factoring was used, employing promax rotation (Kappa = 4), which reflected a one-factor solution. The value of the Kaiser-Meyer-Olkin test was .83 and all loadings ranged from .70 to .90. 3 The Global Orientation Scale (Chen, Lam et al., 2016) measures individual differences in the psychological process of acculturating to cultural globalization. The measure consists of two subscales: multicultural acquisition (13 items, e.g., "I am curious about traditions of other cultures") as a proactive response, and ethnic protection (12 items, e.g., "My own culture is much superior to other cultures") as a defensive response to globalization. All items were answered using a 7-point Likert scale ranging from strongly disagree to strongly agree.

Analyses
For this and the second study, all statistical analyses were conducted in SPSS and Mplus 8 (Muthén & Muthén, 1998-2011. The hypothesized indirect effects model was examined using structural equation modeling (SEM) employing the ML estimator with bootstrapping (5000 resamples). Kline's (2015) criteria for evaluating model fit was used, suggesting that the following fit index cut-offs should be satisfied: Comparative Fit Index (CFI) ≥ .90, Standardized Root Mean Square Residual (SRMR) ≤ .08, and Root Mean Square Error of Approximation (RMSEA) ≤ .08. For Study 2 where the participants were experimentally primed in regard to all variables, we compared the fit of an unconstrained model where all paths were free to vary against the fit of a constrained model with all relevant paths set equal across experimental groups. The null hypothesis of invariance is evaluated by comparing the fit indices between the constrained and unconstrained models using Little's (2013) criteria (ΔCFI ≤ .010 and ΔRMSEA ≤ .010).

Experimentally examining reactions to direct and indirect globalization-based acculturation
Comparing global orientations across direct and indirect experimental priming did not indicate any significant differences. That is, the indirect intercultural contact group did not score higher on multicultural acquisition (M = 5.56, SD = 0.78, t(400), p = .857, d = 0.01) or lower on ethnic protection (M = 2.96, SD = 0.72, t(400), p = .780, p = .780, d = 0.03) as compared to the direct intercultural contact group (M multicultural acquisition = 5.57, SD = 0.79; M ethnic protection = 2.94, SD = 0.80). These results did thus not support our first (H1) and second (H2) hypotheses. Because the experimental priming failed to differentiate between the reactions to direct vs. indirect intercultural contact, the two groups were collapsed in further analysis, while controlling for possible effects of the experiment using a dummy variable for experimental condition.

Intercultural dynamics of direct and indirect globalization-based acculturation
To examine the intercultural dynamics of direct and indirect globalization-based acculturation, we estimated an indirect effects model using SEM (see Table 1 for bivariate correlations and means). The SEM model utilized fully disaggregated/latent (Interculutrual Contact Scale, Intercultural Threat Scale, & Local-Global Identity Scale) or partially aggregated (Global Orientation Scale) selfreported measurement (factor loadings ranged from .44 to .90). We controlled for the effects of experimental group, gender, age, and SES on the mediating and outcome variables. The indirect effects model provided an satisfactory fit to the data, χ 2 (391) = 857.42, .00]).

Study 1 discussion
In our first study, we did not find an experimental effect prompting more defensive reactions of ethnic protection to direct intercultural contact and greater levels of multicultural acquisition in relation to indirect intercultural contact, refuting H1 and H2. In support of H3, we found participants' frequency of both direct and indirect intercultural contact to be positively associated with multicultural acquisition through a stronger global identity. However, we only found partial support for H4, as local identification did not yield a significant effect in our model; only direct intercultural contact was associated negatively with ethnic protection through perceived intercultural threat.
These findings suggest that the proposed outcomes of the contact hypothesis pertain to both direct and indirect intercultural contact. However, our study was limited in that it could not produce an experimental effect of direct versus indirect intercultural priming. No manipulation check was included, limiting our ability of explaining this lack of experimental effect. Furthermore, the study was limited by the use of English measures for a Danish bilingual population and a sample consisting of university students-constituting a homogenous sample of well-educated youth from high SES backgrounds.

Method
In the second study, we addressed the limitations of the first study by including a more heterogeneous sample, translating the measures into Danish, and employing a more comprehensive experimental setup requiring participants to recall and write about actual intercultural experiences. Because negative contact experiences can result in negative responses (Meleady & Forder, 2019), these descriptions were expected to provide the opportunity to examine the character and quality of the reported contact experiences. Furthermore, we included an inert control group.

Participants
Using G*Power (Faul et al., 2009), we conducted a power analysis for an ANOVA to detect small effects (d = .20) with a power of .80 (Cohen, 1992). This power analysis revealed a minimum required sample size of N = 246 to detect significant effects (alpha level of .05). Again, we recruited additional participants for SEM analysis (Kline, 2015).
Participants were 424 Danish citizens with both parents born in Denmark. The majority (71.0 %) of participants were female and 29.0 % were male. The mean participant age was 46.30 (SD = 13.82). In terms of perceived socioeconomic status, 25.5 % indicated an income below average, 11.1 % reported just below average, 24.1 % reported average, 21.2 % indicated just above average, and 18.2 % reported an income above average.

Procedure
Participants were recruited through local community Facebook groups representing various Danish cities. Participants were randomly assigned to one of three experimental groups. Within the first group (direct intercultural contact; n = 135, M age = 45.0, 70.4 % female), participants were asked to reflect upon and spend at least two minutes on writing a few sentences about their actual experiences of new cultural influences in Denmark brought about by immigration. Participants in the second group (indirect intercultural contact; n = 136, M age = 45.9, 72.1 % female) were asked to reflect and write about new cultural influences in Denmark brought about by media and the Internet. Finally, in the third group (control; n = 153, M age = 47.8, 70.6 % female), participants were asked to reflect upon and write about daily chores. 4 After completing the experimental procedure, participants completed the measures listed below.

Measures
To evaluate the effects of the experimental priming, the following measures from Study 1 were translated from English to Danish using a forward-back translation procedure (see Table 2 for Cronbach's alphas): Local-global Identity Scale, The Intercultural Threat Scale, and The Global Orientation Scale.

Study 2 results
First, we examined the effects of our experimental manipulation on all our measures (see Table 2 for mean comparisons of all variables). These analyses indicated a significant effect of experimental condition on both multicultural acquisition, F(2,397) = 3.40, p = .034, η 2 = .017 and ethnic protection, F(2,394) = 3.28, p = .039, η 2 = .016. Post-hoc analyses indicated that participants in the indirect intercultural contact group (M = 4.81, SD = 1.33) reported less multicultural acquisition than did participants in the direct intercultural contact condition (M = 5.18, SD = 0.99, p = .010, d = 0.32). Moreover, participants in the indirect intercultural contact group (M = 3.44, SD = 1.03) reported greater ethnic protection than did participants in the direct intercultural contact (M = 3.16, SD = 0.84, p = .020, d = 0.31) or in the control conditions (M = 3.20, SD = 0.97, p = .037, d = 0.25). Overall, the indirect contact condition was associated with more ethnic protection and less multicultural acquisition. Although the priming for indirect cultural contact was operationalized to capture cultural representations influencing the participants through media, we found that 51 participants Note. *p < .05; **p < .01; Within this model, local identity was allowed to covary with both global identity (r = -.41**) and cultural threat (r = .37**). Furthermore, global identity was allowed to covary with cultural threat (r = -.27**). Finally, direct and indirect intercultural contact was allowed to covary (r = .09).
described experiences with media discussions of immigration and new cultural influences in Denmark. Accordingly, we found that, within the indirect contact condition, participants referring to such discussions reported significantly lower multicultural acquisition (M = 4.45, SD = 1.43), t(100) = -3.64, p < .001, d = 0.07) and higher intercultural threat (M = 3.57, SD = 1.26), t(100) = 2.13, p = .036, d = 0.04) as compared to other participants in the same indirect contact condition (M multicultural acquisition = 5.39, SD = 1.16; M ethnic protection = 3.14, SD = 0.75), potentially driving the group differences observed earlier. Indeed, when the 51 participants describing how immigrants and new cultural influences are portrayed in media were excluded from analysis, no significant differences emerged between the three experimental conditions in regard to multicultural acquisition (F(2,350) = 0.93, p = .463, η 2 = .004) and ethnic protection (F(2,347) = 1.39, p = .250, η 2 = .008). Consequently, the overall difference in global orientation appeared to be driven by a broad conception of indirect intercultural contact, including a negative portrayal of new cultural influences in the local media.
Building on Study 1, we next examined a path model in which intercultural threat, together with local and global identity, was associated with multicultural acquisition and ethnic protection (see Fig. 2). This model was estimated through SEM, including all participants across the three experimental groups. The SEM model used fully disaggregated (Intercultural Threat Scale & Local-Global Identity Scale) or partially aggregated (Global Orientation Scale) self-report measurement with all factor loadings ranging from .65 to .94. Experimental group, age, gender, and SES were entered as covariates. Multigroup SEM was then conducted to examine model equivalence across the three experimental manipulations by comparing constrained (parameters constrained to be equal across groups) versus unconstrained (parameters freely estimated for all groups) models. Examining the measurement model across the three groups did not yield significant differences in factor loadings across groups, Δχ 2 (70) = 92.35, p = .038; ΔCFI = .003; ΔRMSEA = .002. Second, examining our path model across groups also did not indicate any significant group differences, Δχ 2 (24) = 26.71, p = .318; ΔCFI = .001; ΔRMSEA = .001. We therefore concluded that only multicultural acquisition and ethnic protection were affected by our manipulation and that the relationships among intercultural threat, global and local identity, and global orientations were equivalent across the three experimental groups. Note. Statistical significance is marked by *p < .05 and **p < .01.

Study 2 discussion
Results did not support H1 and H2 as our experimental effects failed to distinguish direct from indirect intercultural contact in regard to multicultural acquisition and ethnic protection. Contrary to H2, indirect intercultural contact appeared to prompt a greater level of perceived intercultural threat as compared to either direct intercultural contact or the control group. Such perceived intercultural threat was further associated with a higher level of multicultural acquisition through global identity. An inspection of participants' written responses, however, suggested that this difference in global orientation was driven by a broader conception of indirect intercultural contact that included negative media descriptions of direct intercultural contact (i.e., through public discourse around immigration). Sensitivity analyses indicated that, excluding participants with these responses, no significant differences emerged between our experimental conditions.
Supplementing and partly replicating our findings from Study 1, our indirect effects model indicated that perceived intercultural threat and global identity, but not local identity, were associated with multicultural acquisition and ethnic protection. This finding suggests that inclusive global identification and cultural threat play key roles in globalization-based acculturation vis-à-vis contemporary intercultural relations and approaches to cultural diversity (Buchan et al., 2011).

General discussion
Our results indicate that, although direct and indirect globalization-based acculturation appear to be dissimilar phenomena, reactions to these phenomena are associated with similar identifications and reactions across the generic nature of intercultural contact. However, although prior direct intercultural contact appeared to reduce perceived cultural threat and endorsement of ethnic protection, this might not be the case for indirect intercultural contact. Both types of intercultural contact were associated with greater global identification, but only indirect intercultural contact was directly linked to multicultural acquisition. Although reactions to globalization-based acculturation appeared to be generally consistent across the two types of intercultural contact, small differences emerged as direct intercultural contact appeared to be strongly and negatively associated with ethnic protection whereas indirect intercultural contact was strongly and positively associated with multicultural acquisition. As such, both types of intercultural contact are likely needed to facilitate adaptive intergroup relations.
Across direct and indirect intercultural contact, the results of our indirect effects model were aligned with other findings regarding majority groups' reactions to new cultural influences and intergroup processes. Friendship and contact with outgroup members have been found to be associated with positive attitudes toward multiculturalism (Pettigrew & Tropp, 2006;Verkuyten & Martinovic, 2006). Additionally, consistent with our findings, a superordinate group identification has been found to moderate the association between value incongruence and negative intergroup attitudes (Guan et al., 2011), suggesting that a global identification could improve intercultural relations (Buchan et al., 2011).

Processes of intercultural relations
In contemporary globalized societies, intercultural relations have become a major area of research within and beyond the discipline of psychology. Understanding the processes and consequences of acculturation, not only among immigrants but also among majority populations exposed to various forms of new cultural influences, is key for improving intercultural relations and collaboration in globalized societies (Ozer, 2019). Overall, our results support an intercultural version of the contact hypothesis stating that intergroup contact would increase intergroup tolerance (Pettigrew & Tropp, 2006). Interestingly, Haugen and Kunst (2017) found that, among a majority sample of Norwegians, those reporting a separated acculturation strategy (maintaining their ethnic Norwegian culture without adopting immigrant cultures) were most likely to live in a multicultural environment that one may think facilitates intercultural contact. In our Study 1, we asked about direct intercultural contact in a way that most probably would reflect voluntary contact (e.g., contact with friends, with classmates, and through trade). Consequently, an important difference between our study and the study by Haugen and Kunst could involve the nature of the intercultural contact (e.g., voluntary/involuntary, public/private domain, intermittent/continuous). The voluntary aspect of direct intercultural contact (e.g., interacting with immigrant friends and colleagues) may be more closely aligned with the nature of indirect intercultural contact as compared to involuntary contact (e.g., interactions with neighbors or with fellow shoppers at a supermarket). Furthermore, voluntary contact can reflect the ways in which various sources of new cultural influences are favored or disfavoredsuch as when Danish citizens choose to socialize with people from Britain and not with people from the Middle East. Accordingly, Cheon and Hong (2020) found negative reactions among Hong Kong residents toward cultural mixing with the disfavored Chinese cultural stream and positive reactions toward cultural mixing with the favored American cultural stream. This pattern is consistent with our findings from Study 2, indicating that the media's negative portrayal of immigrants may prompt aversive reactions to global orientations reflecting disfavor toward specific immigrant cultural streams.
Acculturation processes have become a significant part of everyday life both for immigrants and for non-immigrants in many Western countries. This increased emphasis on acculturation may set in motion processes of identity negotiation, cultural threat, and intercultural inclusion/exclusion (Ozer, 2019). Experiences of cultural threat may represent an inevitable part of adapting to increasing intercultural contact, at least for some non-immigrant individuals. In our results, perceived cultural threat was associated with less prior direct intercultural contact and positively linked with ethnic protection (in Study 1) and negatively linked with multicultural acquisition (in both studies). Similarly, majority groups' perceived identity threat has been found to be significantly higher among separated individuals (i.e., those who reject immigrant cultures) compared to the integrated and marginalization clusters (Haugen & Kunst, 2017). Experiences of direct cultural contact could inform people about cultural similarity and dissimilarity between groups. Such interactions can prompt greater cultural understanding, decrease perceived cultural threat, and promote greater openness toward new cultural influences. Furthermore, such prior intercultural experiences can differentiate actual intercultural encounters from the often negative imagery reflected in media portrayals of immigration (Fryberg et al., 2012). Likewise, decreased concern about threats to one's ethnic culture could increase multicultural acquisition and potentiate intercultural contact. Such simultaneous processes could facilitate an inclusive intercultural identity as a way of transcending ingroup-outgroup distinctions around immigration and associated cultural changes (Kim, 2015). Accordingly, although defensive reactions to cultural globalization are evident, a long-term development toward global awareness, intercultural openness, and polycultural configurations could represent a likely outcome of intercultural interactions. Our results highlight the importance of a shared global identity in linking prior intercultural contact with acceptance of cultural diversity.

Similarities and differences between direct and indirect intercultural contact
Within the context of Denmark, direct intercultural contact through immigration has been a highly publicly discussed topic, especially with regard to anti-immigration discourses and intolerance toward Muslim cultural streams (Lindekilde, 2014). Although these discussions have occurred with increasing frequency at the national level, less concern has been dedicated to the effects of remote acculturation. In this regard, remote acculturation is often characterized as a voluntary process and includes less cultural distance with a primarily cultural influence from other Western countries. Such remote acculturation is often regarded as a natural societal development and as a gradual development toward a fused global culture. Such a development does not necessarily prompt oppositional attitudes toward, for example, globalization versus localization and intergroup bias (Gaertner & Dovidio, 2000;Reese et al., 2019). These differences in collectively perceived threats associated with direct versus indirect globalization-based acculturation could be reflected in our results, suggesting that perceived intercultural threat and ethnic protection may be influenced by prior direct cultural contact as well as by the media's portrayal of immigrants. In turn, media portrayals of immigrants might lead people's perceptions of intercultural contact to be shaped by assumptions rather than by actual experiences with immigrants. Our student sample in Study 1 reported a high frequency of indirect intercultural contact, suggesting that indirect intercultural contact may represent an everyday normalized experience among young Danes. Haugen and Kunst (2017) found that the primary acculturation domains influencing Norwegian students were school, food, and work. Indirect acculturation would most likely be associated with categories such as entertainment, social media, and sport, reflecting the voluntarily and selectively acquired aspects of remote acculturation (Ferguson & Bornstein, 2012).
Consistent with our results, Ward and Masgoret (2006) found that multicultural ideology and intercultural contact with immigrants were associated with attitudes toward immigrants indirectly through perceived threat. We found this association with regard to direct, but not indirect, intercultural contact (when excluding participants responding to the media's portrayal of immigration). It is possible that this finding could reflect differences regarding abstract (broad) versus concrete (specific) construals of multiculturalism. Indeed, Yogeeswaran and Dasgupta (2014) found that concrete, rather than abstract, construals of multiculturalism were more strongly associated with prejudice toward ethnic minorities and with unwillingness to engage in intergroup contact. These associations were mediated by perceived cultural identity threat and moderated by perceived cultural distance (Mahfud, Badea, Verkuyten, & Reynolds, 2018). Similarly, indirect intercultural contact might appear more abstract and associated with proximate cultural streams as compared to direct contact with immigrants originating from non-Western cultural contexts reflecting concrete intercultural experiences.
Research has found higher ingroup identification to be associated with lower endorsement of multiculturalism (Verkuyten & Martinovic, 2006). In our results from Study 1, local identification was positively associated with multicultural acquisition yet unrelated to both direct and indirect intercultural contact. In both of our studies, local identity was unrelated to ethnic protection, suggesting that identification with the local context-even in relation to perceived cultural threat-does not necessarily initiate defensive reactions to cultural globalization (Kunst et al., resubmitted). Overall, local identification did not emerge as a key variable in relating intercultural contact with reactions to cultural globalization.

Limitations
The results of our study should be considered in the light of at least four important limitations. First, given that our experimental manipulation did not operate as expected in Study 1, the analyses for that study were correlational, and consequently we cannot determine causality in our path models. Although the effects of intercultural contact on prejudice have been found to be largely unidirectional in prior work (Pettigrew & Tropp, 2006), longitudinal and experimental data are needed to draw conclusions regarding the directionality of effects within the Danish context. Second, our results may be culturally bound within the Danish context, such that other contexts with different histories of intercultural contact might yield different results. Future research could investigate how societies with more or less contested intercultural dynamics compared to those in Denmark might differ in terms of effects of direct versus indirect intercultural contact on attitudes toward multiculturalism and diversity. For example, in non-Western societies, indirect globalization-based acculturation might be experienced as a challenge to local traditions, perhaps promoting a defensive stance. Similarly, immigration might be perceived in a more negative manner in many Western societies as compared to non-Western societies. This could in part be due to perceived cultural similarity and differences, with a greater differences prompting perceptions of intercultural threat.
Third, the majority of participants in both studies were women. Although gender was controlled for in the analyses, research indicates that women report slightly higher levels of support for multiculturalism than men do (Chen, Lam et al., 2016;Chen, Leung et al., 2016). Moreover, samples collected through Facebook are likely not representative of the larger population. Recruiting participants through Facebook can result in subjects who share similar traits and characteristics, given that previous research has found significant differences in the psychological profiles of those who use (versus do not use) Facebook (Brailovskaia & Margraf, 2016). Consequently, future research is needed to identify whether the findings obtained in the present studies can be generalized using representative samples and other methods of collecting data. Fourth, our experimental manipulation was unsuccessful in differentiating direct versus indirect intercultural contact. Future research could focus on developing more effective approaches to discern possible differences between these types of intercultural contact. Such efforts could, for example, be pursued through a more comprehensive and explicit definition of indirect intercultural contact or by exposure to specific visual cultural priming reflecting both direct and indirect intercultural contact. Additionally, we found significant effects associated with negative media portrayals of immigration, suggesting that such negative discourses may prompt a more defensive stance toward cultural diversity. Future research is needed to further examine the effects of how media present intercultural relations within the local context.

Conclusion
In the present research, direct and indirect intercultural contact similarly predicted more multicultural acquisition and less ethnic protection. Path models indicated that higher endorsement of global cultural identity mediated these effects. Directbut not indirectintercultural contact was negatively associated with ethnic protection both (a) directly and (b) indirectly through lower levels of perceived cultural threat. Moreover, direct intercultural contact was indirectly and positively associated with multicultural acquisition through lower levels of perceived cultural threat. Our results hence suggest that, although reactions to cultural globalization are shaped by dissimilar intercultural processes regarding the nature of intercultural contact, endorsement of a global identity, multicultural acquisition, and ethnic protection appear to represent key constructs underlying the effects of both direct and indirect cultural contact. This finding reflects the importance of contemporary intercultural processes and supports the general study of globalizationbased intercultural ramifications among majority populations. We hope that the present study will inspire further research in this direction.