Which Partners Are More Human? Monogamy Matters More than Sexual Orientation for Dehumanization in Three European Countries

Compared to monogamous relationships, consensual non-monogamous (CNM) relationships are stigmatized. Similarly, compared to heterosexual individuals, gay men are perceived to have more promiscuous and less committed romantic relationships. Hence, CNM and same-sex male relationships are potential targets of dehumanization (i.e., denied traits considered unique of human beings). We tested the impact of monogamy and sexual orientation on dehumanization, and examined if CNM (vs. monogamous) and same-sex (vs. heterosexual) partners are dehumanized through the attribution of primary (non-uniquely human) and secondary (uniquely human) emotions. A sample of heterosexual young adults (N = 585, 455 women; Mage = 25.55, SD = 7.48) in three European countries—Croatia, Italy, and Portugal—attributed primary and secondary emotions to four groups: (a) CNM same-sex male partners, (b) CNM heterosexual partners, (c) monogamous same-sex male partners, and (d) monogamous heterosexual partners. Results showed that uniquely human emotions were attributed less to CNM than to monogamous partners, and this happened regardless of sexual orientation. Furthermore, CNM same-sex and CNM heterosexual partners were evaluated similarly. This pattern of results was consistent across countries. The implication of these findings for social policies and sexual rights is discussed.

In this study, we extend these findings by examining if being in a consensually non-monogamous (CNM) relationship implies being also dehumanized. Dehumanizing means perceiving the target as belonging to a lower order of humanity (Haslam, 2006;Leyens et al., 2000), and emerges when the target is perceived as not sharing normative values (i.e., ingroup values; Schwartz & Struch, 1989). Hence, because non-monogamous individuals are judged as deviant from mononormativity (Grunt-Mejer & Campbell, 2016;Hutzler et al., 2016), we argue they also may be targets of dehumanization.
We also consider sexual orientation of the partners. Nonmonogamy occurs among both heterosexual and same-sex partners (Conley, Ziegler, et al., 2013), and researchers have called for the inclusion of the lesbian, gay, and bisexual (LGB) lens in research dealing with non-monogamy (e.g., Hegarty, 2013). We specifically examined same-sex male couples, because gay men are often perceived as sexually permissive and non-monogamous (Peplau & Fingerhut, 2007), and also as deviating from moral and normative standards (Kimmel, 1997). As a consequence, same-sex male partners in CNM relationships may be at higher risk of being dehumanized than heterosexual partners, because they are perceived to violate both norms of monogamy and heterosexuality.
To broaden our scope of analysis, we examined the dehumanization effect in three European countries-Croatia, Italy, and Portugal-that share a similar cultural and religious background. Results from the European Values Survey 2008 (EVS, 2011) suggest that these countries share norms about faithfulness and monogamy. For example, extremely high percentages of Croatians (99%), Italians (98%), and Portuguese (97%) think that faithfulness is Bvery important^or Brather important^for a successful marriage. Also, beside bigamy and polygamy being legally forbidden in these countries, monogamy is the most common and the only practice tolerated by the main religion (i.e., Catholicism). On the other hand, these countries differ in the level of homophobia, with Croatia being characterized by more homophobic attitudes and Portugal being the least homophobic (Eurobarometer, 2015;European Union Agency for Fundamental Rights, 2009).

Perception of Romantic Relationships: Monogamy and Sexual Orientation
Monogamous relationships are defined by a mutual agreement of sexual exclusivity with one partner, whereas CNM relationships can be broadly defined by a mutual agreement of either independent sexual (e.g., open relationships) or romantic relationships (e.g., polyamory; Cohen, 2016;Conley, Ziegler, et al., 2013;Rubel & Bogaert, 2015). CNM relationships are perceived as an immoral departure from the ideal monogamous relationship typically conveyed by society, are often disapproved, and are targets of negative attitudes and prejudice (Hutzler et al., 2016;Treas & Giesen, 2000). Compared to monogamous relationships, CNM relationships are negatively appraised across different attributes (e.g., less committed, intimate, sexually satisfactory, and socially accepted), as are individuals who practice them (e.g., less satisfied with life, less caring, less kind; Conley, Moors, Matsick, & Ziegler, 2013). Notably, this negative perception of individuals in CNM relationships extends to competences and personality attributes (e.g., lower cognitive abilities and morality; Grunt-Mejer & Campbell, 2016). In direct contrast with these appraisals are the experiences of individuals who choose to have CNM relationships. For instance, individuals often indicate that CNM relationships allow them greater freedom to have new experiences and to sexually satisfy themselves with other partners (Cohen, 2016;Moors, Matsick, & Schechinger, 2017). Importantly, this translates directly into relationship quality experiences, with research showing that heterosexual and gay men in CNM relationships do not differ from their monogamous counterparts regarding relationship satisfaction, commitment, and intimacy (Hosking, 2013;Mogilski, Memering, Welling, & Shackelford, 2017;Rodrigues, Lopes, & Smith, 2016;Rubel & Bogaert, 2015).
A possible explanation for CNM stigmatization derives from the strength of the monogamous norm. Usually, monogamous relationships are perceived as better than all other types of relationships (Conley, Moors, et al., 2013). Positive attitudes towards relationships come along with the perception of the partners as committed, satisfied, and loving with each other (Grunt-Mejer & Campbell, 2016). In contrast, extradyadic sex is usually associated with perception of problems in the primary relationship (e.g., poor relationship adjustment or satisfaction; Rubel & Bogaert, 2015;Shaw, Rhoades, Allen, Stanley, & Markman, 2013). Hence, CNM relationships are equated to infidelity (Burris, 2014), and as a consequence, non-monogamous individuals are negatively appraised (Hutzler et al., 2016).
From these findings, the negative perception surrounding CNM relationships seems to derive mostly from its association with sexual permissiveness, and from shared beliefs that sexual acts should be accompanied by an emotional involvement with the partner (Grunt-Mejer & Campbell, 2016;Matsick et al., 2014). Similar to individuals in CNM relationships, sexually permissive individuals are negatively appraised in several characteristics (e.g., less moral, more promiscuous) and considered undesirable romantic partners and friends (Crawford & Popp, 2003;Oliver & Sedikides, 1992;Vrangalova, Bukberg, & Rieger, 2014). This effect parallels the stigma surrounding CNM relationships (Conley, Moors, et al., 2013;Conley, Ziegler, et al., 2013;Moors et al., 2013).
However, it remains unclear whether these negative effects are a reaction to sexual orientation per se, or a reaction to sexually promiscuous behaviors. Indeed, one can question if same-sex male partners invite negative social judgments because people are judging promiscuity, rather than a romantic relationship between two men. There is some evidence suggesting this to be the case. For example, Mak and Tsang (2008) found that students were equally likely to help a gay and a heterosexual person, but were less likely to help a sexually promiscuous than a celibate person, regardless of sexual orientation. Also, Wilkinson and Roys (2005) found that heterosexual men reacted more negatively towards a gay man described as often engaging in noncommittal sexual behavior, when compared to a gay man who just fantasied or desired other men.
Examining the interaction between monogamy and sexual orientation, Moors et al. (2013) showed that individuals in CNM relationships were more negatively perceived than those in monogamous relationships, regardless of their sexual orientation. Yet, whereas gay individuals in monogamous relationships were perceived more negatively than their heterosexual counterparts, gay individuals in CNM relationships were rated as having slightly better relationship quality and sexual satisfaction than their heterosexual counterparts. Taken together, these studies suggest that not adhering to the socially conveyed monogamous norm has a more negative impact than being a gay individual. However, the mixed findings from this research need further investigation. Examining monogamy and sexual orientation in judgments other than negative evaluation allows shedding new light and examining if such evaluations translate into stigmatization at different levels. One such example is dehumanization, that is, the perception of humanness of others.
Literature suggests two types of dehumanization (Haslam, 2006;Leyens et al., 2000). On the one hand, people conceptualize humanness based on the characteristics that are uniquely human (e.g., cognitive capacity, civility) and that differentiate humans from animals. On the other hand, humanness is based on characteristics that represent the core of Bhuman nature^(e.g., emotionality, warmth). If individuals are denied the former attributes, then there is evidence of animalistic dehumanization; when they are denied the latter, mechanistic dehumanization is involved.
Emotion-based attributions can be further subdivided into primary and secondary emotions, as specific ways to express dehumanization towards a given target (see Haslam & Loughnan, 2014). Primary emotions (e.g., anger, happiness) are exhibited by both humans and animals. Thus, they are non-uniquely human (Demoulin et al., 2004). In contrast, secondary emotions (e.g., love, embarrassment) are exhibited only by humans, because they require a higher level of cognition. Thus, they are uniquely human and distinguish humans from other species. By denying secondary emotions, the target is perceived as lacking in humanness and as being more animallike (Leyens et al., 2000).
Animalistic dehumanization has been mainly studied in the context of intergroup relations. Typically, these investigations show that individuals attribute secondary emotions to ingroup and outgroup members differently (for an overview, see Leyens et al., 2007). More secondary emotions are attributed to the ingroup than to the outgroup, whereas primary emotions are usually equally attributed to both groups (Leyens et al., 2000). This effect has been found to be independent of negative attitudes, or ingroup favoritism (Cortes, Demoulin, Rodriguez, Rodriguez, & Leyens, 2005;Leyens et al., 2007). Therefore, dehumanization is not a mere form of negative appraisal and is associated with concrete outcomes or implications, such as lack of helping (e.g., Andrighetto, Baldissarri, Lattanzio, Loughnan, & Volpato, 2014) or doing harm (e.g., Kteily, Bruneau, Waytz, & Cotterill, 2015).
To our knowledge, no research has investigated if non-monogamous individuals, compared to monogamous individuals, are dehumanized. Moreover, there is mixed evidence as to whether dehumanization occurs in the case of sexual orientation. For instance, Brown and Hegarty (2005) found that heterosexual individuals associated secondary emotions according to stereotypes, and not according to sexual orientation. Specifically, secondary emotions were attributed more to gay than to heterosexual men, possibly because the emotions used were perceived as more typical of women and related to femininity (see gender inversion theory, Kite & Deaux, 1987). In another study, MacInnis and Hodson (2012) found that homosexuals were perceived as experiencing more uniquely human emotions than heterosexuals, bisexuals, and asexuals, and that the latter group was specifically subject of dehumanization. In a follow-up study, however, the same authors found no differences in emotion attribution between homosexuals and heterosexuals. More recently, it has been shown that gay dehumanization by heterosexuals may emerge under certain circumstances. For instance, while extended contact with homosexuals decreases dehumanization (Capozza et al., 2014), exposure to homophobic epithets enhances it (Fasoli et al., 2016).
All in all, research shows that dehumanization of homosexuals per se does not emerge, evidencing that this phenomenon only occurs in the presence of specific variables or contexts. Nevertheless, research in this domain always approached homosexual dehumanization considering individuals or the whole group, but never considered same-sex and heterosexual romantic partners and their sexual norms.

Current Study
In the present study, we extended previous work by going beyond the mere evaluation of romantic partners. Specifically, we examined if same-sex and heterosexual partners in CNM or monogamous relationships are dehumanized by being perceived as experiencing less secondary (i.e., uniquely human) emotions. Hence, we investigated an important process that is associated with stigmatization and discrimination.
Compared to monogamous relationships, CNM relationships are perceived as more promiscuous and deviant from mononormativity (Conley, Moors, et al., 2013;Grunt-Mejer & Campbell, 2016;Matsick et al., 2014). Hence, we predicted that CNM partners will be subjects of dehumanization, such that participants should attribute less secondary emotions to CNM partners than to monogamous partners, whereas no differences in the attribution of primary emotions were expected (H1).
Past research has shown that direct dehumanization towards gay men does not occur (Brown & Hegarty, 2005;MacInnis & Hodson, 2012). However, this has never been tested towards same-sex male partners. Therefore, we explored whether dehumanization, namely a lower attribution of secondary emotions, would occur for these relationships (H2).
We also explored the interaction between monogamy and sexual orientation by examining two competing hypotheses. Because research has shown that CNM individuals are negatively evaluated regardless of their sexual orientation (Moors et al., 2013), we can expect no difference in dehumanization between same-sex and heterosexual CNM partners (H3a). However, research has also shown that dehumanization can be induced by highlighting the deviant status of the target (Fasoli et al., 2016). Therefore, we can also expect same-sex CNM partners to be dehumanized more than heterosexual C N M p a r t n e r s , b e c a u s e t h e y d e p a r t f r o m b o t h mononormativity and heteronormativity (H3b).
We explored these hypotheses across three European countries-Croatia, Italy, and Portugal. These countries differ in their level of homophobia, with Croatia being the most homophobic (Eurobarometer, 2015;European Union Agency for Fundamental Rights, 2009). In turn, homophobia is associated with negative attitudes towards homosexuals (Kuntz, Davidov, Schwartz, & Schmidt, 2015;Lopes et al., 2016). Moreover, all countries share mononormativity, that is, bigamy and polygamy are legally forbidden and monogamy is the most common practice (EVS, 2011).

Method Participants and Design
A total of 585 heterosexuals (455 women; M age = 25.55, SD = 7.48) took part in a cross-national study (Croatia n = 307; Italy n = 111; Portugal n = 167). Table 1 presents a summary of the demographics across counties. There were no differences in gender or religion across the samples. Minor differences emerged in terms of education, age, and regional backgrounds. Around half of the total sample indicated that they know men in same-sex couples (56.6%), and this was especially evident in the Portuguese and Italian samples. Although there were differences in the attitudes towards gay individuals across countries, these differences were very low in magnitude (η p 2 = .011). A 2 (emotion: primary vs. secondary) × 2 type of relationship (CNM vs. monogamous) × 2 sexual orientation of the partners (same-sex vs. heterosexual) × 3 (country: Croatia vs. Italy vs. Portugal) factorial design was used, with the first three factors as within-subjects.

Procedure and Measures
The study was in agreement with the ethics guidelines of each hosting institution. Heterosexual individuals were invited through mailing lists and social networking sites to take part in a web survey about the perception of romantic partners. Individuals were given a full description of their rights and duties. After providing informed consent, participants were presented with demographic questions (e.g., age, education, regional residence, religion). Following this, participants were randomly presented with four descriptions of romantic partners: (a) BJohn and Michael are both highly committed in a longtermed sexually non-monogamous romantic relationship (i.e., each partner can have sexual encounters with other people)( CNM same-sex male partners), (b) BJohn and Mary are both highly committed in a long-termed sexually non-monogamous romantic relationship (i.e., each partner can have sexual encounters with other people)^(CNM heterosexual partners), (c) BJohn and Michael are both highly committed in a long-termed sexually monogamous romantic relationship (i.e., each partner cannot have sexual encounters with other people)^(monogamous same-sex male partners), and (d) BJohn and Mary are both highly committed in a long-termed sexually monogamous romantic relationship (i.e., each partner cannot have sexual encounters with other people)^(monogamous heterosexual partners).
After each description, participants were asked to indicate to what extent both partners of the couple experience each of the 12 emotions presented (taken from Paladino et al., 2002 and used also by Brown & Hegarty, 2005). These included three positive (i.e. lust, desire, pleasure) and three negative (i.e., fear, pain, anger) primary emotions, as well as three positive (i.e. happiness, compassion, love) and three negative (i.e., guilt, embarrassment, remorse) secondary emotions. Responses were given on 7-point scales (1 = not at all; 7 = a lot). The order in which each emotion was presented was also randomized. Lastly, participants were asked if they know gay men in same-sex relationships (yes/no) and to indicate their global attitude towards gay individuals (0 = extremely negative; 100 = extremely positive). At the end, participants were thanked and debriefed about the purpose of the study.

Results
Because dehumanization is assumed to be independent of valence (Leyens et al., 2007;Viki, Osgood, & Phillips, 2013), primary and secondary emotions were collapsed across valence. An index of primary emotions (α = .81 across descriptions) and secondary emotions (α = .82 across descriptions) was computed by averaging the respective items (see MacInnis & Hodson, 2012).

Hypotheses Testing
Results were analyzed in a 2 (emotion: primary vs. secondary) × 2 type of relationship (CNM vs. monogamous) × 2 sexual orientation of the partners (same-sex vs. heterosexual) × 3 (country: Croatia vs. Italy vs. Portugal) repeated-measures ANOVA. The first three factors were within-participants.
The interaction between emotion and sexual orientation was non-significant, F(1, 582) = 1.20, p = .274, η p 2 = .002, such that primary and secondary emotions were similarly attributed to same-sex partners and to heterosexual partners (see Fig. 2). Again, country did not qualify these results, F(2, 582) = 0.75, p = .475, η p 2 = .003, suggesting that same-sex partners were not dehumanized in the three countries.
Results also showed that the three-way interaction between emotion, type of relationship, and sexual orientation was nonsignificant, F(1, 582) = 0.98, p = .323, η p 2 = .002. Once more, this result was not qualified by country, F(2, 582) = 0.16, p = .853, η p 2 = .001. This suggests that primary and secondary emotions were similarly attributed to CNM and monogamous same-sex and heterosexual partners, regardless of country.
There was also a significant interaction between type of relationship and country, F(2582) = 13.79, p < .001, η p 2 = .045. Pairwise comparisons with Bonferroni correction showed that Croatians (M = 3.92, SD = 0.64) attributed more emotions to monogamous partners than did both Portuguese (M = 3.62, SD = 0.68) and Italian participants (M = 3.43, SD = 0.60), all p < .001. In contrast, no differences between countries were found in emotions attributed to CNM partners, all p = 1.00.

Controlling for Contact with and Attitudes Towards Gay Individuals
Controlling for contact with gay men in romantic relationships and attitudes towards gay individuals, the interaction between emotion and type of relationship remained significant, F(1, 571) = 20.69, p < .001, η p 2 = .035; the interaction between emotion and sexual orientation remained non-significant, F(1, 571) = 0.38, p = .536, η p 2 = .001; and the interaction between emotion, type of relationship, and sexual orientation remained non-significant, F(1, 571) = 0.08, p = .777, η p 2 = .001. Neither of these effects were qualified by country, all F < 2.26, p > .105, η p 2 < .008.

Discussion
In a cross-national study, we examined if dehumanization of CNM and same-sex partners is promoted by deviations from mononormativity and/or heteronormativity, and whether it occurs to a similar extent across three European countries. In line with our predictions, our findings suggest that CNM partners are a target of dehumanization, because they were perceived as experiencing less uniquely human emotions (i.e., secondary emotions) than monogamous partners. Hence, they were relegated to a lower order of humanity. In contrast to Leyens and colleagues' theory (2001), and to our own prediction, CNM partners were also associated with more primary emotions than monogamous partners (for over-attribution of primary emotions to the outgroup, see also Marcu & Chryssochoou, 2005;Rohmann, Niedenthal, Brauer, Castano, & Leyens, 2009;Viki & Calitri, 2008). This may reflect the fact that CNM partners are not merely perceived as less human but are also Banimalized,^in that they are strongly related to their animalistic nature, expressed by primary emotions (see also Bharj & Hegarty, 2015). Denying humanness and ascribing animalistic characteristics can be two non-mutually exclusive processes, especially in a context where sexual behavior is involved. Because sexual desires are biological needs shared between humans and animals, CNM partners may be perceived as more animal-like as they are not able to limit their animal sexual instincts, a criterion that represents the basis of monogamy.
Results also showed no differences in the emotions attributed to same-sex and heterosexual partners in a romantic relationship. Despite the small effect size, this finding replicates previous research on gay individuals (Brown & Hegarty, 2005;MacInnis & Hodson, 2012), and shows a lack of dehumanization of same-sex male partners. Moreover, our results suggest that CNM same-sex partners did not elicit stronger dehumanization than CNM heterosexual partners, converging with past findings on the stigmatization of non-monogamy, regardless of sexual orientation (Moors et al., 2013). Furthermore, our findings indicate that, although gay men are perceived as having less committed and less invested romantic relationships (Doan et al., 2015;Testa et al., 1987) and to be targets of prejudice (e.g., Lopes et al., 2016), dehumanization may be more aligned with a stereotypical view of promiscuous behaviors (Mak & Tsang, 2008;Wilkinson & Roys, 2005), rather than a stereotypical view of sexual orientation per se. This suggests that, in eliciting dehumanization, the deviation from mononormativity-typical of CNM partners-is stronger than the deviation from heteronormativity-typical of gay men-at least concerning romantic partners. Interestingly, the same pattern of results emerged in three European countries-Croatia, Italy, and Portugal-that share a similar background concerning mononormativity and differ in attitudes towards gay individuals. The fact that dehumanization occurred for CNM partners in all countries, but not for same-sex partners, further supports the idea that sexual behaviors matter more than sexual orientation in the dehumanization of romantic partners (e.g., MacInnis & Hodson, 2012). Indeed, even in countries showing more negative beliefs and attitudes towards gay individuals, dehumanization was not driven nor enhanced by sexual orientation of both partners. Results also showed that dehumanization was independent of personal attitudes and contact with gay individuals (Cortes et al., 2005;Leyens et al., 2007). This suggests that dehumanization of same-sex male partners does not emerge when they are represented as highly committed romantic partners, and that same-sex male partners' dehumanization may be independent of cultural and national environments supporting lesbian, gay, bisexual, and transgender rights. However, this is not to say that overall same-sex partners are targets of less prejudice and social discrimination than individuals in CNM relationships. In fact, gay men can experience discrimination (e.g., verbal or physical aggression) solely based on their physical appearance or sound of voice (e.g., Fasoli, Maass, Paladino, & Sulpizio, 2017;Rule, Bjornsdottir, Tskhay, & Ambady, 2016), in contrast to heterosexual individuals (either monogamous or not).

Possible Mechanism for CNM Dehumanization
Our findings suggest that the typical negative judgments directed at same-sex partners are not driven by sexual orientation, but rather by assumptions regarding sexual behavior. In this regard, MacInnis and Hodson (2012) have suggested a link between dehumanization and sexual desire, rather than sexual orientation. For these authors, sexuality and sexual desire may be characteristics that define humanness, as those who do not have it (i.e., asexuals) are perceived as less human. However, sexual desire represents a need of both animals and humans, while the combination of sexual and romantic desire is specific of humans. Thus, being involved in extradyadic sex may equate CNM partners to animals and lead to perceiving them as lacking in those emotions (e.g., Btypical^love or guilt associated with infidelity) that define human uniqueness.
Furthermore, humans are involved in romantic relationships that stand on cultural and moral norms (e.g., monogamy vs. polygamy), and deviations from these norms are linked with moral exclusion. Moral disengagement is related to denial of humanness (Bastian & Haslam, 2010;Brandt & Reyna, 2011;Castano & Giner-Sorolla, 2006), and dehumanization overlaps with sexual immorality (Haslam, Bastian, Laham, & Loughnan, 2012). Therefore, if CNM partners are perceived as more sexually promiscuous and as failing to engage in mononormativity, they may be judged as immoral (Grunt-Mejer & Campbell, 2016;Hutzler et al., 2016). Indirect support for Bdeviant^and/or Bimmoral^perceptions as the mechanism underlying CNM dehumanization is found in research showing that victims of sexual assault are dehumanized only if described as having promiscuous history, and thus blamed for the assault (Gillmor, Bernstein, & Benfield, 2014; see also Mak & Tsang, 2008).

Implications for Social Policies
CNM relationships have received an increased visibility in recent years (e.g., Conley, Moors, Matsick et al., 2013), and researchers have shown these relationships to span different developmental stages (e.g., Rodrigues, Lopes, & Pereira, 2016;Rodrigues, Lopes, & Smith, 2016;Sizemore & Olmstead, 2016). This puts forth the need not only to revise legal rights and laws for CNM relationships, but also to have debates about family policies, rights, and protections of individuals who choose to have such type of relationship (Blaney & Sinclair, 2013). As an example, when civil partnerships were approved in Italy by the end of 2016, a debate surrounding partners' fidelity arose. Indeed, while marriages legally involve the requirement of fidelity, civil partnerships do not. This led people to believe that civil partnerships represented a lower form of couples' recognition, because it was assumed that same-sex couples were not involved in fidelity and, thus, potentially discarded from adoption rights. At the same time, individuals stated that fidelity is a legacy of an outdated view of marriage and should not be involved in the legal recognition of relationships. This example testifies to the need of recognizing nonmonogamy by social policies, at both social and legal levels.
Policymakers could devise intervention programs directed at raising awareness to CNM relationship in the public domain, for instance using empirical data showing that individuals in CNM relationships are as committed, satisfied, and intimate with their partner as individuals in monogamous relationships (e.g., Mogilski et al., 2017;Rodrigues, Lopes, & Smith, 2016;Rubel & Bogaert, 2015). This can influence the perceptions that others have of individuals in CNM relationships, including issues such as parenting concerns, time and financial constraints, children, and privacy issues with family members and friends (e.g., Brandon, 2016). Indeed, one of the main areas of intervention of policymaking could be the reduction of discrimination and prejudice of non-monogamous individuals, which might be achieved by changing mentalities and informing the public in general.
Equally important, policymakers could also devise intervention programs directed at raising awareness of practitioners in several domains. For instance, it is important for professionals in clinical and counseling psychology to understand how different views of sexuality can help couple's counseling and promote intimacy in relationships. Also, it is important for public health professionals not to be derogatory against non-monogamous individuals, due to beliefs that they are promiscuous or that they make less normative sexual choices (e.g., perception that non-monogamous individuals use condoms less consistently). In fact, this is not the case given that recent research shows that these individuals have less unprotected sex than individuals in monogamous relationships engaging in extradyadic sex (e.g., Conley, Moors, Ziegler, & Karathanasis, 2012).
More broadly, and related to findings showing that CNM relationships are stigmatized and dehumanized, it is to be expected they will also be targets of violence and at risk of discrimination. Because recent findings showed that social policies that promote equality based on sexual orientation are associated with decreased crime incidences (Levy & Levy, 2017), reforms in social policies for CNM relationships might be fruitful in the long run to reduce stigma and dehumanization.

Limitations and Future Research
The current study has a few limitations that should be overcome in future research. Although the descriptions of the couples were randomized, the fact that all participants were presented with the four descriptions might have led to ratings on the first description serving as an anchor against which to compare subsequent descriptions. To disentangle this possibility, future research should seek to devise a between-subjects design. Also, the emotions used to assess positive primary emotions might not have been ideal, because these items overlapped with sexuality. However, this was the case of positive but not of negative primary emotions that were both attributed more to CNM than monogamous partners. Also, if the type of primary emotions created confound, the dehumanization of CNM partners was still occurring through denial of secondary emotions. Nevertheless, future research should seek to examine CNM dehumanization using a different measure.
Given the descriptions of the relationships presented in the current study and the possible association between dehumanization and sexual desire, our findings may be restricted to CNM relationships that primarily allow for extradyadic sexual encounters, and do not necessarily generalize to CNM relationships that allow for extradyadic affective bonds (e.g., polyamory). This is particularly relevant because polyamory relationships are perceived more positively than sexually open relationships . Likewise, our findings may be restricted to same-sex male partners and do not necessarily generalize to same-sex female partners. Hence, future research should seek to generalize our pattern of results.
Furthermore, although we took care to include only heterosexual participants, it would be necessary to test the phenomenon in both heterosexuals and LGB individuals, and in individuals involved in both monogamous and CNM romantic relationships. So far, literature has only showed that sexual minority men and women both have similar attitudes and a similar desire to have CNM relationships . Another possible caveat of our work, but also of other studies (Conley, Moors, et al., 2013;Moors et al., 2013), is that we do not know whether our participants were by any chance in a CNM relationship. Given the strength of the monogamous norm in the three countries under examination in this study (EVS, 2011), it is plausible that most of our sample represent the monogamous majority. Moreover, as effects of CNM dehumanization are strong, even if any CNM individuals were excluded from our analyses, we would have only obtained even stronger effects. Future research should address these issues by recruiting more representative samples across countries. Because the majority of our cross-national samples were women, future studies should also seek to examine gender differences in dehumanization, and include measures tapping individual differences such as traditional values (Callahan & Vescio, 2011), gender role beliefs (Whitley, 2001), religiosity (Whitley, 2009), authoritarianism and social dominance orientation (Whitley, 1999), and conservative political orientation (Schwartz, 2010).

Conclusion
CNM partners are not only negatively viewed but also dehumanized. This comes at risk of severe implications because dehumanization is linked to consequences such as reduced empathy and higher tolerance for violence towards the target (Haslam & Loughnan, 2014;Vaes et al., 2012). This is far beyond what negative evaluation implies. Hence, by being perceived as less human and more animal-like, CNM partners can be targets of social exclusion and harsh discrimination. By highlighting that individuals in CNM relationships are stigmatized and dehumanized, this research raises important questions regarding discrimination and social policies in legal rights surrounding these types of relationships, especially when this stigmatization is directed not only at same-sex partners but also at heterosexual partners.

Compliance with Ethical Standards
Conflict of Interest The authors declare that they have no conflict of interest.
Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed Consent Informed consent was obtained from all individual participants included in the study.
Funding This research was funded by the grant SFRH/BPD/73528/ 2010, awarded by Fundação para a Ciência e Tecnologia (FCT) to the first author.