Bias at the intersection of identity: Conflicting social stereotypes of gender and race augment the perceived femininity and interpersonal warmth of smiling Black women

Bias at the intersection of identity: Conflicting social stereotypes of gender and race augment the perceived femininity and interpersonal warmth of smiling Black women

Journal of Experimental Social Psychology 74 (2018) 43–49 Contents lists available at ScienceDirect Journal of Experimental Social Psychology journa...

646KB Sizes 0 Downloads 20 Views

Journal of Experimental Social Psychology 74 (2018) 43–49

Contents lists available at ScienceDirect

Journal of Experimental Social Psychology journal homepage: www.elsevier.com/locate/jesp

Case Report

Bias at the intersection of identity: Conflicting social stereotypes of gender and race augment the perceived femininity and interpersonal warmth of smiling Black women

MARK

Erin Cooley⁎, Hannah Winslow, Andrew Vojt, Jonathan Shein, Jennifer Ho Colgate University, United States

A B S T R A C T Research indicates that Black women are socially ignored given that they are neither the prototypical Black person nor the prototypical woman. We build from augmentation principle to propose that factors that increase the salience of Black women's gender identity may lead to particularly positive social expectations given countervailing associations of Blackness/threat. First, Study 1 demonstrates that smiles increase the salience of Black women's gender identity as indicated by fewer categorization errors in a speeded gender categorization task. Next, Study 2 demonstrates that, consistent with augmentation principle, the expression of a smile increases the perceived femininity of a Black woman to a greater degree than a smile expressed by a White woman. Moreover, smiles increase positive expectations for an interaction with a Black woman more so than they do for a White woman. We conclude that Black women navigate a precarious balance between social invisibility and social hypervisibility based on shifts in the salience of their gender identity.

When trying to understand people's lived experiences, attitudes toward race cannot be isolated from attitudes toward the myriad intersecting identities that people have. However, psychological research examining racial prejudice toward Black people often focuses on attitudes toward Black men. Similarly, research examining sexism often focuses on attitudes toward White women (Cole, 2009; Goff & Kahn, 2013; Goff, Thomas, & Jackson, 2008; Purdie-Vaughns & Eibach, 2008). Thus, researchers often ignore the experiences of Black women who are neither the prototypical Black person nor the prototypical woman (Goff et al., 2008; Hooks, 1981; Purdie-Vaughns & Eibach, 2008). The fact that Black women are relatively ignored within the literature on racial prejudice is mirrored by their experiences in daily interactions. Sesko and Biernat (2010) found that the faces of Black women, as well as their statements within a group conversation, were less likely to be remembered than faces or statements made by Black men or White people of either gender. This distinct prejudice experienced by Black women has been termed “intersectional invisibility” (Beale, 1970; Crenshaw, 1991; Purdie-Vaughns & Eibach, 2008) and is thought to be driven by the fact that Black women are neither the default when people imagine women, nor the default when people imagine Black people (Goff et al., 2008; Purdie-Vaughns & Eibach, 2008; Thomas, Dovidio, & West, 2014). Consistent with this reasoning, people experience greater difficulty in categorizing Black women than White



Corresponding author. E-mail address: [email protected] (E. Cooley).

http://dx.doi.org/10.1016/j.jesp.2017.08.007 Received 24 April 2017; Received in revised form 3 August 2017; Accepted 29 August 2017 0022-1031/ © 2017 Elsevier Inc. All rights reserved.

women according to their gender; and greater difficulty in categorizing Black women than Black men according to their race (Goff et al., 2008; Thomas et al., 2014). While these data indicate that the dual identities of Black women can lead them to experience social isolation, might there also be circumstances in which the dual identities of Black women lead them to be sought out socially? In some ways, the experiences of Black women are likely to be similar to the experiences of Black men given their shared racial identity. For example, both male (Goff, Jackson, Di Leone, Culotta, & DiTomasso, 2014) and female (Blake, Keith, Luo, Le, & Salter, 2017) Black children are perceived as older and less innocent than their White peers—with tragic implications for their legal treatment (Epstein, Blake, & González, 2017). Similarly, Black women are often stereotyped as high in agency and strength, much like Black men (Donovan & West, 2015; West, Donovan, & Daniel, 2016). However, unlike Black men, Black women simultaneously navigate conflicting gender stereotypes that emphasize warmth and social approachability (Fiske, Cuddy, Glick, & Xu, 2002). Because these race and gender stereotypes are in direct conflict with one another, they may make behaviors that are consistent with one of these identities (e.g., a behavior that emphasizes femininity) seem particularly diagnostic of a Black woman's general social disposition. Such a prediction is consistent with classic work on attribution theory (Duncan, 1976; Heider, 1958; Kelley & Michela, 1980)—most notably,

Journal of Experimental Social Psychology 74 (2018) 43–49

E. Cooley et al.

1.2. Procedure

augmentation principle (Kelley, 1971). Augmentation principle predicts that behaviors that occur despite some countervailing force are perceived as particularly diagnostic of an underlying disposition (Kelley, 1971). For example, if a woman makes a generous donation, people perceive that behavior as more meaningful when they learn that she also had the opportunity to invest the money in a lucrative fund (i.e., a countervailing force) than when this alluring alternative was not available (Kruglanski, Schwartz, Maides, & Hamel, 1978). Applying this reasoning to the social experiences of Black women, behaviors stereotypically associated with femininity, when enacted by a Black woman, may accentuate her gender identity and associated gender stereotypes. If so, attributions associated with femininity—such as interpersonal warmth—may be augmented given countervailing stereotypes that link Blackness to masculinity and threat (Goff et al., 2008; Shields, 2008). Building from this logic, we propose that behaviors—such as the flash of a smile—that accentuate the gender identity of Black women may play an important role in the types of expectations non-Black people have for interactions with Black women. We chose facial expressions as a manipulation of gender salience because of the strong association of smiles with femininity and because facial expressions are a frequent social cue (Ekman, 2001; Jack, Sun, Delis, Garrod, & Schyns, 2016) that play an important role in interracial perception (Kunstman, Tuscherer, Trawalter, & Lloyd, 2016; Lloyd, Kunstman, Tuscherer, & Bernstein, 2017). First, in Study 1, we reasoned that smiles (versus neutral facial expressions) would accentuate the gender identity of Black women as indicated by fewer errors in a speeded gender categorization task. Next, in Study 2 we hypothesized that if smiles accentuate Black women's gender identity, smiles should also increase the perceived femininity of Black women to a greater degree than White women given conflicting stereotypes associating Blackness with masculinity (i.e., an augmentation effect). Finally, given associations of femininity with interpersonal warmth, we predicted that this augmentation effect may lead to preferences for interactions with Black (versus White) women—but only when Black women are smiling. All measures are disclosed. Data and materials for all studies are available upon request.

After signing an informed consent, participants were told that they were going to engage in a concentration task. The stimulus set contained 24 photos each of Black women, Black men, White women, and White men. For each stimulus, we obtained two photos, one with the person expressing a smile and one with the person expressing a neutral expression for a total of 192 photos. Photos were selected from the Chicago Face Database (CFD 2.0; Ma & Wittenbrink, 2015) and closely matched (based on CFD norming data) on attractiveness (MBlackWomen = 3.33, SD = 1.36; MWhiteWomen = 3.33, SD = 1.31; MBlackMen = 3.32, SD = 1.43; MWhiteMen = 3.29, SD = 1.37). The method for the categorization tasks was derived from previous research (Thomas et al., 2014). In the gender categorization task, participants categorized the target photos by gender (“man” or “woman”); in the race categorization task, participants categorized the same photos by race (“White” or “Black”). Participants made categorizations by pressing the “Q” and “P” computer keys, and the order of the two tasks was counterbalanced. For each categorization task, participants completed 8 practice trials and then 96 critical trials composed of 12 smiling photos and 12 neutral photos of each photo type (i.e., Black women, Black men, White women, White men) in a random order. Participants were told to respond as quickly and accurately as possible. Participants were also randomly assigned to one of two stimuli conditions so that no individual saw the same face both smiling and neutral, but across participants all faces appeared with a smiling and neutral facial expression. Participants concluded by reporting demographic information, explicit racial attitudes (Sears & Henry, 2002), a measure of the degree to which they prioritize positivity (Catalino, Algoe, & Fredrickson, 2014), and motivations to control prejudice (Plant & Devine, 1998). Results were not moderated by any of these scales (see Supplementary Materials). 1.3. Results and discussion We predicted that smiles (versus neutral expressions) would facilitate categorization of Black women by gender, as indicated by fewer categorization errors. To test our hypotheses, we ran a 2 (Target Race: Black vs. White) × 2 (Target Gender: woman vs. man) × 2 (Target Facial Expression: smile vs. neutral) × 2 (Categorization Dimension: gender vs. race) repeated-measures ANOVA. Overall there were two main effects, a three-way interaction, and the predicted four-way interaction. First, there was a main effect of target gender such that nonBlack people made more errors categorizing women (M = .038, 95% CI [.031, .045]) than men (M = .031, 95% CI [.026, .037), F(1, 99) = 5.66, p = .019, ηp2 = .05. Additionally, there was a main effect of categorization dimension, such that non-Black people made more errors when categorizing gender (M = .038, 95% CI [.032, .044]) than race (M = .032, 95% CI [.032, .044]), F(1, 99) = 6.58, p = .012, ηp2 = .06. There was also an unpredicted three-way interaction of Target Race × Target Gender × Categorization Dimension, F(1, 99) = 6.09, p = .015, ηp2 = .06. Critically, this three-way interaction was qualified by the predicted four-way interaction of Target Race × Target Gender × Target Facial Expression × Categorization Dimension, F(1, 99) = 5.29, p = .024, ηp2 = .05. We next probed the predicted four-way interaction by running separate 2 (Target Gender: woman vs. man) × 2 (Target Facial Expression: smile vs. neutral) × 2 (Categorization Dimension: gender vs. race) repeated-measures ANOVAs for Black and White targets. For White targets, this analysis revealed no significant main effects, no twoway interactions, nor a three-way interaction (see Supplementary Materials). For Black targets, however, there were two main effects, two two-way interactions, and the predicted three-way Target Gender × Target Facial Expression × Categorization Dimension interaction, F(1, 99) = 6.49, p = .012, ηp2 = .06 (see Supplementary Materials for all descriptives). We next probed this three-way

1. Study 1 method Previous research indicates that people struggle to categorize Black women (who are expressing neutral facial expressions) by gender in speeded categorization tasks due to stereotypes associating Blackness with masculinity (Thomas et al., 2014). In Study 1, we tested whether the expression of smiles facilitates categorization of Black women by their gender. To test this, participants completed two speeded categorization tasks: one for race and one for gender. We predicted that there would be more errors categorizing Black women by gender than race when they had neutral expressions. However, when expressing smiles, we expected categorization of Black women by gender to be facilitated (i.e., fewer errors).

1.1. Participants We recruited a sample of 100 students from our psychology participant pool based on a power analysis (Faul, Erdfelder, Lang, & Buchner, 2007) that indicated that we needed a sample of 92 to have adequate power (1-β ≥ .80) to detect a small effect (f = .15) in our fully within-subjects design. Participants were 104 students (81.7% White; 3% Black; 1% Native American/Pacific Islander; 12% Asian; 1.9% Other; participants could select more than one option) who completed the study for course credit. Given our interest in interracial perception of Black people, we conducted all analyses on the 100 nonBlack participants (24 men, 75 women, 1 other). 44

Journal of Experimental Social Psychology 74 (2018) 43–49

E. Cooley et al.

Fig. 1. Target Facial Expression × Categorization Dimension interactions predicting error rates in categorization of Black women (statistically significant) and Black men (statistically nonsignificant), Study 1.

this augmentation effect (Kelley, 1971) may lead to particularly positive expectations for interactions with smiling Black women. First, we selected images of Black and White women from the Minear and Park (2004) and NimStim databases (Tottenham et al., 2009). Given that smiles can vary in how genuine they appear (Frank & Ekman, 1993) and because we were interested in perceptions of social approachability in this study, a researcher trained in facial action coding (FACS) coded all smiling faces for the presence of Duchenne, or genuine, smiles (Ekman & Friesen, 1978). Faces coded as displaying Duchenne smiles were submitted to pretesting. Our final set of images were 6 Black women and 6 White women who did not vary in perceived typicality of their race, attractiveness, or how friendly they appeared (see Supplementary Materials for pretest data).

interaction by testing the two-way interaction of Target Facial Expression × Categorization Dimension separately for Black men and Black women (see Fig. 1). This interaction was not significant for Black men, F(1, 99) = .28, p = .598, ηp2 = .00. However, as predicted, it was significant for Black women, F(1, 99) = 9.19, p = .003, ηp2 = .09, and in the predicted patterns (see Fig. 1). To probe this significant, predicted Target Facial Expression × Categorization Dimension interaction for Black women, we first broke down the interaction by examining the effect of target facial expression separately for each task. Non-Black people made significantly fewer categorization errors when categorizing the gender of Black women when they were smiling (M = .041, 95% [.028, .054]) versus expressing neutral expressions (M = .064, 95% [.045, .083]), F (1, 99) = 5.38, p = .022, ηp2 = .05. Conversely, there was no difference in categorization errors when categorizing the race of Black women when they were smiling (M = .037, 95% [.025, 048]) versus expressing neutral expressions (M = .028, 95% [.018, .037]), F(1, 99) = 1.82, p = .180, ηp2 = .02. Thus, smiles facilitate gender categorization but do not affect racial categorization of Black women. Next we broke down the interaction by examining the effect of categorization dimension separately by target facial expression. When Black women were expressing neutral expressions, participants made significantly more errors on the gender categorization task (M = .06, 95% CI [.05, .03]) than on the race categorization task (M = .028, 95% CI [.012, .04]), F(1, 99) = 17.32, p < .001, ηp2 = .15. However, when Black women were expressing smiles, there were no differences in errors on the gender categorization task (M = .041, 95% CI [.028, .048]) versus the race categorization task (M = .037, 95% CI [.028, .054]), F (1, 99) = .32, p = .571, ηp2 = .00. Together these results indicate that facial expressions play an important role in the salience of Black women's intersecting race and gender identities. Because of associations of Blackness with masculinity, non-Black people have greater difficulty categorizing Black women by gender than race when Black women are expressing neutral expressions (as indicated in other research; Thomas et al., 2014). However, the flash of a smile eliminates differences in error rates between gender and race categorization tasks. These patterns are consistent with our proposal that smiles facilitate categorization of Black women by their gender. Interestingly, smiles decreased errors in gender categorization but did not increase errors for race categorization. Thus, smiles seem to facilitate categorization by gender but not impede categorization by race. In Study 2, we build from these findings to test the role of augmentation principle (Kelley, 1971) in expectations of the femininity, and thus, social approachability of smiling (versus neutral) Black women.

2.1. Participants We recruited a sample of 250 workers from Amazon Mechanical Turk based on a power analysis (Faul et al., 2007) that indicated that we needed a sample of 250 to have adequate power (1-β ≥ .80) to detect a medium effect (f = .25). Participants were 249 (82.3% White; 7.2% Black; 1.6% Native American/Pacific Islander; 7.2% Asian; 1.6% Other) workers on Amazon Mechanical Turk. As in previous studies, we conducted all analyses on non-Black participants. We also omitted 15 participants who failed an attention check. Our final sample was 216 (135 women, 81 men) people.

2.2. Procedure Participants learned we were interested in expectations of interactions with people based only on looking at their photos. Of the 6 Black women and 6 White women identified in pretesting, participants were randomly assigned to see 3 Black women and 3 White women expressing Duchenne smiles and 3 Black women and 3 White women with neutral expressions. There were two stimuli conditions so that no participant made a rating of the same woman with a smile and neutral expression, but across all participants all stimuli were rated with both facial expressions. For each face, participants were asked to imagine engaging in an interaction with the person depicted. Then participants reported their expectations of shared positive experiences (SPE) in the context of this interaction (e.g., “I would think our interaction was inspiring;” “We would have a lot of fun;” Fredrickson, 2017; See Supplementary Materials). Next participants rated perceptions of feminine stereotypes (i.e., compassionate, kind, gentle), masculine stereotypes (i.e., aggressive, dominant, assertive), and overall femininity/masculinity of the faces on 0 (not at all) to 100 (extremely) scales. Finally, participants completed an attention check (Oppenheimer, Meyvis, & Davidenko, 2009), demographic information, and the same scales as Study 1. As in Study 1, none of these scales moderated our findings (see Supplementary Materials).

2. Study 2 Method In Study 2, we reasoned that if smiles accentuate the gender identity of Black women, then smiles, versus neutral expressions, may also increase the perceived femininity of Black women significantly more than they do for White women due to countervailing associations of Blackness with masculine stereotypes of aggression and threat. If so, 45

Journal of Experimental Social Psychology 74 (2018) 43–49

E. Cooley et al.

Fig. 2. Perceived femininity based on gender and facial expression, Study 2.

Fig. 3. Expectations of shared positive experiences based on face race and facial expression, Study 2.

2.3. Results and discussion femininity was significantly stronger for Black women (i.e., an augmentation effect).

In Study 2 we predicted that smiles (versus neutral expressions) would augment that perceived femininity of Black women more than they would for White women. As a result, we predicted that non-Black people would have more positive expectations of interactions with Black versus White women, but only when they are expressing smiles.

4. Expectations of shared positive experiences Next we examined whether this augmentation effect also extended to expectations of shared positive experiences (SPE) in an interracial interaction. We reasoned that if smiles augment the perceived femininity of Black women, then they may also augment perceived social approachability of Black women. In particular, we predicted scores on the SPE scale with a 2 (Face Race: Black vs. White) × 2 (Facial Expression: Duchenne smile vs. neutral expression) × 2 (Participant Gender: woman vs. man) mixed-model ANOVA. Results revealed one main effect and the predicted two-way interaction. First, a main effect of facial expression indicated that smiles elicited more positive expectations (M = 3.66, 95% CI [3.60, 3.72]) than neutral facial expressions (M = 2.72, 95% CI [2.65, 2.79]), F(1, 214) = 459.82, p < .001, ηp2 = .68. However, there was no main effect of face race, F (1, 214) = 2.51, p = .115, ηp2 = .01. Critically, there was the predicted Face Race × Facial Expression interaction, F(1, 214) = 3.96, p = .048, ηp2 = .02 (See Fig. 3; See Supplementary Materials for descriptives and two-way interactions with participant gender). There was no three-way interaction, F(1, 214) = 1.15, p = .285, ηp2 = .01. To break down the predicted Face Race × Facial Expression interaction we first examined the effect of face race separately for smiling and neutral faces. For neutral faces, expectations for interactions with Black women were no more positive (M = 2.72; 95% CI [2.63, 2.80]) than those with White women (M = 2.72, 95% CI [2.64, 2.79]), F(1, 214) = .00, p = .983, ηp2 = .00. However, as predicted, when faces were expressing smiles, non-Black people had significantly more positive expectations for interactions with Black women (M = 3.70, 95% CI [3.63, 3.77]) than White women (M = 3.62, 95% CI [3.56, 3.69]), F(1, 214) = 7.63, p = .006, ηp2 = .03. Another way to break down the interaction is to look at the effect of facial expression separately by face race. For White faces, non-Black people had significantly more positive expectations for interactions with smiling White women as compared to neutral White women, F(1, 214) = 335.22, p < .001, ηp2 = .61. For Black faces, non-Black people also had significantly more positive expectations for interactions with smiling Black women than neutral Black women, F(1, 214) = 451.60, p < .001, ηp2 = .68. Critically, the effect of a smile in boosting expectations of shared positive experiences was stronger in response to Black as compared to White women (i.e., an augmentation effect). Moreover, this augmentation effect in expectations of shared positive experiences was mediated by the augmentation effect (reported above) in perceived femininity (see Supplementary Materials).

3. Perceptions of Femininity First we examined whether—consistent with augmentation principle—smiles boosted perceptions of femininity when expressed by Black women significantly more than when expressed by White women. To test this, we conducted a 2 (Face Race: Black vs. White) × 2 (Facial Expression: Duchenne smile vs. neutral expression) × 2 (Participant Gender: woman vs. man) mixed-model ANOVA predicting perceived femininity. All factors were within-subjects except participant gender. We included participant gender as a factor because research indicates that men and women interpret smiles differently (Kowalski, 1993). Results revealed two main effects and the predicted two-way interaction. A main effect of face race indicated that non-Black people perceived White people as more feminine (M = 72.80; 95% CI [70.85, 74.76]) than Black people (M = 69.82, 95% CI [67.88, 71.76]), F(1, 214) = 22.18, p < .001, ηp2 = .09. There was also a main effect of facial expression such that smiles elicited greater perceptions of femininity (M = 75.60, 95% CI [73.59, 77.61]) than neutral expressions (M = 67.03, 95% CI [65.02, 69.04]), F(1, 214) = 111.32, p < .001, ηp2 = .34. Critically, there was the predicted Face Race × Facial Expression interaction, F(1, 214) = 8.56, p = .004, ηp2 = .04 (See Fig. 2; See Supplementary Materials for descriptives). To break down the significant two-way interaction, we first examined the effect of face race separately for smiling and neutral faces. For neutral faces, Black women were perceived as less feminine (M = 64.46, 95% CI [62.22, 66.69]) than White women (M = 69.60, 95% CI [67.25, 71.94]) consistent with racial stereotypes that link Blackness with masculinity, F(1, 214) = .00, p < .001, ηp2 = .09. However, consistent with our hypotheses, the expression of smiles eliminated racial biases in perceptions of femininity such that smiling Black women were rated as just as feminine (M = 75.19, 95% CI [73.10, 77.27]) as smiling White women (M = 76.01, 95% CI [73.77, 78.25]), F(1, 214) = 1.04, p = .309, ηp2 = .00. Another way to break down the interaction is to look at the effect of facial expression separately by face race. For White faces, non-Black people perceived smiling White women as significantly more feminine than White women with neutral expressions, F(1, 214) = 27.98, p < .001, ηp2 = .12. Critically, to an even greater degree, non-Black people perceived smiling Black women as significantly more feminine than Black women with neutral expressions, F(1, 214) = 122.62, p < .001, ηp2 = .36. Thus, while smiles increased perceptions of femininity for both Black and White women, the effect of a smile on 46

Journal of Experimental Social Psychology 74 (2018) 43–49

E. Cooley et al.

Fig. 4. Mediation depicting the relationship between expectations for interactions with Black women and image type (1 = smiling, 0 = neutral expression), mediated by perceived femininity (M1), and feminine stereotypes (M2), Study 2. Note: ** p < .01 *** p < .001.

5. Serial mediation by perceptions of femininity and stereotypes of warmth and dominance

only accentuates Black women's gender (rather than racial) identity, but that smiles also augment perceptions of their femininity and positive expectations for social interactions—more so than when expressed by White women. This finding is consistent with augmentation principle, which would predict that smiles, in the presence of a countervailing force (i.e., Blackness-threat), might be perceived as particularly diagnostic of one's underlying disposition (Kelley, 1971). Second, mediation analyses indicated that increased perceptions of the femininity of smiling (versus neutral) Black women led to increased positive expectations for interactions with those women in two ways: by increasing the application of warm stereotypes associated with femininity and decreasing the application of threat stereotypes associated with masculinity/Blackness.

Finally, we tested whether smiles (X) led to augmented expectations of shared positive experiences with smiling (versus neutral) Black women (Y) due to changes in perceived femininity (M1) and associated increases in the application of feminine stereotypes (M2; i.e., warmth, compassion, gentleness; reliability was calculated separately for each type of stimulus and ranged from α = .84 to .91). To test this, we ran a two-condition within-participant serial mediation analyses with 10,000 bootstraps (MEMORE; Montoya & Hayes, 2016). Variables were entered in their raw scale form. Results of this mediation model were consistent with serial mediation (see Fig. 4). Critically, this mediation pattern did not extend to a model that switched the order of the mediators (see Supplementary Materials). Next we ran the same model, but this time, included masculine stereotypes (i.e., ratings of aggression, dominance, assertiveness; reliability was calculated separately for each type of stimulus and ranged from α = .83 to .93) to test whether smiles (X) also led to augmented expectations of shared positive experiences with smiling (versus neutral) Black women (Y) due increases in perceived femininity (M1) resulting in the decreased application of masculine stereotypes associated with Blackness (M2). Indeed, results were consistent with serial mediation (see Fig. 5) and did not extend to a model that switched the order of the mediators (see Supplementary Materials). Together these findings build from Study 1 in two ways. First, they suggest that, for non-Black participants, the expression of smiles not

6. General discussion Research examining racial prejudice toward Black women has historically focused on the “intersectional invisibility” of Black women (Goff et al., 2008; Purdie-Vaughns & Eibach, 2008; Sesko & Biernat, 2010; Thomas et al., 2014). In the present work, we qualify this existing narrative to demonstrate that situational shifts in the salience of Black women's gender identity can lead to particularly positive social evaluations of Black women—more so than for White women. First, we demonstrate that smiles increase the salience of Black women's gender (rather than racial) identity as indicated by more accurate categorizations in a speeded gender categorization task. Next, we reason that if smiles accentuate Black women's gender identity (and associated

Fig. 5. Two-mediator sequential pattern depicting the relationship between expectations for the interaction and image type (1 = smiling, 0 = neutral expression), mediated by perceived femininity (M1), and masculine stereotypes (M2), Study 2. Note: ** p < .01 *** p < .001.

47

Journal of Experimental Social Psychology 74 (2018) 43–49

E. Cooley et al.

7. Conclusion

stereotypes) that the simultaneous presence of conflicting racial stereotypes may lead to an augmentation effect in perceptions of their disposition. Indeed, in Study 2, smiles augmented the perception of Black women's femininity and augmented positive expectations for interactions with those women to a greater degree than comparable smiles expressed by White women. Together these findings indicate that augmentation principle may help predict the social experiences of people who have identities associated with conflicting social stereotypes (e.g., Black women, Asian men, etc.). Existing work on augmentation effects has emphasized the importance of people having the motivation and ability to take nuances of the situation into account for these effects to emerge (Gilbert, Pelham, & Krull, 1988). Similarly, research on attitudes at the intersection of multiple group memberships indicates that integrating multiple group memberships requires greater processing—especially when those different categories carry conflicting social information (Kang & Bodenhausen, 2015). Thus, in Study 2, we tested our hypothesized augmentation effects for smiling Black women by evaluating controlled evaluations of Black women. However, we should also note that Study 1 demonstrated that smiles shifted categorizations of Black women to emphasize their gender (rather than racial) identity in a task that assessed automatic responding. Thus, future research should explore whether augmented perceptions of femininity and desires to interact with smiling Black women may also emerge automatically. These findings also have interesting implications for the ways that race and gender may interact for Black women in positions that require stereotypically masculine traits—such as leadership positions. White women often face backlash in leadership because stereotypical expectations linked to their gender (e.g., warmth) conflict with behaviors that are associated with effective leadership (e.g., assertiveness; Fiske et al., 2002; Brescoll, Dawson, & Uhlmann, 2010). Perhaps, however, Black women, and especially Black women who emphasize their gender identity through a smile, may be perceived as having an ideal combination of assertiveness (associated with their race) and yet warmth (associated with their gender). Such a finding would be consistent with research that finds that Black men with “babyfaceness” experience leadership benefits (as compared to “babyfaced” White men) because these facial features attenuate perceived threat that is usually associated with Blackness (Livingston & Pearce, 2009). While we think that smiles may also affect perception of Black men, we intentionally focused on the experience of Black women for a couple of reasons. Most critically, the mechanism that we propose behind the present effects is specific to Black women. In particular, we reason that particularly positive expectations for interactions with Black women are driven by smiles shifting focus to Black women's gender identity—an identity that has distinct social stereotypes from their racial identity and an identity that is not shared with Black men. Indeed, in Study 1, while smiles reduced errors in speeded categorizations of the gender identity of Black women, they did not affect how people categorized the race nor gender of Black men. Second, we are concerned by the relative dearth of research on prejudice experienced by Black women given their intersecting marginalized identities. Future research should work to understand how facial expressions may also affect social expectations for interactions with Black men. Finally, while the present findings suggest that smiles may increase non-Black people's positive social evaluations of Black women, and may provide some benefits to Black women in particular situations (e.g., leadership), it is also possible that these positive evaluations could strain interracial interactions. In particular, disproportionally positive expectations for interactions with smiling Black women may place an unreasonable pressure on Black women to behave in overtly warm and welcoming ways. Such a possibility is consistent with other research that has explored the pernicious effects of “positive” stereotypes (Cheryan & Bodenhausen, 2000; Czopp, 2008; Kay, Day, & Zanna, 2013).

Because people often fear rejection when thinking about interracial interactions (Shapiro, Baldwin, Williams, & Trawalter, 2011), nonverbal affiliative cues may be an important factor that can shift whether people ignore or approach a potential interaction partner. Consistent with augmentation principle, the present research indicates that the expression of a smile by a Black woman may be perceived as particularly meaningful by increasing the salience of stereotypes associated with her gender identity, which are in conflict with stereotypes of her racial identity. As a result, Black women may be walking a thin line between feelings of social invisibility (Purdie-Vaughns & Eibach, 2008; Sesko & Biernat, 2010) and particularly positive expectations for their behavior in interracial interactions—inferences that may shift as rapidly as the flash of a smile. Appendix A. Supplementary Materials Supplementary Materials to this article can be found online at http://dx.doi.org/10.1016/j.jesp.2017.08.007. References Beale, F. (1970). Double jeopardy: To be Black and female. In T. Cade (Ed.). The Black woman: An anthology (pp. 90–100). New York: Signet. Blake, J. J., Keith, V. M., Luo, W., Le, H., & Salter, P. (2017). The role of colorism in explaining African American females' suspension risk. School Psychology Quarterly, 32(1), 118. Brescoll, V. L., Dawson, E., & Uhlmann, E. L. (2010). Hard won and easily lost: The fragile status of leaders in gender-stereotype-incongruent occupations. Psychological Science, 21(11), 1640–1642. Catalino, L. I., Algoe, S. B., & Fredrickson, B. L. (2014). Prioritizing positivity: An effective approach to pursuing happiness? Emotion, 14(6), 1155. Cheryan, S., & Bodenhausen, G. V. (2000). When positive stereotypes threaten intellectual intellectual performance: The psychological hazards of “model minority” status. Psychological Science, 11(5), 399–402. Cole, E. R. (2009). Intersectionality and research in psychology. American Psychologist, 64(3), 170–180. Crenshaw, K. (1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43(6), 1241–1299. Czopp, A. M. (2008). When is a compliment not a compliment? Evaluating expressions of positive stereotype. Journal of Experimental Social Psychology, 44(2), 413–420. Donovan, R. A., & West, L. M. (2015). Stress and mental health: Moderating role of the strong black woman stereotype. Journal of Black Psychology, 41(4), 384–396. http:// dx.doi.org/10.1177/0095798414543014. Duncan, B. L. (1976). Differential social perception and attribution of intergroup violence: Testing the lower limits of stereotyping of Blacks. Journal of Personality and Social Psychology, 34(4), 590–598. Ekman, P. (2001). Facial expressions. In Blakemore, C. & Jennett, S. (Eds.), Oxford companion to the body. London: Oxford University Press. Ekman, P., & Friesen, W. V. (1978). The facial action coding system. Palo Alto, CA: Consulting Psychological Press. Epstein, R., Blake, J., & González, T. (2017). Girlhood interrupted: The erasure of black Girls' childhood. Washington, DC: The Georgetown Law Center on Poverty and Inequality. Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191. Fiske, S. T., Cuddy, A. J., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82(6), 878. Frank, M. G., & Ekman, P. (1993). Not all smiles are created equal: The differences between enjoyment and nonenjoyment smiles. Humor-International Journal of Humor Research, 6(1), 9–26. Fredrickson, B. L. (2017). Expectations of shared positive experiences scale. Unpublished raw data. Gilbert, D. T., Pelham, B. W., & Krull, D. S. (1988). On cognitive busyness: When person perceivers meet persons perceived. Journal of Personality and Social Psychology, 54(5), 733. Goff, P. A., Jackson, M. C., Di Leone, B. A. L., Culotta, C. M., & DiTomasso, N. A. (2014). The essence of innocence: Consequences of dehumanizing Black children. Journal of Personality and Social Psychology, 106(4), 526. Goff, P. A., & Kahn, K. B. (2013). How psychological science impedes intersectional thinking. Du Bois Review: Social Science Research on Race, 10(02), 365–384. Goff, P. A., Thomas, M. A., & Jackson, M. C. (2008). “Ain't I a woman?”: Towards an intersectional approach to person perception and group-based harms. Sex Roles, 59, 392–403. Heider, F. (1958). The psychology of interpersonal relations. New York, NY: Wiley. Hooks, B. (1981). Ain't I a woman? Black women and feminism. Boston (MA): South End665.

48

Journal of Experimental Social Psychology 74 (2018) 43–49

E. Cooley et al.

Montoya, A. K., & Hayes, A. F. (2016). Two-condition within-participant statistical mediation analysis: A path-analytic framework. Psychological Methods, 22(1), 6–27. Oppenheimer, D. M., Meyvis, T., & Davidenko, N. (2009). Instructional manipulation checks: Detecting satisficing to increase statistical power. Journal of Experimental Social Psychology, 45, 867–872. Plant, E. A., & Devine, P. G. (1998). Internal and external motivation to respond without prejudice. Journal of Personality and Social Psychology, 75(3), 811–832. Purdie-Vaughns, V., & Eibach, R. P. (2008). Intersectional invisibility: The distinctive advantages and disadvantages of multiple subordinate-group identities. Sex Roles, 59, 377–391. Sears, D. O., & Henry, P. J. (2002). The symbolic racism 2000 scale. Political Psychology, 23(2), 253–283. Sesko, A. K., & Biernat, M. (2010). Prototypes of race and gender: The invisibility of black women. Journal of Experimental Psychology, 46(2), 356–360. Shapiro, J. R., Baldwin, M., Williams, A. M., & Trawalter, S. (2011). The company you keep: Fear of rejection in intergroup interaction. Journal of Experimental Social Psychology, 47(1), 221–227. Shields, S. A. (2008). Gender: An intersectionality perspective. Sex Roles, 59(5–6), 301–311. Thomas, E. L., Dovidio, J. F., & West, T. V. (2014). Lost in the categorical shuffle: Evidence for the social non-prototypicality of Black women. Cultural Diversity and Ethnic Minority Psychology, 20(3), 370–376. Tottenham, N., Tanaka, J. W., Leon, A. C., McCarry, T., Nurse, M., Hare, T. A., ... Nelson, C. (2009). The NimStim set of facial expressions: Judgments from untrained research participants. Psychiatry Research, 168(3), 242–249. West, L. M., Donovan, R. A., & Daniel, A. R. (2016). The price of strength: Black college women's perspectives on the strong Black woman stereotype. Women & Therapy, 39(3–4), 390–412. http://dx.doi.org/10.1080/02703149.2016.1116871.

Jack, R. E., Sun, W., Delis, I., Garrod, O. G., & Schyns, P. G. (2016). Four not six: Revealing culturally common facial expressions of emotion. Journal of Experimental Psychology: General, 145(6), 708. Kang, S. K., & Bodenhausen, G. V. (2015). Multiple identities in social perception and interaction: Challenges and opportunities. Annual Review of Psychology, 66547–66574. http://dx.doi.org/10.1146/annurev-psych-010814-015025. Kay, A. C., Day, M. V., & Zanna, M. P. (2013). The insidious (and ironic) effects of positive stereotypes. Journal of Experimental Social Psychology, 49, 287–291. Kelley, H. H. (1971). Attribution in social interaction. New Jersey: General Learning Press. Kelley, H. H., & Michela, J. L. (1980). Attribution theory and research. Annual Review of Psychology, 31, 457–501. Kowalski, R. M. (1993). Inferring sexual interest from behavioral cues: Effects of gender and sexually relevant attitudes. Sex Roles, 29(1–2), 13–36. Kruglanski, A. W., Schwartz, J. M., Maides, S., & Hamel, I. Z. (1978). Covariation, discounting, and augmentation: Towards a clarification of attributional principles. Journal of Personality, 46(1), 176–189. Kunstman, J. W., Tuscherer, T., Trawalter, S., & Lloyd, E. P. (2016). What lies beneath? Minority group members' suspicion of Whites' egalitarian motivation predicts responses to Whites' smiles. Personality and Social Psychology Bulletin, 42(9), 1193–1205. Livingston, R. W., & Pearce, N. A. (2009). The teddy-bear effect: Does having a baby face benefit black chief executive officers? Psychological Science, 20(10), 1229–1236. Lloyd, E. P., Kunstman, J. W., Tuscherer, T., & Bernstein, M. J. (2017). The face of suspicion: Suspicion of Whites' motives moderates mental representations of whites. Social Psychological and Personality Science, 1948550617699251. Ma, C., & Wittenbrink (2015). The Chicago face database: A free stimulus set of faces and norming data. Behavior Research Methods, 47, 1122–1135. Minear, M., & Park, D. C. (2004). A lifespan database of adult facial stimuli. Behavior Research Methods, Instruments, & Computers, 36(4), 630–633.

49