https://orcid.org/0000-0002-1509-9037
Correction
2 Jan 2024: The PLOS One Staff (2024) Correction: (Not) part of the team: Racial empathy bias in a South African minimal group study. PLOS ONE 19(1): e0296581. https://doi.org/10.1371/journal.pone.0296581 View correction
Figures
Abstract
Minimal Group Paradigm (MGP) research suggests that recategorization with an arbitrarily defined group may be sufficient to override empathy biases among salient social categories like race. However, most studies utilizing MGPs do not consider sufficiently the socio-historical contexts of social groups. Here we investigated whether the recategorization of White participants into arbitrarily defined mixed-race teams using a non-competitive MGP would ameliorate racial empathy biases towards ingroup team members in the South African context. Sixty participants rated their empathic and counter-empathic (Schadenfreude, Glückschmerz) responses to ingroup and outgroup team members in physically painful, emotionally distressing, and positive situations. As anticipated, results indicated significant ingroup team biases in empathic and counter-empathic responses. However, mixed-race minimal teams were unable to override ingroup racial empathy biases, which persisted across events. Interestingly, a manipulation highlighting purported political ideological differences between White and Black African team members did not exacerbate racial empathy bias, suggesting that such perceptions were already salient. Across conditions, an internal motivation to respond without prejudice was most strongly associated with empathy for Black African target individuals, regardless of their team status. Together, these results suggest that racial identity continues to provide a salient motivational guide in addition to more arbitrary group memberships, even at an explicit level, for empathic responding in contexts characterized by historical power asymmetry. These data further problematize the continued official use of race-based categories in such contexts.
Citation: Deist M, Fourie MM (2023) (Not) part of the team: Racial empathy bias in a South African minimal group study. PLoS ONE 18(4): e0283902. https://doi.org/10.1371/journal.pone.0283902
Editor: Ashley Weinberg, University of Salford School of Health and Society, UNITED KINGDOM
Received: December 3, 2021; Accepted: March 20, 2023; Published: April 6, 2023
Copyright: © 2023 Deist, Fourie. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data that support the findings of this study are openly available at Open Science Framework, https://osf.io/9jqvy/.
Funding: This work was supported by a grant MF received from Stellenbosch University, Subcommittee A. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study.
Competing interests: The authors have declared that no competing interests exist.
Introduction
An enduring legacy of apartheid South Africa is the artificial racial categories that were engineered with the objective to benefit White people while discriminating against Black people (including Black African, Coloured, and Indian people) (Population Registration Act, 1950). While the power dynamics have shifted since 1994, with a ruling party that is now representative of the Black majority, these socio-historical race categories remain in use as part of government policy to address historically unjust racial inequalities [1]. Yet measures of redress have not rebalanced the scales, as the White minority continue to hold significant economic power and privilege, while wealth and income inequalities continue to be rampant within Black communities [2–4].
In some ways, apartheid South Africa could be viewed as a large, albeit ethically deeply inhumane, social laboratory—forging racial categories that maximized differences and response biases between groups that continue to be resistant to change today [5]. Racial categorization is socially constructed, however, and serves political goals, rather than carrying inherent or biological meaning [6]. As with other social categories, racial categories are therefore always in flux and amenable to manipulation. Indeed, abundant research in social psychology suggests that recategorization with an arbitrarily defined group may be sufficient to override racial response biases across perceptual, emotional, cognitive, and behavioral domains [7–9]. The question we grapple with here, however, is whether recategorization via the Minimal Group Paradigm is sufficient to negate deeply-rooted response biases inherent to contexts with protracted conflict and historical social hierarchies?
Empathy and the Minimal Group Paradigm
The Minimal Group Paradigm (MGP) is arguably one of the most influential strategies to impact intergroup relations in the absence of realistic conflicts of interest [10]. It involves assigning participants to novel groups based on some arbitrary criteria (e.g., whether they over or underestimate the number of dots on a screen), after which participants then typically report heightened ingroup identification and respond with ingroup biases on a variety of tasks. The success of the strategy could be attributed to the fact that social group biases (including racial biases), appear to depend heavily on the perceiver’s social motivation, which is context sensitive and can be modulated [11, 12] (for a review, see [13]). Membership in a mixed-race team thus appears to align people’s motivation in favor of the ingroup, regardless of its (racial) make-up. Some evidence suggests that the MGP is effective also in modulating implicit or neural responses, overriding visually salient race cues in favor of the ingroup team [14–17] (however, see also [18]).
A salient finding in intergroup contexts is that people respond with greater empathy when they share a social category with the person in distress, and stifled empathy when the person does not belong to their ingroup [19]. This phenomenon has variously been described as “an intergroup empathy gap”, “intergroup empathy bias”, or “parochial empathy” [20–22], and consistently shapes both behavioral and neural indices of empathic responding [23–25]. Impaired empathy for a social outgroup may mediate a host of negative outcomes, including reduced prosocial action and passive harm, which exacerbates intergroup conflict [26, 27]. Ameliorating intergroup empathy bias is therefore of great import.
Arbitrary group categorization by way of the MGP also appears effective in modulating empathic responses [28]. For example, the mere assignment of participants into non-relevant social groups appears sufficient to facilitate an ingroup bias in empathy for physical pain [29]. Similarly, across a series of studies, Cikara and colleagues [30] demonstrated that pitting novel teams against each other not only decreased empathy for the outgroup relative to the ingroup team, but also increased counter-empathic responses. Hence, compared to ingroup team members, participants responded with more pleasure when outgroup team members experienced adversity (Schadenfreude), and more displeasure when outgroup team members experienced success (Glückschmerz). Moreover, a recent study showed that ingroup activation biases in empathy-related brain regions associated with established social categories (e.g., religion) could be extended also to categories formed more arbitrarily (e.g., novel religious teams at war) [31], providing evidence that neural empathic responses can be modulated flexibly by group affiliation.
Not many studies employing the MGP in empathic responding have included race as a variable, however, and results from those that have are inconsistent. In two studies examining Chinese students’ neural empathic responses to physical pain expressed by Chinese and Caucasian faces, for example, the temporary group relationship formed through mixed-race teams seemed sufficient in reducing racial bias in neural empathic response to ingroup team members [32, 33]. Other studies, however, showed that racial biases in neural empathic responses were unaffected by minimal forms of group categorization [34, 35]. Specifically, whereas participants showed clear team biases (unaffected by race) on explicit group identification and implicit affective priming tasks, neural empathic response was indicative of an own-race bias [35]. A working hypothesis for this broader area of study might therefore be that the MGP is more effective in overriding racial response biases on measures over which one has some degree of conscious control (i.e., behavioral, implicit) than those measures beyond the scope of conscious control (i.e., neural activation).
Contextual factors
Historical power dynamics.
In contexts where race plays a significant role in group distinction, racial identity—and the multiple features of intergroup relations that covary with racial identity, such as majority/minority status—may provide an equally salient motivational guide for perception and behavior (including empathic responses) as group membership [36, 37]. Yet, despite the efficacy of MGPs in modulating racial biases under certain conditions, most studies utilizing the MGP do not sufficiently take into account the social contexts in which social groups exist [32–36]. Historically-embedded social hierarchies, for example, may continue to impact current intergroup relations and interfere with MGP categorization [37, 38].
Racialized hierarchies tend to shape social identities, which, in turn, influence attitudes and behaviors towards racial outgroups [36, 39]. Because groups at the top of hierarchical social structures often enjoy better educational, financial, occupational, and health outcomes, they have a direct stake in maintaining the hierarchical status quo [40–42]. Lower-status groups may therefore be perceived as potential threats to societal structures that safeguard ingroup privilege, which could bring forth status keeping behavior that reinforce hierarchical group boundaries [36, 39, 43].
Given the above, it is unclear whether arbitrarily-defined mixed-race teams (as per the MGP) are a reliable means to override racial empathy biases, even at an explicit level, in social contexts with an extended history and continuing legacies of asymmetric structural relations. In South Africa, the role of race in historical social hierarchies and contentious power dynamics may contribute to shifting group categorization from team membership back to race, thus eliciting racial empathy bias.
Racialized political ideologies.
Another contextual factor that could limit the efficacy of recategorization via the MGP, is the public salience of racialized political ideologies. Previous work has demonstrated the influence of ideological salience on intergroup relations and the importance of understanding cognitive processes within larger ideological contexts [44–46]. Opposing political ideologies have also been shown to modulate empathic responses, with greater empathy reported between individuals with similar political ideologies [47, 48].
In the South African context, political narratives are often racialized and shared in ways that arouse, rather than inform, their audience [49, 50], potentially increasing intergroup threat [36, 51, 52] and reinforcing racialized group boundaries. Two salient examples include the sensationalistic manner in which mainstream and social media regularly report on heated political debates surrounding Broad-based Black Economic Empowerment (BBBEE) and land expropriation without compensation. Whereas the BBBEE Act of 2013 aims to redress the apartheid legacies of Black economic exclusion [53, 54], the land redistribution program centers around the equitable transfer of underutilized, White-owned land to previously disadvantaged population groups in response to the historic forced removals of Black South Africans from their land [55]. Both political policies thus aim to redress past imbalances, yet are often depicted in inflammatory ways that incite racial polarization and intergroup threat (e.g., “If you see a beautiful piece of land, take it–Malema”, News24, 28-02-2017, [56], see also [57, 58]). Moreover, political parties tend to weaponize these policies to sow racial division and gain political traction [49, 50].
It may thus be reasonable to predict that when such racialized political ideologies are afforded greater saliency, it will contribute to greater racial empathy bias, also under MGP conditions.
Individual differences.
Beyond the social structures that might influence empathic responding, individual beliefs and attitudes can also influence racial empathy bias [59]. Notably, converging evidence suggests that racial response bias could be influenced by personal motivations to appear non‐prejudiced [60, 61]. In the South African context, White participants motivated by external societal pressures to respond without prejudice were associated with reduced neural activation in key empathy areas in response to racial outgroup members [62], whereas in another study, participants with greater internal motivation to appear non‐prejudiced responded with greater prosocial helping behavior toward racial outgroup members [63].
Acknowledging the extent of historical suffering experienced by the outgroup may also attenuate cross-racial empathic response bias [64]. Hughes and colleagues [65], for example, found that European American children expressed less racial bias after learning about historical racism experienced by African Americans. Indeed, various studies show that acknowledging the impact of historical suffering on continuing racialized inequality fosters support for social change and transformational redress [25, 66, 67].
Study aims
Here we investigated whether findings from empathy research using the MGP replicate within the South African context, given its oppressive socio-political history and continuing asymmetric structural relations. Specifically, we adopted a non-competitive MGP to examine whether the recategorization of historically advantaged White individuals (perceived racial ingroup, and racial group with whom the authors identify—see also Positionality Statement) into artificial mixed-race teams with Black African individuals (perceived racial outgroup) would be effective in overriding potential ingroup racialized response biases in empathy for physically painful, emotionally distressing, and positive events [68, 69]. We distinguished between empathy for these event types because previous research suggests they may be separable capacities with distinct affective, behavioral, and physiological markers [70–73]. In addition, we examined whether cross-racial empathic responding within arbitrarily-defined teams is affected when racial groups’ political ideological differences are made salient, as is often done in the media.
The participant group in the present study consisted of second-generation White students enrolled at a historically White university. Despite significantly greater diversification in the demographic composition of the student body at such higher education institutions post-apartheid, they often remain highly racialized spaces in which power inequalities, exclusionary practices, and racism persist [74–76]. Exploring the flexibility of group categorization and resulting empathic responding in such settings is therefore of great interest. In using racial categories, our intention is not to essentialize or reify race [77], but rather to unmask the influence of apartheid on specific groups of people and to accentuate the continued social implications of racial identity [78, 79]. In this spirit, race labels are employed to refer to groups of people whose lives have been substantially marked by the meanings of these labels, in full recognition of the complexities and heterogeneities that they represent [80].
We constrain our use of the word empathy to refer to an other-oriented affective reaction elicited by and congruent with the perceived well-being of someone experiencing either a negative or positive event, and counter-empathic responses as being incongruent with that person’s welfare [81]. Based on previous research, we hypothesized that (i) White participants will exhibit an ingroup team bias, such that empathy ratings will be greater (and counter-empathy ratings will be reduced) for ingroup compared to outgroup team members. We further anticipated that (ii) the MGP will reduce (but not override) racial empathy bias for ingroup team members, but not for outgroup team members, and that (iii) a racial political salience manipulation, where politically divisive ideologies are made salient, will have the effect of reinstating racial empathy bias amongst ingroup team members.
We also examined associations between participants’ individual difference characteristics and their degree of empathic responding towards racial outgroup members, regardless of team status. Here we hypothesized that (iv) a higher internal motivation (IMS) to respond without prejudice will be associated with greater empathy for racial outgroup members, whereas a higher external motivation (EMS) to respond without prejudice will be associated with reduced empathy for racial outgroup members. Finally, we predicted that (v) greater perceptions of Black historical suffering will be associated with greater empathy for racial outgroup members.
Materials and procedures
Participants
A power analysis using G*Power [82] indicated that a sample size of 58 participants would be sufficient to detect a moderate to low effect size (0.3) with reasonably good power (0.8) at an α level of .05. Sixty self-identified White, undergraduate students (50% female, Mage = 20.35 years, SDage = 1.62 years) participated in the present study and received ZAR50 (~ 4 USD) as compensation for their time (see S1 Text for further sampling information). Participants were randomly assigned to the control (n = 30) or manipulation (n = 30) conditions. All study procedures were approved by the University of Stellenbosch’s Humanities Research Ethics Committee and the research was carried out according to these guidelines.
Experimental design
We employed a 2 (racial political salience: control vs manipulation) x 2 (target team: Eagles vs Leopards) x 2 (target race: Black African vs. White) repeated measures multi-factorial design. The MGP was both blinded and non-competitive, and adapted from Cikara and colleagues [30]. A flow diagram of the data collection procedure is illustrated in Fig 1.
*All participants were assigned to the Eagles team. ** These included the (i) Internal and External Motivation to Respond Without Prejudice scales (IMS/EMS) [83]), and (ii) perceptions of Black African and White historical suffering.
First, to examine whether the MGP would override racial empathy bias for ingroup team members, participants were ostensibly assigned to one of two mixed-race teams, Eagles or Leopards. In reality, all participants were assigned to the Eagles team. They then completed a task measuring their empathic and counter-empathic reactions to members of both teams in physically painful, emotionally distressing, and positive situations.
Second, half of participants were also assigned to a racial political salience manipulation. Participants in this condition were asked to rate their support for five online petitions, some of which were linked to highly racialized political policies. They were then presented with diagrams supposedly representing the polarized petition support of White and Black African members of the Eagles and Leopards teams. These diagrams were fictitious, however, and did not reflect the opinions (or images) of any study participants. Rather, the diagrams were designed to highlight purported political ideological differences between White and Black African team members, making these identities more salient. We hypothesized that spotlighting these differences would exacerbate racial empathy bias amongst ingroup team members.
Team assignment.
A dot-estimation task (adapted from Montalan et al. [29]) was used ostensibly to assign participants into two teams (i.e., Eagles or Leopards). Participants were shown a random display of dots for five seconds and were asked to write down the estimated number of dots (see stimulus in S1 Fig). Each participant was then assigned a cognitive profile supposedly based on whether they overestimated or underestimated the number of dots. In actual fact, all participants were assigned to the Eagles team. The results of the dot-estimation task were not recorded, given that the task simply served to give rise to a minimal group effect.
Team identification.
Participants rated their identification with the teams by responding to two statements regarding each team (“I identify with the [Eagles/Leopards]” and “I feel connected to the [Eagles/Leopards]”) on a scale ranging from 1 (strongly disagree) to 9 (strongly agree). These items were highly correlated for each team (rs > .50, ps < .001), and thus averaged to create a composite identification score for the ingroup (Eagles) and outgroup (Leopards) team.
Racial political salience manipulation.
In this manipulation we increased the salience of potential political ideological differences between White and Black African people to examine its effect on empathic responding, particularly with respect to the newly-formed racially-mixed ingroup (i.e., Eagles team).
Participants in the manipulation condition were presented with five supposedly real online petitions: one petition sought support against a progressive policy that would benefit historically disadvantaged racial groups (White-biased petition; “Urge parliament to tone down the Broad-based Black Economic Empowerment (BBBEE) Act, so that qualified White people stand a better chance of employment”); two petitions sought support towards more aggressive policies that would benefit historically disadvantaged racial groups at the expense of White people (Black-biased petitions; “Urge Stellenbosch University to expand free tertiary education for Black students at the expense of White students” and “Urge parliament to make urgent provision for the expropriation without compensation of White-owned land”), and two petitions sought support towards more general, politically neutral, campus issues, namely water saving efforts and drinking and driving (arbitrary petitions—see stimulus in S1 Text). These petitions were presented to participants using E-Prime 3.0 software [84] and participants were asked to rate their support for each petition using a visual analogue scale (VAS) ranging from 1 (not at all) to 7 (very much).
Once they completed each rating, participants were shown a diagram supposedly representing petition support amongst other Eagle and Leopard team members (see Fig 2A–2C). The diagrams indicated that White individuals were largely in support of “White-biased petitions” and Black African individuals were largely in support of “Black-biased petitions”, irrespective of their team membership, whereas all individuals supported the arbitrary petitions.
A) The diagram indicated that White team members were largely in support of a petition that urged parliament to revoke a progressive policy that would benefit historically disadvantaged groups. B) The diagram indicated that Black African team members were largely in support of petitions that urged parliament to implement aggressive policies at the expensive of previously advantaged White people. C) The diagram indicated broad support from all team members towards arbitrary petitions about general campus matters. Note. None of the individuals in these images participated in the current study.
Empathy task.
Empathy and counter-empathic responses were assessed using a self-reported reaction task that measured how good/bad participants felt about the fortunes/misfortunes of others. Each participant was presented with 24 events (8 physically painful events, e.g., “James’s fingers got slammed in a car door”/“Anele cut her foot on broken glass”; 8 emotionally distressing events, e.g., “Nicola and her best friend haven’t spoken in months”/“Thandi hasn’t seen her mother in a year”; and 8 positive events, e.g., “Siyanda scored a goal for his soccer team”/“Nico found a R50 note on the floor”) using E-Prime 3.0 software [84] (for a full list of stimuli, see S1 Text).
Each event type was randomly paired with the photograph of the target individual who supposedly experienced the event. These photographs were not photos of actual participants in the study, but were collected with prior consent from random students on campus and included 12 Black African and 12 White individuals (six males and six females in each racial group). To make the group identity salient, each photograph was accompanied by a background color (red or blue) and team logo (Eagle or Leopard) signifying supposed team membership. The team membership, event type, and team logo color were counterbalanced for each target individual across participants and the presentation order was randomized.
After each event, participants were asked to indicate: (1) “How sorry does this make you feel?” and (2) “How good does this make you feel?” on continuous VAS sliders ranging from 1 (not at all) to 9 (extremely). Empathic responses for physically painful and emotionally distressing events were assessed with “sorry” ratings, whereas counter-empathic responses were assessed with “good” ratings. For positive events, empathic responses were assessed with “good” ratings, whereas counter-empathic responses were assessed with “sorry” ratings.
Motivation to respond without prejudice.
The Internal Motivation Scale (IMS) and External Motivation Scale (EMS) [83] were used to determine internal (e.g., “I attempt to act in nonprejudiced ways toward Black African people because it is personally important to me”) and external motivations (e.g., “Because of today’s PC (politically correct) standards I try to appear nonprejudiced toward Black African people”) to respond without prejudice. Each item was measured on a Likert-type scale, ranging from 1 (strongly disagree) to 9 (strongly agree). Higher IMS scores indicate a higher likelihood that a participant’s reason for responding without prejudice is due to personal motivations, whereas higher EMS scores indicate a higher likelihood that a participant’s reason for responding without prejudice is due to external social motivations.
Perceptions of historical suffering.
To assess perceptions of historical suffering, participants were asked to respond to the questions “Historically, how much suffering has Black African people in South Africa endured?” and “Historically, how much suffering has White people in South Africa endured?” These questions were rated on Likert-type scales ranging from 1 (not much at all) to 9 (very much). Higher scores thus indicated greater perceptions of historical suffering.
Procedure
During recruitment, the study was presented as a problem-solving challenge. Similar to the cover story used by Cikara and colleagues [30], prospective participants were told that the study purpose was to examine whether knowledge about team members enhances team performance in subsequent problem-solving tasks. Participants were thus blinded to the true focus of the present investigation to minimize socially desirable responses.
All procedures were conducted in private computer rooms on campus. Upon arrival, experimental procedures were explained and written informed consent was obtained from all participants.
Participants were first asked to complete a short demographic questionnaire reporting their age, gender, racial group they identify with (inclusion criteria), and home language.
Next, they were told the following cover story: “The goal of our study is to assess cognitive functioning in teams and whether knowledge about team members enhances team performance in subsequent problem-solving challenges. We will be assigning you to a team (either Eagles or Leopards) with a similar cognitive processing profile in a moment. Each week, for the duration of the data collection period, these two teams participate in a series of cognitive problem-solving challenges. Though they are not in competition with each other (thus independent), we want to see whether knowing something personal about your fellow team members will enhance your performance. So, some participants will read about random news events, and some will read about team members’ recent experiences. You are in the latter category. When a team successfully completes 50 tasks before the end of the semester, all members of that team will win extra cash. You will be contacted in due course if this happens.”
After this briefing, participants completed the dot-estimation task to supposedly divide them into two teams (Eagles or Leopards) based on their cognitive profile (see S1 Fig). They were either identified as under-estimators (individuals who supposedly underestimated the number of dots in the dot-display) or over-estimators (individuals who supposedly overestimated the number of dots in the dot-display). Participants were assured that neither of the two cognitive profiles were deemed better than the other and that it simply influenced how they processed information. Unbeknownst to them, however, all participants were assigned to the Eagles team.
After team assignment, participants completed the computerized team identification task. Furthermore, in line with the cover story, participants were asked to write down a recent experience (either positive or negative) that would supposedly be read by future team members.
This was followed by the manipulation task for the 30 participants randomly assigned to this condition, with the supporting cover story: “As part of getting to know the Eagles and Leopards team members, you will be presented with a list of petitions on recent social issues that investigate social attitudes. You will be asked to indicate the extent to which you support each statement from 1 (not at all) to 7 (very much). After indicating the extent to which you support each petition, you will be shown a network diagram of the extent to which other Eagles and Leopards team members have supported each petition.” The three petitions of interest (i.e., 1 White-biased petition and 2 Black-biased petitions) were followed by diagrams supposedly indicating the racially polarized manner in which other Eagles and Leopards team members had responded to the petitions, whereas the arbitrary petitions were followed by diagrams indicating broad support from all team members (see Fig 2). Those participants in the control condition neither participated nor received any feedback on this task.
Next, participants across both conditions were presented with the computerized empathy task and asked to rate the extent to which they felt “sorry” and/or “good” about recent physically painful, emotionally distressing, and positive experiences of other ingroup (Eagles) and outgroup (Leopards) team members. Each participant was randomly presented with 24 scenarios—8 representing physically painful events (4 ingroup team members: 2 Black African, 2 White, and 4 outgroup team members: 2 Black African, 2 White), 8 representing emotionally distressing events, and 8 representing positive experiences.
Following these measures, participants were presented with timed problem-solving tasks not analyzed here, which supported the cover story (see S1 Text). We emphasized that the problem-solving tasks was not a competition between teams and that each team that successfully completed 50 tasks in the assigned time frame will win extra cash. Finally, participants completed measures assessing motivation to respond without prejudice and perceptions of historical suffering.
All participants were probed for suspicion before being debriefed fully. They were granted the opportunity to ask questions and allowed to withdraw their data should they wish to (none did). Participants were asked not to discuss the study and data collection methods with other students.
Results
Manipulation checks
Team manipulation.
The efficacy of the team manipulation was assessed by analyzing the extent to which participants identified with the ingroup (Eagles) versus the outgroup (Leopards) team. A paired-samples t-test indicated that participants’ ingroup team identification score (M = 7.16, SD = 0.73) was significantly greater than their outgroup team identification score (M = 2.15, SD = 0.43), t(59) = 40.54, p < .001, d = 5.23. The team manipulation thus appeared successful.
Racial political salience manipulation.
In the manipulation condition, participants responded to the five petitions as expected: the petition that would benefit White people, yet negatively impact historically disadvantaged people (i.e., White-biased petition) was strongly supported, whereas those petitions that would advance equality (i.e., Black-biased petitions) went unsupported. Specifically, support for the White-biased petition (BBBEE act: M = 5.17, SD = 1.18) was significantly above the scale midpoint (4), t(29) = 5.43, p < .001, d = 0.99. By contrast, support for the Black-biased petitions (Tertiary education: M = 1.87, SD = 1.01; Land reform: M = 1.77, SD = 1.01) was significantly below the scale midpoint, t(29) = -11.59, p < .001, d = -2.12 and t(29) = -12.16, p < .001, d = -2.22, respectively. Support for the arbitrary petitions (Water saving: M = 6.10, SD = 0.96; Drunk driving: M = 6.40, SD = 0.93) were significantly above the scale midpoint, t(29) = 11.99, p < .001, d = 2.19 and t(29) = 14.10, p < .001, d = 2.57, respectively. Given that participants’ responses were in line with the diagrams showing racial polarization for the petitions, it could be deduced that the manipulation afforded greater salience to racialized political identities.
Minimal group empathic responses
Mixed factorial ANOVAs tested the influence of the between-group racial political salience manipulation (control vs manipulation), and the repeated-measures target team (Eagles vs Leopards) and target race (Black African vs White) variables on self-reported empathy (see S1 Table for extended ANOVA results). Empathy and counter-empathic responses are presented in Table 1.
Self-reported empathic responses.
A 2 (racial political salience: control vs manipulation) x 2 (target team: Eagles vs Leopards) x 2 (target race: Black vs White) mixed-factorial ANOVA detected significant main effects for target team in response to physically painful, F(1, 58) = 89.97, p < .001, ƞ2 = .61, emotionally distressing, F(1, 58) = 20.25, p < .001, ƞ2 = .26, and positive events, F(1, 58) = 30.68, p < .001, ƞ2 = .35. As anticipated, participants felt more “sorry” for ingroup team members (Eagles) in response to physically painful and emotionally distressing events, and more “pleasure” when ingroup team members (Eagles) experienced positive events, than for outgroup team members (Leopards) (see Fig 3A).
Empathy ratings for (A) ingroup (Eagles) versus outgroup (Leopards) teams, for (B) White versus Black African team members, and for (C) Team x Race interactions. Empathy ratings in response to physically painful and emotionally distressing events were determined by how “sorry” the events made participants feel, whereas empathy ratings in response to positive events were determined by how “good” the events made participants feel. Error bars indicate standard error of the mean. ***p < .001.
We also detected significant main effects for target race in response to physically painful, F(1, 58) = 43.91, p < .001, ƞ2 = .43; emotionally distressing, F(1, 58) = 65.90, p < .001, ƞ2 = .53; and positive events, F(1, 58) = 73.64, p < .001, ƞ2 = .56. On average, participants felt more “sorry” for White individuals in response to physically painful and emotionally distressing events, and more “pleasure” when White individuals experienced positive events, than for Black African individuals (see Fig 3B). Contrary to our predictions, we did not observe significant interaction effects between target team and target race, indicating that the racial bias in empathic responding was similar in size in both ingroup and outgroup teams.
Finally, the mixed-factorial ANOVAs did not detect any significant main effects for the racial political salience manipulation in response to physically painful, F(1, 58) = 0.53, p = .469, ƞ2 = .01, emotionally distressing, F(1, 58) = 0.00, p = .952, ƞ2 = .00, or positive events, F(1, 58) = 0.63, p = .430, ƞ2 = .01. Contrary to our expectations, these results suggest that our manipulation did not exacerbate any racial empathy biases amongst participants. We did not observe any other significant interaction effects (see Fig 3C).
Self-reported counter-empathic responses.
A 2 (racial political salience: control vs manipulation) x 2 (target team: Eagles vs Leopards) x 2 (target race: Black vs White) mixed-factorial ANOVA on counter-empathic responses detected a significant main effect for target team in response to physically painful, F(1, 58) = 103.09, p < .001, ƞ2 = .64, emotionally distressing, F(1, 58) = 16.08, p < .001, ƞ2 = .22, and positive events, F(1, 58) = 28.02, p < .001, ƞ2 = .33. On average, participants expressed more Schadenfreude in response to the physical pain and emotional distress of outgroup team members (Leopards), and more Glückschmerz when outgroup team members (Leopards) experienced positive events, compared to ingroup team members (Eagles) (see Fig 4A).
Counter-empathic ratings for (A) ingroup (Eagles) versus outgroup (Leopards) teams, for (B) White versus Black African team members, and for (C) Team x Race interactions. Counter-empathic ratings in response to physically painful and emotionally distressing events were determined by how “good” the events made participants feel, whereas counter-empathic ratings in response to positive events were determined by how “sorry” the events made participants feel. Error bars indicate standard error of the mean. ***p < .001.
We again observed significant main effects for target race in response to physically painful, F(1, 58) = 81.74, p < .001, ƞ2 = .58, emotionally distressing, F(1, 58) = 68.56, p < .001, ƞ2 = .54, and positive events, F(1, 58) = 149.57, p < .001, ƞ2 = .72. On average, participants expressed more Schadenfreude in response to Black African individuals’ physical pain and emotional distress, and more Glückschmerz when Black African individuals experienced positive events, compared to White individuals (see Fig 4B).
We also detected a significant target team x target race interaction in response to physically painful events, F(1, 58) = 11.35, p = .001, ƞ2 = .16. In line with predictions, post-hoc comparisons indicated that participants’ “pleasure” ratings for Black African compared to White team members were significantly greater for the outgroup (Leopards) compared to the ingroup (Eagles) team (see Fig 4C).
Finally, we again did not detect a significant main effect for the racial political salience manipulation in response to physically painful, F(1, 58) = 0.00, p = .972, ƞ2 = .00, emotionally distressing, F(1, 58) = 0.18, p = .673, ƞ2 = .00, or positive events, F(1, 58) = .20, p = .653, ƞ2 = .00. These results suggest that ingroup race biases with regard to counter-empathic responses were not exacerbated by the manipulation.
Individual differences
We examined the extent to which individual difference characteristics, including motivation to respond without prejudice (as measured by IMS/EMS scores) and perceived Black and White historical suffering, contributed to empathy in response to Black African target individuals. For these analyses we collapsed empathy ratings across the control and manipulation conditions (for a comparison of individual difference scores between the control and manipulation conditions, see S2 Table). Furthermore, given that we observed racial empathy biases similar in magnitude for both ingroup and outgroup teams, we collapsed empathic responses across the Eagles and Leopards teams for each event type. Descriptive statistics and variable intercorrelations are presented in Table 2 (separate correlations for each team are presented in S3A and S3B Table).
As anticipated, higher IMS scores were associated with greater empathy in response to Black African targets in physical pain and emotional distress, whereas higher EMS scores were associated with less empathy in response to Black African targets in physical pain and emotional distress, and less pleasure when they experienced positive events. In turn, greater perceptions of Black historical suffering were positively associated with empathy for Black African targets in physical pain, whereas greater perceptions of White historical suffering were negatively associated with empathy for Black African targets in physical pain.
Next, we performed a series of simultaneous regressions with (i) IMS and EMS scores and (ii) perceptions of Black and White historical suffering as predictors of empathy in response to Black African targets experiencing physically painful, emotionally distressing, and positive events (see Table 3; for extended regression results, see S4 Table). Here IMS emerged as the only significant predictor of empathy for both physically painful and emotionally distressing events (βs > .58, ps < .01). Results of these analyses thus show that White participants’ internal motivation to respond without prejudice contributes significantly to greater empathy in response to the misfortunes experienced by Black African individuals.
It should be noted that due to the high correlations between IMS and EMS scores (-.89) and Black and White Historical suffering (-.72), the estimated regression parameters could be affected negatively by collinearity. We therefore examined collinearity statistics for these variables in our sample (see Table 3). Our analysis showed that collinearity for our measures of historical suffering were negligible, whereas the IMS and EMS variables showed moderate collinearity, with a tolerance of .20 for both and a variance inflation factor (VIF) of 5.13 and 5.02, respectively. Although these conditions are not ideal, they do fall within the acceptable guidelines of tolerance (> 0.1) and VIF values (< 10) [85, 86].
Discussion
In the present study, we examined whether recategorization into arbitrarily defined groups would be effective in bridging empathy biases constructed under generations of socio-historical and asymmetric conflict. Specifically, we investigated whether the recategorization of White individuals into mixed-race (Black African and White) teams using a non-competitive MGP design would ameliorate racial empathy biases towards ingroup team members in the South African context. As expected, we observed a team effect, with participants reporting significantly greater empathy and reduced counter-empathic responses for ingroup team members experiencing physically painful, emotionally distressing, and positive events. However, recategorization through team assignment did not override racial empathy bias towards ingroup team members for any event type: We observed significant racial biases in empathy and counter-empathic responses towards both ingroup and outgroup team members, with an internal motivation to respond without prejudice predicting greatest empathy for Black African target individuals. Interestingly, a manipulation that highlighted racialized political ideological differences between racial groups did not further exacerbate racial empathy bias in our sample, suggesting that such perceptions were already salient. These results imply that the MGP may be ineffective in overriding racial empathy bias, even at an explicit/behavioral level, in contexts characterized by overt power asymmetry and historical social hierarchies.
Findings from MGP studies report that recategorization with an arbitrarily-defined group may be sufficient in overriding behavioral and even automatic racial biases [14–17], also when it comes to empathy [32–35]. Given the significantly greater empathy we observed for racial ingroup versus outgroup members, even in the ingroup team, our results do not support these findings. What makes our results especially noteworthy is that racial empathy bias was observed despite task demands that emphasized attention to targets’ emotional state and psychological experience. In previous studies, active perspective taking not only increased the reported saliency of pain experienced by others, but also attenuated explicit and implicit expressions of racial empathy bias [33, 87].
Previous neuroimaging work suggests that racial empathy bias may persist within a minimal group context during the early perceptual stages of neural processing, but dissipates at the later cognitive stages of neural processing and subsequent explicit outcomes [32, 34, 35]. A possible reason for this finding is that humans may automatically detect and encode race as a by‐product of an adaptation to identify fellow group members in early perceptual stages of neural response [88, 89]. However, this bottom-up bias in neural perception can be modulated by top-down, cognitive evaluation that takes into account more complex social motivations and contexts [12, 35, 90]. Indeed, in our study we examined responses at this latter, explicit, level of evaluation.
In trying to interpret our findings, one could argue that beliefs regarding how much life hardship/pain Black African people have previously endured may have impacted empathic responses. Deska and colleagues [69] and Trawalter and colleagues [68] found that White participants empathized less with Black individuals when they assumed a priori that Black people experience less pain than White people. By this account, racial empathy bias is rooted in the perceived lower social status of Black people, and the stereotypical beliefs of White participants that Black people “toughen up” (perhaps as a coping mechanism) due to the various hardships typically associated with lower status [68, 69]. Our results do not support this interpretation, however, and suggests the opposite: greater perceived historical suffering of Black African people was associated with greater empathy in response to Black African target individuals experiencing physical pain.
Alternatively, racial empathy bias observed in our study might be rooted in the salience of social identity. Social identity theory [91] holds that group categorization shifts one’s self-concept from the individual to the collective, which often manifests as intergroup bias [51, 92]. While minimal groups have the potential to generate stronger group biases than naturally occurring social groups (e.g., gender, race) [93], these biases tend to be context sensitive. Moreover, ingroup favoritism tends to increase within naturally-occurring groups if subjects show a deeper attachment to those groups [94]. In South Africa, the hierarchical race categories forged by the apartheid government is still considered a primary constituent of identity [1, 5, 95, 96]. Given this racialized and asymmetric socio-political context, it is reasonable to assume that the minimal groups created in our study were not meaningful enough to shift saliency away from White identity towards the new ingroup.
Another factor that may exacerbate racial empathy bias is status threat. Intergroup threat theory [43] suggests that a threat to one’s group’s social status could increase the salience of ingroup identity, resulting in higher levels of intergroup bias [36, 51, 52]. In the South African post-apartheid context, White people lost their political power to the Black majority, yet largely continue to hold economic power and privilege [95, 96]. Such a shift in power may destabilize White status and challenge the legitimacy of the existing status quo, thereby augmenting intergroup threat [39, 97]. Highlighting racialized differences in political ideologies as we did in our manipulation condition, especially those pertaining to economic prosperity (e.g., affirmative action legislation and land expropriation without compensation), was thus anticipated to exacerbate racial empathy bias, potentially through increased intergroup status threat.
Yet, Black African team members’ supposed support of policies promoting their interests at the expense of White people appeared to have little effect on White participants’ expressed racial empathy bias. All this while participants clearly supported the White-biased petition (toning down of the BBBEE Act) and opposed the two Black-biased petitions (expansion of free tertiary education and increased land expropriation). Given the polarizing race-based narratives that are often shared in sensationalistic media reports (e.g., [56, 98, 99]), we believe it is possible that the ideologies highlighted by our manipulation were assumed by participants already, thereby muting the impact thereof. Future research should consider whether adapting the racial political salience manipulation to include views of racial outgroups that align with the political and ideological identities of the ingroup might mitigate against racial bias.
Our results further showed significant associations between individual differences and racial empathy bias. Here, an internalized motivation to be free of prejudice (as measured by IMS scores) held the most sway in the expression of empathy towards Black African individuals’ physical and emotional misfortunes. Additionally, we observed significant associations between expressed racial empathy and an external motivation to respond without prejudice (as measured by EMS scores) and perceived Black and White historical suffering.
Group position theory suggests that intergroup conflict is rooted in perceptions of entitlement and status threat [100]. Accordingly, perceptions of entitlement are fostered by dominant group members to justify unfair advantages begotten through an unequal social system, and anyone challenging this entitlement is consequently perceived as a threat [66, 101]. A sense of entitlement may explain why greater perceptions of White historical suffering in our sample were associated with lower empathy towards Black African individuals in physical pain, whereas greater acknowledgement of Black historical suffering was associated with greater empathy for their pain. Relatedly, perceived disenfranchisement may explain why White participants who felt compelled to act in non‐prejudiced ways by external pressures (as measured by the EMS) were more likely to hold greater perceptions of White historical suffering.
Some limitations of the present research are worth observing. First, while the presented results provide a glimpse into the complexities of cross-racial empathic responding in the South African context, the data are not necessarily representative of the broader population. Participants in the current study were young, White, educated, university students. Considering that only 7% of the South African population between 25 and 34 years old has a tertiary education [102], university students are considered a super elite demographic that typically enjoys a more privileged lifestyle compared to most South Africans. Young and educated White South Africans have also been associated with more progressive racial perceptions and motivations [67], so that the racial empathy biases we observed might be more substantial in the general population.
It should further be noted that the current study focused on the views and responses of a self-identified sample of White individuals as they pertained to Black African individuals, without intending to represent race as a binary construct. During apartheid, the hierarchical power division embedded in the system was not a simple, two-tier matrix that only distinguished “Whites” from “non-Whites”, but a complex system involving four racial groups (White, Black African, Coloured, and Indian people) [1]. This hierarchical system effectively created division between the different racial groups that continue to find expression today [103]. To better understand the complex social dynamics within the South African context, future studies should examine minimal group empathic responding also amongst historically disadvantaged groups.
Finally, in line with the original MGP boundary conditions [10, 91], the current study adopted a non-competitive MGP to examine intergroup empathy bias. Recently some MGP studies have introduced a salient competitive condition to examine its impact on intergroup bias [30, 104]. Across these studies, greater competitive threat from outgroup teams seemed to enhance ingroup identity, thereby increasing intergroup bias in empathic responding [23, 51, 105, 106]. Competition between minimal teams might thus, ironically, reduce racial bias within one’s own team. Future MPG research should consider including a competitive condition to investigate whether it would be effective in reducing racial empathy bias in contexts emerging from long-standing and asymmetrical social conflict.
Conclusion
Despite significant progress in racial integration, particularly in higher education, post-apartheid South Africa continues to suffer from racial division [1, 96]. The present results suggest that racial identity continues to play a pivotal role in empathic responding in South Africa: non-competitive mixed-race minimal groups were unsuccessful in overriding explicit ingroup racial empathy biases in a sample of White South Africans, regardless of the event type. Our study further highlights the impact of individual beliefs and attitudes, with an internal motivation to respond without prejudice being most strongly associated with cross-racial empathy. These findings underscore the complexities of intergroup dynamics in contexts of historical power asymmetry. It further calls for more empirical attention devoted to understanding the impact that the continued use of race-based labels, fraught with historical preconceptions, has in such contexts [107]. Given increasing racial diversity in social and educational contexts, it is of intense interest to devote resources into the development of strategies aimed at bridging empathy divides in present-day South Africa.
Positionality statement
Mindful that our social identities can influence our research [108], the authors wish to provide the reader with potentially relevant information about our social positioning: both MD and MF identify as White and South African. We believe referring to racial groups is warranted to unmask the continued social implications of race in the South African context, in full recognition of the intersectional complexities and heterogeneities within these artificial categories.
Supporting information
S1 Fig. Sample stimulus for dot-estimation task.
https://doi.org/10.1371/journal.pone.0283902.s002
(TIF)
S1 Table. Empathic and counter-empathic responses: Extended ANOVA results.
https://doi.org/10.1371/journal.pone.0283902.s003
(DOCX)
S2 Table. Variations in individual difference scores for control and manipulation conditions.
https://doi.org/10.1371/journal.pone.0283902.s004
(DOCX)
S3 Table. Individual difference measures: Descriptive statistics and variable intercorrelations.
A) Ingroup team (Eagles) targets. B) Outgroup team (Leopards) targets.
https://doi.org/10.1371/journal.pone.0283902.s005
(DOCX)
S4 Table. Individual difference measures: Extended results of simultaneous regressions predicting self-reported empathy for different event types.
https://doi.org/10.1371/journal.pone.0283902.s006
(DOCX)
References
- 1. Hino H, Leibbrandt M, Machema R, Shifa M, Soudien C. Identity, inequality and social contestation in the post-apartheid South Africa 2018.
- 2. Chatterjee A. Measuring wealth inequality in South Africa: An agenda. Dev South Afr. 2019;36(6):839–59.
- 3. Francis D, Webster E. Poverty and inequality in South Africa: Critical reflections. Dev South Afr. 2019;36(6):788–802.
- 4. Statistics South Africa. Inequality trends in South Africa: A multidimensional diagnostic of inequality. Pretoria, South Africa: Statistics South Africa, 2019 03-10-19.
- 5. Posel D. Race as common sense: Racial classification in twentieth-century South Africa. Afr Stud Rev. 2001;44(2):87–114.
- 6. Richeson JA, Sommers SR. Toward a social psychology of race and race relations for the twenty-first century. Annu Rev Psychol. 2016;67:439–63. pmid:26361050
- 7. Ratner KG, Dotsch R, Wigboldus DH, van Knippenberg A, Amodio DM. Visualizing minimal ingroup and outgroup faces: Implications for impressions, attitudes, and behavior. J Pers Soc Psychol. 2014;106(6):897–911. pmid:24841095
- 8. Paluck EL, Green DP. Prejudice reduction: What works? A review and assessment of research and practice. Annu Rev Psychol. 2009;60:339–67. pmid:18851685
- 9. Young SG, Hugenberg K. Mere social categorization modulates identification of facial expressions of emotion. J Pers Soc Psychol. 2010;99(6):964–77. pmid:20919774
- 10. Otten S. The Minimal Group Paradigm and its maximal impact in research on social categorization. Curr Opin Psychol. 2016;11:85–9.
- 11. Amodio DM, Cikara M. The social neuroscience of prejudice. Annu Rev Psychol. 2021;72:439–69. pmid:32946320
- 12. Amodio DM. The neuroscience of prejudice and stereotyping. Nat Rev Neurosci. 2014;15(10):670–82. pmid:25186236
- 13.
Van Bavel JJ, Hackel LM, Xiao YJ. The group mind: The pervasive influence of social identity on cognition. In: Decety J, Christen Y, editors. New Frontiers in Social Neuroscience. Switzerland: Springer International Publishing; 2014. p. 41–56.
- 14. Ratner KG, Amodio DM. Seeing “us vs. them”: Minimal group effects on the neural encoding of faces. J Exp Soc Psychol. 2013;49(2):298–301.
- 15. Simon JC, Gutsell JN. Effects of minimal grouping on implicit prejudice, infrahumanization, and neural processing despite orthogonal social categorizations. Group Process Intergroup Relat. 2020;23(3):323–43. pmid:33981179
- 16. Van Bavel JJ, Cunningham WA. Self-categorization with a novel mixed-race group moderates automatic social and racial biases. Pers Soc Psychol Bull. 2009;35(3):321–35. pmid:19098257
- 17. Van Bavel JJ, Packer DJ, Cunningham WA. The neural substrates of in-group bias: A functional magnetic resonance imaging investigation. Psychol Sci. 2008;19(11):1131–9. pmid:19076485
- 18. Yan Z, Schmidt SN, Saur S, Kirsch P, Mier D. The effect of ethnicity and team membership on face processing: A cultural neuroscience perspective. Soc Cogn Affect Neurosci. 2019;14(9):1017–25. pmid:31680173
- 19. Zaki J, Cikara M. Addressing empathic failures. Curr Dir Psychol Sci. 2015;24(6):471–6.
- 20. Gutsell JN, Inzlicht M. Intergroup differences in the sharing of emotive states: Neural evidence of an empathy gap. Soc Cogn Affect Neurosci. 2012;7(5):596–603. pmid:21705345
- 21. Bruneau EG, Cikara M, Saxe R. Minding the gap: Narrative descriptions about mental states attenuate parochial empathy. PLoS ONE. 2015;10(10):e0140838. pmid:26505194
- 22.
Fourie MM, Subramoney S, Gobodo-Madikizela P. A less attractive feature of empathy: Intergroup empathy bias. In: Kondo M, editor. Empathy: An Evidence-based Interdisciplinary Perspective. London: INTECH; 2017. p. 45–61.
- 23. Cikara M, Bruneau EG, Saxe R. Us and them: Intergroup failures of empathy. Curr Dir Psychol Sci. 2011;20(3):149–53.
- 24. Hein G, Silani G, Preuschoff K, Batson CD, Singer T. Neural responses to ingroup and outgroup members’ suffering predict individual differences in costly helping. Neuron. 2010;68(1):149–60. pmid:20920798
- 25. Fourie MM, Moore-Berg SL. We cannot empathize with what we do not recognize: Perceptions of structural versus interpersonal racism in South Africa. Frontiers in Psychology. 2022;13. pmid:36248600
- 26.
Hoffman ML. Empathy and prosocial behavior. In: Lewis M, Haviland-Jones J, Barrett L, editors. Handbook of Emotions. 3. New York, NY: Guilford Press; 2008. p. 440–55.
- 27. Bruneau EG, Cikara M, Saxe R. Parochial empathy predicts reduced altruism and the endorsement of passive harm. Soc Psychol Personal Sci. 2017;8(8):934–42. pmid:29276575
- 28. Masten CL, Gillen-O’Neel C, Brown CS. Children’s intergroup empathic processing: The roles of novel ingroup identification, situational distress, and social anxiety. J Exp Child Psychol. 2010;106(2–3):115–28. pmid:20202649
- 29. Montalan B, Lelard T, Godefroy O, Mouras H. Behavioral investigation of the influence of social categorization on empathy for pain: A minimal group paradigm study. Front Psychol. 2012;3(389). pmid:23087657
- 30. Cikara M, Bruneau E, Van Bavel JJ, Saxe R. Their pain gives us pleasure: How intergroup dynamics shape empathic failures and counter-empathic responses. J Exp Soc Psychol. 2014;55:110–25. pmid:25082998
- 31. Vaughn DA, Savjani RR, Cohen MS, Eagleman DM. Empathic neural responses predict group allegiance. Front Hum Neurosci. 2018;12(302). pmid:30108493
- 32. Shen F, Hu Y, Fan M, Wang H, Wang Z. Racial bias in neural response for pain is modulated by minimal group. Front Hum Neurosci. 2018;11(661). pmid:29379429
- 33. Sheng F, Han S. Manipulations of cognitive strategies and intergroup relationships reduce the racial bias in empathic neural responses. NeuroImage. 2012;61(4):786–97. pmid:22542636
- 34. Contreras-Huerta LS, Baker KS, Reynolds KJ, Batalha L, Cunnington R. Racial bias in neural empathic responses to pain. PLoS ONE. 2013;8(12):e84001. pmid:24376780
- 35. Contreras-Huerta LS, Hielscher E, Sherwell CS, Rens N, Cunnington R. Intergroup relationships do not reduce racial bias in empathic neural responses to pain. Neuropsychologia. 2014;64:263–70. pmid:25281885
- 36. Kteily N, Richeson JA. Perceiving the world through hierarchy-shaped glasses: On the need to embed social identity effects on perception within the broader context of intergroup hierarchy. Psychol Inq. 2016;27(4):327–34.
- 37. Xiao YJ, Coppin G, Van Bavel JJ. Clarifying the role of perception in intergroup relations: Origins of bias, components of perception, and practical implications. Psychological Inquiry. 2016;27(4):358–66.
- 38. Fourie MM, Deist M, Moore-Berg SL. Hierarchies of being human: Intergroup dehumanization and its implication in present-day South Africa. Peace and Conflict: Journal of Peace Psychology. 2022;28(3):345–60.
- 39. Fiske ST, Dupree CH, Nicolas G, Swencionis JK. Status, power, and intergroup relations: The personal is the societal. Curr Opin Psychol. 2016;11:44–8. pmid:27453923
- 40. Kraus MW, Onyeador IN, Daumeyer NM, Rucker JM, Richeson JA. The misperception of racial economic inequality. Perspect Psychol Sci. 2019;14(6):899–921. pmid:31505132
- 41.
Piff PK, Kraus MW, Keltner D. Unpacking the inequality paradox: The psychological roots of inequality and social class. In: Olson JM, editor. Adv Exp Soc Psychol. Burlington: Academic Press; 2018. p. 53–124.
- 42. Jost JT, Banaji MR, Nosek BA. A decade of system justification theory: Accumulated evidence of conscious and unconscious bolstering of the status quo. Polit Psychol. 2004;25(6):881–919.
- 43. Riek BM, Mania EW, Gaertner SL. Intergroup threat and outgroup attitudes: A meta-analytic review. Pers Soc Psychol Rev. 2006;10(4):336–53. pmid:17201592
- 44. Costa‐Lopes R, Pereira CR, Judd CM. Categorisation salience and ingroup bias: The buffering role of a multicultural ideology. Int J Psychol. 2014;49(6):508–12. pmid:25355674
- 45. Todd AR, Galinsky AD. The reciprocal link between multiculturalism and perspective-taking: How ideological and self-regulatory approaches to managing diversity reinforce each other. J Exp Soc Psychol. 2012;48(6):1394–8.
- 46. Lowery BS, Unzueta MM, Knowles ED, Goff PA. Concern for the in-group and opposition to affirmative action. J Pers Soc Psychol. 2006;90(6):961–74. pmid:16784345
- 47. Hasson Y, Tamir M, Brahms KS, Cohrs JC, Halperin E. Are liberals and conservatives equally motivated to feel empathy toward others? Pers Soc Psychol Bull. 2018;44(10):1449–59. pmid:29739293
- 48. Stevens SM, Jago CP, Jasko K, Heyman GD. Trustworthiness and ideological similarity (but not ideology) promote empathy. Pers Soc Psychol Bull. 2020. pmid:33284730
- 49. Marx Knoetze H. Romanticising the “Boer”: Narratives of White victimhood in South African popular culture. J Lit Stud. 2020;36(4):48–69.
- 50. Mathe L. Political discourses on race and social inequalities through social media and live parliamentary debates in South Africa: A content analysis. Communicare. 2020;39(2):1–24.
- 51. Cikara M, Van Bavel JJ. The neuroscience of intergroup relations: An integrative review. Perspect Psychol Sci. 2014;9(3):245–74. pmid:26173262
- 52. Craig MA, Richeson JA. More diverse yet less tolerant? How the increasingly diverse racial landscape affects white Americans’ racial attitudes. Pers Soc Psychol Bull. 2014;40(6):750–61. pmid:24625658
- 53. Horwitz FM, Jain H. An assessment of employment equity and Broad Based Black Economic Empowerment developments in South Africa. Equality, Diversity and Inclusion. 2011;30(4):297–317.
- 54.
Vilakazi T, Bosiu T. Black economic empowerment, barriers to entry, and economic transformation in South Africa. In: Andreoni A, Mondliwa PA, Roberts S, Tregenna F, editors. Structural transformation in South Africa: The challenges of inclusive industrial development in a middle-income country: Oxford University Press; 2021.
- 55. Nkomo S. Land redistribution: South Africans prioritize land taken in forced removals, support ’willing seller’ approach. Afrobarometer, 2018.
- 56. Herman P. If you see a beautiful piece of land, take it–Malema. News24. 2017.
- 57. Lowman S. BEE–current version of Apartheid. Robbing the ‘Rainbow Nation’. BizNews. 2016.
- 58.
News24. Whites have owned economy for too long—Ramaphosa. News24. 2016.
- 59. Zaki J. Empathy: A motivated account. Psychol Bull. 2014;140(6):1608–47. pmid:25347133
- 60. Devine PG, Plant EA, Amodio DM, Harmon-Jones E, Vance SL. The regulation of explicit and implicit race bias: the role of motivations to respond without prejudice. Journal of personality and social psychology. 2002;82(5):835. pmid:12003481
- 61. Richeson JA, Trawalter S. The threat of appearing prejudiced and race-based attentional biases. Psychol Sci. 2008;19(2):98–102. pmid:18271854
- 62. Fourie MM, Stein DJ, Solms M, Gobodo-Madikizela P, Decety J. Empathy and moral emotions in post-apartheid South Africa: An fMRI investigation. Soc Cogn Affect Neurosci. 2017;12(6):881–92. pmid:28338783
- 63. Meiring L, Subramoney S, Thomas KG, Decety J, Fourie MM. Empathy and helping: Effects of racial group membership and cognitive load. S Afr J Psychol. 2014;44(4):426–38.
- 64. Bonam CM, Nair Das V, Coleman BR, Salter P. Ignoring history, denying racism: Mounting evidence for the Marley hypothesis and epistemologies of ignorance. Soc Psychol Personal Sci. 2019;10(2):257–65.
- 65. Hughes JM, Bigler RS, Levy SR. Consequences of learning about historical racism among European American and African American children. Child Dev. 2007;78(6):1689–705. pmid:17988315
- 66. Durrheim K, Dixon J, Tredoux C, Eaton L, Quayle M, Clack B. Predicting support for racial transformation policies: Intergroup threat, racial prejudice, sense of group entitlement and strength of identification. Eur J Soc Psychol. 2011;41(1):23–41.
- 67. Fourie MM, Verwoerd WJ. Covid-19 as natural intervention: Guilt and perceived historical privilege contributes to structural reform under conditions of crisis. Affect Sci. 2021:1–12. pmid:34608456
- 68. Trawalter S, Hoffman KM, Waytz A. Racial bias in perceptions of others’ pain. PLoS ONE. 2012;7(11):e48546. pmid:23155390
- 69. Deska JC, Kunstman J, Lloyd EP, Almaraz SM, Bernstein MJ, Gonzales J, et al. Race-based biases in judgments of social pain. J Exp Soc Psychol. 2020;88:103964.
- 70. Morelli SA, Lieberman MD, Zaki J. The emerging study of positive empathy. Soc Personal Psychol Compass. 2015;9(2):57–68.
- 71. Andreychik MR, Migliaccio N. Empathizing with others’ pain versus empathizing with others’ joy: Examining the separability of positive and negative empathy and their relation to different types of social behaviors and social emotions. Basic Appl Soc Psych. 2015;37(5):274–91.
- 72. Stellar JE, Anderson CL, Gatchpazian A. Profiles in empathy: Different empathic responses to emotional and physical suffering. J Exp Psychol. 2020;149(7):1398. pmid:31804122
- 73. Nordgren LF, Banas K, MacDonald G. Empathy gaps for social pain: Why people underestimate the pain of social suffering. J Pers Soc Psychol. 2011;100(1):120. pmid:21219077
- 74. Cornell J, Kessi S. Black students’ experiences of transformation at a previously “white only” South African university: a photovoice study. Ethn Racial Stud. 2017;40(11):1882–99.
- 75. Walker M. Race is nowhere and race is everywhere: Narratives from black and white South African university students in post‐apartheid South Africa. Br J Sociol Educ. 2005;26(1):41–54.
- 76. Walker M. Rainbow nation or new racism? Theorizing race and identity formation in South African higher education. Race Ethnicity and Education. 2005;8(2):129–46.
- 77. Gillespie A, Howarth CS, Cornish F. Four problems for researchers using social categories. Culture and Psychology. 2012;18(3):391–402.
- 78. Msimang P. Lessons in our faults: Fault lines on race and research ethics. S Afr J Sci. 2020;116(9–10):1–3.
- 79. Jansen J, Walters C. Fault lines: A primer on race, science and society. Stellenbosch: Sun Media; 2020.
- 80. Dolby N. White fright: The politics of White youth identity in South Africa. Br J Sociol Educ. 2001;22(1):5–17.
- 81.
Batson CD, Lishner DA, Stocks EL. The empathy-altruism hypothesis. In: Schroeder DA, Graziano WG, editors. The Oxford Handbook of Prosocial Behavior. New York, NY: Oxford University Press; 2015. p. 259–81.
- 82. Faul F, Erdfelder E, Lang A-G, Buchner A. G* Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39(2):175–91. pmid:17695343
- 83. Plant EA, Devine PG. Internal and external motivation to respond without prejudice. Journal of personality and social psychology. 1998;75(3):811.
- 84.
Psychology Software Tools Inc. E-Prime 3.0. Pittsburgh, PA: Psychology Software Tools, Inc.; 2018.
- 85.
Vittinghoff E, Glidden DV, Shiboski SC, McCulloch CE. Regression methods in Biostatistics: Linear, logistic, survival, and repeated measures models. 2nd ed. New York, NY: Springer; 2012.
- 86.
Quinn GP, Keough MJ. Experimental design and data analysis for biologists. New York, NY: Cambridge University Press; 2002.
- 87. Sheng F, Liu Q, Li H, Fang F, Han S. Task modulations of racial bias in neural responses to others’ suffering. NeuroImage. 2014;88:263–70. pmid:24135167
- 88. Kurzban R, Leary MR. Evolutionary origins of stigmatization: The functions of social exclusion. Psychol Bull. 2001;127(2):187–208. pmid:11316010
- 89. Voorspoels W, Bartlema A, Vanpaemel W. Can race really be erased? A pre-registered replication study. Front Psychol. 2014;5(1035). pmid:25278922
- 90. Chiao JY, Mathur VA. Intergroup empathy: How does race affect empathic neural responses? Curr Biol. 2010;20(11):R478–R80. pmid:20541493
- 91.
Tajfel H, Turner JC. An integrative theory of intergroup conflict. In: Austin WG, Worchel S, editors. Organizational Identity: A Reader. Monterey, CA: Brooks-Cole; 1979.
- 92. Everett JA, Faber NS, Crockett M. Preferences and beliefs in ingroup favoritism. Front Behav Neurosci. 2015;9:1–15. pmid:25762906
- 93. Yang X, Yang F, Guo C, Dunham Y. Which group matters more: The relative strength of minimal vs. gender and race group memberships in children’s intergroup thinking. Acta Psychologica. 2022;229:103685. pmid:35870236
- 94. Eckel CC, Wilson RK, Youn S. In-group favoritism in natural and minimal groups. Economics Letters. 2022;219:110794.
- 95. Verwey C, Quayle M. Whiteness, racism, and Afrikaner identity in post-apartheid South Africa. Afr Aff. 2012;111(445):551–75.
- 96. Seekings J. The continuing salience of race: Discrimination and diversity in South Africa. J Contemp Afr Stud. 2008;26(1):1–25.
- 97. Hornsey MJ, Spears R, Cremers I, Hogg MA. Relations between high and low power groups: The importance of legitimacy. Pers Soc Psychol Bull. 2003;29(2):216–27. pmid:15272949
- 98.
Makinana A. Land reform: Zuma moves for expropriation with no compensation. City Press. 2017.
- 99. Gladwin-Wood C, Combrinck T. Expropriation without compensation? Times Live. 2017.
- 100. Blumer H. Race prejudice as a sense of group position. Pacific Sociological Review. 1958;1(1):3–7.
- 101. Bobo LD. Prejudice as group position: Microfoundations of a sociological approach to racism and race relations. Journal of Social Issues. 1999;55(3):445–72.
- 102.
OECD. Education at a glance 2019: OECD indicators 2019.
- 103. Dixon J, Elcheroth G, Kerr P, Drury J, Al Bzour M, Subašić E, et al. It’s not just “us” versus “them”: Moving beyond binary perspectives on intergroup processes. Eur Rev Soc Psychol. 2020;31(1):40–75.
- 104. Falk CF, Heine SJ, Takemura K. Cultural variation in the minimal group effect. J Cross Cult Psychol. 2014;45(2):265–81.
- 105. Cikara M, Botvinick MM, Fiske ST. Us versus them: Social identity shapes neural responses to intergroup competition and harm. Psychol Sci. 2011;22(3):306–13. pmid:21270447
- 106. Richins MT, Barreto M, Karl A, Lawrence N. Empathic responses are reduced to competitive but not non-competitive outgroups. Soc Neurosci. 2019;14(3):345–58. pmid:29633906
- 107. Erasmus Z. Apartheid race categories: Daring to question their continued use. Transformation: Critical Perspectives on Southern Africa. 2012;79(1):1–11.
- 108. Roberts SO, Bareket-Shavit C, Dollins FA, Goldie PD, Mortenson E. Racial inequality in psychological research: Trends of the past and recommendations for the future. Perspect Psychol Sci. 2020;15(6):1295–309. pmid:32578504