Violent Assault on a Chinese Man: COVID-19 Psychosocial Resource Loss Diminishes Right Wing Authoritarianism Variability in Societal Reactions.

We examined whether prejudice-related personality characteristics (i.e., right wing authoritarianism [RWA]) and COVID-19-driven psychological resource loss might predict perpetrator-directed punitive responding (i.e., support for criminal charges) to a COVID-19-related attack on a Chinese victim by a White male. Across two studies, participants completed an RWA measure and reported the extent they had experienced COVID-19-related psychosocial resource loss. They then read a passage describing the COVID-19-related physical assault. For both studies, at low resource loss levels, low RWA participants reported greater punitive responding toward the perpetrator than high RWA participants. This RWA-punitive responding association was mediated by greater victim-directed suffering sensitivity (i.e., empathy) for Study 1 and greater anti-perpetrator reactions (i.e., hate crime perceptions) for Study 2. The RWA association with the relevant outcome variables (i.e., suffering sensitivity, anti-perpetrator bias, and punitive responding) was eliminated at high psychological resource loss levels. Specifically, low and high RWA participants reported similar reactions. While previous research has demonstrated that high RWA individuals tend to report greater outgroup-directed prejudicial responses due to COVID-19-driven perceptions of threat, our findings demonstrate that the COVID-19 pandemic can also elicit feelings of resource loss that can diminish the egalitarian reactions typically reported by low RWA individuals. In sum, we demonstrate that experiencing difficult life circumstances such as COVID-19 psychosocial resource loss can diminish supportive reactions toward victimized minority group members even among low RWA participants who are typically expected to be more sensitive to the struggles of those who are disadvantaged.

Introduction

Given the reports of the Chinese origin of the COVID-19 pandemic, it is not surprising that there have been extreme negative reactions toward Asians across the world. For example, complaints of anti-Asian discrimination filed with the New York City Commission on Human Rights increased by 92% from March to May of 2020 relative to the same period in 2019 (Han, 2020). Most relevant to the current examination, there has been a greater incidence of violent assaults of Asians (Gover et al., 2020; Tessler et al., 2020). However, there has been minimal empirical examination of factors that might influence societal reactions to such violence. This lacuna in the literature is striking when one considers the extensive historical (Greene et al., 2001; Hovland & Sears, 1940) and contemporary (Johnson et al., 2020; Johnson & Lecci, 2020) concerns regarding unwarranted physical violence against innocent minority group members. Thus, across two studies, we were specifically interested in the extent to which right wing authoritarianism (RWA) and COVID-19–related psychosocial resource loss (i.e., loss of hope, optimism, and motivation) might influence our participants’ punitive reactions (i.e., support for criminal charges) toward the perpetrator of a COVID-19–related assault of a Chinese victim. We also examined the mediating role of sensitivity to the victim suffering (i.e., greater empathy) in Study 1 and anti-perpetrator reactions (i.e., greater hate crime perceptions) in Study 2.

Right Wing Authoritarianism and Prejudice: New Empirical Directions

RWA is commonly explained as “authoritarian submission, authoritarian aggression, and conventionalism” (Jost et al., 2003; Saunders & Ngo, 2017; Son Hing et al., 2007). There is a broad body of evidence demonstrating that RWA is strongly related to outgroup-directed prejudice and prejudice-related attitudes and beliefs (Duckitt & Sibley, 2007; Hodson & Costello, 2007). We sought to extend the RWA–prejudice association literature by examining the relationship between RWA and punitive reactions toward the perpetrators of minority-directed violence. An assessment of this issue is relevant because the public’s reactions or lack thereof may reinforce or reduce the occurrence of such violent acts. Specifically, perpetrators of hate crimes might interpret any “societal leniency” toward these insidious and injurious criminal actions as an “endorsement” of their hateful cause (Monroe, 2008; Staub, 2002). Moreover, public perceptions and reactions are important in the context of legal prosecution, as members of the public can serve on juries that make important decisions, including determining whether to indict (formally charge) the perpetrator and/or financially compensate the victim.

Previous research has also shown that link between RWA and prejudice tends to be “amplified” when individuals are under various forms of threat. For example, Cohrs and Ashbrock (2009) found that the effects of RWA on prejudice were significantly elevated when the outgroup was portrayed as a threat to the participants’ way of life. Similarly, Lindén et al. (2016) showed that the link between RWA and support for torturing outgroups was elevated when the threat of terrorism was made salient. Finally, and most relevant to the present examination, Hartman et al. (2021) demonstrated that the association between RWA and anti-immigration sentiments was increased when individuals felt the threat posed by the COVID-19 pandemic. We extend the extant research on the moderation of the RWA–prejudice association in two important ways. First, unlike perceived threat which amplifies the RWA–prejudice association, we believe that debilitating feelings of COVID-19–related resource loss will diminish the RWA and prejudice association. Second, we provide a moderated-mediational analysis to identify factors that might play a mediating role at low levels of resource loss.

Conservation of Resources Theory

The Conservation of Resource (COR theory, Hobfoll, 1989; Hobfoll et al., 2006) provides theoretical leverage to our fundamental proposition that emotional duress will diminish the RWA–prejudice association. The basic tenet of the theory is that individuals strive to obtain, retain, and protect that which they value. One central category of valued “resources” involves psychosocial elements (e.g., hope and optimism). More broadly, a major component of the model is the idea that the loss of resources can have a significant deleterious impact on one’s everyday life. Consistent with this perspective, resource loss has been a strong predictor of serious mental health outcomes such as post-traumatic stress disorder (Sattler et al., 2018; Schumm et al., 2004), depression (Freedy et al., 1994), and other forms of stress in a variety of settings (Hobfoll et al., 2006; Sattler et al., 2014). Notably, the COVID-19 pandemic has facilitated an array of “debilitating” mental health outcomes linked to feelings of loss in populations across the world (Brooks et al., 2020; Griffiths & Mamun, 2020; Orrù et al., 2021; Williams et al., 2020; Zhai & Du, 2020). Thus, our examination of the deleterious “downstream” consequences of such feelings of resource loss for intergroup processes seems critically important.

In the present research, we aim to show that psychosocial resource loss will likely impact low RWA participants because they tend to report greater racially sensitive and justice driven reactions (Duckitt & Sibley, 2007; Hodson & Costello, 2007). More specifically, there is strong empirical evidence suggesting that such egalitarian reactions require both motivation (Major et al., 2013; Monteith, 1993; Monteith et al., 2002) and cognitive resources (Spears & Haslam, 1997; Von Knippenberg et al., 1999). We believe that COVID-19–driven psychosocial resource loss should deplete both the motivational and cognitive resources necessary for low RWA participants to maintain high levels of racial sensitivity and justice driven ideals.

We propose that, when our participants have experienced low levels of COVID-19–related psychosocial resource loss (e.g., loss of hope, optimism, motivation, etc.), RWA should have a negative association with (a) favorable victim-directed responses and (b) positive perpetrator-directed responses. However, when our participants have experienced high resource loss, the link between RWA and subsequent reactions should be reduced to non-significance. We tested these predictions across two studies.

Study 1: Victim-Directed Mediators

Participants completed an RWA measure and reported the extent to which they had experienced COVID-19–related psychosocial resource loss (e.g., loss of hope, optimism, and motivation). They then read a passage describing a COVID-19–related physical assault by a White male on a Chinese male in the United States. After reading the passage, participants completed measures that assessed their (a) victim-directed empathic responding and (b) perpetrator-directed punitive responding (i.e., the extent to which they supported criminally charging the perpetrator).

We chose victim-directed empathy as a primary mediator for three reasons. First, empathy has been shown to be a critical motivator for prosocial reaction toward those who are suffering (Batson, 2016). Second, empathy for the victim has been linked to greater punitive responding toward their harm-doer (Deitz et al., 1982) and sympathy for their families (Paternoster & Deise, 2011). Third, diminished empathy has been shown to mediate the association between RWA and prejudice (Nicol & Rounding, 2013).

Our specific hypotheses are given below:

At low resource loss levels, we posit a negative association between RWA and victim-directed empathy (H1) and perpetrator-directed punitive responding (H2). However, at high resource loss levels, the association between RWA and our outcome measures should be reduced to non-significance. We further posit that moderated-mediational analysis would reveal that at low (but not high) resource loss levels, victim-directed empathy would mediate the association between RWA and perpetrator-directed punitive responding (H3).

Method

Participants

The participants were 143 (89 male and 54 female) who were recruited using MTurk (2022) in the United States. The sample included White (N = 106), Black (N = 28), and Hispanic (N = 9) individuals (age: M = 37.68, SD = 11.17, range: 23–70). Their highest level of education completed was secondary school (N = 12), some college (N = 28), completed college (N = 79), or completed postgraduate degree (N = 24). The nature of the sample in both studies demonstrates a certain level of diversity among the participants. This diversity increases the likelihood that we obtain the perspectives of different cultural groups. To ensure the quality of the data, participants had to report an MTurk reputation score of 0.90 or greater. MTurk respondents tend to be more representative of the US population relative to in-person convenience samples but less representative relative to national probability samples (Berinsky et al., 2012). In addition, the requirement that all participants have a high reputation score is typically sufficient to ensure a high-quality participant pool (e.g., Chandler & Shapiro, 2016). The research was approved by the Institutional Review Board of the host institution.

Because previous studies have not tested the specific interaction between RWA and resource loss, we were unable to conduct a priori power analysis to estimate the needed sample size. Our sample size was based on the following factors: (a) it is similar to sample sizes utilized in past research on reactions to minority-directed physical violence (Cohrs & Asbrock, 2009; Dru, 2007), and (b) it was within our budget. We used G*Power to conduct a post hoc power analysis. The results indicated that given the effect sizes of the interaction effect found in the current study, the achieved statistical power ranged from 0.77 to 0.90.

Procedure

During the early phases of the COVID-19 pandemic (July, 2020), participants were recruited for a study titled “Decision Making” and received financial compensation (US $4) for their participation (see Supplemental Appendix). As a portion of the “background information,” they completed the Resource Loss Scale and the RWA Short Scale. Participants were then informed through the instructions that the study focused on how people make decisions about various life situations. Next, they read a “newspaper story” that described a situation involving a White male who attacked a Chinese passenger on a train. The ethnicity of both parties and a picture of them were presented. The story stated that the White male (the assailant) approached the Chinese passenger and shouted loudly to him, “Why are you wearing the mask—you must be carrying the corona virus! I saw you sneeze earlier so you shouldn’t be out!” The victim calmly replied that he has allergies and that is why he sneezed. The assailant persisted in making belittling remarks and then kicked the victim in the chest three times, which resulted in the victim falling on the floor. He then stood directly over the victim and said, “You had better get off at the next stop because you are putting us all in danger! I do not want to get sick!”

After reading the passage, participants completed several measures including victim-directed empathic responding and perpetrator-directed punitive responding.

Measures

Psychosocial resource loss was assessed by having participants complete items from the Resource Loss Scale (Freedy et al., 1994) which showed strong internal consistency (α = .95), modified by the current authors for COVID-19 (Sattler et al., 2014). Using a 5-point scale (1 = no loss to 5 = great loss), participants reported the amount of COVID-19 pandemic-induced loss that they felt in areas such as hope, sense of optimism, and motivation to get things done. Participants then completed the RWA Short Scale (Manganelli Rattazzi et al., 2007; α = .92) using a 5-point scale (1 = strongly disagree to 5 = strongly agree). Low resource loss participants (1 SD below the mean) reported a resource loss mean score of 1.58, moderate loss participants (at the mean) reported a resource loss mean score of 2.60, and high resource loss participants (1 SD above the mean) reported a resource loss mean score of 3.63. RWA mean scores were 2.25 (low RWA), 3.12 (moderate RWA), and 3.99 (high RWA).

After reading the passage, participants completed several measures using a 5-point scale (1 = not at all to 5 = very much) to indicate their responses. Victim-directed empathy, which is an other-oriented emotional response reflecting the perceived welfare of another (see Batson, 1991; Batson et al., 1995), was assessed by averaging (α = .85) the extent to which participants experienced five emotions (compassion, moved, soft-hearted, warmth, and sympathy) for the victim. Perpetrator-directed punitive responding was measured by averaging the items that assessed (a) the extent to which the participants felt that the assailant should be arrested; and (b) the extent to which the participants felt that the assailant should be criminally charged (i.e., indicted) for his actions if they were on a grand jury (α = .74).

Statistical Analysis Plan

Analysis of the hypothesized interaction

Because RWA and resource loss are continuous variables, we used PROCESS Model 1 (Hayes, 2013) to test the significance of the interaction between RWA and resource loss on our relevant outcome measures (i.e., victim-directed empathy and perpetrator-directed punitive responding). PROCESS is a SPSS macro that executes path analysis-based moderation and mediation analysis using bootstrapping measures with 10,000 samples.

Probing the interactions

PROCESS Macro Model 1 provides assessments of the effects of the independent variable (RWA) on our relevant outcome measures at low (1 SD below the mean), moderate (at the mean), and high (1 SD above the means) levels of the continuous moderator (resource loss).

Moderated mediation

To test our expectation that victim empathy would mediate the RWA–punitive responding association at low (but not high) resource loss levels, we utilized the PROCESS Macro Model 7 (Hayes, 2013).

Results

Victim-Directed Empathic Responding

RWA was negatively related to empathy, B = −0.18, SE = 0.08, β = −.19, p = .021. Psychosocial resource loss was not related to empathy, B = −0.06, SE = 0.06, β = −.08, p = .398.

In support of H1 (see Figure 1), the RWA × resource loss interaction for empathic responding reached significance,  = .032, F(1, 139) = 4.90, p = .028, B = 0.17, 95% CI [0.02, 0.32]. A probe of the interaction provides further confirmation to our expectations. At 1 SD below the mean (B = −0.35, SE = 0.11, t = −3.12, p = .002), and at the mean of resource loss (B = −0.18, SE = 0.08, t = −2.13, p = .034), lower RWA was associated with greater empathic responding. However, at 1 SD above the mean, there was no association between RWA and empathic responding (B = −0.001, SE = 0.12, t = −0.012, p = .990).

Figure 1.

Victim-directed empathy as a function of RWA and psychosocial resource loss.

Note. Greater values indicate greater RWA and empathy. RWA = right wing authoritarianism.

Perpetrator-Directed Punitive Responding

RWA was negatively related to punitive responding, B = −0.35, SE = 0.10, β = −.29, p < .001. Psychosocial resource loss was also negatively related to punitive reactions, B = −0.19, SE = 0.09, β = −.19, p = .027.

In support of H2 (see Figure 2), the RWA × resource loss interaction for punitive responding reached significance,  = .074, F(1, 139) = 12.30, p < .001, B = 0.35, 95% CI [0.15, 0.55]. Consistent with our expectations, at 1 SD below the mean (B = −0.66, SE = 0.14, t = −4.49, p < .001), and at the mean of resource loss (B = −0.29, SE = 0.10, t = −2.75, p = .006), lower RWA was associated with greater punitive responding. However, at 1 SD above the mean, there was no association between RWA and punitive responding, B = 0.06, SE = 0.15, t = 0.42, p = .669.

Figure 2.

Perpetrator-directed punitive responding as a function of RWA and psychosocial resource loss.

Note. Greater values indicate greater RWA and punitive responding. RWA = right wing authoritarianism.

Moderated-Mediational Effects

In support of H3, a moderated-mediation effect test demonstrated that empathic responding mediated the association between RWA and perpetrator-directed punitive responding at low resource loss levels, B = −0.17, BootSE = 0.09, 95% CI [−0.3430, −0.0337], but not at high resource loss levels, B = 0.006, BootSE = 0.07, 95% CI [−0.1021, 0.1675]. The index of moderated mediation was also significant, B = 0.08, BootSE = 0.05, 95% CI [0.0023, 0.2078].

Brief Discussion

The results of Study 1 supported our major expectations. There was an interaction effect between RWA and psychosocial resource loss such that at low levels of resource loss, there was greater victim-directed empathy. A moderated-mediational analysis revealed that at low (but not high) resource loss levels, victim-directed empathy mediated the association between RWA and perpetrator-directed punitive responding.

Study 2: Perpetrator-Directed Mediators

The central purposes of Study 2 were to replicate and extend the findings of Study 1 by assessing the mediating role of anti-perpetrator reactions in the RWA and punitive responding association. Our anti-perpetrator reaction measure was hate crime perceptions (i.e., the perspective that the perpetrator’s violent attacks were not driven by racial prejudice) toward the perpetrator. Importantly, Leander et al. (2020) contend that one form of “sympathy” for perpetrators who commit violent acts against outgroup members is the hesitancy in believing that such actions are hate crimes. In support of their assertions, they found that feelings of disempowerment among majority group members predicted less certainty that perpetrators of well-known attacks on minority group members were driven by racial prejudice. Thus, for the current study, increased hate crime perceptions should reflect greater negative feelings toward the perpetrator.

To test the mediating roles of the hate crime perceptions, participants completed the same RWA measure and COVID-19–related psychosocial resource loss measures from Study 1 and the same violent assault passage concerning the Chinese victim. After reading the passage, participants completed measures that assessed (a) their hate crime perceptions (i.e., perpetrator was driven by prejudice) and (b) perpetrator-directed punitive responding (i.e., the extent to which they supported criminally charging the perpetrator).

Our specific hypotheses are given below:

At low resource loss levels, we posit a negative association between RWA and hate crime perceptions (H4) and perpetrator-directed punitive responding (H5). However, at high resource loss levels, we posit that the association between RWA and our outcome measures should be reduced to non-significance. We further posit that moderated-mediational analysis would reveal that at low (but not high) resource loss levels, hate crime perceptions would mediate the association between RWA and perpetrator-directed punitive responding (H6).

The participants were 250 (152 male, 96 female, and 2 non-binary) people who were recruited using MTurk in the United States. The sample included White (N = 191), Black (N = 30), Hispanic (N = 27), and other (N = 2) individuals (age: M = 36.9, SD = 10.23, range: 19–65). Their highest level of education completed was some secondary school (N = 2), secondary school (N = 21), some college (N = 49), completed college (N = 136), or completed postgraduate degree (N = 42). To ensure the quality of the data, participants had to report an MTurk reputation score of 0.90 or greater (Chandler & Shapiro, 2016). For this power sensitivity analysis, we used the lm.beta package (Behrendt, 2022) in R Core Team (2022). Setting alpha at .05, we simulated 10,000 times a model wherein RWA interacts with resource loss (with punitive responding as the dependent variable). We used the Study 1 parameter estimates as coefficients for the main effects. The minimum detectable effect size for the interaction effect of interest was the one that yielded 80% power (i.e., it yielded a significant p-value 8,000 out of 10,000 times). The research was approved by the Institutional Review Board of the host institution.

The procedure was identical to the Study 1 procedure. After reading the passage, participants completed several measures including hate crime perceptions and perpetrator-directed punitive responding.

Participants first completed the Resource Loss Scale (α = .96) and the RWA Short Scale (Manganelli Rattazzi et al., 2007; α = .92) using a 5-point scale (1 = strongly disagree to 5 = strongly agree). Resource loss mean scores were 1.26 (low resource loss), 2.26 (moderate resource loss), and 3.73 (high resource loss). RWA mean scores were 1.66 (low RWA), 2.93 (moderate RWA), and 3.80 (high RWA).

After reading the passage, participants completed several measures using a 5-point scale (1 = not at all to 5 = very much) to indicate their responses. Since attacks on minority group members are defined as “hate crimes” if they manifest evidence of prejudice toward the minority group that was the target (Berk, 1990; Gover et al., 2020; Morgan et al., 2011), hate crime perceptions were measured by averaging (α = .70) participant certainty that: (a) the attack on the victim was due to “group-based” discrimination and (b) the perpetrator held group-based prejudices toward the victim. Perpetrator-directed punitive responding was measured by averaging the same two items (α = .61) as in Study 1. We utilized the same statistical analysis plan as Study 1.

Hate Crime perceptions

RWA (B = −0.32, SE = 0.06, β = −.33, p < .001) and resource loss (B = −0.21, SE = 0.05, β = −.26, p < .001) were negatively related to hate crime perceptions.

In support of H4 (see Figure 3), the RWA × resource loss interaction for hate crime perceptions reached significance,  = .019, F(1, 228) = 5.08, p = .025, B = 0.14, 95% CI [0.02, 0.26]. A probe of the interaction provides further confirmation to our expectations. At 1 SD below the mean (B = −0.37, SE = 0.09, t = −4.21, p < .001), and at the mean of resource loss (B = −0.22, SE = 0.07, t = −3.06, p = .002), lower RWA was associated with greater hate crime perceptions. However, at 1 SD above the mean, there was no association between RWA and hate crime perceptions, B = −0.07, SE = 0.11, t = −0.69, p = .490.

Figure 3.

Hate crime perceptions as a function of RWA and psychosocial resource loss.

Note. Greater values indicate greater RWA and hate crime perceptions. RWA = right wing authoritarianism.

RWA (B = −0.27, SE = 0.05, β = −.28, p < .001) and resource loss (B = −0.22, SE = 0.05, β = −.27, p < .001) were negatively related to punitive reactions.

In support of H5 (see Figure 4), the RWA × resource loss interaction for perpetrator-directed punitive responding reached significance,  = .06, F(1, 228) = 16.64, p < .001, B = 0.25, 95% CI [0.13, 0.37]. At 1 SD below the mean (B = −0.42, SE = 0.09, t = −4.93, p < .001), at the mean of resource loss (B = −0.16, SE = 0.07, t = −2.27, p = .023), lower RWA was associated with greater punitive responding. However, at SD above the mean, B = 0.10, SE = 0.11, t = 0.96, p = .335, there was no association between RWA and punitive responding.

Figure 4.

Perpetrator-directed punitive responding as a function of RWA and psychosocial resource loss.

Note. Greater values indicate greater RWA and punitive responding. RWA = right wing authoritarianism.

In support of H6, hate crime perceptions mediated the association between RWA and perpetrator-directed punitive responding at low resource loss levels, B = −0.13, BootSE = 0.04, 95% CI [−0.2306, −0.0625], but not at high resource loss levels, B = −0.03, BootSE = 0.04, 95% CI [−0.1006, 0.0675]. The index of moderated mediation was also significant, B = 0.05, BootSE = 0.03, 95% CI [0.0045, 0.1128].

General Discussion

In one of the few direct assessments of factors that influence societal reactions to COVID-19–related anti-Asian violence, the findings of the current study demonstrated that contextual factors (i.e., COVID-19 psychological resource loss) and prejudice-related personality factors (i.e., RWA) independently and jointly predicted perpetrator-directed punitive reactions in an instance of COVID-19 assault of a Chinese man. Higher levels of both resource loss and RWA predicted less support for punitive actions against the perpetrator. More critically, when they were experiencing minimal resource loss, participants scoring low on RWA were more likely to report greater victim-directed empathy (Study 1) and greater anti-perpetrator negative reactions (Study 2) relative to high RWA participants. This greater sensitivity to victim suffering and greater anti-perpetrator responding among low RWA participants, in turn, predicted stronger support for perpetrator-directed punitive responding (i.e., criminal indictment). However, as psychological resource loss increased, the differential victim-directed and perpetrator-directed reactions as a function of RWA were significantly diminished and even reduced to non-significance. Under high levels of resource loss, all participants, irrespective of their RWA level, cared less about the suffering of the victim.

COVID-19–Driven Resource Loss

At the onset of the pandemic, a team of leading mental health scholars and practitioners warned that the “psychological burden of the COVID-19 pandemic” would likely be associated with widespread catastrophic mental health outcomes (Campion et al., 2020). One crucial issue involves the potential deleterious downstream consequences of such disease-driven emotional duress. The limited direct research in this area has demonstrated that COVID-19 psychological burden predicted greater anti-systemic attitudes (dissatisfaction with the fundamental social and political order) and political violence (Bartusevičius et al., 2021). To extend the research in this area, the present study focused on the reactions to a completely innocent Chinese victim who suffered a barrage of disparaging remarks, endured three kicks to the chest, was knocked to the floor, and faced further threats while on the floor. Despite the brutality of this unjustified attack, COVID-19 psychological resource loss was a direct predictor of an array of reactions that reflected reductions in the recognition of the victim’s civil and human rights. In sum, COVID-19 psychological resource loss was linked to diminished hate crime perceptions and more lenient reactions toward the violent perpetrator. Notably, it is likely that such sympathetic and lenient reactions toward the perpetrator can serve to create a “climate of social tolerance” toward mistreatment of the disadvantaged (Gracia & Herrero, 2007).

RWA and Outgroup-Directed Prejudice

The physical assault incident that we presented to our participants reflects the high incidence of dominant group members inflicting pain and suffering on less advantaged members of society (Monroe, 2008; Staub, 2002). Across two studies, we showed that RWA would relate to greater supportive reactions toward the perpetrators of such minority-directed violence. Perpetrators of hate crimes might interpret these types of favorable reactions as an indication of support for their hateful causes (Monroe, 2008; Staub, 2002). This issue is very important because, according to the Federal Bureau of Investigations annual Hate Crime Statistics Report (Bates, 2020), hate crimes in America have recently been at their highest rates since the US government began tracking the numbers in the early 1990s. It would be interesting for future research to assess whether the RWA link to greater perpetrator violence would generalize to other forms of minority-directed violence. For example, recent analyses of police shooting statistics indicate that Black unarmed males are seven times more likely than White males to be killed by police gunfire (Somashekhar et al., 2015). More broadly, there is recent evidence that Black individuals are at the greatest risk of being an unarmed victim of a police shooting (Ross, 2015). An assessment of whether RWA would predict greater supportive reactions toward the police officer perpetrator of such minority-directed violence would provide an interesting extension to the extant RWA–prejudice research.

Moreover, the bulk of research on moderators of the RWA–prejudice association has tended to focus on factors that might elevate prejudicial reactions among high RWA participants (like threat, e.g., see Cohrs & Asbrock, 2009; Lindén et al., 2016). The current findings suggest that greater attention should be given to factors that might diminish the more egalitarian reactions among low RWA individuals. Moreover, the limited research on the RWA–prejudice association in the context of the COVID-19 pandemic has focused on the negative intergroup outcomes from feelings of threat. However, our findings demonstrate that deleterious intergroup consequences of COVID-19–driven feelings of loss should also be considered. Specifically, resource loss was shown to “stifle” the oft-cited goodwill of those who typically report minority-directed favorable reactions (i.e., low RWA individuals). These results are consistent with a number of theoretical perspectives that suggest racially sensitive reactions among majority group members require both motivation (Major et al., 2013; Monteith, 1993; Monteith et al., 2002) and cognitive resources (Spears & Haslam, 1997; Von Knippenberg et al., 1999). It seems that psychosocial resource loss may deplete both these resources necessary to maintain high levels of racial sensitivity among low RWA participants. Notably, there have been recent concerns about the modern-day tendency for majority group members to “appear sympathetic” (i.e., “virtue signaling”; Zaki & Cikara, 2020) to the struggles of minority group members. Our findings suggest that this tendency may be “fragile” and contingent on favorable life conditions. Future research that provides a direct examination of the issue certainly seems warranted.

Limitations and Conclusion

There are a number of limitations with the current research that must be acknowledged. For example, survey and vignette responses are necessarily more limited when researchers examine actual real-world events and their consequences. Thus, it is possible that current results may not generalize to video depictions or other more realistic simulations. In addition, demand effects can play a significant role in social psychological experiments, especially those involving racially charged issues. However, the fact that we reported consistent two-way interactions across both studies diminished the concern about demand effects, as the participants would have to know to adjust their reactions as a function of RWA and resource loss.

We note the critical importance of all members of a diverse society to be able to live, thrive, and not be penalized, in any way, for their ethnic membership. There is clear evidence that the COVID-19 pandemic has elevated the risks of violent discrimination among Asians around the world. The present findings demonstrated that challenging life experiences (i.e., psychosocial resource loss) linked to this pandemic can diminish the sensitivity to the suffering of those victimized individuals even among those who are typically expected to be more racially sensitive and supportive of cultural diversity (i.e., low right wing authoritarians). It is hoped that greater awareness of these insidious pandemic-driven intergroup consequences will lead to critical intervention measures and policies.

Click here for additional data file.

Supplemental material, sj-doc-1-jiv-10.1177_08862605221123301 for Violent Assault on a Chinese Man: COVID-19 Psychosocial Resource Loss Diminishes Right Wing Authoritarianism Variability in Societal Reactions by James Johnson, David Sattler, Alain Van Hiel, Kim Dierckx, Shanhong Luo and Loris Vezzali in Journal of Interpersonal Violence