Abstract
Empathy is critical for social functioning, but it often wanes when it is needed most. Resulting empathic failures precipitate and worsen social conflict. Accordingly, conflict-reduction interventions prioritize developing empathy in order to achieve harmony. Recent research has indicated that such interventions can benefit from a more nuanced understanding of empathy. First, empathy is a multidimensional construct, including understanding, sharing, and feeling concern for others’ emotions. The expression of these empathic processes is further influenced by psychological factors that “tune” people toward or away from empathy. Interventions must therefore diagnose the specific nature and precursors of empathic failures and tailor interventions appropriately. Second, empathy alone may be insufficient to produce prosocial behavior, especially when parties differ in status or power. In these cases, interventions should promote equitable goals and norms in addition to empathy. By understanding its component processes and boundary conditions, practitioners can work to promote empathy in maximally effective ways.
Empathy—people’s sharing and understanding of each other’s emotions—bolsters relationships, improves individuals’ well-being, and promotes prosocial behavior. Despite its benefits, empathy is neither a universal nor an automatic response. For instance, encounters across racial, political, cultural, and even laboratory-created social boundaries diminish behavioral, neural, and physiological expressions of empathy for the other (Cikara, Bruneau, & Saxe, 2011). These empathic failures represent cases in which an individual could conceivably feel empathy but does not because of salient social and psychological factors.
Empathic failures predict discrimination, neglect, and overt aggression (Cikara, 2015). Their effect is perhaps most pervasive—and most costly—in the context of active intergroup conflicts, such as in our increasingly polarized political system (Prior, 2013) or longstanding ethnic and religious clashes (Cikara et al., 2011). Empathic failures also arise in more quotidian settings—for instance, when people fail to understand the emotions of other-race social targets (Adams et al., 2010), when adolescents bully each other (Gini, Albiero, Benelli, & Altoè, 2007), and when doctors underestimate their patients’ suffering (Haque & Waytz, 2012).
One common assumption is that exercises designed to increase empathy should eliminate empathic failures and their consequences. Armed with this philosophy (and following classic theoretical precursors; cf. Allport, 1954), a slew of programs employ empathy-building techniques, such as perspective taking, in attempts to improve interpersonal and intergroup relations, for example, in schools and medical practices (Riess, Kelley, Bailey, Dunn, & Phillips, 2012; Şahin, 2012; Todd & Galinsky, 2014). In many cases, these strategies do foster empathy and improve interpersonal and intergroup relationships. For instance, instructing people to imagine a stigmatized target’s feelings can increase empathic concern and reduce antipathy toward that target and even toward other members of the target’s group (e.g., Batson et al., 1997). However, mounting evidence from across the psychological sciences indicates that fostering empathy among parties in active conflict requires a more nuanced approach than is employed by most interventions.
Here, we highlight emerging insights about the nature of empathy and the implications of these insights for practitioners. Empathy is a multidimensional construct comprising several cognitive and affective processes. Furthermore, people’s emotions and beliefs prior to entering empathy-inducing situations can predispose them to experience or avoid empathy within those situations. Therefore, interventions must diagnose the nature of particular empathic failures and their precursors and treat idiosyncratic features of each failure accordingly. Finally, empathy alone does not always produce positive social outcomes, particularly when parties in conflict differ in their power or status. Interventions in such cases should expand their focus on cultivating empathy and positive regard to include an emphasis on building conditions that recognize asymmetries between parties and foster equitable norms and behavior.
Diagnosis
Although it is a useful overarching term, “empathy” refers to a suite of related but distinct phenomena. These include mentalizing, or inferring others’ mental states; experience sharing, or vicariously taking on others’ internal states; and compassion, or feeling concern for others’ well-being (Zaki & Ochsner, 2012). These component processes of empathy diverge along multiple dimensions, including subjective experience, behavior, and associated neural architecture (Davis, 1994; Singer & Klimecki, 2014; Zaki & Ochsner, in press). The dissociability of these processes suggests that empathic failures may reflect the absence of any one (or more) of these.
A taxonomy of empathic failures
In some cases, people may not accurately mentalize about the experiences of others (Adams et al., 2010). For example, people who viewed images of Hurricane Katrina victims attributed fewer secondary emotions (e.g., anguish) to racial out-group relative to in-group members; decreased attribution of secondary emotions to out-group members predicted decreased intent to volunteer in hurricane relief efforts (Cuddy, Rock, & Norton, 2007). In other cases, people may easily understand each other’s perspectives and experiences but fail to share each other’s emotional states or feel no concern for those states. For instance, individuals with psychopathy are capable of reporting on targets’ mental states but do not feel congruent emotions (Blair, 2005). Another example of this dissociation in a much different context occurs among professional helpers, who sometimes experience “compassion fatigue.” Over time, they find that their ability to feel for their clients declines, though they still understand their clients’ perspectives (Batson, Ahmad, & Stocks, 2004). Finally, there are cases in which people understand what another person is feeling and experience personal distress in response, but this distress inhibits their capacity for concern (Batson, Fultz, & Schoenrade, 1987).
Active conflict engenders a different empathic-response profile altogether. Both parties must consider the other side’s internal states (i.e., mentalize with opponents). However, in these cases, experience sharing and empathic concern are replaced with overt antipathy—for instance, pleasure in response to others’ suffering (Cikara, Bruneau, Van Bavel, & Saxe, 2014). Antipathy inhibits helping behavior and promotes harm. For example, in one study, behavioral and neural indices of pleasure in response to the suffering of a rival sports fan predicted decreased willingness to later help that rival (Hein, Silani, Preuschoff, Batson, & Singer, 2010).
Psychologists have continued to map out how different combinations of the subcomponents of empathy predict specific emotional and behavioral responses (approach vs. avoidance: Decety, 2011; Zaki, 2014; help vs. harm: Cikara, 2015), some of which may be more consequential for conflict reduction than others. For example, although some interventions emphasize experience sharing, mentalizing and compassion are significantly better predictors of sensitivity to injustice (Decety & Yoder, 2015) and thus might be more useful empathic components to cultivate when promoting justice-related concerns.
A wellspring of recent evidence further indicates that people’s expectations and emotions prior to social encounters can affect their tendency to engage empathically within those encounters. These precursors often motivate people to approach or avoid engagement with others’ mental states and emotions (see Zaki, 2014, for review). Therefore, practitioners should be aware of the precursors to empathy that might exacerbate empathic failures.
Emotions prior to encounters
In conflict-intervention contexts, preexisting negative emotions (e.g., anger, suspicion) toward another person or group motivate empathy avoidance and hamper instructed efforts to empathize (Bar-Tal & Halperin, 2011). By contrast, above and beyond positive attitudes, positive states such as trust are critical for building cooperative interactions across group boundaries (Balliet & Van Lange, 2013). In fact, in the absence of trust, expressions of empathy from out-group members fail to promote reconciliation between groups in conflict (Nadler & Liviatan, 2006).
Beliefs about empathy
One common assumption is that empathy is uncontrollable—something that automatically happens or does not happen to perceivers when they encounter others in distress (Zaki, 2014). People who share this assumption might interpret difficulty empathizing with one person as evidence of a boundary condition on their empathic ability in general. Like other “fixed” mind-sets (Dweck & Leggett, 1988), this belief can cause people to avoid situations (and people) that might challenge their empathic abilities (Schumann, Zaki, & Dweck, 2014).
Norms
People often adjust their behaviors and preferences to match the norms of their group: a choice that sometimes propagates antisocial behaviors and opinions. At the group level, prejudice and stereotyping closely track group norms. Individuals who believe—or are experimentally led to believe—that their in-group holds hostile attitudes toward an out-group are more likely to express and tolerate antisocial attitudes toward that out-group (Crandall, Eshleman, & O’Brien, 2002; Stangor, Sechrist, & Jost, 2001). Furthermore, intergroup contact exerts less of a positive effect on the attitudes of people who believe that others close to them (e.g., family members) do not support intergroup harmony (Ata, Bastian, & Lusher, 2009). At the individual level, adolescents’ perceptions of normative attitudes predict their responses to bullying. Teenagers who believe that their peers support bullying often encourage bullies or fail to intervene on behalf of victims (Salmivalli & Voeten, 2004). This effect is even more pernicious given that children and adolescents tend to believe that their peers hold more pro-bullying attitudes than they actually do (Sandstrom, Makover, & Bartini, 2013).
Treatment
Understanding the precise nature of empathic failures and their potential precursors yields concrete recommendations about how to address these failures. Interventions should target the specific empathic process that is missing in a given context. In addition, several techniques can be used prior to interactions between relevant parties to motivationally “tune” and prepare individuals to engage empathically with others.
Targeting specific empathic failures
When people fail to understand the experiences of out-group members (e.g., Adams et al., 2010), perspective-taking exercises can improve accuracy (Eyal & Epley, 2010; Todd, Hanko, Galinsky, & Mussweiler, 2011) and aid in building intergroup rapport. In cases of overt conflict, different issues arise. Parties in conflict may understand each other well but feel incongruent emotions and antipathy toward one another. These situations are unlikely to benefit from perspective taking (but see Batson et al., 1997).
Managing emotions prior to encounters
Because conflict-related emotions are characterized by both the absence of compassion and the presence of antipathy, parties in active conflict should focus on regulating these emotions prior to social encounters. In one experiment, Israeli participants received training in cognitive-reappraisal strategies in which they were asked to respond to anger-inducing stimuli in a cold and detached manner (cf. Richards & Gross, 2000) or were given no such training. They then read material about the Israel-Palestine conflict and were asked about their opinions on the conflict. Participants who were trained in reappraisal, compared to those who were not, were subsequently more supportive of policies designed to resolve the conflict (Halperin, Porat, Tamir, & Gross, 2013). It is worth noting that Halperin et al. (2013) taught participants reappraisal strategies in response to stimuli unrelated to the conflict, suggesting that general reappraisal ability can mitigate antisocial affect in difficult intergroup contexts. Similar efforts should be directed not only at diminishing negative conflict-related emotions but also at building positive emotions such as trust (Nadler & Liviatan, 2006).
Encouraging malleable lay theories
Interventions should also dispel preconceptions of individuals’ empathic capacity as unchangeable. Consistent with this suggestion, convincing individuals that empathy is malleable increases their willingness to empathize in challenging contexts. People who were induced to hold theories of empathy as something malleable, relative to something fixed, were more willing to spend time listening to out-group members’ emotional stories and to volunteer for empathy training after failing at an interpersonal-accuracy task (Schumann et al., 2014).
Highlighting empathy-positive norms
Although norms can encourage hostility, they can just as powerfully encourage prosocial behaviors, even in the midst of conflict (Paluck, 2011; Raymond, Weldon, Kelly, Arriaga, & Clark, 2013). For example, people who learn that other members of their social group feel empathy for stigmatized social targets report greater empathy themselves and act more prosocially toward those targets (Nook, Ong, Morelli, Mitchell, & Zaki, 2015). Norms can also foster empathy in intergroup settings. Highlighting empathy as a normative characteristic of one’s own group can enhance the empathy individuals from that group subsequently report for out-group targets (Tarrant, Dazeley, & Cottom, 2009).
Moving Beyond Empathy Building in Interventions
The ultimate goal of most interventions is to promote peaceful, cooperative interactions between parties in conflict. There are many interpersonal and intergroup situations in which different components of empathy foster not only positive attitudes but also positive relational and social behavior (Todd & Galinsky, 2014). That said, practitioners should be aware of boundary cases in which empathy on its own either backfires or fails to achieve these laudable goals. For example, at the interpersonal level, empathy—specifically mentalizing—can produce antisocial outcomes when it brings unsavory social information into focus. Individuals engaged in zero-sum negotiations who take the perspective of their negotiation partner subsequently act less ethically, presumably because they foresee their partner’s motivation to do the same (Pierce, Kilduff, Galinsky, & Sivanathan, 2013).
Likewise, empathy alone may be insufficient for mitigating conflict between groups. Recent evidence indicates that in intergroup contexts, the gap between in-group and out-group empathy predicts out-group helping better than either trait empathy or out-group empathy alone; larger gaps predict less helping (Bruneau, Cikara, & Saxe, 2015). The limitations of empathy are particularly stark when groups differ in social or economic status. For example, perspective taking improves high-power group members’ (e.g., Israelis, Americans) feelings about low-power groups (e.g., Palestinians, Mexican immigrants) but does not improve low-power group members’ feelings about high-power groups (Bruneau & Saxe, 2012). Relatedly, perpetrators and victims express divergent needs in the wake of a violation: Whereas perpetrators want their morality restored, victims prioritize restoring their sense of power (Shnabel & Nadler, 2008).
Even when empathy does improve intergroup relations, it may have the ironic effect of masking structural sources of the conflict, such as unequal access to education, health care, and other resources. The effects of softening discord between groups can backfire by perpetuating the status quo. For instance, exercises that induce attitudinal harmony between groups also delegitimize lower-status groups’ entitlement to concrete change and reduce motivation to engage in collective action (Dixon, Tropp, Durrheim, & Tredoux, 2010).
Finally, interventions that emphasize commonalities between groups—as many empathy interventions do—may set up overly optimistic expectations among low-power groups. In one experiment, participants were assigned to either a low-power group or a high-power group that decided how many points were allocated to the low-power group and were set to the task of discussing either the groups’ commonalities or their differences. Commonality-focused as compared to difference-focused contact created higher expectations for equitable resource allocation among the low-power-group members but did not actually elicit more egalitarian behavior from the high-power group (Saguy, Tausch, Dovidio, & Pratto, 2009), generating a potentially dangerous sense of betrayal between groups.
Applied efforts should reflect current insights about how and when empathy can effectively foster positive outcomes and complement empathy building with other approaches. For instance, prior to negotiations between individuals, interventions might focus not only on accurately understanding one’s competitor but also on framing interactions with that competitor in a way that promotes ethical norms (e.g., committing to ethical behavior prior to the interaction; Shu, Mazar, Gino, Ariely, & Bazerman, 2012).
In intergroup contexts, interventions should seek to minimize the distance between empathy for “us” and “them” rather than universally increasing empathy (Bruneau, Cikara, & Saxe, under review). In cases involving groups of unequal status, conflict resolution should move beyond mere contact and harmony building and support opportunities for collective action among low-power groups. As noted above, perspective taking does not lead to more positive attitudes among low-power groups; however, in line with this idea, “perspective giving”—communicating the in-group’s grievances to a member of the high-power group—improves intergroup attitudes among low-power groups (Bruneau & Saxe, 2012). Similarly, feeling that another has successfully taken one’s own perspective promotes prosocial behavior (Goldstein, Vezich, & Shapiro, 2014).
Conclusions
Empathy often appears to be in short supply, especially during interpersonal and intergroup conflict. Resulting apathy and antipathy stand in the way of peace building and conflict resolution. However, simply fostering more empathy may not always facilitate positive change. Therefore, addressing lapses in empathy—as well as the boundary conditions of empathy itself—should constitute a key mission not only of social scientists but of practitioners and policymakers as well. It is our hope that integrating these evidence-based insights—and the contemporary science of empathy more broadly—will help individuals and programs that aim to improve collective social outcomes do so in a more effective and informed way.
Footnotes
Acknowledgements
We thank Adam Waytz and members of the Stanford Social Neuroscience Lab and the Harvard Intergroup Neuroscience Lab for helpful comments on earlier versions of this manuscript.
Authors’ Note
Authors contributed equally; authorship order was determined by a coin flip.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.