Compassion & Empathy Research: An Annotated BibliographyBy Jennifer Goetz | May 1, 2004 | 0 comments
What has scientific inquiry revealed about compassion? Read summaries of the most intriguing research to date.
II. Compassionate Outcomes
III. Encouraging Compassion
The goal of this bibliography is threefold: 1) to provide references to broad definitions of compassion and related constructs, 2) to summarize behaviors and outcomes related to compassion, and 3) to review the situational, developmental, and cultural processes that may lead to compassion. The bibliography is by no means exhaustive, but readings have been chosen on the basis of their relevance to the subject and noteworthiness to the field. While many of the articles are highly technical in nature, many readings are included because of their readability and accessibility to general audiences.
I. What is compassion?
The construct of compassion is not clearly defined in psychological literature. Our first step was to form a working definition that would allow us to explore related constructs. We define compassion as a feeling of sorrow or concern for another person's suffering or need accompanied by a subsequent desire to alleviate the suffering. This phrasing focuses on compassion as an emotion: a short-lived feeling that anyone may experience. We expect, however, that there are specific conditions in which people will be more likely to feel compassion, that there are differences in individual propensities to feel compassion, and that many people and cultures may view compassion as a basic human value.
Strict dictionary definitions have a hard time separating the feelings of empathy, sympathy, compassion, and pity. Often these words are used to define each other. However research on the concepts has begun to pull them apart. Empathy is considered a mirroring or vicarious experience of another's emotions, whether they be sorrow or joy. Sympathy, on the other hand, is a feeling of sorrow associated specifically with the suffering or need of another. Literally, it is fellow-feeling, and requires a certain degree of equality in situation or circumstances. This is in contrast to pity, which regards its object not only as suffering, but weak, and hence as inferior. Compassion is much like sympathy in that it stems from the suffering of another, but it also includes the need or desire to alleviate suffering (Eisenberg, 2002).
Psychological research by Batson and his colleagues has identified two distinct emotions: empathic concern and personal distress (Batson, Fultz, & Schoenrade, 1987). Empathic concern is other-focused congruent emotion produced by witnessing another person's suffering that involves such feelings as sympathy, compassion, softheartedness, and tenderness. Personal distress, on the other hand, is the negative self-focused emotion that may be the result of the uncomfortable situations that often produce empathic concern. There is recent evidence for facial and physiological components to these emotions (Zhou, Valiente, & Eisenberg, 2003).
Similar work has found personality correlates of empathic concern and distress. With their scale, Davis and colleagues were able to predict people who were likely to feel empathic concern, personal distress, and also those who were more likely to look at situations from another person's perspective (Davis, 1983). Finally, research on basic human values has identified constructs that are related to our conception of compassion. Benevolence is concerned with the preservation and enhancement of the welfare of people with whom one is in frequent and personal contact, while universalism reflects a focus on understanding, appreciation, tolerance and protection for the welfare of all people and for nature (Schwartz et al, 2001).
Batson, C. D., Fultz, J., & Schoenrade, P. A. (1987). Distress and empathy: Two qualitatively distinct vicarious emotions with different motivational consequences. Journal of Personality, 55(1), 19-39.
In a review of emotional evidence, Batson and colleagues make a case for the distinction between empathic concern and personal distress. This distinction is relevant to the study of compassion because the authors' definition is interchangeable with that of compassion: the other-focused congruent emotion produced by witnessing another person's suffering that involves such feelings as sympathy, compassion, softheartedness, and tenderness. Presenting factor analyses from six studies, the authors show that two consistent factors occur in emotional responses to observing another in distress: empathic concern and distress. In addition, results from 5 studies support the assertion that empathic concern consistently motivates people to act altruistically, whereas distress motivates them to act egoistically.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality & Social Psychology, 44(1), 113-126.This article describes the Interpersonal Reactivity Index (IRI), which measures the multifaceted aspects of empathy. Four subscales tap into interrelated personality constructs of empathy: perspective-taking, empathic concern, personal distress, and fantasy. The article describes each of the constructs and their relationship to measures of social competence, self-esteem, emotionality, sensitivity to others, and intelligence.
Eisenberg, N. (2002). Empathy-related emotional responses, altruism, and their socialization. In R. J. Davidson & A. Harrington (Eds.), Visions of compassion: Western scientists and Tibetan Buddhists examine human nature (pp. 131-164). London: Oxford University Press.This chapter summarizes the history of research to date on sympathy and prosocial action. It is a good introduction to the study of compassion. The chapter includes the necessary distinction between sympathy (what Batson calls empathy) and personal distress, the first of which will lead to helping behavior, the second of which will not necessarily do so. Eisenberg illustrates how individual differences in emotionality and regulation may lead to tendencies to feel sympathy or personal distress. The chapter summarizes socialization that may lead to individual differences, including quality and style of parent-child relationship, disciplinary style, parental responses to emotions, and parental expressiveness.
Schwartz, S. H., Melech, G., Lehmann, A., Burgess, S., Harris, M., & Owens, V. (2001). Extending the cross-cultural validity of theory of basic human values with a different method of measurement. Journal of Cross-Cultural Psychology, 32(5), 519-542.In this recent article, Schwartz and colleagues follow-up on longstanding research showing that human values are basically the same across cultures. This is not to say these are valued to the same extent by peoples in all cultures, but that they exist and are recognized as values in all cultures. Values are defined as "desirable, trans-situational goals, varying in importance, that serve as guiding principles in people's lives". The 10 basic values include: power, achievement, hedonism, stimulation, self-direction, universalism, benevolence, tradition, conformity, and security. The two values most relevant to the concept of compassion are benevolence and universalism. Benevolence is the value of preservation and enhancement of the welfare of people with whom one is in frequent and personal contact. Universalism has a broader application - the understanding, appreciation, tolerance and protection for the welfare of all people and for nature. Both benevolence and universalism are shown to increase slightly with age in Italy and South Africa. Benevolence increased with education in South Africa. Finally, Isreali students who supported liberal political parties were more likely to value universalism than nationalist and centrist students.
Zhou, Q., Valiente, C., & Eisenberg, N. (2003). Empathy and its measurement. In S. J. Lopez & C. R. Snyder (Eds.), Positive psychological assessment: A handbook of models and measures (pp. pp. 269-284). Washington, DC, US: American Psychological Association.This is the most up to date review on empathy and the related phenomena of sympathy and personal distress. The authors provide definitions for each of these constructs, as well as summaries of the measurements and methods related to each for children and adults. The authors recommend the use of questionnaires that can differentiate empathy from personal distress (e.g. Davis' Interpersonal Reactivity Scale, 1983) as opposed to those that lump them together (e.g. Mehrabian and Epstein, 1972). The use of other-report for children and adults can increase reliability. Evidence that suggests that measurements of facial, gestural, and vocal behavior may also differentiate between sympathy and personal distress is cited. Finally, physiological indices such as skin conductance and heart rate are believed to be able to distinguish between sympathy and personal distress. Sympathy may be connected to decreased heart rate and skin conductance, whereas personal distress may be connected to increased heart rate and skin conductance.
II. What good things does compassion do?
Once we have identified the emotional, personality, and value constructs of compassion, it is helpful to examine the outcomes associated with them. Our definition includes not only feelings of sympathy and concern, but also the desire to alleviate suffering. The following articles cite the outcomes of feeling compassion in the domains of helping, stigma, relationships, forgiveness, negotiation, and justice.
From the outset, feelings of empathic concern have been positively related to helping and prosocial behavior (Batson et al, 1987). Feelings of personal distress, however, have only related to helping behavior when it was the easiest means of reducing distress. This has been shown even in such self-serving economic games such as the prisoner's dilemma (Batson & Moran, 1999). Additionally, personality differences in empathic concern and personal distress predict the types of prosocial behavior people choose to engage in (Davis et al, 1999). People who are more likely to feel empathic concern are also more likely to choose volunteer opportunities in which they expect to feel sympathy. Finally, although sympathy is generally considered to be an emotion that is directed towards individuals, research has shown that feeling and compassion for an individual from a stigmatized group can induce helping for the group as a whole (Batson et al, 2002).
In the domain of personal relationships, personality correlates of sympathy like empathic concern and perspective taking are positively related to long-term relationship satisfaction (Davis & Oathout, 1987). This is perhaps because of the close association between feeling compassion and forgiveness (McCullough et al, 1997) as well as those who are able to look at things from the other person's perspective (McCullough, 2001).
Forgiveness is not easy, however, and when we evaluate an individual to be responsible for negative outcomes, we tend to feel very little compassion and a lot of anger (Weiner, 1993). Research shows that this combination of emotions can be detrimental in negotiation situations because it reduces individuals desire to work together and their ability to identify win-win situations (Allred et al, 1994). In addition, feeling a lot of anger and little sympathy is related to choosing punitive, retaliatory outcomes in classroom settings (Reyna & Weiner, 2001) and public policy (Gault & Sabini, 2000).
Allred, K. G., Mallozzi, J. S., Matsui, F., & Raia, C. P. (1997). The influence of anger and compassion on negotiation performance. Organizational Behavior & Human Decision Processes, 70(3), 175-187.
Batson, C. D., Chang, J., Orr, R., & Rowland, J. (2002). Empathy, attitudes and action: Can feeling for a member of a stigmatized group motivate one to help the group. Personality & Social Psychology Bulletin, 28(12), 1656-1666.Although empathy is an emotion that is felt directly for another individual, it may be possible that these feelings can have a more generalized effect. It is important to remember that Batson's definition of empathy is very similar to our definition of compassion: an other-oriented emotional response that includes sympathy, compassion, and tenderness. In this study, empathy was measured by ratings of adjectives: sympathetic, compassionate, soft-hearted, warm, tender, and moved. In the first study, female participants felt more empathy for a young woman with AIDS when they were asked to take her perspective rather than be objective. This effect was lessened, however, when they were led to believe the young woman was responsible for having contracted AIDS. In addition, feelings of empathy led to better attitudes towards people with AIDS, except when the target group was extremely similar to the participants (young women) and responsible for having AIDS.
The second experiment looked at a more stigmatized group: the homeless. Feelings of empathy for a homeless man increased attitudes for the homeless in general. Again, responsibility of the homeless man for his condition lessened feelings of empathy. Finally, the third study used convicted murderers as the target group. Short-term laboratory results showed that empathic feelings were related to more favorable attitudes towards murderers. A week later, a telephone interviewer called participants and asked about their attitudes towards prison reform. The table below (from the article) shows that attitudes were higher when participants were induced to feel more empathy towards a killer.Attitude Toward Murderers in Experiment 3:
At Laboratory Session and in Telephone Interview 1-2 Weeks Later
Time of measurement of attitudes
|Empathy condition||Laboratory session||Telephone interview|
Batson, C. D., & Moran, T. (1999). Empathy-induced altruism in a prisoner's dilemma. European Journal of Social Psychology, 29(7), 909-924.Over 2000 prisoner's dilemma studies have shown that 30 to 50% of people cooperate with their partner without any coaching. This study sought to show that this proportion could be increased with empathy-induced altruism. Empathy was induced by giving participants a note from a "partner" that indicated that she had just broken up with her boyfriend. Additional instructions encouraged participants to read the note either objectively or with their partner's perspective in mind. People who both read the note and took their partner's perspective indicated more empathy. They also cooperated more than did those who received no note or who read the note objectively. There was also an additional manipulation that led participants to believe the experiment was a business experiment. This manipulation showed that when empathy was not induced, cooperation decreased. When empathy was induced, however, cooperation was almost as strong as the non-business condition. This supports the authors' argument that empathic altruism is separate and distinct from moral motivation.
Davis, M. H., Mitchell, K. V., Hall, J. A., Lothert, J., Snapp, T., & Meyer, M. (1999). Empathy, expectations, and situational preferences: Personality influences on the decision to participate in volunteer helping behaviors. Journal of Personality, 67(3), 469-503.This series of studies shows that personality differences in empathic concern and personal distress affect the choices people make about volunteer activities. In three separate studies, they show that people who are high on empathic concern tend to choose volunteer opportunities in which they expect to feel sympathy. Likewise, people who are high on personal distress are likely to avoid situations in which they expect to feel more personal distress. This was true in a study with hypothetical, volunteering in a lab situation, and real-life volunteering.
Davis, M. H., & Oathout, H. A. (1987). Maintenance of satisfaction in romantic relationships: Empathy and relational competence. Journal of Personality & Social Psychology, 53(2), 397-410.In a correlational study of 264 heterosexual couples, the authors measured dispositional empathy to predict romantic relationship satisfaction via positive and negative behaviors. They used the IRI (see Davis, 1983), which has separate facets of empathy: perspective taking, empathic concern, and personal distress. The first two components predicted positive behaviors such as good communication, warmth, even temperedness, and positive outlook. Personal distress was negatively related to these behaviors, but positively related to untrustworthiness, insensitivity, and possessiveness. Overall, personal dispositions towards perspective taking and empathic concern were predicted more relationship satisfaction in long term rather than short term relationships.
Gault, B. A., & Sabini, J. (2000). The roles of empathy, anger, and gender in predicting attitudes toward punitive, reparative, and preventative public policies. Cognition & Emotion. Special Issue: Emotion, cognition, and decision making, 14(4), 495-520.This article investigated the possibility that gender differences in political attitudes and policy preferences may be the result of state and trait emotional differences. The first three studies provide evidence that women prefer victim-oriented solutions and men prefer perpetrator-punishment policies. The fourth study is most relevant to our interest in compassion. Although it failed to reproduce all of the gender differences found in the earlier studies, the fourth study showed that empathy and anger account for some of the gender differences that exist. Overall, women reported more trait empathy than men. Trait empathy was related positively to ratings of a victim-support organization. In women only, state empathy was also positively related to support for a victim-focused organization. In contrast, the amount of anger participants felt was related to their support of a perpetrator-punishing organization. The authors suggest that an empathic disposition may have more effect on political action or attitude than does immediate emotional feeling.
McCullough, M. E. (2001). Forgiveness: Who does it and how do they do it? Current Directions in Psychological Science, 10(6), 194-197.This article is a quick and comprehensive summary of the forgiveness literature to date. It is written by the leading researcher in this field. People who are inclined to forgive their transgressors tend to be more agreeable, more emotionally stable, and perhaps more spiritually or religiously inclined. Perhaps most relevant to the study of compassion, feeling empathy for a transgressor is highly related to forgiveness, as are generous attributions and appraisals regarding the transgression.
McCullough, M. E., Worthington, E. L. J., & Rachal, K. C. (1997). Interpersonal forgiving in close relationships. Journal of Personality & Social Psychology, 73(2), 321-336.In two studies of forgiving in close relationships, the McCullough and colleagues show that people who feel empathy for the person who hurt or offended them are more likely to forgive them. In the first study, the researchers used Batson's measure of empathic concern (which we liken to compassion) to show that apologies tended to be associated with more empathy and concern for the offending person. In turn, forgiveness was associated with increased conciliatory behavior and decreased avoidance behavior. In other words, apologies led people to see that the offender was feeling remorse and then to reach out to them. The second study tested interventions by comparing two seminars: one that promoted empathy as a precursor to forgiving (empathy seminar) and one that simply encouraged forgiving. Results showed that the empathy seminar encouraged more empathy and more forgiving. Indeed the individual reports of empathy mediated the effects of seminar on forgiving.
Reyna, C., & Weiner, B. (2001). Justice and utility in the classroom: An attributional analysis of the goals of teachers' punishment and intervention strategies. Journal of Educational Psychology, 93(2), 309-319.In this recent study, Reyna and Weiner provide evidence showing that punitive actions of teachers are often closely linked to emotional reactions to students. When students are perceived as responsible for negative outcomes, teachers tend to feel anger and this leads to more punitive responses (the top process in figure 1, below). When students are perceived as not responsible for negative outcomes, however, teachers tend to show more sympathy and this leads to less punitive responses (the lower process in figure 1). The outcomes in these studies of teachers are simply an example of Weiner's work on generally punitive philosophies and behavior.
III. How do we encourage compassion?
I have defined compassion as an emotion we feel when we perceive others in need and want to alleviate that need or suffering. But does everyone feel compassion when they see another person in need? Research reviewed above indicates stable personality differences in tendencies to feel empathic concern and personal distress (Davis, 1983; Davis et al, 1999). The articles reviewed below suggest that individual differences in compassion have genetic, cultural, and socialization factors. In addition, research shows that despite individual differences, situational and target characteristics strongly influence when and how much we feel compassion.
In the first article in this section, Batson (1983) argues that empathy (which can lead to compassion) has evolved to support prosocial behavior towards those who are genetically close to us. He also argues that by using familial symbolism, religion encourages us to expand our prosocial acts to a larger group. This view suggests that there should be both genetic and situational factors leading to differences in compassion. In a study of twins, Zahn-Waxler and colleagues (1992) found that young monozygotic twins showed more similarity in empathic response than did dizygotic twins, indicating that at least some differences empathic responses are genetically based. In addition, the work of Zhou and colleagues (2002) provides strong evidence that empathic tendencies are passed from parent to child. Parental warmth and positive expression predicted more empathic responses in children, which may only underlines the interrelatedness of biological pathways and socialization processes. Finally, research on early adulthood has shown that empathy, sympathy, and prosocial behaviors in childhood strongly related to sympathy and prosocial behavior in early adulthood (Eisenberg et al, 2002).
Apart from genetics and upbringing, situational and target factors affect when and how much we feel compassion. Batson and colleagues (1996) found that having experienced a need ourselves can affect how much we empathize or de-empathize with someone else. For women, this increased the likelihood that they would empathize with the target person. For men, however, it did not have such an affect. In addition, there may be certain emotions which are more likely to elicit empathy and compassion. Duan (2000) found that participants showed more empathy for pleasant emotions and sadness than they did for shame and anger, indicating that the unpleasantness of the target emotion may have an effect on empathy and compassion. Finally, a long line of research by Weiner and colleagues has shown that judgments of responsibility are key to distinguishing when we feel sympathy instead of anger. A negative act that is judged to be the target person's responsibility elicits anger. The same negative act that is judged outside of the target's responsibility elicits sympathy. This sort of analysis applies to judgments related to achievement (failure and success), to stigma (e.g. paraplegia, obesity), and to peer rejection (Weiner, 1993). As discussed earlier, it also predicts teacher's punitive responses in the classroom (Reyna & Weiner, 2001).
Research included in the bibliography (and much not included) provides ample evidence that emotional responses such compassion, sympathy, and empathy motivate prosocial behavior. However, there are clearly other possible motivators which should not be overlooked. Self-interest, valuing the group as a whole, and moral principles are all non-emotional ways to motivate people to work towards the public good (Batson, Ahmad, & Tsang, 2002).
Batson, C.D. (1983). Sociobiology and the role of religion in promoting prosocial behavior: An alternative view. Journal of Personality and Social Psychology, 45, 1380-1385.
Batson, C.D., Ahmad, N., & Tsang, J. (2002). Four motives for community involvement. Journal of Social Issues, 58, 429-445.This article provides a conceptual analysis that differentiates four types of motivation for community involvement: egoism, altruism, collectivism, and principlism. Each of these motives has a different ultimate goal, but each can contribute to the public good. The altruism motive relates to our emotional analysis of compassion. However, the article provides a strong analysis of the broad range of motives for prosocial behavior.
Batson, C.D., Sympson, S.C., Hindman, J.L., Decruz, P., Todd, R.M., Weeks, J.L., Jennings, G., & Burris, C.T. (1996). "I've been there, too": Effect on empathy of prior experience with a need. Personality and Social Psychology Bulletin, 22, 474-482.This article examines the effect of prior experience on empathic reactions to someone in need. Essentially, it asks "must we have been there to care?" Interestingly, the two studies presented in this article find that only women show more empathy when they have had prior experience. In one study, the researchers made participants believe that they would be shocked, and then had them watch someone else get mild shocks. Women showed more empathic reactions when they thought it could have been them compared to having no prior experience. In another study looking at prior experience with break-ups and severe acne, women again showed more empathy if they had prior experience than if they did not. The question remains, why don't men show the same effect? It could be that they are distancing themselves from the person in need. Could it be a form of regulating threats to esteem? Hopefully, future research will address this question. Overall, however, the article shows that both men and women showed high levels of empathy even without prior experience. Apparently, prior experience is not necessary for feeling empathy.
Duan, C. (2000). Being empathic: The role of motivation to empathize and the nature of target emotions. Motivation & Emotion, 24(1), 29-50.This article investigates two situational influences on likelihood to empathize: motivational factors and target emotions. Participants read excerpts from a diary in which a break-up was described and then indicated their emotion and attribution responses. Participants felt more emotional empathy when the target person expressed pleasant emotions or sadness than when the target person expressed anger or shame. The author suggests that this is because anger and shame are both very unpleasant emotions. In addition, the participants' motivation to empathize increased their intellectual empathy when the target person was sad. This was shown by their increased perspective-taking with the target person. Motivation increased empathic emotion when the target person was happy.
Eisenberg, N., Guthrie, I.K., Cumberland, A., Murphy, B.C., Shepard, S.A., Zhou, Q., & Carlo, G. (2002). Prosocial development in early adulthood: A longitudinal study. Journal of Personality & Social Psychology, 82, 993-1006.This article provides one of the only longitudinal studies on adult development of prosocial motivations and behavior. It provides ample evidence showing that the consistent links between empathy, sympathy and prosocial behavior found in children continue into adulthood. Also important, they show that interindividual differences in childhood predict similar differences in adulthood, providing evidence for prosocial personality.
Weiner, B. (1993). On sin versus sickness. American Psychologist, 48, 967-985.This article is written by the leading researcher in attribution analysis (the study of how we judge & evaluate why things happen). He reviews the literature in support of the argument that sympathy and anger result from very similar situations - those with negative outcomes. The main difference is that when we judge a negative act to be someone's responsibility, we feel anger; when we judge it not to be someone's responsibility, we feel sympathy. This sort of analysis applies to judgments related to achievement (failure and success), to stigma, and to peer rejection. In one study (Weiner, Perry & Magnussen, 1988), people distinguished examples of stigmata they felt were uncontrollable (Alzheimer's disease, blindness, cancer, heart disease, paraplegia, and Vietnam War syndrome) and elicited sympathy, as opposed to those considered controllable (AIDS, child abuse, drug addiction, and obesity) and elicited anger.
Zahn-Waxler, C., Robinson, J.L., & Emde, R.N. (1992). The development of empathy in twins. Developmental Psychology, 28, 1038-1047.Studies of children in the 2nd year of life indicate that they have the cognitive, affective, and behavioral capacities needed to display integrated patterns of concern for others in distress. In this study of twins, 94 monozygotic (MZ) and 90 dizygotic (DZ) pairs were observed during 2nd year of life. Moms and experimenters both pretended to hurt themselves and the response of the child was coded from videotape. There were small but consistent gender differences - girls showed more empathic concern and did more prosocial acts (e.g. patting victim or getting a band-aid), and showed more personal distress than did boys. Boys showed more unresponsive behavior. These gender differences increased with age. It is unclear from this study whether these differences are biologically or environmentally based. MZ twin pairs were more similar in their displays of empathic concern than were DZ twin pairs. They were also more similar in their tendency not to respond to the mother. The high similarity in MZ twin pairs suggests that empathic concern has a modest genetic component to it.
Zhou, Q., Eisenberg, N., Losoya, S. H., Fabes, R. A., Reiser, M., & Guthrie, I. K. et al. (2002). The relations of parental warmth and positive expressiveness to children's empathy-related responding and social functioning: A longitudinal study. Child development, 73(3), 893-915.This longitudinal study examined the effects of parenting on children's empathic responses and to their social competence. They found that parents who were warm and supportive tended to express more positive emotions in their child's presence (though not just towards their child). Children whose parents were warm and supportive and who showed more positive emotions showed more empathic responses. In turn, children who showed empathic responses, particularly to negative stimuli, were more socially competent and showed less externalizing of problems as rated by their teachers and parents. It is important to note that parental positive expressivity and warmth predicted child empathy and competence better than the other way around. In addition, children's empathy for negative emotions (possibly sympathy) and not positive emotions was predictive of social competence and less externalizing.
Greater Good wants to know:
Do you think this article will influence your opinions or behavior?