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

AbstractExisting work linking empathy with social behavior has focused overwhelmingly on empathy for the negative emotions of others. But recent research suggests that feeling along with others' negative emotions is a capacity distinct from feeling along with others' positive emotions. In Study 1, we demonstrate the separability of positive and negative empathy by showing that although both relate to some of the same foundational empathic processes, each has a number of distinct correlates. In Study 2 we take an experimental approach and show that encouraging participants to empathize with the positive versus negative emotions of a suffering yet hopeful social group results in distinct patterns of vicarious emotion. Finally, Study 3 shows that although both positive empathy and negative empathy are associated to a similar degree with helping behavior directed toward others in need, positive—but not negative—empathy is related to "everyday" prosocial behaviors aimed specifically at increasing others' positive emotions (e.g., random acts of kindness). Together, these results provide what to our knowledge is the first demonstration of the causal potency of positive and negative empathy as well as the first evidence that positive and negative empathy relate to different types of social behaviors. ACKNOWLEDGMENTSWe thank Michael Gill, Susan Rakowitz, three anonymous reviewers, and editor David Trafimow for their thoughtful feedback. We also thank Michelle Cusumano, Stephanie Gutkin, Laura Ivy, Alex Kimble, Sam Krivensky, Eliza Lewis, Ariana Philbin, Nick Restivo, Francesca Romano, and Elizabeth Sheerin for their assistance with data collection.Notes1Given criticisms surrounding the utility of hierarchical regression as a tool to effectively isolate the unique variance in criterion variables accounted for by predictor variables (e.g., Trafimow, Citation2004), we also note that the key predicted relationships between positive empathy, negative empathy, and the criterion variables employed in both Studies 1 and 3 emerged whether looking at the zero-order correlations or the results of the hierarchical regressions.2We replicated this factor structure for a third time in a separate sample of 236 participants.aFrom Mehrabian and Epstein (Citation1972).eCreated by the authors.bAdapted from Davis (Citation1980).cFrom Spreng et al. (Citation2009).dFrom Mayer et al. (Citation1999).