In everyday life, people mostly empathize with very close others, and they empathize with positive emotions 3 times as frequently as with negative emotions

The Experience of Empathy in Everyday Life. Gregory John Depow, Zoë Francis, Michael Inzlicht. Psychological Science, July 9, 2021. https://doi.org/10.1177/0956797621995202

Abstract: We used experience sampling to examine perceptions of empathy in the everyday lives of a group of 246 U.S. adults who were quota sampled to represent the population on key demographics. Participants reported an average of about nine opportunities to empathize per day; these experiences were positively associated with prosocial behavior, a relationship not found with trait measures. Although much of the literature focuses on the distress of strangers, in everyday life, people mostly empathize with very close others, and they empathize with positive emotions 3 times as frequently as with negative emotions. Although trait empathy was negatively associated only with well-being, empathy in daily life was generally associated with increased well-being. Theoretically distinct components of empathy—emotion sharing, perspective taking, and compassion—typically co-occur in everyday empathy experiences. Finally, empathy in everyday life was higher for women and the religious but not significantly lower for conservatives and the wealthy.

Keywords: emotions, interpersonal relationships, motivation, personality, sex differences, social cognition, theory of mind, well-being, open data, open materials, preregistered

Discussion

The majority of research on empathy has focused on negative emotions—typically of strangers and typically in laboratory settings. However, in everyday life, empathy was more often reported in response to positive emotions, not negative emotions, and participants empathized to a greater extent as emotions became more positive. Although these results may be influenced by reporting biases, they are consistent with the relative frequency of emotions experienced in daily life; positive emotions, such as excitement and enthusiasm, are experienced approximately 3 times more frequently than negative emotions, such as disgust, anger, and fear (Zelenski & Larsen, 2000).

Empathy was most often reported in response to close others. This effect may be due to mere availability because many people likely have more social interactions within existing relationships than with strangers. However, empathy also may be biased in favor of close others (Cikara et al., 2011). Supporting this possibility, our results showed that individuals empathized to a greater extent as closeness increased.

These findings have implications for our understanding of empathy as a motivated phenomenon (Zaki, 2014). The emerging work on lack of motivation to empathize might be related, in some part, to most lab studies of empathy involving the negative emotions of strangers. Whereas some work suggests that people might avoid both positive and negative empathy (Cameron et al., 2019), empathy for strangers might be especially unmotivating (Ferguson et al., 2020). In future lab studies on motivation to empathize, researchers should consider including a range of expressed emotions and including empathy targets who are close to the participant, not just strangers.

In prior work, researchers have suggested that dissociable components of empathy interact in real-life interactions (Morelli et al., 2014; Zaki & Ochsner, 2012). The current study shows that emotion sharing, compassion, and perspective taking are reported together almost all of the time and are rarely reported in isolation. Although these components of empathy can be theoretically differentiated, our data suggest that they are typically experienced together by most people in most daily situations.

Finally, we examined various established demographic findings about empathy to see which findings hold across people’s experiences in daily life. Some established effects were replicated—for example, both women and religious participants tended to report experiencing empathy more often than men and the nonreligious. However, other relationships did not replicate in the context of daily life; we found a weak relationship between compassion and income and little to no relationship between empathy and political orientation—although the true effects of income and politics on empathy may be smaller than we were able to detect given our statistical power.

Limitations

Ground truth

Experience sampling allows us to get closer to the temporal, emotional, and social context of empathy, but it is still self-report data. The number of empathy opportunities and true ratios of positive and negative opportunities may vary from those reported—indeed, participants appeared selective in which observed emotions were perceived as empathy opportunities, consistent with motivational accounts of empathy (Cameron et al., 2019; de Vignemont & Singer, 2006; Zaki, 2014). However, truly objective measures of empathy opportunities would be difficult to obtain, given that empathy cues in the environment may not be attended to. Furthermore, feeling empathy may be best captured via self-report, given that it is an internal and subjective phenomenon.

Representativeness

Our sample was quota matched to census data on six key demographics, making our results more representative than is typical. Because the sample is not random, representativeness cannot be assumed on other demographics. For example, to join the study, participants were required to have a smartphone. However, 81% of U.S. adults had a smartphone as of 2019 (Pew Research Center, 2021), and the results from our lowest income participants mirrored those from the entire sample. The generalizability of our findings to other populations, especially non–Western, educated, industrialized, rich, and democratic (WEIRD) populations, remains to be demonstrated.

Training and fatigue

Potentially, repeatedly responding to surveys on empathy might have trained participants to notice empathy opportunities. However, there was only one significant difference on trait empathy questionnaires at baseline compared with the final experience-sampling survey (all ps > .05), and empathy itself did not become more prevalent over the course of the week. Empathy opportunities were reported slightly less frequently as the week progressed, suggesting that individuals may have become decreasingly inclined to report empathy. However, because we nested results within survey day, this shift should minimally affect our results.

Definitions and demand

We chose to supply a definition of empathy to our participants to reduce noise from varying lay theories of empathy (Hall et al., 2021). By defining empathy as a process involving three related but distinct experiences, however, we may have introduced demand, resulting in participants inflating reports of the co-occurrence of all three processes or otherwise influencing participants’ reports (although similar results were reported for participants’ own empathy and empathy received). Future work should replicate these findings with different definitions of empathy.