What is compassion connected to, and does it differ by gender and age? A network analysis of compassion, self-compassion, mindfulness, empathy and mental health
Su-gyeong Lee, Jang-won Seo
BMC Psychology November 5, 2026 DOI: 10.1186/s40359-026-05930-3 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Observational cross-sectional network analysis Peer reviewed |
|---|---|
| Sample size | 1,020 |
| Population | Korean adults |
| Topics | Meditation |
| Key findings | Compassion was closely connected to empathy, whereas self-compassion was more strongly linked to mindfulness and well-being; the capacities to feel, tolerate, and alleviate suffering were central nodes. Gender and age subgroup networks showed differing organization: men's compassion and self-compassion tended to center on cognitive factors, women showed relatively immediate behavioral motivation for compassion without emotional responding, suffering tolerance was most influential in middle-aged adults, and empathy's role diminished while mindfulness's expanded in older adults. |
Abstract
Abstract Compassion is an emotional state experienced in response to witnessing the suffering of others, accompanied by an internal motivation to help. When directed toward the self, it is referred to as self-compassion. Previous research suggests that compassion and self-compassion are closely related to mindfulness and empathy, contributing to reductions in anxiety and depression and to the enhancement of positive affect. Accordingly, this study employed network analysis to explore the complex structure of interactions among these psychological variables. A total of 1,020 Korean adults completed measures of compassion, self-compassion, mindfulness, empathy, well-being, and negative affect (depression, anxiety, and stress). An EBICglasso network was estimated in R (version 4.5.1). Centrality indices were computed, and subgroup networks were constructed by gender and age to compare structural differences. The results showed that compassion was closely connected to empathy, whereas self-compassion was more strongly linked to mindfulness and well-being; the capacities to feel, tolerate, and alleviate suffering emerged as central nodes. In the gender-based networks, compassion and self-compassion among men tended to be organized primarily around cognitive factors, whereas women exhibited a pattern in which the behavioral motivation for compassion was activated relatively immediately, even in the absence of emotional responding. In the age-based networks, the influence of suffering tolerance was most pronounced in middle-aged adults, while the role of empathy diminished and the role of mindfulness expanded in older adults. By examining compassion at the level of its constituent elements and visualizing the interaction structure among related variables, this study provides an empirical foundation for extending theoretical understanding and for developing tailored intervention strategies.
Comparable studies
Other cross-sectional and survey studies on meditation, most cited first.
| Study | Year | Design | Participants |
|---|---|---|---|
| Construct Validity of the Five Facet Mindfulness Questionnaire in Meditating and Nonmeditating Samples Experienced meditators and nonmeditating comparison groups | 2008 | Cross-sectional study | |
| Neural Correlates of Dispositional Mindfulness During Affect Labeling Adult participants | 2007 | Observational cross-sectional | n = 27 |
| Assessing mindfulness in children and adolescents: Development and validation of the Child and Adolescent Mindfulness Measure (CAMM). School-aged children and adolescents | 2011 | Cross-sectional validation study | n = 1,413 |
| Mindfulness and self-compassion as predictors of psychological wellbeing in long-term meditators and matched nonmeditators Experienced meditators and demographically matched nonmeditators | 2012 | Cross-sectional study | n = 152 |
| Mechanisms of Action in the Inverse Relationship Between Mindfulness and Psychological Distress Undergraduates | 2008 | Cross-sectional, correlational |