The study recruited 175 Chinese mothers whose children with autism were receiving rehabilitation services. Each mother completed validated self-report measures covering five domains: parenting stress, positive self-compassion, self-coldness, anxiety symptoms, and depressive symptoms. The researchers then used hierarchical regression analyses, a statistical technique that allows investigators to test whether the relationship between two variables changes depending on the level of a third variable, to examine whether the two facets of self-relating moderated the link between stress and psychological symptoms. This moderation approach is crucial because it moves beyond simply asking whether self-compassion correlates with wellbeing and instead asks whether it changes the slope of the stress-to-symptom pathway itself.

The conceptual foundation of the study rests on an increasingly influential distinction in the self-compassion literature. When Kristin Neff first formalized self-compassion in 2003, she described it as a single construct comprising self-kindness, common humanity, and mindfulness, with their opposites, self-judgment, isolation, and over-identification, treated as the low poles of the same dimensions. Subsequent psychometric work, including factor-analytic studies of the Self-Compassion Scale, has suggested that the positive and negative items form two separable factors. Positive self-compassion reflects an actively warm, supportive stance toward oneself in moments of failure, while self-coldness captures self-criticism, rumination, and a sense of being alone in one’s suffering. Treating these as mirror images, the new findings imply, can obscure which of them is actually doing the psychological work.

The results were unambiguous on the first point. Higher parenting stress was associated with more severe anxiety symptoms and more severe depressive symptoms, consistent with a large body of international research showing that caregivers of autistic children carry elevated rates of psychopathology compared with parents of neurotypical children. Meta-analytic work has repeatedly documented heightened depression and anxiety prevalence in this population, and studies in Chinese samples specifically have linked caregiving demands, behavioral problems in children, and affiliate stigma, the internalized shame that comes from association with a stigmatized condition, to poorer maternal mental health. The Guangzhou study replicates this stress-symptom association in a context where caregiving expectations weigh heavily on mothers.

What makes the new paper noteworthy is the asymmetry in its moderation findings. Self-coldness significantly moderated both associations: the links between parenting stress and anxiety, and between parenting stress and depression, were stronger among mothers who reported higher levels of self-coldness. In practical terms, two mothers experiencing identical levels of parenting stress faced different psychological trajectories depending on how coldly they related to themselves. For the mother high in self-coldness, stress translated more readily into clinical symptoms; for the mother low in self-coldness, the same stress exerted a weaker pull toward anxiety and depression. This is the classic signature of a vulnerability amplifier, a psychological characteristic that does not cause stress but determines how damaging stress becomes once it arrives.

Positive self-compassion, by contrast, did not significantly moderate either association. The absence of a buffering effect from the warm, kind facet of self-relating is the study’s most provocative result, and it aligns with a growing meta-analytic literature suggesting that the negative components of self-compassion are often more strongly and consistently related to psychopathology than the positive components. One influential meta-analysis of the differential effects of self-compassion components found that self-coldness tends to show robust associations with distress, while positive self-compassion’s protective contributions are smaller and more variable. The new findings extend this pattern into a high-stress caregiving population and into a cultural context where self-critical tendencies may be particularly consequential.

The Chinese setting is not incidental. The authors note that mothers in China often face strong expectations regarding caregiving responsibility and social evaluation, pressures rooted in longstanding cultural concepts of face and the moral weight placed on maternal devotion. Qualitative research on Chinese mothers of autistic children describes intense scrutiny from extended family and community, feelings of guilt and shame, and the burden of negotiating what it means to be a good mother when a child’s development diverges from social norms. In such an environment, self-coldness, the internalized voice of judgment and isolation, may interact powerfully with external stigma, converting everyday caregiving stress into internalizing symptoms. The study’s design cannot test that mechanism directly, but the pattern is consistent with theoretical models in which self-criticism and rumination amplify the emotional impact of stressors.

The researchers also examined whether child gender and subjective socioeconomic status, measured with a subjective social status scale, moderated the stress-symptom links. Neither interaction was significant, with those results reported in supplementary materials. This null finding sharpens the central message: among the moderators tested, self-coldness stood out as the factor that changed how stress was experienced psychologically, while demographic characteristics did not. The measures themselves were well-established instruments, including Chinese-validated versions of the Parenting Stress Index, the Self-Compassion Scale, the seven-item Generalized Anxiety Disorder scale, and the nine-item Patient Health Questionnaire for depression, giving the findings a solid psychometric footing.

The clinical implications are potentially significant. If self-coldness amplifies the psychological toll of parenting stress, then interventions that specifically target self-critical self-relating, such as compassion-focused therapy, which was developed by Paul Gilbert precisely for people with high shame and self-criticism, may be especially valuable for this population. Mindfulness- and self-compassion-based programs for parents of autistic children have already shown promise in randomized trials, reducing stress and improving wellbeing and resilience. The new findings suggest that such programs should not assume that building warmth is sufficient; directly dismantling self-coldness, the harsh internal narrative that turns stress into suffering, may be the more potent therapeutic lever. Screening for self-coldness could also help identify mothers at greatest risk of developing anxiety and depression under caregiving strain.

The authors are careful about the limits of their evidence. The study is cross-sectional, meaning the data capture a single moment in time, so it cannot establish whether self-coldness causes symptom amplification or whether emerging anxiety and depression deepen self-critical habits; longitudinal designs are needed to disentangle these directions. The sample of 175 mothers drawn from rehabilitation service settings is also geographically constrained, and the authors explicitly call for studies with more diverse samples to confirm the findings. Nonetheless, the study makes a methodological argument that reaches beyond autism research: self-compassion is not one thing, and the field’s habit of measuring it as a single global score may have hidden the very mechanisms that matter most. For mothers carrying the daily weight of raising autistic children, the difference between self-kindness and self-coldness may be the difference between stress that is weathered and stress that becomes illness.

Subject of Research: Moderating roles of positive self-compassion and self-coldness in the association between parenting stress and anxiety and depressive symptoms among Chinese mothers of children with autism spectrum disorder

Article Title: Parenting Stress, Anxiety, and Depressive Symptoms in Chinese Mothers of Children With ASD: Distinct Moderating Roles of Positive Self-Compassion and Self-Coldness

Article References: Wang, J., Su, X., Li, Y., & Zeng, X. (2026). Parenting Stress, Anxiety, and Depressive Symptoms in Chinese Mothers of Children With ASD: Distinct Moderating Roles of Positive Self-Compassion and Self-Coldness. Journal of Autism and Developmental Disorders. https://doi.org/10.1007/s10803-026-07535-6

Image Credits: AI Generated

DOI: 10.1007/s10803-026-07535-6

Keywords: autism spectrum disorder, parenting stress, anxiety, depression, self-compassion, self-coldness, self-criticism, Chinese mothers, caregiver mental health, moderation analysis, affiliate stigma, compassion-focused therapy