Why does non-suicidal self-injury develop, and what might sustain it? In their review, Kaess et al. (2021) examine biological findings alongside biological findings alongside psychological and social processes. They distinguish vulnerabilities that may precede self-injury, characteristics associated with ongoing behaviour, and temporary changes surrounding individual episodes. Their framework brings together research on emotion processing, stress and pain, while emphasising uncertainties about cause and consequence. It supports attention to stress and emotion regulation in care, with personalised biological treatments remaining a future research goal.
Understanding the timing of biological processes
Non-suicidal self-injury (NSSI) refers to deliberate damage to one’s own body tissue without suicidal intent and outside socially sanctioned practices. The review focuses on this behaviour, rather than every form of self-harm. Absence of suicidal intent in an episode does not remove its clinical significance, since findings highlight the association between NSSI and subsequent suicidal outcomes (Kaess et al., 2021). Distal biological traits are vulnerabilities present early in life or developing over time. Proximal traits are relatively stable characteristics found among people currently engaging in NSSI. Biological states are shorter-lived changes immediately before, during or after an episode. Separating these categories matters because a difference detected after self-injury begins cannot establish whether it contributed to onset or developed subsequently (Kaess et al., 2021). Adolescence is particularly relevant because NSSI commonly emerges during a period of substantial brain development. The authors describe this developmental flexibility as both a source of vulnerability and a potential opportunity for intervention, since understanding when biological processes change, and how they interact with psychological and social experiences, could help researchers identify the timing and targets of future interventions more precisely (Kaess et al., 2021).
Early vulnerability and childhood adversity
Research on genetic influences was limited and inconsistent in the evidence reviewed. Twin studies suggested a heritable contribution, but studies of individual genetic variants generally involved samples too small to establish reliable effects. The authors also discuss possible interactions between genetic variation and stressful environments, without identifying a genetic explanation that accounts for NSSI (Kaess et al., 2021). Childhood adversity is associated with later self-injury, although its biological pathways remain uncertain. The review considers changes in stress regulation, brain development and inflammatory processes as possible links. It also distinguishes a direct pathway to NSSI from a broader vulnerability to psychological difficulties that may subsequently increase risk. The authors emphasise the need to follow young people before self-injury begins; associations measured afterwards cannot resolve these developmental questions (Kaess et al., 2021).
Emotion processing and social experience
Among the more consistent findings reviewed are differences in fronto-limbic systems, which connect regions involved in emotional responses with regions supporting regulation. An early study led by Plener found greater activation in the amygdala and anterior cingulate cortex when adolescent participants with NSSI viewed emotional material. The review interprets this as evidence relevant to heightened emotional reactivity, while recognising the limitations of the developing imaging literature (Kaess et al., 2021; Plener et al., 2012). Additionally, social experiences are closely connected to these processes. Studies involving social exclusion, including work by Brown et al. (2017), identified differences in neural responses among participants with NSSI. Across this literature, the review describes patterns consistent with increased sensitivity to rejection and difficulties regulating the emotions it evokes. Biological findings therefore sit alongside the interpersonal difficulties that can surround self-injury (Kaess et al., 2021). Ho et al. (2021) also identified altered connectivity within networks involved in internally focused thought and detecting salient information in depressed adolescents with NSSI. Including adolescents with depression but without NSSI helped examine whether differences extended beyond depression alone. Nevertheless, the review stresses that psychiatric comorbidity remains a major challenge when interpreting biological associations (Kaess et al., 2021). Evidence concerning cognitive control and reward was less consistent. Studies did not establish a clear pattern of impaired cognitive control, and reward experiments produced differing results. The authors therefore caution against treating either impulsivity or altered reward processing as a settled biological account of NSSI (Kaess et al., 2021).
Stress responses depend on what is measured
The review examines both the autonomic nervous system and the hypothalamic–pituitary–adrenal axis, which regulates cortisol release. Several studies found a reduced cortisol response to laboratory psychosocial stress among adolescents with NSSI. Klimes-Dougan et al. (2019), for example, investigated this pattern in adolescents with depression, including comparisons between those with and without NSSI. This finding does not establish that cortisol is universally low in people who self-injure. The review describes differences between responses to acute stress, changes after awakening and longer-term measures. Findings concerning heart rate, heart rate variability and skin conductance were also mixed. These measures capture different aspects of regulation, and their relationships with NSSI may depend partly on co-occurring difficulties and previous adversity (Kaess et al., 2021).
Pain sensitivity and the opioid hypothesis
Reduced sensitivity to laboratory pain is one of the better-replicated findings discussed. A meta-analysis found that people engaging in self-injury generally had higher pain thresholds and tolerance, and reported lower pain intensity, than healthy controls (Koenig et al., 2016). However, Kaess et al. (2021) emphasise that the direction of the relationship remains unresolved: reduced sensitivity might contribute to vulnerability, develop through repeated self-injury, or reflect interacting processes. Endogenous opioids, including beta-endorphin, are one proposed explanation. In an adolescent study, van der Venne et al. (2021) found lower resting plasma beta-endorphin alongside reduced pain sensitivity. Crucially, beta-endorphin levels were not significantly associated with pain threshold. The review also notes that concentrations measured in blood may not adequately represent activity in the brain (Kaess et al., 2021). The authors therefore regard explanations centred on opioid deficiency as premature. Lower resting opioid concentrations do not straightforwardly explain reduced pain sensitivity, and stress itself can alter opioid activity. Understanding these relationships requires repeated measurements across time and around episodes, rather than assuming that one biological difference explains the behaviour (Kaess et al., 2021).
Changes surrounding an episode
Research on biological states asks what changes when distress is followed by self-injury. Laboratory studies reviewed by Kaess et al. (2021) suggest that pain-related stimulation can coincide with reduced subjective arousal and changes in interactions between sensory and emotion-processing systems. Much of this evidence came from adults with borderline personality disorder, limiting its direct applicability to adolescents. Stress responses also varied with the experimental context. Koenig et al. (2017) found a greater cortisol response to pain in adolescent girls with NSSI, contrasting with the reduced responses reported during psychosocial stress. The review proposes that these differing responses might help explain how self-injury becomes maintained as a means of regulating distress. This remains a hypothesis requiring investigation during everyday experiences (Kaess et al., 2021).
Implications for clinical understanding and research
The authors identify stress and emotion regulation as treatment priorities supported by the evidence they reviewed. They also propose using biological findings in education for therapists, young people and caregivers to improve understanding of possible mechanisms. Their account recognises that people may engage in NSSI for different functions, and that brief experimentation and longstanding patterns may involve different biological profiles (Kaess et al., 2021). Further progress requires larger samples, replication and longitudinal studies that distinguish pre-existing vulnerability from changes accompanying self-injury. Measurements collected in everyday life could clarify how biological states develop around episodes. The authors also call for intervention research to test whether changes in proposed biological targets correspond to clinical improvement. Personalised treatment based on biological subgroups is presented as a future possibility, dependent on a stronger understanding of development, individual differences and treatment response (Kaess et al., 2021).
Conclusion
Neurobiology offers a framework for understanding NSSI within its psychological, social and developmental context. Kaess et al. (2021) distinguish early vulnerabilities, biological characteristics associated with ongoing self-injury, and transient changes surrounding individual episodes. Findings on emotion processing, stress responses and pain support a clinical focus on stress and emotion regulation, while uncertainty remains about causal pathways and opioid mechanisms. The authors emphasise that larger, longitudinal studies are needed before these insights can reliably guide biologically personalised treatment. Their framework can nevertheless inform psychoeducation by explaining possible mechanisms while acknowledging differences between individuals.
Where next?
Join us on 1 December 2026 for the most important conference on self-harm this year, ‘Understanding and Responding to Self-Harm: Latest Evidence and Clinical Practice. Judy Dunn International Online Conference 2026‘, with Professor Nav Kapur, Professor Paul Plener, Professor Rose McCabe, Professor Caroline Richards, Associate Professor Johan Bjureberg and Professor Joan Asarnow.
Bringing together leading international experts, this conference will explore current evidence and key developments in the field, from emerging insights into the neurobiology of self-harm to the role of emotion regulation, assessment and engagement in care.
References
- Brown, R. C., Plener, P. L., Groen, G., Neff, D., Bonenberger, M., & Abler, B. (2017). Differential neural processing of social exclusion and inclusion in adolescents with non-suicidal self-injury and young adults with borderline personality disorder. Frontiers in Psychiatry, 8, Article 267. https://doi.org/10.3389/fpsyt.2017.00267
- Ho, T. C., Walker, J. C., Teresi, G. I., Kulla, A., Kirshenbaum, J. S., Gifuni, A. J., Singh, M. K., & Gotlib, I. H. (2021). Default mode and salience network alterations in suicidal and non-suicidal self-injurious thoughts and behaviors in adolescents with depression. Translational Psychiatry, 11, Article 38. https://doi.org/10.1038/s41398-020-01103-x
- Kaess, M., Hooley, J. M., Klimes-Dougan, B., Koenig, J., Plener, P. L., Reichl, C., Robinson, K., Schmahl, C., Sicorello, M., Westlund Schreiner, M., & Cullen, K. R. (2021). Advancing a temporal framework for understanding the biology of nonsuicidal self-injury: An expert review. Neuroscience & Biobehavioral Reviews, 130, 228–239. https://doi.org/10.1016/j.neubiorev.2021.08.022
- Klimes-Dougan, B., Begnel, E., Almy, B., Thai, M., Westlund Schreiner, M., & Cullen, K. R. (2019). Hypothalamic-pituitary-adrenal axis dysregulation in depressed adolescents with non-suicidal self-injury. Psychoneuroendocrinology, 102, 216–224. https://doi.org/10.1016/j.psyneuen.2018.11.004
- Koenig, J., Rinnewitz, L., Warth, M., Hillecke, T. K., Brunner, R., Resch, F., & Kaess, M. (2017). Psychobiological response to pain in female adolescents with nonsuicidal self-injury. Journal of Psychiatry & Neuroscience, 42(3), 189–199. https://doi.org/10.1503/jpn.160074
- Koenig, J., Thayer, J. F., & Kaess, M. (2016). A meta-analysis on pain sensitivity in self-injury. Psychological Medicine, 46(8), 1597–1612. https://doi.org/10.1017/S0033291716000301
- Plener, P. L., Bubalo, N., Fladung, A. K., Ludolph, A. G., & Lulé, D. (2012). Prone to excitement: Adolescent females with non-suicidal self-injury (NSSI) show altered cortical pattern to emotional and NSS-related material. Psychiatry Research: Neuroimaging, 203(2–3), 146–152. https://doi.org/10.1016/j.pscychresns.2011.12.012
- van der Venne, P., Balint, A., Drews, E., Parzer, P., Resch, F., Koenig, J., & Kaess, M. (2021). Pain sensitivity and plasma beta-endorphin in adolescent non-suicidal self-injury. Journal of Affective Disorders, 278, 199–208. https://doi.org/10.1016/j.jad.2020.09.036
