Five hundred fifty-seven adolescents and young adults with a history of non-suicidal self-injury fell into two distinct groups when researchers analysed their depression, anxiety and self-efficacy scores. One cluster showed high distress and low confidence in their ability to cope. The other reported lower distress and higher self-efficacy. The groups differed sharply in how often they self-harmed, how they managed emotions, and how frequently intrusive mental images interrupted their thoughts.
The finding, published in Frontiers in Child and Adolescent Psychiatry in 2026, challenges the assumption that all young people who self-harm are experiencing the same psychological struggle. Elisa Schmied-Weggler and colleagues at the University of Ulm in Germany used a person-centered statistical approach to look for naturally occurring subgroups rather than treating self-injury as a single uniform behaviour.
What the clusters looked like
The High-Risk cluster — characterised by elevated depression and anxiety scores alongside low self-efficacy — reported self-harming significantly more often over their lifetime than the Low-Risk group. They also scored higher on measures of expressive suppression, meaning they were more likely to bottle up emotional reactions rather than express them. At the same time, they used cognitive reappraisal less often, a strategy that involves reinterpreting a situation to change its emotional impact.
Intrusive prospective imagery — vivid, unwanted mental pictures of future self-harm — was markedly higher in the High-Risk cluster. These images are thought to form part of the pathway from emotional distress to the act of self-injury itself, according to the Integrated Motivational-Volitional model the researchers were testing.
The Low-Risk cluster, despite also having a history of self-injury, looked different on nearly every measure. Lower depression and anxiety, higher belief in their own coping ability, less suppression, more reappraisal, and fewer intrusive images of future harm.
Age of onset mattered independently
Young people who started self-harming before age twelve reported higher lifetime frequency of self-injury and more intrusive imagery, regardless of which psychological cluster they belonged to. The effect was additive: early onset and High-Risk profile together predicted the most severe pattern, but early onset alone still raised risk even in the Low-Risk group.
The study found no interaction between cluster type and age of onset, meaning the two factors operated independently rather than amplifying each other. A person in the Low-Risk cluster who started self-harming early did not suddenly resemble someone in the High-Risk group. The risks stacked rather than multiplied.
The researchers tested this using bootstrapped general linear models with 1,000 resamples, a technique that provides more stable estimates when data distributions are not perfectly normal.
Why emotion regulation differed between groups
Suppression and reappraisal are two common strategies for managing emotions, measured in this study using the Emotion Regulation Questionnaire. Suppression involves inhibiting the outward signs of emotion — keeping a neutral face when you’re upset, not letting your voice shake. Reappraisal involves changing how you think about the situation before the emotion fully forms.
The High-Risk cluster relied more on suppression and less on reappraisal. That pattern aligns with what the Integrated Motivational-Volitional model predicts: when people feel unable to manage distress and lack confidence in their coping skills, they’re more likely to suppress feelings rather than rethink them. Suppression tends to maintain or even increase emotional intensity over time, while reappraisal often reduces it.
The Low-Risk group’s higher use of reappraisal may partly explain their lower self-injury frequency, though the study design — a cross-sectional online survey — cannot prove that reappraisal caused the reduction. The association could run the other way, or both could stem from a third factor not measured here.
What intrusive imagery means
Intrusive prospective mental imagery refers to vivid, uninvited mental pictures of self-harm that haven’t happened yet. The images come unbidden and are experienced as more concrete and compelling than ordinary thoughts. The research team measured this using the Intrusive Future Experiences Scale.
In the Integrated Motivational-Volitional framework, these images are considered part of the volitional phase — the transition from thinking about self-harm to actually doing it. The High-Risk cluster’s higher imagery scores suggest they experience this bridging mechanism more intensely, which may help explain their higher frequency of actual self-injury.
Whether reducing intrusive imagery through targeted therapy would reduce self-harm behaviour is a question this study cannot answer, but the authors note that Imagery Rescripting — a technique that involves changing the content or outcome of distressing mental images — might be worth testing in this population.
What this study can’t tell you
The sample was recruited online and self-selected, meaning it likely overrepresents people comfortable discussing self-harm in a survey format and underrepresents those without internet access or those too unwell to participate. All 557 participants had a history of non-suicidal self-injury, so the study says nothing about young people who never self-harm or about how common these two profiles are in the general adolescent population.
The design was cross-sectional — all measures taken at one point in time — so it cannot establish whether low self-efficacy leads to more self-injury, whether self-injury erodes self-efficacy, or whether both are consequences of something else. The clusters describe how variables grouped together in this sample in 2026, not fixed categories that individuals permanently belong to.
The study did not include information about trauma history, family functioning, social support or other factors known to influence self-injury risk. The psychological profiles are based solely on depression, anxiety and self-efficacy scores.
Where this points for intervention
The authors write that the findings support tailoring interventions to psychological profile. For young people in the High-Risk cluster, treatment might prioritise building self-efficacy and teaching cognitive reappraisal alongside addressing depression and anxiety. For those who started self-harming before age twelve, regardless of current profile, early and sustained support may be particularly important given the higher lifetime frequency observed in that group.
The distinct profiles suggest that a one-size-fits-all approach to self-injury prevention may miss important variation in needs. A teenager in the Low-Risk cluster who self-harms occasionally under acute stress likely needs different support than someone in the High-Risk cluster experiencing daily intrusive imagery and chronic emotional suppression.
The study does not provide a method for clinicians to assign individuals to clusters outside the research context, and attempting to categorise a young person based on three questionnaire scores would be a misuse of the findings. The value is in recognising that self-injury in adolescence is not psychologically uniform, and in pointing towards mechanisms — self-efficacy, emotion regulation, intrusive imagery — that might be worth targeting in treatment.
If you’re a parent or professional working with a young person who self-harms, this research does not tell you what to do. What it does suggest is that the internal experience varies considerably, and that asking about how someone manages emotions and how confident they feel in handling distress may matter as much as asking how often they self-harm.
If you need support
If you are in immediate danger, contact your local emergency number.
- India — Tele-MANAS: 14416 (free, 24/7)
- India — AASRA: +91 98204 66726 (24/7)
- United States — 988 Suicide & Crisis Lifeline: call or text 988
- United Kingdom — Samaritans: 116 123 (free, 24/7)
- International — findahelpline.com
Sources
- Schmied-Weggler, E., Sosic-Vasic, Z., Connemann, B., & Kroener, J. (2026). Psychological risk profiles are associated with severity and motivational-volitional mechanisms of non-suicidal self-injury in adolescents and young adults. Frontiers in Child and Adolescent Psychiatry. https://doi.org/10.3389/frcha.2026.1878269
Published for general education. This article is not medical advice, diagnosis or treatment. If you are struggling, speak to a qualified professional.
