The Dark Triad is a well-established trio of socially aversive personality traits: narcissism, characterized by grandiosity and entitlement; Machiavellianism, marked by manipulation and cynical worldviews; and psychopathy, defined by callousness, impulsivity, and shallow emotional responses. Decades of research have linked these traits to aggression, risk-taking, and interpersonal problems, but far less attention has been paid to whether they predict self-directed harm. The study’s authors note that the relationship between negative personality traits and non-suicidal self-injury has rarely been examined, a gap they set out to close using a multi-method approach that combines traditional regression modeling with modern network analysis.

The research team recruited 843 students from a secondary school in Chengdu, China, a sample large enough to support the fine-grained statistical techniques the study required. Personality was measured with the Dirty Dozen, a concise and widely validated twelve-item questionnaire that yields separate scores for narcissism, Machiavellianism, and psychopathy; the Chinese version of the instrument has been validated in prior research. Self-injury was assessed with the Chinese Adolescent Non-Suicidal Self-Injury Assessment Questionnaire, an instrument developed to capture the full topography of self-harm among adolescents, including behaviors such as cutting, scratching, bruising, stabbing or pricking the skin, and hitting oneself. The regression analyses were conducted in SPSS 22.0, while the network analyses were performed in R 4.4.2, and the study received ethics approval from the Psychological Research and Counseling Center at Southwest Jiaotong University.

The regression results delivered an unambiguous verdict on which of the three dark traits matters most. When narcissism, Machiavellianism, and psychopathy were entered together as predictors of self-injury, psychopathy emerged as the most dominant predictor, outperforming its two siblings. This finding aligns with a broader literature on psychopathy’s distinctive features: individuals high in psychopathic traits tend to show reduced pain perception and tolerance, heightened impulsivity, and diminished emotional responsiveness, all of which could plausibly lower the psychological barriers to harming one’s own body. Previous work has also connected psychopathic traits to suicidal behavior through emotional and behavioral dysregulation, making the new finding a natural extension of that research into non-suicidal forms of self-harm.

But the study’s most novel contribution comes from its network analysis, a technique that has transformed how psychologists think about mental health. Instead of assuming that a disorder or trait is a single underlying entity that causes its symptoms, network analysis treats symptoms and behaviors as nodes in a graph, with statistical associations drawn as edges between them. In this framework, the structure of the connections themselves carries the scientific information: some nodes sit at the center of dense clusters, some act as hubs that activate many neighbors, and some serve as bridges that connect otherwise separate communities of nodes. The approach, pioneered by researchers such as Denny Borsboom and Eiko Fried, has been applied extensively to depression and anxiety, and this study extends it to the intersection of personality and self-harm.

Within the network of specific self-injury behaviors, three nodes showed the highest strength centrality, meaning they had the strongest direct connections to the other behaviors: stabbing or pricking, scratching, and bruising. In practical terms, adolescents who engage in these behaviors are most strongly connected to the wider web of self-harm, suggesting these forms may act as core symptoms around which other behaviors cluster. Even more striking was the betweenness result. Stabbing or pricking showed the highest betweenness centrality, meaning it sits on the shortest paths between many other nodes in the network, functioning as a gateway through which activation can spread from one behavior to another. If the network perspective is right that problems propagate along these edges, then a behavior with high betweenness is a particularly consequential target for intervention, because disrupting it could sever the pathways along which self-harm spreads.

The bridge analysis took this logic one step further by asking which node connects the Dark Triad community to the self-injury community. Using bridge centrality methods developed by Payton Jones and colleagues, the researchers identified psychopathy as the strongest bridge node linking the two clusters. This means that within the combined network, psychopathy is the trait most directly and densely connected to the self-injury behaviors, serving as the primary channel through which dark personality characteristics relate to self-harm. Narcissism and Machiavellianism, by contrast, played far weaker bridging roles. The convergence of the regression and network findings is notable: two statistically independent methods, one asking which trait predicts self-injury variance and the other asking which trait anchors the connection between communities, both singled out psychopathy.

The authors argue that this convergence provides valuable insight into the mechanisms underlying adolescent self-injury and offers concrete directions for targeted intervention. If psychopathy, with its signature combination of callous affect and impulsive action, is the trait that most readily translates dark personality into bodily harm, then prevention programs might achieve more by addressing the cognitive-affective dysfunctions associated with psychopathic traits, such as emotion regulation deficits and altered pain processing, than by treating the Dark Triad as an undifferentiated whole. Similarly, the centrality of stabbing or pricking, scratching, and bruising suggests that screening and safety planning in school settings could prioritize adolescents who report these specific behaviors, since they occupy the most connected positions in the self-harm network and may be at greatest risk of engaging in additional forms of injury.

The study also carries implications for how researchers conceptualize the link between personality and psychopathology. Traditional approaches might ask whether dark traits predict self-injury in aggregate, but the network approach reveals that the answer depends on which trait and which behavior you examine. The finding that a single trait can serve as the bridge between two symptom communities echoes broader debates in the field about comorbidity and about whether psychological problems spread through direct activation between symptoms rather than through a shared latent cause. For adolescents, whose personalities are still developing and whose self-harm behaviors often first emerge in the teenage years, identifying the specific pathways of connection could shape earlier and more precise identification of those at risk.

Caveats remain. The study is cross-sectional, drawn from a single secondary school in one Chinese city, so it cannot establish that psychopathy causes self-injury, only that the two are systematically connected in this sample’s network structure. Self-report questionnaires, however well validated, depend on honest and accurate responding, particularly for sensitive behaviors like self-harm. Still, the dataset has been made publicly available on the Open Science Framework, allowing other researchers to verify and extend the findings, and the sample of 843 students provides a solid foundation for the network estimation techniques, which are known to be sensitive to sample size. As non-suicidal self-injury continues to rise as a global public health challenge, and as meta-analyses document its prevalence and its role as a precursor to suicidal thoughts and behaviors, studies like this one move the field closer to answering a deceptively simple question: not just who is at risk, but which specific psychological characteristics connect to which specific harmful behaviors, and where a well-aimed intervention might break the chain.

Subject of Research: The relationship between Dark Triad personality traits, especially psychopathy, and non-suicidal self-injury in adolescents

Article Title: Psychopathy as a key predictor and bridge: a network and regression analysis of the Dark Triad and non-suicidal self-injury

Article References: Hou, M., Xu, X., Zheng, L., & Tang, S. (2026). Psychopathy as a key predictor and bridge: a network and regression analysis of the Dark Triad and non-suicidal self-injury. Current Psychology, 45(19), Article 1567. https://doi.org/10.1007/s12144-026-10125-5

Image Credits: AI Generated

DOI: 10.1007/s12144-026-10125-5

Keywords: psychopathy, Dark Triad, non-suicidal self-injury, network analysis, adolescents, narcissism, Machiavellianism, bridge centrality, regression analysis, personality traits, self-harm, Current Psychology