Neural Correlates of Non-Suicidal Self-Injury Functions: Neurobiological Vulnerabilities and Psychosocial Risks in Psychiatric Disorders
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Non-suicidal self-injury (NSSI) is common in adolescents and young adults, especially among psychiatric disorders. It arises from neurobiological and psychosocial factors, yet its underlying mechanisms remain unclear.
Methods
A total of 304 inpatients (aged 16–25 years) with NSSI and psychiatric diagnoses (major depressive disorder [MDD], bipolar disorder [BD], and eating disorders [ED]) were assessed using the Ottawa Self-Injury Inventory. Resting-state functional magnetic resonance imaging were collected from 163 of them. Functional factors, representing motivations of NSSI, were identified using orthogonal projection non-negative matrix factorization. These factors were then linked to brain functional connections (FC) via canonical correlation analysis and used to classify patients into subtypes through fuzzy C-means clustering. Differences in FC and psychosocial characteristics were analyzed across subtypes.
Results
Two NSSI functional factors were identified: self-related and social-related. Self-related factor was associated with an amygdala-centered cortico-limbic network and prevalent in affective disorders (MDD and BD). Social-related factor correlated with frontoparietal and frontotemporal networks, prominent in ED. Clustering identified three NSSI functional subtypes. Subtype 1 was primarily driven by self-related functions. Subtype 2 was influenced by both self- and social-related functions with greater exposure to psychosocial risks. Subtype 3 remained undifferentiated. No subtype exclusively driven by social-related function was observed.
Conclusion
This study establishes the neural correlates of a “self-social” functional model of NSSI. Self-related functions are underpinned by emotional regulation networks, while social-related functions synergize with psychosocial risks via cognitive-emotional regulation pathways. These findings highlight the need for tailored psychological and neuromodulatory interventions targeting specific functional subtypes.
