Recent psychopathology classification systems such as the Diagnostic and Statistical Manual, Fifth Edition (DSM-5) have recognized the importance of non-suicidal self-injury (NSSI) and attempted to consider this behavior as a separate state independent of other mental disorders (Glenn and Klonsky, 2013). According to the criteria, NSSI referred a condition that “the individual has, on five or more days, engaged in intentional self-inflicted damage to the surface of own body, with the expectation that the injury will lead to only minor or moderate physical harm (i.e., there is no suicidal intent), in the last year” (Association, 2013). NSSI is not done with the intention or purpose of completing suicide but is reported as a predictor of subsequent suicide attempts (Horwitz et al., 2015).

NSSI is most common in adolescence or young adulthood. The onset of behavior is around 12–14 years of age (Nock et al., 2006). The prevalence of NSSI is approximately 7.5–46.5% in the adolescent population, 38.9% in the college student population, and 4–23% in the adult population (Cipriano et al., 2017). According to a longitudinal study of 1,800 adolescents, NSSI decreases with adulthood (Moran et al., 2012), and another systematic literature review suggested that the behavior of NSSI peaked in adolescents aged 15–17 and then gradually decreased (Plener et al., 2015). Regarding gender, females showed higher values than males. This difference was more pronounced in the clinical than in the community group (Bresin and Schoenleber, 2015).

Individuals who have experienced self-harm are diagnostically heterogeneous and experience a wide range of psychological disorders (Nock et al., 2006). Abundant evidence indicates that the group engaging in self-harm exhibits more symptoms of borderline personality disorder (BPD) than the group that does not (Andover et al., 2005). Depression and anxiety disorders were also associated with self-harm, which, like BPD, is thought to be due to negative emotionality and emotional dysregulation (Seeley et al., 2015). Although diagnostically heterogeneous, self-harm generally exhibits prominent features of negative emotions and self-derogation (Klonsky and Muehlenkamp, 2007), and self-harm behavior is used as a dysfunctional coping strategy for emotional regulation in NSSI patients (Plener et al., 2018). In addition, abnormal stress processing and elevation of the pain threshold have been observed in NSSI patients (Osuch et al., 2014; Schmahl and Baumgärtner, 2015). Recently, various pieces of evidence of neurobiological abnormalities related to NSSI have been discovered, and changes in brain circuits or abnormalities in physiological factors (hypothalamus-pituitary-adrenal axis, endogenous opioid hypothesis, and others) have been reported (Groschwitz and Plener, 2012; Liu, 2017; Plener, 2019). Understanding the underlying neurobiological mechanisms of NSSI provides a better explanation of how emotion regulation, pain processing, and response to stress work. Brain imaging research is an area of interest in the effort to find the clinical association between self-harmful behavior, emotional dysregulation, and neurobiological abnormalities. In functional MRI studies, when processing negative emotional stimuli, BPD patients showed high activity in the left amygdala and posterior cingulate cortex (PCC), whereas the bilateral dorsolateral prefrontal cortex (DLPFC) showed a blunt reaction (Schulze et al., 2016). Similarly, in NSSI adolescents, hyperarousal patterns in the amygdala, hippocampus, and anterior cingulate cortex (ACC) have been reported (Plener et al., 2012). Another fMRI study in BPD patients observed hyperarousal in the amygdala, insula, and ACC during neutral and negative stimuli (Niedtfeld et al., 2010). In addition, compared with healthy controls, atypical amygdala-frontal connectivity and functional connectivity of the right orbitofrontal cortex-ACC were weakened in patients with BPD and NSSI (Osuch et al., 2014; Schreiner et al., 2017a). In the study of structural MRI, changes in brain volume in the insula and ACC were observed in female NSSI adolescents, and changes in ACC were associated with a history of suicide attempts (Ando et al., 2018). Another study confirmed a decrease in gray matter in the bilateral insular cortex and right inferior frontal gyrus in female NSSI adolescents. This decrease was correlated with emotional dysregulation on the self-report scale, and some similarities with those of adult BPD patients were confirmed for these brain volume changes (Beauchaine et al., 2019b). Studies on the neuroanatomical and functional correlations with NSSI confirm that emotional dysregulation has a neurological association (Beauchaine et al., 2019a). Previous studies were conducted in the existing psychiatric patient group because NSSI was listed among symptoms in earlier diagnostic systems (DSM-IV-TR or International Classification of Diseases, Tenth Revision). Despite the increasing prevalence of the syndrome (Swannell et al., 2014), NSSI has been overlooked and dismissed as a temporary phenomenon expressing rebelliousness in adolescence or classified as a pre-suicidal stage or a component of personality disorder. Although the recently revised DSM suggested NSSI as an independent disease entity, studies on NSSI, not as part of symptoms, are insufficient compared to their prevalence, severity, and social cost (Domínguez-Baleón et al., 2018). Therefore, we attempted to evaluate neurobiological changes in NSSI patients through brain imaging studies. We investigated subcortical structure volumes (i.e., the hippocampus, amygdala, caudate, putamen, and thalamus). We hypothesized that structural differences would exist between NSSI patients and healthy controls and that these changes would be associated with the pathology of NSSI.