Non-suicidal self-injury and digital ostracism: a novel smartphone-based paradigm capturing real-time responses to social media exclusion

Ostracism on social media is a common and emotionally salient stressor. However, its moment-to-moment psychological and behavioral consequences, remain poorly understood in individuals with Non-Suicidal Self-Injury (NSSI) and Borderline Personality Disorder (BPD). Using an Instagram-mimicking smartphone platform (Socializapp) coupled with Ecological Momentary Assessment (EMA), we examined responses to experimentally manipulated peer feedback across 7 days in three groups: healthy Controls (HC; n = 20), NSSI without BPD (NSSI; n = 18), and comorbid BPD+NSSI ( n = 17). Participants were exposed to Inclusion, Partial Exclusion, and Exclusion conditions. Linear and generalized linear mixed-effects models tested within-person effects on perceived exclusion and approval, affective reactivity, objective engagement (Likes and Comments) and NSSI urges, with borderline traits (BPQ) modeled dimensionally. Compliance was high, supporting feasibility. The manipulation was perceived as intended, with perceived exclusion increasing and perceived approval decreasing stepwise from Inclusion to Partial and Full Exclusion across participants. Exclusion also increased negative affect, reduced positive affect, and suppressed effortful interaction (fewer comments). In exploratory analyses, Exclusion was associated with more than a threefold increase in the odds of reporting an NSSI urge, after adjusting for borderline severity. While the data more clearly supported a baseline negative appraisal bias associated with borderline severity (even under objectively favorable feedback) evidence for cross-level moderation effects remained inconclusive. Social media ostracism emerges as a relevant interpersonal stressor for individuals with NSSI, producing detectable shifts in online engagement and, in exploratory analysis, increases in self-harm urges and changes. Borderline severity tracked a sustained negative interpretation of social feedback, although potential moderation effects regarding exclusion reactivity remained inconclusive. These findings establish Socializapp as a feasible paradigm and highlight everyday digital social experiences as meaningful targets for understanding short-term fluctuations in NSSI vulnerability.