EUVIMEDEUVIMED entdecken
Crossref

LightGBM Prediction of Digital Self-Harm from Self-Criticism, Emotion Dysregulation, and Online Disinhibition

Deborah Cohen, Rachel Higgins, Ariel Paredes de Jesús, Javier Galvez-Martos

Journal of Assessment and Research in Applied Counseling · 2026 · 8 (3) · 1-15

DOI: 10.61838/kman.jarac.5824 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Assessment and Research in Applied Counseling, 2026. Verfasst von Deborah Cohen und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

This study aimed to predict digital self-harm severity among young adults in Chile using a LightGBM model based on self-criticism, emotion dysregulation, and online disinhibition. Methods and Materials: The study used a cross-sectional, correlational, and predictive design. The statistical population consisted of young adults living in Chile who were active users of digital communication platforms. The final sample included 512 participants aged 18 to 29 years. Data were collected using a Digital Self-Harm Behavior Checklist, the Forms of Self-Criticizing/Attacking and Self-Reassuring Scale, the Difficulties in Emotion Regulation Scale, the Online Disinhibition Scale, and a demographic and digital behavior questionnaire. Data analysis included descriptive statistics, reliability assessment, Pearson correlation coefficients, and predictive modeling using Light Gradient Boosting Machine. The dataset was divided into training and test sets, and model performance was evaluated using cross-validation, root mean square error, mean absolute error, and coefficient of determination. SHAP values were used to interpret the relative contribution of predictors.

Findings

Digital self-harm was positively correlated with inadequate self (r = 0.43, p < 0.01), hated self (r = 0.57, p < 0.01), emotion dysregulation total score (r = 0.52, p < 0.01), benign online disinhibition (r = 0.21, p < 0.01), and toxic online disinhibition (r = 0.61, p < 0.01), and negatively correlated with reassured self (r = -0.32, p < 0.01). The LightGBM model showed the strongest predictive performance, with test RMSE = 3.96, MAE = 2.98, and R² = 0.59. SHAP analysis identified toxic online disinhibition, hated self, limited access to emotion regulation strategies, impulse control difficulties, and nonacceptance of emotional responses as the most important predictors of digital self-harm severity.

Conclusion

The findings indicate that digital self-harm is best understood as a multidimensional phenomenon shaped by hostile self-criticism, impaired emotion regulation, and toxic online disinhibition.

Zitieren

APA

Cohen, D., Higgins, R., Jesús, A. P. D., & Galvez-Martos, J. (2026). LightGBM Prediction of Digital Self-Harm from Self-Criticism, Emotion Dysregulation, and Online Disinhibition. Journal of Assessment and Research in Applied Counseling, 8(3), 1-15. https://doi.org/10.61838/kman.jarac.5824

Vancouver

Cohen D, Higgins R, Jesús APD, Galvez-Martos J. LightGBM Prediction of Digital Self-Harm from Self-Criticism, Emotion Dysregulation, and Online Disinhibition. Journal of Assessment and Research in Applied Counseling. 2026;8(3):1-15. doi:10.61838/kman.jarac.5824

BibTeX
@article{Cohen2026,
  author = {Cohen, Deborah and Higgins, Rachel and Jesús, Ariel Paredes de and Galvez-Martos, Javier},
  title = {LightGBM Prediction of Digital Self-Harm from Self-Criticism, Emotion Dysregulation, and Online Disinhibition},
  journal = {Journal of Assessment and Research in Applied Counseling},
  year = {2026},
  volume = {8},
  number = {3},
  pages = {1--15},
  doi = {10.61838/kman.jarac.5824}
}
Kostenlos starten →