Emotion dysregulation as a marker in adolescent mental health with EEG-based prediction model
Emotion dysregulation as a marker in adolescent mental health with EEG-based prediction model

Emotion dysregulation as a marker in adolescent mental health with EEG-based prediction model

Sci Rep. 2025 Oct 31;15(1):38178. doi: 10.1038/s41598-025-22067-2.

ABSTRACT

This study comprehensively tackles the critical challenge of understanding and mitigating adolescent violent crime by integrating advanced insights from psychological and environmental research with cutting-edge digital public health tools. Current methods for examining adolescent aggression often fail to provide a holistic framework that effectively accounts for the intricate interplay of emotional dysregulation, environmental influences, and relational dynamics, thereby limiting the scope and efficacy of intervention strategies. In response to these limitations, we propose a comprehensive approach that leverages EEG-based emotion analysis in combination with a novel Psycho-Social Risk Interaction Model (PRIM), designed to uncover latent variables and dynamic interactions underlying violent behavior in adolescents. PRIM is a robust framework that encapsulates psychological vulnerabilities such as impulsivity and aggression, environmental stressors like socioeconomic pressures, and relational influences within peer and family networks, offering a nuanced understanding of the multifaceted factors contributing to violent tendencies. Building upon the PRIM framework, we introduce the Targeted Intervention and Risk Reduction Strategy (TIRRS), an innovative system that translates theoretical insights into actionable, personalized, and adaptive interventions. TIRRS dynamically modulates the interaction of psychological, environmental, and relational factors by employing real-time monitoring tools and resource optimization frameworks, ensuring that interventions are both responsive and impactful. Experimental results demonstrate that our approach improves the prediction accuracy of violent tendencies to 87.5%, representing a 21.3% increase compared to traditional statistical models (which averaged 66.2% accuracy). Moreover, the intervention success rate improved by 18.7% relative to standard counseling-based approaches. These outcomes enable the development of cost-effective, scalable, and sustainable prevention strategies.

PMID:41173991 | DOI:10.1038/s41598-025-22067-2