INVESTIGATING EMOTION REGULATION DIFFICULTIES IN INDIVIDUALS WITH MENTAL HEALTH DISORDERS USING ADVANCED AUDIO ANALYSIS TECHNIQUES

Abstract: Emotion regulation is essential for psychological well-being, particularly in individuals with Major Depressive Disorder (MDD) and Post-Traumatic Stress s Disorder (PTSD), who often struggle with managing their emotions. Traditional methods like self-reports and behavioral observations have limitations in fully capturing the complexities of emotion regulation. This research work addresses these gaps by employing advanced audio analysis techniques to explore the difficulties in emotion regulation experienced by individuals with mental health disorders. The proposed model uses advanced audio analysis techniques on audio data from participant-interviewer interactions and self-reported questionnaires. Logistic regression classifiers are employed to predict emotion regulation difficulties and shows strong correlations between DERS subscales and severity of disorders. The model’s robustness is validated through Receiver Operating Characteristic (ROC) curve analysis and by calculating Area Under the Curve (AUC) values for each subscale of the Difficulties in Emotion Regulation Scale (DERS).