LAUSR

Research has demonstrated a relationship between anger and suicidality, while real-time authentic emotions behind facial expressions could be detected during advising hypothetical protagonists in life dilemmas. This study aimed to investigate the predictive validity of anger expressions during advising for suicide risk. Besides advising on life dilemmas (a friend's betrayal, a friend's suicide attempt), 130 adults completed the suicidal scale of the Mini-International Neuropsychiatric Interview. Participants' anger during advice-giving was measured 29 times/s by artificial intelligence (AI)-based software FaceReader 7.1. The results showed that anger was a significant predictor of suicide risk. Increased anger during advising was associated with higher suicide risk. In contrast, there was no significant correlation between suicide risk and duration or length of advising. Therefore, measuring micro expressions of anger with AI-based software may help detect suicide risk among clinical patients in both traditional and online counseling contexts and help prevent suicide.