Developing CrisisBot: A Tool to Help Train Suicide Prevention Helpline Counselors
Crisis helplines fill a critical need by giving distressed individuals access to counselors experienced in de-escalation strategies during times of crisis. However, training counselors can be challenging for helplines that often operate with limited time and human resources. In addition to resource limitations, a significant challenge is that novice counselors must build experience in realistic environments without putting distressed individuals in danger. In an effort to provide a safe, no-risk environment for novice counselors to learn how to generate appropriate responses and practice counseling, we are building an interactive, automated tutoring system that novice counselors can use to learn and, more importantly, practice de-escalation strategies. This system includes a chat interface to practice counseling, real-time feedback, and real-time suggestions on improving counseling techniques. With such a system, new counselors could feel more confident starting out and crisis helplines could expand their services to help more people in need.
Orianna DeMasi is a postdoctoral scholar in Computer Science at the University of California, Davis working with Zhou Yu on applied dialogue systems. She completed her PhD in Computer Science at the University of California, Berkeley where she spent time as a Data Science Fellow at the Berkeley Institute of Data Science. Her research focuses on using computational tools to improve mental health, wellbeing, and care delivery.