TAMIDS Tech Talk: Tim Althoff: Human-AI Collaboration Enables More Empathic Conversations in Mental Health Support

Biography: Tim Althoff is an assistant professor in the Paul G. Allen School of Computer Science & Engineering at the University of Washington, directing the Behavioral Data Science Group. His research advances computational methods that leverage large-scale behavioral data to extract actionable insights about our lives, health and happiness through combining techniques from data science, social network analysis, and natural language processing. Tim holds a Ph.D. degree from Stanford University. He has received several fellowships and awards including the WWW 2021 Best Paper Award, two ICWSM 2021 Best Paper Awards, the SIGKDD Dissertation Award 2019, a Best Paper Award by the International Medical Informatics Association, and an NSF CAREER Award. Tim’s research has been covered internationally by news outlets including BBC, CNN, The Economist, The Wall Street Journal, and The New York Times.

Abstract: Computing technology has connected people across the world in unprecedented ways. Going forward, how might computing help us to connect with each other in meaningful ways? In this talk, I will present our recent work on computational approaches to model, understand and facilitate more empathic conversations online. We apply our models to analyze hundreds of thousands of mental health interactions and show that while online mental health support users appreciate empathy, they do not improve, or self-learn empathy over time, revealing opportunities for empathy training and feedback. In order to facilitate higher expression of empathy in online mental health support, we develop a language generation task and reinforcement learning model for “empathic rewriting,” in which a computational model provides suggestions to transform low-empathy conversational posts to higher empathy. Learning such transformations is challenging and requires a deep understanding of empathy while maintaining conversation quality through text fluency and specificity to the conversational context. Building on this work, we examine how AI can collaborate with humans to facilitate peer empathy during textual, online supportive conversations. We develop an AI-in-the-loop agent that provides just-in-time feedback to help peer supporters respond more empathically to those seeking help. Through a randomized controlled trial with real-world peer supporters on Talklife, a large online peer-to-peer support platform we demonstrate that our Human-AI collaboration approach leads to a significant increase in conversational empathy between peers. We systematically analyze the Human-AI collaboration patterns and find that peer supporters are able to use the AI feedback both directly and indirectly without becoming overly reliant on AI, while reporting improved self-efficacy post-feedback. Our findings demonstrate the potential of feedback-driven, AI-in-the-loop writing systems to empower humans in open-ended, social, and high-stakes tasks such as empathic conversations.

You can also click this link to join the seminar

This Tech Talk is organized by TAMIDS Data Science Ambassador Scholarship Program. If you have any questions, please contact Diego Rodriguez, Assistant Director for Educational Programs, TAMU Institute of Data Science.