ICAT Playdate: 10/13 Human-AI Interaction presented by CHCI

October 9, 2023

Eugenia Rho • Virginia Tech Computer Science • Society + AI & Language Lab

Eugenia Rho will talk about her collaborative research with Angela Scarpa-Friedman on empathy and human-AI interaction at the ICAT Playdate on Friday, October 13 (8:30-9:30 AM). This work is funded by a planning grant from the Center for Human-Computer Interaction (CHCI).

Title: Supporting Mental Health in Neurodiverse Communities: Enhancing LLM Accessibility through Multimodal Human-AI Interactions

Location - LLC Building, Community Assembly, 185 Kent St (near VT bookstore)

Time - 8:30-9:30 AM (donuts and coffee at 8:30; talk starts at 9 AM)

Talk Abstract

Recent advancements in Large Language Models (LLMs), exemplified by the likes of Generative Pre-trained Transformers (GPT)-4, offer tremendous potential in transforming how we can address the mental health and communication needs of neurodiverse individuals, especially those on the autism spectrum. Autistic individuals often face significant challenges in communication and social interactions, which can heighten their risk of depression and anxiety. A notable concern is the prevalence of negative self-talk (NST), which can intensify existing mental health challenges in this community. Compared to the general population, traditional therapy may not be as effective for those with limited verbal communication skills. Hence there is a need for innovative technologies tailored to offer support to this specific community of users. Recently, with the advent of LLMs, there has been an increase in the development of mental health chatbots. However, developing AI technologies powered by LLMs, especially for neurodiverse users, are fraught with open challenges and risks. This short talk will introduce the motivation behind an on-going project that examines such challenges and aims to understand how LLM powered tools can be empathetically designed to help autistic individuals cope with negative self-talk.

About the Researchers

Eugenia Rho

Eugenia Rho is an Assistant Professor of Computer Science at Virginia Tech, where she leads the SAIL (Society + AI & Language) lab. Prior to joining Virginia Tech, she was a postdoctoral scholar at the NLP group in the Department of Computer Science at Stanford University. She earned her PhD in Informatics from the University of California, Irvine.

Her research focuses on exploring the intersection of Natural Language Processing (NLP) and Human-Computer Interaction (HCI), and she is particularly interested in how AI-mediated systems impact interactions across people and machines.

Angela Scarpa-Friedman

Angela Scarpa received her Ph.D. in Clinical Psychology from the University of Southern California in 1993. She is currently a Professor of Psychology at Virginia Tech, Founder and Director of the Virginia Tech Autism Clinic & Center for Autism Research. She is a licensed clinical psychologist with specialized training in individual and group behavioral and cognitive-behavioral therapy for children, families, and adults. Before coming to Virginia Tech, she held faculty positions at the University of Georgia and Eastern Washington University and obtained postdoctoral training at Western Psychiatric Institute and Clinic of the University of Pittsburgh Medical Center.

Her general interest is in child and adolescent mental health, with over 30 years of experience in the study of developmental psychopathology. Currently her work is focused on children, adolescents, and young adults on the autism spectrum.