Candidate: Keith Rebello
Date: July 13, 2026
Time: 10:00 AM
Location: In-person
Supervisor: Kerstin Dautenhahn
Co-Supervisor: Chrystopher L. Nehaniv
All are welcome!
Abstract:
Adolescents are particularly susceptible to self-conscious emotions, such as embarrassment, shyness, and social anxiety, which can negatively impact their development if not appropriately supported. Therapeutic engagement with adolescents can be challenging, often marked by a higher likelihood of “ruptures” in the therapeutic alliance, underscoring the critical need for empathy in fostering effective interventions. This thesis proposal outlines research into the use of empathic social robots in roleplay and exposure therapy settings to help adolescents manage mild social anxiety (embarrassment), with empathy posited as a medium for improving engagement and therapeutic outcomes. The research addresses three core empirical questions. First, can robots explicitly simulate empathy? This will be explored by developing and modeling robotic empathic expressions (nonverbal and verbal) and their contextual usage. Second, do adolescents want to interact with empathic robots? Adolescent co-design methodologies, including interviews, storyboarding, and prototyping with a Furhat robot as an empathic peer mentor, will be employed. This includes exploring adolescent perspectives on role-playing with single or multiple Furhats in exposure therapy scenarios. Third, can robots help adolescents specifically with embarrassment? This will be investigated by designing interaction scenarios where adolescents roleplay embarrassing situations with multiple Furhat robots, drawing from resources like The READY toolkit. The usability of the empathic Furhat in reducing adolescents’ fears of embarrassment will be studied, considering key usability factors.
Ultimately, this research aims to demonstrate the potential of empathic robots as engaging tools for managing adolescent social anxiety, focusing on non-clinical forms of anxiety, within a roleplay and exposure therapy framework.