Name: Sarah Kim
School of Psychological and Brain Sciences – Ph.D. Dissertation Proposal Meeting
Date: Thursday, October 29th, 2026
Time: 2:00 - 4:00PM
Location: https://gatech.zoom.us/j/3889029032
Dissertation Committee Chair/Advisor:
Hsiao-Wen Liao, Ph.D. (Georgia Tech)
Dissertation Committee Members:
Mengyao Li, Ph.D. (Georgia Tech)
Paul Verhaeghen, Ph.D. (Georgia Tech)
Tammy Thanh Tran, Ph.D. (Georgia Tech)
Sarah Barber, Ph.D. (Georgia State University)
Title: Differences in the Content and Function of Emotional Autobiographical Memory Sharing With Human and Robot Listeners
Abstract: Autobiographical memories are often shared within a social context. Studies have shown that memory sharing can be influenced by who the listener is. This research extends prior work by exploring potential differences in the content and socioemotional functions of emotional autobiographical memory sharing with robot and human listeners. A mixed-methods study was conducted using a 2 (memory sharing condition: human vs. robot) × 3 (memory valence: neutral, positive, negative) within-subjects design. Undergraduate participants (N = 58; 51.7% female; age range: 18–26 years) recalled and orally shared one neutral, one positive, and one negative memory each with a human listener and with a NAO robot in counterbalanced order. Within each listener condition, the neutral memory was shared first, followed by positive and negative memories in counterbalanced order. The NAO robot was operated through a Wizard-of-Oz procedure in which participants were led to believe the robot responded autonomously. After sharing each memory, participants reported socioemotional experiences (i.e., felt closeness, the listener’s trustworthiness and responsiveness, and interaction quality), memory qualities, and their thoughts about their experience with the human and robot listeners by filling out questionnaires. Content analysis of memory narratives (n = 348) will be performed to capture memory length, event type, internal and external details, use of emotion words, and affective meaning making. Prior to memory sharing, baseline individual difference factors were assessed (e.g., mood, prior experience with AI tools and robots, general trust in people and automated technology, attitudes toward robots, mental health issues, and personality). Mixed-effects models will be conducted to investigate the effects of listener type and memory valence and the interaction effect of these two factors on socioemotional experiences, perceived interaction quality, and self-rated and coded episodic and affective qualities of the memory shared. Differences in event type and memory length will also be examined. Based on the interpersonal process model and existing research, the human listener was expected to be rated higher on closeness, trustworthiness, and responsiveness than the robot listener. Given limited research and mixed findings in available studies, differences in the interaction quality and memory content and qualities between the two listener conditions are exploratory, as are the interaction effects of listener type and memory valence. Thematic analysis will be conducted to analyze open-ended responses to understand listener preferences, reasons behind preferences and memory selection, and perceived purposes of memory sharing. The potential influence of prior experience with and attitudes toward relevant technology will be explored. Together, this research will enhance a better understanding of the common and respective socioemotional functions of emotional memory sharing with human and embodied robot listeners, advancing the literatures of functional use of autobiographical memory and human-robot interactions.