Date of Award

Spring 6-2024

Document Type

Thesis (Master's)

Department or Program

Computer Science

First Advisor

Lorie Loeb

Second Advisor

SouYoung Jin

Third Advisor

Elizabeth Murnane

Abstract

Embodied conversational agents (ECAs) have significantly enhanced human-machine interactions and show considerable potential in various industries such as customer service, education, healthcare, entertainment, and finance [1, 2]. This study explores the impact of similarities in gender and physical appearance between ECAs and users on the perceptions of trustworthiness, empathy, and service evaluation within the context of counselor ECAs. We conducted a within-subject experiment (n=50), using a 2x2 factorial arrangement, that varied the gender and the physical appearance of four distinct AI avatars. Participants interacted with each avatar, completing a post-experiment survey and participating in semi-structured interviews. Our findings indicate that the perceptions of trustworthiness, empathy, and service evaluation are closely correlated with the similarities in gender and physical appearance between ECAs and users. AI Avatars matching the participants’ gender and appearance were rated significantly higher in trustworthiness and service evaluation. Conversely, AI avatars depicting a different gender and physical appearance from users received higher empathy scores than their counterparts. This study provides insights into developing and deploying ECAs in sectors requiring high levels of personal interaction such as in language learning contexts. Additionally, we indicated that female ECAs received approximately five times higher scores in empathy than male ECAs, regardless of similarities in gender and physical appearance to users. Furthermore, these findings suggest how to design ECAs to be more engaging and acceptable to users; therefore, fostering inclusive technologies that accommodate a wider range of user demographics, enhance user experience, and improve satisfaction in technological interactions.

Download

Share

COinS