Gendering the Machine: Preferred Virtual Assistant Gender and Realism in Self-Service

A virtual agent is a human-like character that is designed to assist users in interactions with technology and virtual worlds. Research into the preferred visual characteristics of a virtual agent has focused on education-based agents, gaming avatars, and online help assistants. However, findings from these studies are not necessarily generalizable to other technologies, such as self-service checkouts (SSCO). This paper describes data from 578 participants, looking at the gender preferences of Virtual Assistants (VA) in a SSCO context and the impact of VA realism depending on user gender. Due to female participants’ preference for female VAs, and an overall preference for threedimensional characters, a realistic, female VA should be used in SSCO. The results are discussed in terms of similarity-attraction theory and social role theory.