Selective Fidelity: Investigating Priorities for the Creation of Expressive Avatars

Abstract

(introduction) Recent works of cyberfiction have depicted a not-so-distant future where the Internet has developed into a fully three-dimensional and immersive datascape simultaneously accessible by millions of networked users. This virtual world is described as having spatial properties similar to the physical world and its virtual cities are populated by digital proxies of people, called avatars. The multisensory sophistication of this shared space is such that it supports interpersonal communication on a level of richness interchangeable with faceto-face interaction. The vision presented encapsulates two of the central goals not only of collaborative virtual environments (CVEs), but also of any communication medium. First, to enable groups of people to collaborate and interact socially in an efficient and enjoyable way, and second, to foster the illusion that people are together when in reality they are in distinct physical locations. CVEs have the makings of a potentially powerful medium of communication that heralds new promises and challenges. It is their inherently spatial property that sets them apart from other collaborative media. Though videoconferencing and groupware systems allow users to interact visually, the 3D context of each person’s physical environment is lost. This can pose difficulties in small group interaction where conversation management can be disrupted by ambiguous eye gaze cues. The loss of 3D context can also be particularly problematic in tasks for which it is essential to preserve spatial relationships, such as remote acting rehearsals. CVEs can begin to address these concerns by placing geographically dispersed users in a shared, computer-generated space where they can interact with the environment and with other users represented by avatars. Immersive interfaces can also offer multimodal, surrounding experiences that can create a strong sense of being inside that artificial space (presence), and sometimes of being there with others (copresence). As mediators of users’ actions and R. Schroeder and A.S. Axelsson (Eds.), Avatars at Work and Play, 17–38. C 2006 Springer. Printed in the Netherlands. 18 Garau appearance, avatars are likely to play a significant role in social interaction in CVEs. One of the central challenges in the development of CVEs is the creation of expressive avatars capable of representing users’ actions and intentions in real time. This chapter focuses on the issue of avatar fidelity, arguing for the need to explore priorities by investigating the impact of avatar appearance and behaviour on the experience of interaction. It presents research on minimal fidelity, and discusses its implications for the future development of CVEs as a viable communications medium.