Assessing Multiple Perspectives in Three Dimensional Virtual Worlds: Eye Tracking and All Views Qualitative Analysis (AVQA)

The Human Computer Interaction community is beginning to wrestle with the challenges of effective interaction design for groups collaborating in 3 dimensional virtual worlds. Understanding of the requirements and design processes necessary to support these new technologies is still emerging. While the gaming community has developed heuristics for game design over a 25 year period, the adaptation of this knowledge for the design of virtual collaboration spaces has been limited. Following a research through design approach, we create novel integrations and applications of Human Computer Interaction research methods in an attempt to understand the collaborative work of a small group of individuals in a 3 dimensional virtual world. Specifically, the fine grained perspective provided by eye-tracking (ET), and a more coarse unit of analysis, provided by All-Views-Qualitative-Analysis (AVQA), are applied to understand non-gaming interaction in 3 dimensional virtual worlds. To formalize an integrated method, we describe three lenses for analyzing eye-tracking data in graphics intensive and animated virtual applications. Then, we combine those outcomes with a single method for integrating and analyzing the points-of-view of all subjects in a small group collaborating on a creative work task in a 3 dimensional virtual world. Through these examples, we illustrate the benefits of this type of research for 3 dimensional virtual world designers.