Wayfinding in large-scale virtual worlds

As important as navigation is to human performance in virtual worlds, it is an often overlooked problem in the design process. This dissertation reports an experiment intended to show that real-world wayfinding and environmental design principles are effective in designing virtual worlds which support skilled wayfinding behavior. The design principles are adapted from both the cognitive psychology literature and urban and architectural design methodologies. There are two categories of design principles; those that guide the organizational structure of the environment, and those that guide the use and presentation of maps.; The study measures subject performance on a complex searching task in a number of virtual worlds with differing environmental cues. The environments are extremely large and consist of open sea, land, and ships which are used as targets for search tasks. The worlds are augmented with either a radial grid, a map, both, or neither. For each trial, the subject's viewpoint position and orientation was sampled approximately once per second. A verbal protocol with accompanying video was used to elicit information about the search strategies employed. A map drawing exercise followed each trial in order to gain insight to the subject's spatial knowledge (i.e. cognitive map) of the environment.; The results show that subjects in the treatment without any additional cues were often disoriented and had extreme difficulty completing the task. The grid was shown to provide superior directional information but both treatments using the map were superior overall due to the geocentric perspective provided. Behaviors associated with each treatment indicate that the cues had a strong effect on both search performance and search strategy. The results suggest that users of large-scale virtual worlds require structure in order to effectively navigate. Augmentations such as direction indicators, maps, and path restriction can all greatly improve both wayfinding performance and overall user satisfaction.