Empirical Modeling of Perceived Presence in Virtual Environments Using Sequential Experimentation Techniques

Abstract

Investigations of perceived presence in a virtual environment (VE) consist primarily of single studies examining only a few independent variables at once. A series of three studies was conducted to test a ratio-scale measure of presence and to examine the first- and second-order effects of eleven VE system parameters on perceived presence. Sequential experimentation techniques were used to build an integrated empirical model using polynomial regression. The resulting primary predictors of perceived presence using a head-mounted display included field of view, sound, and head-tracking. Secondary predictors included visual display resolution, texture-mapping, stereopsis, and scene update rate. The procedures and limitations of sequential experimentation, and its application to the examination of perceived presence in a VE are discussed. It is concluded that sequential experimentation is a useful tool for examining perceived presence, but the subjective nature of this phenomenon and individual differences can make data bridging across sequential studies problematic.