The Effects of Varying Levels of Immersive Technologies on Subject's Ability to Perform a 3D Construction Task

Abstract

Previous studies of the use and effectiveness of virtual reality environments have suggested that the VR leads to most effective performance when the subject is physically immersed in the environment. This prior research has suggested that the use of environments that are not highly immersive tend to result in lower user effectiveness. Recent studies by a team of researchers from BGSU’s Computer Science and Psychology departments have noted that another facet, the ability of a user to “offload” some demands on user cognition, may be just as important a factor in the effectiveness of VR as the variable of immersiveness.

In this study, four levels of immersive mediums were compared through an experiment to convey instructions to an inherently 3D construction task. The dependent variable was the accuracy with which the subjects completed the construction task. The construction task involved the folding of an origami whale, and the instructions were delivered in a variety of formats, based on VRML with a standard web browser. Subjects in all conditions received instructions on paper in both pictorial and textual form as well as the computerized form of the directions. The four mediums that were considered, in order from lowest to highest levels of immersion, follow:

• Non-interactive animation. [Lowest level of immersion]

• Interactive VR on a desktop presentation with a mouse as interactive input device

• Interactive VR on a PDA presentation with a stylus as interactive input device

• Interactive VR on a desktop presentation with a touch screen as interactive input device

It was found that the dependent variable, accuracy, was significantly different among the four conditions. Once significance had been established, tests were performed on the number of correct folds and the number of folds in error individually; it was discovered that there is a significant difference among the four conditions with respect to both kinds of folds. From there, tests were performed to find which conditions were better at delivering instructions, so that the number of correct folds was high and the number of folds in error was low. The results of these tests showed that, with respect to the number of correct folds, the touch screen compared to animation conditions, the touch screen compared to the workstation conditions, and the PDA compared to the animation conditions were all significantly different. When the same tests were run using the number of folds in error, it was concluded that there is a significant difference in the number of folds in error between the touch screen/ animation conditions and the touch screen/workstation conditions.