Date of Award

Winter 2001

Document Type


Degree Name

Doctor of Philosophy (PhD)



Committee Director

Mark W. Scerbo

Committee Member

Glynn D. Coates

Committee Member

David A. Dryer

Committee Member

Thomas W. Mastaglio


The key to effectively using the immense body of data on the Internet is an efficient method of organizing relevant information. Researchers and designers are beginning to promote the advantages of three-dimensional (3D) models of information storage and retrieval; however, the potential benefits of perceptual depth cues have not been systematically studied.

The present study used a computer task to examine the effectiveness of three types of virtual desktops. A two-dimensional (2D) virtual desktop display, lacking in the cues that give the illusion of depth, was compared to two different 3D virtual desktops, both of which used perceptual cues to convey a sense of depth. One of the 3D desktop conditions conveyed motion parallax through an automatic rotation. It was expected that performance would increase as the number of perceptual cues increased.

The present study also examined the potential benefits of organizing and retrieving documents from a subjectively organized versus a preconstructed, or fixed, information space. An organization that individuals create for their own use may be difficult for others to use. Thus, subjective organization of documents was expected to promote better performance than a fixed organization scheme, which is exactly what the data showed. There was a very strong performance benefit to those who organized their own desktops.

Contrary to the other hypothesis, the 2D arrangement was more beneficial to users than either the 3D or 3D with motion arrangements. The 2D advantage may be the result of a number of factors. First, although people live in a 3D world they navigate more on 2D planes. Also, people may naturally encode spatial information in a descriptive or symbolic manner, as opposed to creating a spatial analog in the mind's eye.

Designers should not blindly attempt to create interfaces that mimic the real world. The choice between a 2D and 3D interface should be based upon the type of task to which the interface will be applied. Information storage/recall tasks, including the present task, will most likely benefit from a 2D interface. Other tasks that make greater use of navigation in 3D space may be better suited to 3D displays.