<title>Some effects on depth-position and course-prediction judgments in 2-D and 3-D displays</title>

Abstract

With increasing use of three-dimensional stereoscopic display systems, there is a need to define optimal user interface guidelines for these systems. This study examines human performance in two tasks which required participants to view 3-D imagery with or without retinal disparity cues. The imagery consisted of an airspace control zone and it was rendered on the display screen in each of three formats: (1) a plan view, (2) a simple perspective view, and (3) an enhanced perspective view. One task required viewers to judge relative depth positions for objects within the 3-D space, while the other task required extrapolation of object motion headings. Retinal disparity cues reduced both the number and degree of errors, but only for the plan view, course prediction task condition. Not only did retinal disparity cues not provide any performance advantages for perspective displays, but performance levels for these displays were no different than those achievable with a plan view in the absence of retinal disparity cues. Although retinal disparity cues had no effect on search time, the enhanced perspective displays formats effectively increased search times relative to plan view and simple perspective formats. Neither retinal disparity cues nor display format had an effect on subjective ratings of viewer confidence.