Julian Hochberg, PhD - US grants

Artists use static two-dimensional depth cues, such as relative size and linear perspective, to represent three-dimensional scenes. Researchers have traditionally considered such cues to be important in understanding how humans perceive the world as three-dimensional, despite its only projecting a two-dimensional image onto the back of their eyes. Recent research has downplayed the importance of these traditional static depth cues, and has emphasized information present in moving displays. These motion cues are believed to allow for less ambiguous perceptions than do static displays. This research will study the interaction between static and motion information about depth. Observers will see a two- dimensional display in which two square objects are located on a road flanked by poles; the objects will begin at different distances from the observer and will approach the latter at different speeds. In one condition, observers will predict which of the objects will first arrive at the "nearest" poles, and press a corresponding button immediately. In the second condition, observers will press the button precisely at the point in time that the chosen object reaches the "nearest" poles. In both conditions, response times will be measured for displays which represent scenes viewed from locations near the ground and from locations well above the ground, and for displays in which there is either a large or small difference between the heights of the "nearest" and "farthest" poles. The research will test the expectation that high viewing locations, and large pole- height differences provide a greater impression of depth, and will thus result in faster, and more accurate response times. The research will attempt to quantify the changes in information about depth provided by the changes in perspective and relative size cues, and compare the relative contribution to response times of the two classes of depth cues. Research which demonstrates that specific configurations of depth cues result in enhanced perceptual-motor performance will permit more effective designs of graphics-driven systems, such as flight simulators and teleorobots. Results will also suggest designs of motor vehicles that will optimize reactions to possible collisions, for example by maximizing the driver's eye height.