Judgments about Collisions in Simulations of Scenes with Textured Surfaces and Self-Motion: Do Display Enhancements Affect Performance?

Abstract

To move through the environment safely, people must make effective judgments about collisions. It has been asserted that most studies of time-to-collision judgments are limited due to a lack of visual realism (Manser & Hancock, 1996). Studies that compared performance among displays which differed in realism provided mixed results. We measured judgments about whether, and when, two objects would have collided with each other. Results from simulations of scenes with colored, textured surfaces and a moving observer were mostly comparable to earlier results from simulations of black-and-white, line-drawn objects and a stationary observer (DeLucia, 1995; DeLucia & Meyer, 1999). Texture and self-motion affected performance in a restricted set of conditions and did not eliminate errors due to misleading depth cues. Increases in realism, which incur more costs and computational time, may not always be justifiable from a performance standpoint. Results have design implications for simulators and virtual reality systems.