Abstract
My talk will begin with a discussion of the advantages and disadvantages of using immersive virtual reality (VR) to study basic research questions relating to visual perception and visually-based spatial behavior. Foremost of the advantages is that one can do research that has more ecological validity than previous laboratory-based research while retaining the same level of experimental control. In particular, this means that one can decouple stimulus variables that normal covary in the real world.
I will then discuss VR-based research in three different areas relating to vision: visual space perception, visual control of locomotion, and visually-based navigation.
VISUAL SPACE PERCEPTION Despite decades of research, there is much we do not understand about visual space perception, even at the functional level. As an example, I will focus of the perception of egocentric distance. A number of researchers have found a nearly two-fold underestimation of distance in immersive virtual reality, in contrast with accurate perception of distance in the real world. I will present some of this research and discuss the possible reasons for non-veridical perception.
VISUAL CONTROL OF LOCOMOTION Immersive virtual reality is especially suited for studying visually-based action because potential controlling sources of stimulation can be modified while the person is behaving. Several topics relating to visually-controlled locomotion will be discussed: aiming toward a point, aligning with a straight path, and steering a curving path.
VISUALLY-BASED NAVIGATION. There are basically two methods of visually-based navigation through large-scale space (i.e., beyond local vistas): landmark-based navigation (piloting) and integration of sensed self-motion (path integration). In the real world it is difficult to separate these methods, but VR readily allows one or the other to be studied in isolation. I will present research on both topics.