|
 |
 |
 |
 |
 |
Spatial Cognition
Spatial cognition is an area of research that has already embraced virtual environment technology (e.g., special 1998 issue of Presence devoted to orientation and navigation). Much of the appeal of virtual environment technology is the ease with which one can create complex environments for studying spatial behavior. There are innumerable questions that lend themselves to study using virtual environment technology, including navigation, cognitive mapping of natural environments, and spatial memory; moreover, the answers to them are important for designing effective immersive virtual environments. Our group is currently using virtual environment technology for the following research projects: 1) understanding how one aggregates and integrates local cognitive maps into a global cognitive map, 2) examining alignment effects, 3) measuring performance when navigating through 3D environments, and 4) understanding how subjective vertical affects perception of shapes.
Current Projects:
Spatial Navigation in Virtual Environments
Currently, Dave Waller and Jack Loomis are investigating the sources of information that people use when they learn about locations in space. In one set of experiments, a participant learns the location of a target object (e.g., the black cylinder above left) relative to several other landmarks (e.g., the other cylinders). During testing (above right), the participant is placed in a similar environment that has no target present and is asked to find his or her way to where the target should be. Using this paradigm in virtual environments allows us to manipulate geometric cues that would be difficult or impossible to manipulate in the real world. Our current experiments examine the role of distance versus angular information in landmark learning. We are also interested in the degree to which mechanisms of landmark learning vary between individuals.
Work with Dan Montello, Mary Hegarty, David Waller, and Tony Richardson examines factors that influence how flexible people's spatial knowledge can be. People are asked to learn the positions of several objects (or people) along a path (shown above). They then are asked to make judgments of relative directions from an imagined perspective. In general, imagined perspectives that are misaligned with the learning perspective are more error-prone; however, several factors can modulate this effect. We are currently exploring some of these factors. By manipulating the way that people learn and are tested on these layouts, we are able to draw conclusions about the factors that influence spatial representations.
Navigation research by Andrew Beall studies the usefulness of optical references when translating across visually arbitrary path. In particular, participants are asked to judge the distance from, and direction toward their source location (starting point) after traveling through a scene that resembles a forest. Evidence suggests that individuals calibrate their source location best when head rotations and body translations are each coupled with their own separate visual cue, but only when both cues are present. Under such conditions, it is hypothesized that maximal spatial information can be integrated into a cognitive map.
Tony Richardson, Dan Montello, and Mary Hegarty are assessing the nature of the spatial representations of an environment acquired from maps, navigation, and virtual environments. In one study, participants first learned the layout of a simple "desktop" virtual environment (VE) and then were tested in that environment. Then, participants learned two floors of a complex building in one of three learning conditions: from a map, from direct experience, or by traversing through a virtual rendition of the building. VE learners showed the poorest learning of the complex environment overall, and results suggest that VE learners are particularly susceptible to disorientation after rotation. However, all conditions showed similar levels of performance in learning the layout of landmarks on a single floor. Consistent with previous research, an alignment effect was present for map learners suggesting they had formed an orientation specific representation of the environment. VE learners also showed a preferred orientation, as defined by their initial orientation when learning the environment. Learning the initial simple VE was highly predictive of learning a real environment, suggesting that similar cognitive mechanisms are involved in the two learning situations.
Richardson, Montello, and Hegarty are continuing to examine performance learning in large maze-type environments. Of particular interest is the nature of performance differences found between learning in desktop and immersive VE's. Individuals tend to become disoriented in desktop VE's, since no vestibular information is available. Specifically, individuals show an inability to align their egocentric orientation with their orientation within the VE. They are also pursuing a second line of research examining orientation specificity replicating the work of Presson, DeLange & Hazelrigg (1989) and Sholl & Nolin (1999) and are interested if learning in a virtual environment produces the same pattern of performance as that in a real environment using carefully controlled experimental procedures. If a true replication can be achieved, manipulation of variables within the VE could lead to a better explanation of this controversial effect.
Advanced Displays and Controls
The goal of this project, conducted by Andrew Beall, is to develop improved displays and controls for orientation and navigation in virtual environements, and particularly in simulated weightlessness, where movement in all six degrees of freedom is possible. The focus is on human visual orientation in microgravity. We are studying the following topics: 1) the rotary component of viewpoint motion on visual path integration ability; 2) actual head movement on ability to infer virtual vehicle movement; and 3) avatar controlled viewpoint control.
Prototype virtual navigation aid for controlling change of viewpoint.
2D navigation paradigm for measure performance at terrestrial path integration.
2D or 3D navigation paradigm that affords motion out of the ground plane
with rotations not restricted to the vertical axis only.
