The Study of Perception and Action Using Virutal Reality Environments
College: Applied Health Sciences
Award year: 2000-2001
The proposed project is designed to examine the contribution of various sources of visual information for the coupling of perception and movement perception of biological motion. A number of tasks in the sport domain, such as hitting a pitched ball or returning a shot in tennis, require the performer to rapidly and accurately determine the speed and direction of ball travel. It has been suggested that visual information from the opponent's movement is used to predict subsequent ball flight direction. The specific goals of this project are 1) to determine the effectiveness of virtual reality environments for examining the perception-action coupling of complex biological motion; and, 2) to determine the specific visual information used in the perception-action coupling of time critical projectile tasks.
The specific activity to be examined is a tennis ground stroke and volley return combination. The performer hitting the volley is close to the net so that the time to react and move to the ball is very short, approximately 400 to 700 ms. Anticipation, based on the movement pattern of the opponent, may be critical for successful performance under these conditions. To determine the effectiveness of virtual reality environments (goal 1 above), the experimental strategy will be to compare anticipation accuracy of highly skilled performers while viewing three types of display. The display types include a virtual reality display, a standard 2-dimentional video display, and a "point light display".
To determine the nature of the visual information used to perceive ball flight direction (goal 2 above), multi segmental animations of skilled performers will be created using a character animation and visual effects system (Maya). The animations will be based on detailed 3-dimensional data obtained from these animations, portions of the motion pattern or ball flight can be eliminated or manipulated to examine their influence on perception and anticipation.
A major portion of the project is the development of animations for presentation in a virtual reality environment such as the CAVE at NCSA. In a planning grant awarded last year, 3-dimensional data have been generated for a skilled performer hitting tennis ground-strokes to left and right sides of the court. These data have been ported and presented on the CAVE. We are currently working on developing and refining the skeleton based model and expect to have this completed during the planning grant period.
Potentially, this research will provide insights into the critical invariants used for the perception of biological motion. The results will also have a number of practical applications. These include the use of various types of displays for observational learning of motor actions, and perceptual training of high level performers, and biomechanical analysis of movement kinematics. Thus, the results from this work will have implications for the use of virtual reality environments for the training of perceptual-motor skills.
For further information, please see http://www.kines.uiuc.edu/faculty/carlton.html.