Dr. Constance Royden

Research Associate, Computer Science Dept., Wellesley College

Computer Science Dept.
Wellesley College
Wellesley, MA 02181
Research/Job interests
Computational vision, visual psychophysics, human motion perception, human heading perception

Vismod Advisor: Ted Adelson

PhD. Neuroscience, UC San Francisco, 1988
B.S. Biology and Engineering, Caltech, 1980

1992 - present, Postdoctoral fellow, Heading Perception, Wellesley College
1990 - 1992, Postdoctoral fellow, Motion Perception, MIT Media Lab
1988 - 1990, Postdoctoral fellow, Heading Perception, UC Berkeley

Research Description

I combine psychophysical experiments with computational modelling to study how the brain processes visual information. I am particularly interested in motion processing, which is a good system to work with because so much is know at the electrophysiological level and there is a wealth of computational theory on the subject. I am currently studying how people judge their heading in the presence of moving objects in a scene. I am using a computer controlled display to simulate observer motion through an environment containing one or more moving objects. I am testing the conditions under which the presence of a moving objects affects observer's abilities to accurately judge heading. I am concurrently analyzing the effects these moving objects would have on various computational models for computing heading from visual information.

Recent Publications

Analysis of misperceived observer motion during simulated eye movements
Royden, C. S. (1994)
Vision Research (in press)

Estimating heading during eye movements
Royden, C. S., Crowell, J.A. and Banks, M.S. (1994)
Vision Research (in press)

The perception of heading during eye movements
Royden, C. S., Banks, M.S. and Crowell, J.A. (1992)
Nature 360: 583-585.

The effects of moving objects on heading perception
Royden, C. S., and Hildreth, E.C. (1994) Annual Meeting Abstract Issue, Assoc. for Res. in Vision and Ophth. 35: 1999.

Research at Vismod

I worked on the perception of transparent motion. I worked on developing a model for how the visual system might process motion information from transparent stimuli to yield an output that signalled transparent motion, i.e. two different motions at the same spatial location. The model used motion energy mechanisms followed by a motion-opponent stage, which was in turn followed by an interpolation stage. At the same time, I worked on psychophysical experiments testing how people judge their heading in the presence of rotations, in collaboration with M.S. Banks and J.A. Crowell at UC Berkeley.

Look at my web page at Wellesley College.

Vismod Home Page