Visualization and Vision Science
5th December 2012, 10:00am - 11:00am
Salle 465, PCRI (how to get there ?)
This talk will discuss three ways in which work in visualization might be assisted by the methodologies of vision science. In the first, it will be shown that basic techniques for measuring the perception of simple properties (such as luminance) can be adapted to study the perception of visualizations such as scatterplots. Results show that correlation perception is described by simple linear and logarithmic functions, and that these laws are relatively general, occurring for a wide variety of visualization designs. The second approach examines the sequential processes involved in the interpretation of diagrams. Results show that the effectiveness of an assembly diagram can be assessed by the use of timing measures, with linear models providing a good approximation of behavior. Finally, discussion will turn to the issue of how stress. It will be shown that stress has measurable effects on various kinds of visual attention, and that these effects can be good as well as bad, raising the possibility of visualizations that are not only stress-resistant, but stress-optimal, supporting better performance than usual when the observer is stressed.
Ronald A. Rensink is associate professor at the Department of Computer Science at the University of British Columbia in Vancouver, Candada. His main research interest is vision—the various ways that humans, animals, and computers use light to see. He believes that vision involves constraints that apply to any system, and that the most successful visual systems are based on very general information-processing strategies. As such, his approach is to examine biological systems (including humans) to see how they operate, and then to look at these mechanisms from a computational point of view to see if they embody more general principles. Among other things, these more general principles can provide a scientific basis for the design of visual interfaces that can interact with human visual systems in an optimal way.