The Eyes Have It

When a new window or webpage pops up on your computer screen, what do you look at first? Michael Byrne, assistant professor of psychology at Rice, might be able to tell you. He has developed a model of human thinking and performance that can predict how well people locate and select various options on a computer screen. The model has implications for designing air-traffic-control monitors, in-car navigation systems, webpages, and other computer displays to present on-screen information most effectively to the viewer.

“Our theory and model allow scientists to make predictions about human performance on tasks that involve computer displays,” says Byrne. “We can tell, for example, where to position information on a computer screen so that the user is more likely to see it quickly.” Byrne conducted the research at Carnegie Mellon University before joining the faculty at Rice, where he analyzed the results that are published in the July issue of the International Journal of Human Computer Studies.

“Understanding the interaction of a user with a designed device like a computer requires a clear understanding of three components,” Byrne says. “The user’s cognitive, perceptual, and motor capabilities; the task; and the device used to accomplish the task can impact the result.”

Byrne showed each of the 11 participants in his study more than 100 questions that involved identifying a particular number or letter in lists of random numbers and letters on a computer screen. He timed how long it took the participants to spot the selected symbols. A camera mounted on a headband worn by the participants tracked their eye movements.

Using a theory of cognition known as ACT-R/PM and an eye-tracking model, Byrne predicted how long it would take the participants to click on the targeted items. He also predicted the other items they would look at en route to the targeted items, based on the characters or numbers they were likely to fixate on and the order in which they were likely to read information on the display. Byrne found that his predictions averaged within 15 percent of participants’ actual response times—“close enough to be useful to designers,” he says.

Engineers who design air-traffic-control monitors, for example, could use the theory and model to position critical information on a screen where the user is most likely to see it first. “Keeping up with the volume of information on a computer display is often a problem,” Byrne said. “A designer can optimize the rate at which people can process information on the screen by using the theory and model we studied. A strategic placement of menu items can make it possible for people to read the most important information fast enough to keep up with their work so they don’t end up in a situation where they’re likely to make errors.”

Byrne’s research was supported by the National Science Foundation, the Office of Naval Research, the National Institute of Mental Health, and NASA.

—B. J. Almond