Visual technology enables brain to learn in new ways

By using a single screen and two projectors, Tufts is able to produce ultra-high resolution images -- including 3-D images -- that appear smoother and without seams. Images projected at a higher resolution reveal fine, minute details that would be imperceptible on a screen with fewer pixels or tiled images. The VisWall's projectors are equipped with Infitec filters to minimize ghosting, in which an image appears to include elements of another image. Ghosting is a common drawback with conventional polarized filters.

In addition, the Tufts system can combine the sense of touch with that of sight through haptic devices that convey varying levels of resistance to the user when he or she touches graphical objects on the display wall. This also allows Tufts researchers to create virtual environments, such as the human body for surgical simulations that can be physically manipulated and transformed.

Order in Chaos

Tufts faculty have already discovered applications of the new technology. Mathematics Professor Boris Hasselblatt made a surprising find while viewing a mathematical model of butterfly populations as they fluctuated through successive generations. The model, used for research in dynamical systems theory, is based on a simple formula and is well-known to anyone familiar with chaos theory.

Visualizing the large population dataset with the 14-foot-wide, high-resolution graphical display enabled Hasselblatt to detect anomalies impossible to perceive with conventional displays: subtle traces of curving lines that he said indicated irregularities in variations in the population. The lines extended over different areas of the model and then converged at one distinct point.

Hasselblatt has looked at smaller images of this classic model many times during the last 20 years but had never recognized this convergence. He has not yet determined the implications of this discovery, but he said the pattern reflects order in what mathematicians have always thought to be a progression of chaotic cycles. "The pattern is so subtle that it's imperceptible but in this rendition the resolution is fine enough that I can easily see it," he said.

Bruce Boghosian, chairman of the mathematics department at Tufts and principal investigator on the NSF grant, said that the VisWall will benefit his study of fluid dynamics. Visualization capabilities can help him and his fellow researchers better understand fluid flow.

"You can go right up to streamlines in a fluid or dig into a reservoir and see which way it's flowing," said Boghosian. "That's the direction we would like to move in. You can imagine all kinds of other uses for something like that."

Virtual Surgery

The VisWall will also aid Mechanical Engineering Assistant Professor Caroline Cao. Her goal is to develop more robust laparoscopic surgical training systems in which 3-D computer simulations enable surgeons in training to feel as well as see.