Researchers at the University of Toronto have discovered how the brain helps us see and interact with objects by filling in missing information, according to a study published in the June issue of Current Biology.
Because most of what people see is often blocked by other objects, the visual information received by the brain is usually incomplete. "People take perception for granted because it seems so instant and automatic to us," says Allison Sekuler, associate professor of psychology at U of T and one of the study's senior authors. "What many people don't realize is that the objects we see are not necessarily the same as the information that reaches our eyes, so the brain needs to fill in those gaps of missing information."
Sekuler and her colleagues believe they have found the first direct evidence to prove this theory. The group of researchers, led by PhD students Jason Gold and Richard Murray, asked people to describe various types of shapes presented on different backgrounds made up of visual "noise" - gray, black and white pixels similar to the snow on a de-tuned television. The square shapes were either real, illusory, blocked or fragmented.
Because the objects were difficult to see, sometimes they appeared fat or thin, depending on the background noise. (The sides of fat objects bend outward while sides of thin objects bend inward.) By averaging the luminance of the visual noises that led to fat or thin responses, the researchers determined which parts of the stimulus were important for these judgments.
Not surprisingly, the researchers say, when there really were contours in the shape that made it thin or fat, people used information around the location of these defining lines in making the shape discrimination. "Amazingly however, we found that people used information from exactly the same locations even when the contours in those locations were hidden or missing altogether. In other words, people relied on contours that were not really there, but that had been constructed by their brains," says Gold, whose thesis is looking at the mechanisms underlying visual perception.
"If you didn't have the brain filling in all of this missing information, every time you looked at an object from a slightly different view, it would be a different object and that would be very confusing and difficult to cope with," says Patrick Bennett, associate professor of psychology at U of T and the study's other senior author. "This filling in gives some consistency and continuity to the world."
In addition to helping us understand how our brains are constantly constructing the visual world, the researchers believe there may be other practical applications for this discovery. "You may not need a key to get into your house one day if a computer recognizes you and lets you in," Sekuler says. "The problem is, if you're wearing glasses or have a fresh scar, the computer won't recognize you as easily as another human would. Once we understand what makes the human brain so efficient at discriminating one shape from another, we can use this knowledge in developing efficient artificial intelligence systems."
The next step for the researchers will be to determine where in the brain this is happening. The group also intends to examine how the information we use differs in different populations, like people who have suffered strokes whose perception of the world is different and who may not therefore fill in the blanks in the same way. This study was funded by the Natural Sciences and Engineering Research Council.
Steven de Sousa
U of T Public Affairs
Materials provided by University Of Toronto. Note: Content may be edited for style and length.
Cite This Page: