Looking at Picasso's cubist portrait "Girl with a Mandolin", one critic might see a jumble of lines and angles, while another perceives a delicate young musician. Such is the difference between two brain areas involved in perceiving visual shapes. Now researchers suggest that one area--the one that sees a musician--can shut down the one that sees only lines.
When light hits the retina, the visual signal zips to a region of the brain called the primary visual cortex, or V1. Neurons here respond to specific elements of an image--a vertical line, for instance--and send that information on to higher brain areas, including one called the lateral occipital complex (LOC). The LOC processes shapes, so that when we see four lines at right angles, we understand it to be a square. Now researchers say that when the LOC detects a pattern, it tells V1 to stop reporting on the component pieces.
Neuroscientists Scott Murray of the University of California, Davis, and Daniel Kersten of the University of Minnesota, Minneapolis, and their colleagues used functional magnetic resonance imaging to monitor the brains of people viewing moving images. The images could be interpreted in multiple ways--for example, two pairs of diagonal lines could appear to be unrelated lines moving vertically or a diamond moving horizontally, as viewed through the bars of a fence. Consistent with earlier studies, the team found that LOC activity increased when subjects perceived a coherent shape, such as a diamond. But V1 does just the opposite, the researchers report online this week in the Proceedings of the National Academy of Sciences. When a subject reported seeing the lines as just lines, V1 activity increased, but surprisingly, it decreased when the person saw the lines as a diamond. The image itself remained the same; only the subjects' perception of it changed. The scientists conjecture that when people perceive the lines as a shape, the LOC sends the message back to V1, saying "I'll take it from here."
"It's an interesting idea that's receiving some of its first empirical support here," says vision researcher Stephen Engel of the University of California, Los Angeles. "The data present intriguing evidence," he says, but cautions that the proposed model "is far from established."