Two new studies appearing in this week's issue of Neuron elucidate the neural mechanisms behind feature-based attention—essentially, the tuning of your visual processing system to specific colors, shapes or motions as a way of formulating an awareness of a scene. The findings could one day be used to better diagnose and treat disorders such as attention deficit / hyperactivity disorder (ADHD) and dyslexia.
"The idea is that this is the mechanism that allows you to find things when you know what they are but you don't know where they are," says John Serences, a cognitive scientist at the University of California, Irvine, and co-author of one of the new reports.
He and Geoffrey Boynton, a system neurobiologist at the Salk Institute for Biological Studies in La Jolla, Calif., had 10 subjects perform a visual attention task as their brains were scanned via functional magnetic resonance imaging (fMRI). During the test, participants faced a screen, the bottom half of which was empty, while the top half showed two clusters of dots stacked one above the other. Each cluster contained a layer of dots that moved upward and to the left superimposed on a second set of dots that moved upward and to the right.
Neuroscientists established in previous work on spatial attention that varying groups of neurons in the visual cortex process different sections of the visual field. A subset of neurons will fire—that is, send electrical signals that convey information—when a person focuses their attention on the upper half of a scene, whereas a different subgroup activates when attention is transferred to the lower half.
From the fMRI data, the researchers could see that, as they had expected, different subpopulations of neurons in the brain were also activated when the subjects were asked to concentrate on stimuli moving up and left versus up and right. The surprise came when they noticed that a pattern of activity among the neurons that process the lower half of the visual field echoed the behavior of the cells for the field's upper half.
"While [the subjects are] attending to a particular direction of motion, the part of the brain that processes that direction of motion becomes more sensitized to detect that motion," Serences says. "The really new thing is the fact that this sensitization spreads across the entire visual field." This indicates that while you are consciously tuning for a particular shape or color—say, your lost keys—in one part of your visual field, you may be subconsciously alerting the entire visual system to that trait, enabling a more efficient search.
In complementary work, Taosheng Liu, a research scientist in New York University's Department of Psychology and at its Center for Neuroscience, put eight subjects through a similar fMRI task. In Liu's experiment, participants viewed a composite image of two sets of superimposed bars. (One set of bars slanted 20 degrees clockwise from the vertical; the other, 20 degrees counterclockwise.) When the image appeared in front of the volunteers, they were told to focus on either set of skewed bars. The results from the subjects' fMRI scans indicated that neurons in the visual cortex attuned to shapes slanting clockwise would fire if the subjects were told to focus on the corresponding bars, and vice versa.
After focusing on a particular group of bars for a short time, the signal from the associated neurons would begin to quiet—indicating an adaptation to the stimulus. Then the researchers would switch the image to a picture containing only one of the two sets of bars.
According to Liu, if the new image was in the same orientation as the bars the subjects had focused on, the already adapted neurons would be less sensitive to the change and show a "relaxed response." If, however, the scientists showed a picture of the opposite orientation, a different subpopulation of neurons would respond with a flurry of activity. "The cool part," Liu says, "is the clockwise and counterclockwise neurons are all intermingled in the same part of the brain," the primary visual cortex.
"We're starting to understand how you can efficiently find things that you're looking for," Serences says. In addition, he adds, the research is part of "figuring out very mechanistic descriptions of how attention works," which can be used in the future "to develop better tools and better objective measures for diagnosing disorders like ADD and dyslexia."