By David F. Salisbury
September 17, 2001

A new test that measures what people see when viewing discordant images in the right and left eyes has produced important new clues about the location of some of the brain activity underlying visual consciousness.

The procedure, described in the Aug. 30 issue of the journal Nature, depends on a phenomenon called binocular rivalry first described in 1838 by Sir Charles Wheatstone. Using a device that he invented, Wheatstone discovered that when people are presented with dissimilar images in each eye, they report seeing first one image and then the other with the two images alternating unpredictably.

"Since this breakdown in binocular vision was discovered, it has been the subject of scientific interest because it involves the switching of visual consciousness without conscious control," says Randolph Blake, professor of psychology at Vanderbilt. He, Hugh R. Wilson, a mathematician from York University in Toronto, and Vanderbilt graduate student Sang-Hun Lee devised the new test.

In normal binocular vision, sensory information from the two eyes is fused into a single, three-dimensional visual impression. Stereopsis, the ability to fuse two, two-dimensional images into a three-dimensional image, is the flip-side of binocular rivalry. Individuals with misaligned eyes can suffer from binocular rivalry. They generally cope with this condition in one of two ways. They either rely on the view from a single eye or they use each eye for a different purpose, such as close and far vision.

The question of which neurons are responsible for this effect is a matter of scientific controversy. Visual information from the eyes is routed to the back of the brain to an area called the primary visual cortex. From there it is processed in an elaborate hierarchy that works its way forward in the brain through the temporal, parietal and frontal lobes. Some vision researchers argue that binocular rivalry must be handled at a low level in the brain's visual processing hierarchy, while others maintain that it must be handled at higher levels. Results from the new test lend weight to the argument that the effect occurs at a low level in the visual cortex.

The procedure that Blake and his colleagues devised allowed them to measure the time that it takes for one monocular image to replace the other in the visual field. Using pairs of different patterns within the same-sized annulus, they determined that one pattern does not instantaneously replace the other. Instead, the suppressed pattern breaks through at a specific location and then spreads in a wavelike fashion until it totally replaces the other pattern. The researchers found a way to trigger the alteration at a specific location on the ring and then measure the time that it takes the change-wave to reach a fixed point at the bottom of the annulus.

"The fact that the transition between the two images spreads gradually is an indication that binocular rivalry occurs at a lower level," Blake says. "If it were handled at a higher level you would expect the change to be instantaneous."
Another result that points toward the primary visual cortex is the trio's observation that the rate at which the transition zone spreads increases with the size of the annular ring. This relationship is consistent with the way in which the eye and visual cortex are wired.

In the fovea, the portion of the retina at the center of the eye, the density of nerve connections is much higher than it is at the periphery. As a result, a signal traveling a given distance in the fovea stimulates more nerves than a signal traveling the same distance in the periphery. In the test, the person looks at the center of the ring. So, as the diameter of the ring increases, it stimulates nerves located further out on the periphery of the retina. Therefore, if the alteration of images takes place in the visual cortex, a signal traveling halfway around a small ring must pass through a larger number of neurons than a signal traveling halfway around a large ring. The time it takes for a signal to propagate from neuron to neuron is roughly constant. So it makes sense that a signal which must pass through fewer neurons will travel faster than one which must pass through a larger number of neurons.

The final result implicating the primary visual cortex is the observation that when the suppressed pattern consists of concentric rings it spreads faster than when it is a radial pattern. Previous studies indicate that the primary visual cortex is specially wired to pick out continuous lines and curves. "This makes evolutionary sense because most of our visual environment consists of continuous features," says Blake. So the observation that the alteration between the two images proceeds faster when the suppressed pattern is made up of continuous, concentric lines than it does with discontinuous, radial lines is also consistent with an origin in the primary visual cortex.

In the past decade, cognitive psychology has joined forces with neuroscience to address the question of the nature of conscious awareness. The new results are an important step in the process of discovering which neuronal processes go with consciousness, and which do not.

This research was funded by a grant from the Eye Institute at the National Institutes of Health.