Using Hands to See

ORLANDO, FLORIDA--In an odd case study presented here 6 November at the Society for Neuroscience meeting, a man blind in part of his visual field can have partial sight there--as long as his hand is in the blind area. The man might owe his ability to neurons, similar to ones known in monkeys, that react to both vision and touch. The find indicates that humans have the same multitalented neurons, and it might lead to the development of compensatory strategies for some forms of blindness.

Until about 20 years ago, researchers thought that most sensory neurons are assigned fairly specific tasks. Neurons in the primary visual cortex, for example, receive and process signals from the eyes; those in the auditory cortex react to sound. But since then, researchers have found so-called bimodal neurons that respond to two senses. Some of these, well studied in primates, are sensitive to being touched on the hand as well as seeing things within several centimeters of the same hand.

This literature came in handy when psychologists Krista Schendel and Lynn Robertson of the University of California, Berkeley, met "WM." He'd suffered a stroke in his right primary visual cortex and couldn't see anything to his left. But he mentioned that he did better on standard ophthalmological tests when his hand was in his blind area.

Intrigued, the researchers tested WM's vision. They had him look at the center of a computer screen while blue lights flashed in various places in his peripheral vision. When his hands were in his lap, he saw almost nothing on the left side. When his left hand was propped up next to the computer screen, however, he saw up to 40% of the flashes in some locations tested in this "blind" area. When the researchers moved the computer screen out of WM's reach--and beyond the centimeters' range of previously studied primate bimodal cells--he was again unable to see the flashes.

"I think it's cool," says clinical neurologist Michael E. Goldberg of Columbia University, who has studied bimodal neurons. He agrees that bimodal cells are probably responsible for allowing WM to detect the lights despite a lack of input from primary visual cortex. The cells likely aren't giving WM too much detail, he says, but they can say "Yes, there's something there."

Related sites
The abstract from the Society for Neuroscience meeting
Lynn Robertson's site