Rare cases of restored vision reveal how the brain learns to see

Results: By testing formerly blind patients within weeks of sight restoration, Sinha and his colleagues found that subjects had very limited ability to distinguish an object from its background, identify overlapping objects, or even piece together the different parts of an object. The patients gradually improved over time, and the new study suggests that dynamic information — that is, input from moving objects — is critical to the brain's ability to learn to segregate objects from their backgrounds (a task known as visual integration).

Why it matters: Doctors have been hesitant to treat older patients because the conventional dogma holds that the brain is incapable of learning to see after age 5 or 6, but these findings support the idea of treating blindness in older children and adults. The results also offer insight into modeling the human visual system, diagnosing visual disorders, creating rehabilitation procedures and developing computers that can see.

Methods: After three patients, ranging in age from 7 to 29, were treated for blindness, they were asked to identify shapes on a computer screen. The patients performed poorly when objects were stationary, but if a shape was put into motion, success rates improved to about 75 percent. During follow-up tests that continued for 18 months after treatment, the patients' performance with stationary objects gradually improved to almost normal.

Next steps: Project Prakash, the foundation Sinha started in 2004 to find and treat blind children in India, is raising money for a new 50-bed clinic and 500-student school in Rishikesh, 250 miles from New Delhi. In India, the rate of childhood blindness is three times that of Western nations "There's a humanitarian need to tackle this problem, and in addressing this humanitarian need we also have the opportunity, as neuroscientists, to understand how the brain learns to make sense of its visual environment," says Sinha.

Source: Massachusetts Institute of Technology