Introducing light to the blind

Roughly half a million children in India suffer from blindness. But given the proper care, a quarter million of these children could be treated and learn to see. 

Dr. Pawan Sinha, a visual neuroscientist at MIT, investigates how the brain learns to see and, in the process, has coupled his numerous research projects with a humanitarian mission. Project Prakash, founded by Sinha in 2003, identifies children in India suffering from treatable blindness and provides them with the opportunity to undergo curative treatment. To date, Project Prakash has screened over 700 people from ages five to 22. Additionally, the corrective surgery provides researchers with subjects perfectly suited to studying how the brain learns to see.  

Sinha initially traveled to India two years after joining MIT’s faculty. On this trip, he first recognized the scope of childhood blindness in India and decided that something should be done. “Even to my untrained eye,” Sinha said, “it was clear that the blindness that these children had was treatable.” Indeed, the children had cataracts, an eye disease that clouds the lens. While cataracts may allow some light into the eye, vision can be blurry to the point of blindness. However, a simple operation is often enough to cure this ailment. Therefore, in response to the issue, Sinha took it upon himself to establish Project Prakash.  

Project Prakash’s efforts fuel Sinha’s scientific research in the field of visual acquisition. When investigating such an area, finding subjects to study is difficult since infants do not have the capacity for language and thus cannot respond to researchers’ questions or complete basic tasks. However, children treated by Project Prakash have the advantage of full language capabilities and can complete the tasks set to them by researchers as well as comply with fMRI scanning, a form of brain imaging. 

Sinha’s work has provided insight into how the brain learns to see.  Once bandages are removed after surgery, Sinha explained, some aspects of vision come about immediately, like the ability to observe moving objects. “But then there are other abilities, like the ability to recognize objects, or the ability to break up an image into distinct objects, that take several months to develop.”  

When first learning to see, it is difficult to determine what features constitute an object and make one entity separate from another. A single object need not be all one color or have uniform depth, yet it is still one item. These details—which seem basic to lifelong sighted people—make object identification difficult for those who have recently gained sight.  

However, movement considerably eases the identification process. “At the start of the visual journey,” Sinha said, “no other cues were as effective as motion. In hindsight, it makes sense because our basic definition of an object is that it is a cohesive entity. Everything that moves together gets bound together in one object, and things that move differently get segregated.”   

As the children learned to interpret the world visually, Sinha used fMRI scans to observe how their brains changed throughout the process. Each brain displayed remarkable plasticity regardless of age. From children as young as five to young adults of 20, the brain showed evidence of reorganization after surgery. “We would have expected that the brain ought to be fairly set in its organization [by age 20], but the brain maintains its plasticity well into late childhood or early adulthood,” Sinha said.  

Furthermore, the brain develops specialization in a matter of months after regaining sight. One region of a typical brain, the fusiform gyrus, responds most strongly to faces. Sinha wondered whether the children learning to see would also develop such a specialized area and, if so, if it would be contained in the same region as in a normal seer’s brain. He found that while there is no clear facial recognition area immediately after surgery, neurons within the fusiform gyrus begin to hone in on faces within five to six months, just as in a normally seeing individual.  

While he will continue his work with Project Prakash, Sinha is currently developing a computerized system called Dylan that will assist in the study of early visual learning. Dylan receives the same visual input as a newborn infant and then tries to understand its meaning. To collect visual data straight from a baby’s perspective, Sinha attached a webcam to his one-month-old son Darius’ forehead. This data was then fed into Dylan’s system, making Dylan’s task similar to that of one of the children treated by Project Prakash.