More than a century ago, a French neurologist suggested that a specific region of the brain processes the visual images of words. Without it, he postulated, people cannot read except by laboriously recognizing letter after letter, rather than whole words. Yet humans have only been able to read for several thousand years--perhaps not enough time for such a trait to evolve, some scientists have argued. New research, however, supports the idea that reading does rely on a localized set of neurons.

Previous imaging studies using functional magnetic resonance imaging (fMRI) or positron emission tomography (PET) showed that a small region buried deep in the left rear of the brain lit up with activity when subjects read, or recognized words, as opposed to perceiving other objects, such as faces or tools. Victims of stroke with damage in this region often reported reading difficulty. But because stroke damage in these patients was never confined to this region alone and imaging studies can only demonstrate correlation, not causation, controversy persisted.

Neurologist Laurent Cohen of the Hopital de la Salpetriere and his colleagues received a rare opportunity to explore this hole in scientific understanding when a 46-year-old epileptic came to them for treatment. His chronic seizures indicated that a small portion of his brain--roughly contiguous with the so-called visual word-form area--should be removed.

Prior to removing the damaged section, the scientists performed a series of tests on the man, including a wide array of reading challenges and the temporary placement of electrodes in his brain. He proved normal in all regards, including his ability to quickly recognize words no matter how many letters they contained.

But two weeks after the operation, though cured of his epilepsy, the patient complained of difficulty reading and tests showed that his ability to comprehend longer words had slowed by half. Even six months later, he needed roughly an additional 100 milliseconds for each additional letter to recognize a word.

The finding seems to support the contention that this region of the brain is critical to reading, but it does not answer questions as to how it developed. "One possibility is that the [visual word-form area] performs a visual processing function that predisposed it to being co-opted for reading," Alex Martin of the National Institute of Mental Health writes in a commentary accompanying the paper in today's Neuron. Nevertheless, the French team has provided more evidence that this region is critical to your ability to read this article.