Written language poses a puzzle for neuroscientists. Unlocking the meaning in a string of symbols requires complex neural circuitry. Yet humans have been reading and writing for only about 5000 years—too short for major evolutionary changes. Instead, reading likely depends on circuits that originally evolved for other purposes. But which ones?
To investigate, cognitive neuroscientist Stanislas Dehaene of the Institut National de la Santé et de la Recherche Médicale in Gif-sur-Yvette, France, teamed up with colleagues in France, Belgium, Portugal, and Brazil to scan the brains of 63 volunteers, including 31 who learned to read in childhood, 22 who learned as adults, and 10 who were illiterate. Those who could read, regardless of when they learned, exhibited more vigorous responses to written words in several areas of the brain that process what we see, the group reports online today in Science. Based on previous work, Dehaene has argued that one of these areas, at the junction of the left occipital and temporal lobes of the brain, is especially important for reading. In literate, but not illiterate, people, written words also triggered brain activity in parts of the left temporal lobe that respond to spoken language. That suggests that reading utilizes brain circuits that evolved to support spoken language, a much older innovation in human communication, Dehaene says.
It makes sense that reading would rely on brain regions that originally evolved to process vision and spoken language, says Dehaene. But this repurposing may have involved a tradeoff. The researchers found that in people who learned to read early in life, a smaller region of the left occipital-temporal cortex responded to images of faces than in the illiterate volunteers. Dehaene suggests that reading may compete with other tasks—such as face perception—for access to this part of the brain. If so, could learning to read make people worse at recognizing faces? Experiments to test this are already under way, but Dehaene says he doesn't expect to see a huge difference.
All in all, the findings support Dehaene's argument that the left occipital-temporal cortex is a region where the brain has made important adaptations for written language, says Brian Wandell, a neuroscientist at Stanford University in Palo Alto, California. Researchers have long thought that the brain becomes less flexible with age, so it's also interesting that this region seems to change even in adults who learn to read, he says. "Showing that this area is responsive to learning throughout life is a real contribution."
"Dehaene and his group have done a remarkable job," adds Manuel Carreiras, scientific director of the Basque Center on Cognition, Brain and Language in Donostia-San Sebastián, Spain. "It is really striking that schooling profoundly refines cortical organization." But Carreiras notes that not everyone accepts Dehaene's hypothesis that the left occipital-temporal cortex is specialized for reading. The suggestion that literacy competes with face perception in this part of the brain is likely to be controversial as well, he says. "The results ... will definitely cause discussion in the scientific community."