Brain imaging reveals new language circuits

The language network of the brain seemed simpler in the past. One brain area was recognized to be critical for the production of language, another for its comprehension. A dense bundle of nerve fibers connected the two.

But there have always been naysayers who pointed to evidence that failed to fit this tidy picture. Now a study employing a powerful variant of magnetic resonance imaging (MRI) confirms these suspicions. The study will be published December 13, 2004 in the online edition of Annals of Neurology. “We were surprised that the two classical language areas were densely connected to a third area, whose presence had already been suspected but whose connections with the classical network were unknown,” said lead author Marco Catani, M.D., of the Institute of Psychiatry at King’s College London.

The authors dubbed this language area “Geschwind’s territory” in honor of the American neurologist Norman Geschwind who championed its linguistic significance decades ago.

Language is generated and understood in the cortex, the outermost covering of the brain. Paul Broca and Carl Wernicke, 19th Century neurologists, noted that damage to specific cortical areas, which came to bear their names, produced primarily language production or language processing disorders, but not both. A large bundle of nerve fibers was found to connect Broca’s and Wernicke’s areas, and damage to this pathway also produced language disorders, or aphasias.

However, even in the 19th Century, there were bits of evidence that other brain areas play some role in language, though these have remained enigmatic, as scientists could not use animal models to probe language networks in the same way they could visual or movement networks in the brain. In the last few decades, advanced brain imaging techniques such as CT, PET, and more recently, MRI have allowed scientists to begin studying these areas in living humans.

Standard MRI, by itself a powerful innovation, shows the major tissue structures of the brain. A variant called “functional” MRI even allows researchers to identify which areas are being used during different tasks, including producing and comprehending language. Diffusion tensor (DT) MRI has gained prominence in the past decade because it reveals in greater detail the nerve fiber connections through which different brain regions form communication networks.

With DT-MRI, Catani and his colleagues found a separate, roundabout route that connects Broca’s and Wernicke’s areas via a region in the parietal lobe of the cortex, which Geschwind had pointed out as an important language region already in the 1960s.

“There are clues that the parallel pathway network we found is important for the acquisition of language in childhood,” said Catani. “Geschwind’s territory is the last area in the brain to mature, the completion of its maturation coinciding with the development of reading and writing skills. An important future line of study will be to examine the maturation of this area and its connections in the context of autism and dyslexia.”

The fact that these pathways appear to exist – in more rudimentary forms – in the brains of monkeys may also have bearing on the search for the evolutionary origins of language. “These data suggest that language evolved, in part, from changes in pre-existing networks, not through the appearance of new brain structures,” said Catani. “This method provides another example of the remarkable versatility of MRI technology,” said Marsel Mesulam, M.D., of Northwestern University in Chicago, Illinois, whose editorial will accompany the print publication of the article.

“It is theoretically possible to combine diffusion tensor imaging with functional MRI so as to reveal the connectivity of brain areas with identified specializations,” said Mesulam. “This method can be applied anywhere in the brain. Revealing the connections of the human brain will constitute the next frontier in the field of cognitive neurology.”