March 20, 2000
Volume 52, No. 25
Berns studies language wiring in human brain
By Lillian Kim
Scientists generally agree that humans are unique in possessing the gift of language, distinguishable from all other animal communication by its rules of grammar and syntax. These rules allow us to express abstract concepts such as the past or the future.
How we learn language is still unclear, however. Some linguists contend that our brains are inherently wired with an understanding of grammatical rules. Others say we develop this understanding through exposure to language during infancy.
A study conducted by researchers at Emory and Georgia Tech contributes fresh fodder to the debate. The study's results, published in the March 1 issue of the Journal of Neuroscience, indicate that Wernicke's area, a small region of the brain's temporal lobe long known to be involved in language processing, also monitors nonverbal predictability. This discovery implies that our ability to acquire language may stem from the more fundamental ability to subconsciously recognize predictable events.
"This study suggests indirectly that the part of the brain that we know is critical for processing language has a more generic function in determining the predictability of events," said neuroscientist Gregory Berns, senior author and principal investigator. The lead author is Amanda Bischoff-Grethe, who was a postdoctoral fellow in neurology.
An assistant professor of psychiatry and behavioral sciences at Emory University and an assistant professor of biomedical engineering at Georgia Tech, Berns specializes in studying the mechanisms of learning in the nonverbal parts of the brain.
Berns and his fellow researchers originally set out to find which part of the brain is active while processing the predictability of events in a sequence. They used a deceptively simple test and magnetic resonance imaging to observe brain activity in people who were shown predictable sequences versus those shown random sequences.
Study participants were shown a monitor screen displaying four squares arranged in a row. The squares lit up one at a time in one of three randomly selected colors: blue, red or yellow. The participants were asked to keep a mental count of the number of blue squares that appeared.
This task actually was a ruse, aimed at distracting the subjects from the fact that sometimes the squares were illuminated in a repeating sequence. For comparison, some people were shown random sequences. Of those who were shown the predictable patterns, some were previously told that a pattern might exist, while others weren't.
The researchers found that Wernicke's area monitored the predictability of the patterns, whether the subjects were aware of it or not. This finding is significant because for over a century, Wernicke's area has been associated only with the processing of spoken and written language.
In 1874, German neurologist Carl Wernicke first noted that people who suffer stroke or other damage to this particular region of the brain consequently display a condition called aphasia: They can neither speak comprehensibly nor understand the speech of others.
The discovery of a predictability-determining role for Wernicke's area suggests that our ability to make sense of language is rooted in our ability to recognize syntax, defined as the orderly arrangement of parts or elements of a system. Those parts or elements don't necessarily have to be words or phrases; the repeating pattern of colored boxes used by the Emory researchers is essentially an artificially created syntax, albeit an extremely simple one.
"Language itself is a very predictable thing, because it is constrained by the rules of grammar and syntax," Berns said.
By contributing to our knowledge of the neurological basis of language, this study also may eventually lead to better understanding and treatment of dyslexia. Wernicke's area appears to function differently in people with this common learning disability.
"There are some hints that dyslexia is more related to being able to separate events in time," Berns said. "This study would support that."