Researchers at Harvard and the University of California, San Diego have discovered that a small region of the brain that has been predominantly associated with language production, is also responsible for language comprehension—blurring the lines on what has been a fundamental concept in psychology.

The findings debunk the long-standing assumption that Broca’s area—which is commonly linked to language production—serves a completely distinct purpose from Wernicke’s area—the region responsible for language comprehension.

“We have long suspected that this classical model of language was too simple,” said Psychology Professor Steven Pinker, who co-authored the study with Nedim T. Sahin, a post-doctoral fellow in the UCSD radiology department and in the Harvard psychology department.

For their study, researchers focused on three epilepsy patients undergoing the rare procedure Intra-Cranial Electrophysiology, in which electrodes are implanted into the brain to gauge which regions are responsible for causing epileptic seizures.

“It was kind of an accidental experiment,” Pinker said. “We were fortunate that neurosurgeons implant electrodes in the brains of patients with epilepsy.”

During this study—the first to use ICE to study how the brain interprets grammatical rules and produces word—researchers had the patients read a series of words and then reproduce them in different grammatical forms: for example, the inversion of a noun into a plural, or the conversion of a verb into the past tense.

Researchers found that three different linguistic computations—lexical (recognizing the word), phonetic (articulating the word), and grammatical (converting the word to a plural or past tense)—all occurred within small regions of Broca’s area. Brain activity indicating the computations occurred at roughly 200, 320, and 450 milliseconds after the word was presented to the patient.

The discovery that the computations occurred sequentially, all within about half a second of each other, refuted the idea that Broca’s region is only involved in language production.

“This might be a final nail on the coffin of the old model,” Sahin said.

Since different parts of language computation occur in different parts of the brain, the researchers plan to further investigate whether there is synchrony, or simultaneous electrical activity, in both Wernicke’s and Broca’s regions.