Three British men who suffered left brain damage that undermined their capacity to speak and understand language still possess a firm grasp of mathematics, a new study finds. This observation dramatically illustrates the presence of separate brain systems for language and numbers, at least in adults, say neuroscientist Rosemary A. Varley of the University of Sheffield in England and her coworkers.
The findings, however, are unlikely to resolve a long-running debate over whether children use language to develop their number sense. Some researchers argue that initial math insights arise from knowledge of the words for numbers or of grammatical rules for arranging words in phrases. Other scientists suspect that, from infancy on, language and math follow different mental and neural paths.
“I believe that dedicated brain mechanisms exist [from the start] for language and mathematics, but others on my team disagree with me,” Varley says.
An account of the new investigation will appear in the March 1 Proceedings of the National Academy of Sciences.
The men who participated ranged in age from 56 to 59. Burst or injured blood vessels had damaged left brain tissue at least 3 years before the researchers tested the men’s skills.
Participants exhibited little facility with language and were especially poor at grammar. They spoke only in single words and sentence fragments. Moreover, they couldn’t distinguish the meanings of simple sentences with subject and object reversed, such as “The lion killed the man” and “The man killed the lion.”
Yet each participant readily solved mathematical problems, including those that require applying number-combination rules that the researchers view as analogous to grammatical rules for combining words and phrases. For example, the men successfully calculated answers to equations with reversed terms, such as 59-13 and 13-59. They also solved problems such as 36/(3 x 2), which have a structure similar to a sentence with dependent clauses.
The new study “provides additional evidence that mathematics and language are functionally and neuroanatomically independent,” remarks neuroscientist Elizabeth M. Brannon of Duke University in Durham, N.C., in an editorial to be published with the study.
Such research, Brannon notes, can’t exclude the possibility that the evolution of grammar in language provided a basis for mathematical rules or that children use early language discoveries as a foundation for learning about math.
Knowledge of mathematical rules by infants or nonhuman animals has yet to be demonstrated.
Investigators of two Brazilian cultures that lack words for numbers greater than 5 have concluded that language is necessary for people to conceive of and manipulate numbers. In tests, members of both cultures could identify large quantities that had been formed from two smaller quantities, say, 30 and 20, even though they couldn’t name them. But precise arithmetic problems stumped these people.
Those findings don’t necessarily show that language is necessary for mathematical thinking, contend two researchers in the January Trends in Cognitive Sciences. Rochel Gelman of Rutgers University in Piscataway, N.J., and Brian Butterworth of University College London say that people who don’t need more than a handful of discrete numbers in their everyday lives may still be using a rudimentary, language-independent counting system.