Studying people who speak 45 languages, neuroscientists found similar patterns of brain activation and language selection.
For decades, neuroscientists have produced a well-defined map of “language network,” Or regions of the brain dedicated to language processing. Found primarily in the left hemisphere of the brain, this network includes regions within Broca’s area and other parts of the frontal and temporal lobes.
However, the majority of such mapping research is done by English speakers as they listen to or read English texts. MIT neuroscientists have now done brain image study of people who speak 45 different languages. The results show that the language network of the speakers seems to be similar to that of native English speakers.
Although not surprising, these findings suggest that the location and key properties of language networks appear to be universal. This work also lays the groundwork for future studies on linguistic factors that would be difficult or impossible for English speakers to study because English does not have those features.
“This research is very fundamental, extending some of the findings from English to a variety of languages,” said Evelina Fedorenko, Frederick A. and Carole J. Middleton, Associate Professor of Neuroscience at MIT and Fellow. of MIT’s McGovern Institute said Research. “The hope is that now that we see that basic properties appear to be common across languages, we can ask about potential differences between languages and language families in the way they are implemented in brain and we can study phenomena that don’t really exist in English. “
Fedorenko is the lead author of the study, which appeared in Natural neuroscience. Saima Malik-Moraleda, a PhD student in the Science of Technology and Speaking and Listening program at Harvard University, and Dima Ayyash, a former research assistant, are the lead authors of the paper.
Language network mapping
The exact location and shape of language regions varies between individuals, so to find the language network, the researchers asked each person to perform a language task while scanning their brains. them by functional magnetic resonance imaging (fMRI). Hearing or reading sentences in one’s native language activates the language network. To distinguish this network from other brain regions, the researchers also asked participants to perform tasks that shouldn’t trigger it, such as hearing a foreign language or solving math problems.
A few years ago, Fedorenko began designing these “localized” tasks for speakers of languages other than English. While most studies of language networks use English speakers as subjects, English does not include many features commonly found in other languages. For example, word order tends to be fixed in English, while in other languages the arrangement of words is more flexible. Instead, many of these languages use the addition of spheres, or segments of words, to convey additional meanings and relationships between words.
“There is a growing awareness of the need to look at more languages over the years, if you want to make statements about how languages work, as opposed to how English works,” said Fedorenko. “We thought it would be very helpful to develop tools that would allow people to rigorously study language processing in the brain in other parts of the world. Access to brain imaging technology is now available in many countries, but the basic models you’ll need to find areas of language response in a person are not available. “
For the new study, the researchers performed brain imaging of two people who speak 45 different languages, representing 12 different language families. Their goal was to see if key attributes of the language network, such as location, degree of left, and selectivity, were similar in those participants to those whose native language was English. You or not.
The researchers decided to use “Alice in Wonderland” as the text that everyone would listen to, because it is one of the most widely translated works of fiction in the world. They selected 24 short passages and 3 long passages, each recorded by a native speaker. Each participant was also listened to nonsensical passages that should not have activated the language network, and asked to perform various other cognitive tasks that should not have activated it.
The team found that the language networks of these study participants were found in roughly the same brain regions and with the same selectivity, as did the linguistic networks of native English speakers.
Malik-Moraleda says: “Language areas are selective. “They won’t respond during other tasks, such as space working memory tasks, and that’s what we found in the 45-language speakers we tested.”
In addition, language regions that are normally activated together in English speakers, such as frontal and temporal regions, are similarly synchronized in speakers of other languages.
The researchers also showed that among all the subjects, the amount of small variation they saw between speakers of different languages was the same as the amount of variation commonly seen among native English speakers. .
Similarities and differences
While the findings suggest that the overall architecture of the language network is similar among speakers of different languages, that doesn’t mean there aren’t any differences, Fedorenko said. For example, researchers can now look for differences in speakers of languages that primarily use morphemes, rather than word order, to help determine the meaning of a sentence.
“There are all sorts of interesting questions you can ask about morphological processing,” says Fedorenko.
Another possibility is investigating whether language speakers who use differences in tone to convey different word meanings will have a linguistic network with stronger links to areas of the auditory brain that code. high chemistry.
Currently, Fedorenko’s lab is conducting a study in which they are comparing the ‘fields of acquisition over time’ of speakers of six different languages in terms of typography, including Turkish , Mandarin and Finnish. The temporal intake field measures the number of words a language processing system can process at a time, and for English it is six to eight words long.
“The language system seems to be working on paragraphs of just a few words long, and we’re trying to see if this restriction is common in other languages we’re testing,” Fedorenko said.
The researchers are also working on creating language localization tasks and looking for study participants who represent languages other than 45 from this study.
Written by Anne Trafton
Source: Massachusetts Institute of Technology
