In this interview, News-Medical talks to Sarah Phillips about new research on how bilingual brains process different languages, and how the same neural mechanism is used for each when interpreting mixed-language expressions.
Please introduce yourself and tell us about your background in neuroscience.
My name is Sarah Phillips, and I am currently a Linguistics Ph.D. student at New York University. Prior to coming to NYU, I had no experience conducting neuroscience research. I studied code-switching—the alternation between languages in discourse—through sociolinguistic and theoretical perspectives. This was and continues to be highly motivated by my experiences growing up as a Korean and African-American English bilingual.
It was not until I was working for an academic publisher that I was exposed to what neuroscientists were saying about the bilingual mind, which seemed more focused on how bilingualism can affect cognition more broadly. I felt like there was a gap in understanding how bilinguals process languages, so that is where my journey in neuroscience began.
How are languages processed by the brain?
I think the field recognizes how many regions of the brain are used to process language, with each region hosting mechanisms responsible for specific tasks. Liina Pylkkänen, senior author of the paper and my advisor, has been systematically investigating the left anterior temporal lobe’s role in how we derive meanings through composing words together. This is just one of many tasks that participate in language processing.
Image Credit: TypoArt BS/Shutterstock.com
What did you find about the combining of words from different languages by the brain, and would does this tell us about the brain and bilingualism?
My work extends much of what we know about how we combine words from a single language to what many bilinguals do, which is to also combine words from different languages.
We found that bilinguals recruited the left anterior temporal lobe to combine words from the same language as well as from different languages without exhibiting effects of language-switching. I think this tells us that both monolinguals and bilinguals use similar neural mechanisms to derive meanings through composition.
How did you study this?
We used magnetoencephalography (MEG) to record Korean/English bilinguals’ brain activity as they were presented with two words, followed by a picture. The two words varied in three ways: they varied by whether they were composable (“icicles melt” vs “jump melt”); they varied by whether the words were in English or Korean (“고드름 melt” vs “뛰어 melt”); they varied by whether the Korean words were in Hangul (the standard writing system for Korean) or in the Roman alphabet (which is the same set of letters used in English).
After seeing both words, our participants saw a picture and were asked to indicate whether what they read matched the picture they saw.
What is magnetoencephalography (MEG) and how was it used in your study?
I should preface by saying that the neurons in our brain emit electrical signals when they communicate with each other. These signals generate an electrical current, and electrical currents generate electromagnetic fields.
MEG is a technique that records changes in electromagnetic fields produced by the neuronal firings in our brain. We used this technique to identify when, in milliseconds, particular brain regions activate in response to the two words that participants were presented.
What parts of the brain were activated when testing bilingualism?
While several regions were activated in response to the mixed-language expressions, the timing of when they were activated is important in this study. If words are presented in a combinatory context (“melt” presented after “icicles”), we would expect that they would elicit increased activity in the left anterior temporal lobe at ~200 ms when compared to the same words presented in a non-combinatory context (“melt” presented after “jump”).
We saw this effect in our bilingual participants, even when words switched languages. While the left anterior temporal lobe did not show sensitivity to switching languages or writing systems during this combinatory stage of processing, we did see robust effects in several regions elicited by a correspondence between switching languages and switching writing systems.
The left anterior temporal lobe, the left inferior frontal gyrus (better known as “Broca’s Area”), the anterior cingulate cortex, and the ventromedial prefrontal cortex all responded more whenever language-switching corresponded to switches in writing systems but at different timings. This leaves open the question of how bilingualism may affect all of these regions.
Image Credit: decade3d - anatomy online/Shutterstock.com
If the same neural mechanism is used for interpreting mixed-language expressions as for interpreting single-language expressions, how does the brain keep different languages separate?
Current ideas about how bilinguals store words in their lexicons (which one can imagine being a mental dictionary) suggest we maintain language distinctions at the perceptual level (i.e., how words sound or are written).
I think the interaction effects between language membership and writing systems speak to this, but I am not sure how we keep languages separate at the perceptual level based on the results of this project.
Why is understanding the nature of bi- and multilingualism important?
Many, if not most, people in the world engage in more than one language on a daily basis, and yet our current understanding of how our brains process language is limited to those who engage in only one language.
I believe that developing models of language processing that are more linguistically inclusive is critical for how successful our models are when applied in clinical settings.
What impacts could this research have when applied to today’s society?
One of the biggest takeaways is that bilingual behavior should not be viewed as deviant or bad because they do not resemble monolingual behavior.
That said, I think this project serves as a starting point for understanding typical bilingual behaviors so that bilinguals are not misperceived as poor language users or, even worse, misdiagnosed for having a language processing deficit.
What is the next step for this research?
Languages can differ widely in how they structure words together into sentences (syntax), but this study used stimuli that kept syntax consistent within each stimulus type. I did this to ensure that switching effects were not confounded by syntactic differences.
However, I think it is important to see what happens when bilinguals switch languages in expressions where the two languages are syntactically different as this is a commonly observed phenomenon.
Where can readers find more information?
Readers can access the paper using the following link: https://doi.org/10.1523/ENEURO.0084-21.2021
About Sarah Phillips
Sarah F. Phillips is currently a Linguistics Ph.D. Student at New York University. Her work focuses primarily on bilingual language processing and bilingual language development, using both neuroimaging and behavioral methods.
More information about Sarah can be found on her website: www.sarahfphillips.com
Professor Nancy Ip: Pioneering New Paths in Neurodegenerative Therapy