Have you ever wondered whether the process of learning a language is different for people who speak multiple languages compared to those who speak only one? Do transferable skills from learning one language make it easier to learn another? If so, does it matter which of the world’s many languages a person speaks? Those questions all pertain to a concept known as bilingual transfer, and PhD student Xin Sun (Combined Program in Education and Psychology) has been investigating them for her dissertation.
Working alongside Dr. Ioulia Kovelman and a team of undergraduate and graduate researchers, Sun conducted three related research studies. Sun explains that while each study focused on different aspects of English language development, all three explore the ways that “bilinguals—especially Spanish/English bilinguals and Chinese/English bilinguals—learn to read and speak English in different ways compared to monolingual English speakers.”
To understand why Sun and her team chose Spanish and Chinese, one must first understand a couple of major differences between the two languages. For the first study, the most pertinent difference relates to the way each language is written: specifically, the difference between alphabetic and logographic writing systems.
Spanish, like English, uses an alphabet of letters to represent the sounds of the spoken language. That is, the individual components of written Spanish—letters—each correspond to a discrete unit of sound. For example, in Spanish, the letter s represents the sound used to begin the Spanish word serpiente (snake). Those letters are then combined in sequences to represent the sounds of complete words. English also shares this system, albeit with more exceptions and inconsistencies (consider, for example, the pronunciation of the English word island compared to its Spanish equivalent isla). In fact, writing in languages like Spanish or English is so closely related to sound that children often learn to read partly by memorizing the sounds of each letter and “sounding out” new words.
Written Chinese does not work that way. Instead, it is a logographic system, meaning that each individual unit—known as a logogram or character—represents a unit of meaning rather than of sound. Because of that, it is comparatively much more difficult to “sound out” written Chinese. Instead, because of the way Chinese characters are constructed, experienced readers are often able to use their knowledge of other characters to help them guess the meaning of an unfamiliar character without knowing what it sounds like at all.
Based on those differences, Sun hypothesized that the thoughts and techniques of English learners may differ depending on whether they also read Spanish, Chinese, or neither. Specifically, she hypothesized that since native Spanish speakers rely so heavily on sound when first learning to read Spanish, they would also rely more on sound when learning to read English. By contrast, she also hypothesized that because native Chinese readers must develop more skill with manipulating meaning, they would rely more on those skills when learning English (and not as much on sound).
With the first study, which focused on English word reading in children aged 5-10, Sun and her team confirmed those hypotheses. For Spanish speakers, the single strongest predictor of English-learning proficiency was their skill with understanding and manipulating sounds. For Chinese speakers, the greatest predictors were their skills with understanding and manipulating meaning. Monolingual English learners appeared to use both skills more evenly. In other words, the study appeared to confirm that a bilingual child’s other language does indeed influence how they learn English.
With that in mind, Sun and the team then wondered if there would also be measurable differences in the children’s brain functions, a question that became the focus of the second and third studies.
To understand those studies, one must now understand another linguistic difference: this time between two categories of words known as compound words and derived words. Compound words are words formed by combining two existing independent words (known as roots): blackberry, for example, combines the words black and berry, while backpack combines back and pack. A derived word, on the other hand, is a word formed through a combination of an independent word (root) and one of more affixes: linguistic alterations or additions that change the meaning of the word but do not themselves function as independent words. For example, the derived word dishonest combines the root word honest with the affix dis-, thus reversing the meaning of the original root. Similarly, a word like unnaturally uses two affixes, one (un-) to reverse the meaning of the root honest, and another (-ly) to convert the resulting word into an adverb.
Chinese and Spanish are very different with regard to their use of compound and derived words. The vast majority of Chinese words are compounds, while Spanish uses a very rich derivational system to generate many of its words. English serves as a useful middle ground: English has fewer compounds but more derived words than Chinese does; and English has fewer derived words than Spanish does.
Based on those differences, Sun and the team wondered: would the brain processes of Spanish/English bilinguals, Chinese/English bilinguals, and monolingual English learners differ? Specifically, given the prevalence of compound words in Chinese, would learning Chinese also make children’s brains more automatic in processing English compounds? And given the prevalence of derivations in Spanish, would the brains of Spanish/English bilinguals find it easier to process English derivations?
To answer those questions, Sun and her team had children perform linguistic tasks while wearing a cap that measures brain activity in several key language-processing regions. In certain parts of the brain, especially parts of the frontal lobe, greater neural activity is thought to indicate greater effort—that is, the more activity that happens there, the more difficult the person finds a task.
Interestingly, when compared to each other, all three groups appeared to show a similar amount of neural activity when processing compound words (works like blackberry or backpack), with a small amount of extra frontal activity seen in Spanish/English bilinguals. Sun speculates that processing compound words may simply be an inherently easier task because compound words are easier to dissect to their word roots; therefore, processing compounds may not require sufficient cognitive effort to result in drastic differences between the groups.
However, the differences were more pronounced for derived words (words like dishonest or unnaturally). Both Chinese/English bilinguals and English monolinguals showed significantly more frontal activations when processing derived words than when processing compounds. Spanish/English speakers, on the other hand, actually showed less activation in those frontal regions and more activation in other regions more associated with more “automated” language processing. It is possible—though not yet proven—that Spanish/English bilinguals’ exposure to and proficiency with the rich and productive derivation system of the Spanish language accounts for that difference.
Finally, in a third study, Sun and her team investigated the relationship between bilingual children’s overall proficiency in their heritage language—either Spanish or Chinese—and the amount of brain activity shown when processing English words. Interestingly, both groups showed the same pattern: greater proficiency in either heritage language correlated with greater activity in a specific region of the temporal lobe that is involved with auditory (and likely more “automated”) language processing. That is, regardless of which of the two very different languages the children learned, greater mastery in their heritage language somehow results in greater activation of that same temporal region of the brain when learning English. This intriguing relationship suggests that the neurocognitive effects of bilingual transfer may not always be related to the specific characteristic of one’s heritage language.
Sun’s research has important potential applications for understanding how to teach English most effectively, both to bilingual and monolingual speakers. Going forward, she plans to expand the research to include a broader sample of learners—in terms of age, reading skills, familiarity with English, and other factors—to “dig into the more complex picture,” she says, “including how bilingual children from many different backgrounds and circumstances become successful language users”.
