You don’t have to be a scientist to help researchers at Princeton study the human mind. In fact, you don’t even have to know how to walk or talk. The Princeton University Baby Lab, studies the psychology and cognitive skills of children from newborns up to the age of 7 — how young children learn to see, talk, and understand the world. On August 7 the group published a study that shows that even infants can learn two languages at once without being confused.
In the study, an international team of researchers, including those from Princeton University, report that bilingual infants as young as 20 months of age efficiently and accurately process two languages.
The study, published in the journal Proceedings of the National Academy of Sciences, found that infants can differentiate between words in different languages. “By 20 months, bilingual babies already know something about the differences between words in their two languages,” says Casey Lew-Williams, an assistant professor of psychology and co-director of the Princeton Baby Lab, and also the co-author of the paper.
The Princeton Baby Lab is led by Lew-Williams and Lauren Emberson, both assistant professors of psychology at the university. Their team of researchers includes postdoctoral researchers, doctoral students, and undergraduates.
“As a developmental scientist, I’m interested in the beginnings of learning in the earliest years of life,” Lew-Williams says. “How young children begin to make sense of the complexities of information coming their way.”
Lew-Williams says that while there are many people at Princeton University and at the Princeton Neuroscience Institute who study learning, there were not many studying the nature of learning in infants, toddlers, and children before the Baby Lab was established in 2014.
To conduct their studies, Baby Lab researchers need plenty of subjects, or in other words, children from throughout the Princeton area whose parents are willing to drive their kids to the lab to take part. At the start, this meant recruiting potential subjects at area preschools, farmers’ markets, community days — anywhere that parents of young children might be on hand to hear the Baby Lab pitch.
The recruiting effort was successful. Today they have around 1,000 children in their database, and they are adding more all the time. Parents who are interested in signing their children up can fill out a simple form on the lab’s website, babylab.princeton.edu, after which someone from the lab will call to ask a few questions and explain how the program works. “If parents are willing and if their children meet the criteria for these studies then we go ahead and include them,” Lew-Williams says.
The Baby Lab has around 12 studies going on at any given time, pursuing questions about early language learning, early visual development, and early communications. The researchers require between 40 and 150 children for any individual study, so a large pool of potential subjects makes it increasingly possible for them to address the questions they are trying to ask about learning across the early years. All center on the question of how infants “break into the social structure,” Lew-Williams says.
Appointments last around a half hour and are designed to be fun and child-friendly. For most studies, children are asked to listen to sounds, view pictures, or engage in play sessions with researchers. For their participation, children receive free toys or T-shirts, and their parents modest monetary compensation.
Lew-Williams says parents often express interest in what the researchers are hoping to accomplish in their sessions and what the experiments can tell them about their children. He and the Baby Lab staff are happy to share whatever information they can, although he makes it clear that they are not a clinic.
“We do not do assessments of children — that’s what pediatricians do, that’s what speech-language pathologists do,” he says. “We don’t compare and contrast different kids who come in. That said, if parents want information about their children’s vocabulary levels, if they want to ask questions about the experiments we’re doing or how their children performed, we will certainly tell them.”
Lew-Williams grew up in Davis, California, where his parents both worked at the University of California, Davis, in various capacities. “I grew up surrounded by a culture that values higher education,” he says. He earned his bachelor’s degree in psychology from Berkeley in 2004 and a Ph.D. in psychology from Stanford University in 2009.
Before joining Princeton’s psychology department in 2014, he was an assistant professor in Northwestern University’s Department of Communication Sciences and Disorders. From 2009 to 2012, he was a postdoctoral researcher at the University of Wisconsin–Madison.
He says one goal of the Baby Lab studies is to understand the different ways that language learning proceeds for children from low-income families versus high-income families. “There are documented divergences in children’s language learning trajectories based on those factors,” Lew-Williams says. “One project we’re doing is using intervention to understand if providing higher quality talk in the home can give rise to accelerated vocabulary levels. If you’re better able to process information coming your way in the beginning of life, it makes sense that you’d be able to make more and more of the knowledge coming your way. Learning is a cascading process, building on itself.”
In the case of the bilingual study, the researchers found that young children do not think that “dog” and “chien” [French] are just two versions of the same thing. “They implicitly know that these words belong to different languages,” Lew-Williams says.
To determine infants’ ability to monitor and control language, the researchers showed 24 French-English bilingual infants and 24 adults in Montreal pairs of photographs of familiar objects. Participants heard simple sentences in either a single language (“Look! Find the dog!“) or a mix of two languages (“Look! Find the chien!”). In another experiment, they heard a language switch that crossed sentences (“That one looks fun! Le chien!”). These types of language switches, called code switches, are regularly heard by children in bilingual communities.
The researchers then used eye-tracking measures, such as how long an infant’s or an adult’s eyes remained fixed to a photograph after hearing a sentence, and pupil dilation. Pupil diameter is an involuntary response to how hard the brain is “working,” and is used as an indirect measure of cognitive effort.
The researchers tested bilingual adults as a control group and used the same photographs and eye-tracking procedure as tested on bilingual infants to examine whether these language-control mechanisms were the same across a bilingual speaker’s life.
They found that bilingual infants and adults incurred a processing “cost” when hearing switched-language sentences and, at the moment of the language switch, their pupils dilated.
However, this switch cost was reduced or eliminated when the switch was from the non-dominant to the dominant language, and when the language switch crossed sentences.
“We identified convergent behavioral and physiological markers of there being a ‘cost’ associated with language switching,” Lew-Williams says.
Rather than indicating barriers to comprehension, the study “shows an efficient processing strategy where there is an activation and prioritization of the currently heard language,” Lew-Williams says. The similar results in both the infant and adult subjects also imply that bilingual people across the lifespan have important similarities in how they process their languages.
“Language is a very complex structure that requires computations over sounds, over words, over sentences, people,” Lew-Williams says. “It’s also very high stakes, because communicating is essential to what it is to be human. Bilinguals are an especially interesting case because young learners take these two streams of information and learn to make sense of them. It’s a puzzle to try to understand how the infant mind could be capable of cleaning up the complexities. Parents everywhere are interested in bringing up bilingual questions.”
ur understanding of bilingual language use in exciting ways — both in toddlers in the initial stages of acquisition and in the proficient bilingual adult,” says Janet Werker, a professor of psychology at the University of British Columbia, who was not involved with the research. She notes that the findings may have implications for optimal teaching in bilingual settings.
“One of the most obvious implications of these results is that we needn’t be concerned that children growing up bilingual will confuse their two languages. Indeed, rather than being confused as to which language to expect, the results indicate that even toddlers naturally activate the vocabulary of the language that is being used in any particular setting,” she says.
Lew-Williams suggests that this study not only confirms that bilingual infants monitor and control their languages while listening to the simplest of sentences, but also provides a likely explanation of why bilingual speakers show cognitive advantages across their lifespan. Children and adults who have dual-language proficiency have been observed to perform better in “tasks that require switching or the inhibiting of a previously learned response,” Lew-Williams says.
“Researchers used to think this ‘bilingual advantage’ was from bilinguals’ practice dealing with their two languages while speaking,” Lew-Williams said. “We believe that everyday listening experience in infancy — this back-and-forth processing of two languages — is likely to give rise to the cognitive advantages in both bilingual children and adults.”