Monolingual infants expect others to understand only one language, an assumption not held by bilingual infants, a study by researchers at New York University and McGill University has found.
"Our results not only offer insight into infants' perception of linguistic abilities, but, more importantly, may help us better understand whom they see as good communication partners," explains Athena Vouloumanos, an associate professor in NYU's Department of Psychology and one of the study's co-authors.
"Knowing who might make a good communication partner may enhance learning about the many aspects of the world that we learn about from others, including our native languages," adds co-author Kristine Onishi, an associate professor at McGill University.
Their findings appear in the journal Cognition.
Adults of course recognize that others can understand multiple languages. However, it's less clear if infants share this type of perception.
To explore this matter, the researchers examined the responses of both monolingual and bilingual 20-month-olds as they observed a series of interactions between adults with whom the infants were unfamiliar. Here, two adult speakers told an adult listener the location of a ball hidden inside cups using either the same (English or Spanish) or two different languages, which included English and another language (French and Spanish).
Following verbal instruction in one language, the adult always found the ball. Then, in one version of the scenario, the adult following the verbal instruction from a second speaker searched correctly for the ball; in a second version, the adult searched incorrectly (the infants had previously seen where the ball was hidden so knew its correct location).
The researchers employed a commonly used method to measure infants' expectations: looking time. Previous research has shown that a longer gaze indicates that infants see something they did not expect and therefore visually engage with it longer.
Their results showed that infants' expectations about whether the unfamiliar adult was monolingual or multilingual varied with the infants' own language background. For instance, after the listener gave evidence of understanding one language (by searching for the ball in the correct location), both monolingual and bilingual infants looked longer when the listener then searched incorrectly after receiving information from a second speaker using this same language. The longer look suggested the infants expected the adult to seek out the ball in the other (i.e., correct) location. However, when information was provided in two different languages, only monolingual infants looked longer when the listener reached correctly; in contrast, bilingual infants looked equally at both outcomes.
That is, monolingual infants, surprisingly, did not expect the adult to understand a second language, even when this second language was the infants' own language--for example, English-speaking monolingual infants who saw an unfamiliar person respond correctly to Spanish did not then expect that person would understand English.
"The monolingual infants assumed that an unfamiliar person would understand only one language while bilingual infants did not, suggesting that infants do not expect all speech to convey information to all people," explains Vouloumanos.
The study's other authors included Casey Pitts, an NYU undergraduate at the time of the study and now a student at Albert Einstein College of Medicine.
The study was supported by New York University's Dean's undergraduate research fund to Casey and the Social Sciences and Humanities Research Council of Canada.