University of Delaware Office of Public Relations The Messenger Vol. 6, No. 1/1996 Out of the mouths of babes They know a lot. The trouble is, they're not talking. Researchers have long suspected that infants are able to understand a tremendous amount of language before they begin speaking. But, to prove these theories, researchers must be able to probe what infants know and when they know it. And, they must do this using notoriously uncooperative subjects. "If you wait until children start to talk, you've missed the boat," says Roberta Golinkoff, interim dean of the College of Education and H. Rodney Sharp Professor of Educational Studies. In fact, her research has shown that infants saying as few as one or two words can comprehend five- or six-word sentences. "Between the ages of 1 and 3, children progress from not speaking at all to speaking in complete sentences," Golinkoff says. "Where does this explosion of language come from?" Like most of her colleagues, Golinkoff says she believes that language learning is the result of both nature and nurture. But, which is more important, and how do the two interact? Golinkoff says she leans toward the ascendancy of nature, believing that our brains are "hard-wired" in some way to learn language. Infants use sound patterns, syntax and meaning to make sense of language, she explains. Despite this predisposition to language learning, infants will not learn language without the "benevolent guidance" of others. "Human infants come into the world prepared to learn language, but human interaction is crucial," Golinkoff says. Infants need to be talked to regularly by parents, siblings and caregivers if they are to learn how to speak. Golinkoff and others have demonstrated that the use of "motherese," simpler vocabulary delivered in a higher pitch and at a slower pace, assists infants in language learning. In a recent study, Golinkoff and Anthony Alioto, Delaware '92, found that the use of motherese even helped English-speaking adults learn Chinese more quickly. Golinkoff has long been a pioneer in the study of infant language development. She created the infant language lab at the University in 1974, when language research with infants was still unusual. Today, when such research has become more common, Golinkoff continues to blaze a trail. While many researchers concentrate on infants' understanding of nouns, Golinkoff studies their understanding of verbs and grammatical structures. To do this, she developed her own method of testing, using two video screens that show images simultaneously. Only two other infant language labs in the U.S. use this method of research- Stanford and Yale universities-and they adapted it from UD's lab. The infant or preschooler who participates in Golinkoff's studies sits on a parent's lap in front of two television sets. While the parent is blindfolded, the child watches short videos that were created in the campus media services center. The video images are accompanied by a recorded voice, making comments about what the child is seeing on one of the screens. If the child focuses for a significantly longer period on the screen that matches the voice-over, the researchers conclude that the child has understood the language. Various experimental controls ensure that external factors, such as differences in lighting or timing between the two videos, do not bias the outcome. By using videos that require only that the child look at a screen to signal comprehension, Golinkoff can study more complex issues than would be possible if an infant had to follow directions and complete an action. Golinkoff has used this method since the early 1980s to study thousands of youngsters ranging in age from 10 months to 5 years. She often runs as many as seven experiments at one time, using UD students as lab assistants. The experiments can range from testing a 15-month-old's comprehension of verbs (does she know the difference between "bouncing" and "blowing"?) to testing a preschooler's understanding of intransitive verbs (does he understand that the sentence "Cookie Monster is bending Big Bird" has a different meaning from "Cookie Monster and Big Bird are bending"?). According to Golinkoff, children use grammatical cues to make sense of language. For instance, "the" signals to infants that a noun will follow, while the ending "ing" signals a verb. Children as young as 16 months can comprehend word order and the relationships they signal. Thus, they can use the subject-verb- direct-object word order to distinguish between the sentences, "Big Bird is tickling Cookie Monster" and "Cookie Monster is tickling Big Bird." Over the years, Golinkoff's research has led to countless papers, presentations, grants and awards, including a coveted Guggenheim Foundation fellowship. The PBS series Human Language highlighted her work in a 1995 broadcast. This year, Golinkoff and her long-time collaborator, Temple University Prof. Kathy Hirsh-Pasek, published their findings in The Origins of Grammar: Evidence from Early Language Comprehension (MIT Press). They plan to follow up this academic text with a book aimed at a more general audience. The National Science Foundation is underwriting Golinkoff and Hirsh-Pasek's current research into the origins of word learning. Twenty years and hundreds of experiments later, Golinkoff says she enjoys her research as much as ever. "It's just so much fun," she says. "I love the ideas, trying to figure out why something happened and what it means." -Theresa Gawlas Medoff, Delaware '94M