Class Act

Talking baby

The new French documentary "Babies" makes a compelling (and cute) case that we're all pretty much alike in the beginning. Tracking the first year of four far-flung infants, it shows that those reared in urban San Francisco or Tokyo are as likely to delight in a discovery or strive mightily toward their first steps as their counterparts in rural Mongolia or Namibia.

Despite their similar early development, each of these babies will soon speak a language that's completely unintelligible to the others, says Tilbe Goksun, a developmental psychology graduate student at Temple University.

"How do they end up learning these very different languages? That is what inspires my research," says Goksun, who is doing her work in the lab of psycholinguistics expert Kathy Hirsh-Pasek, PhD, and in collaboration with Roberta Golinkoff, PhD, of the University of Delaware.

Goksun, a native of Istanbul, Turkey, became interested in the area while earning her master's degree at Istanbul's prestigious Koc University. As she studied grammar acquisition in children, she was compelled to dig deeper. "I wanted to go backwards and see which prelinguistic foundations impact upon language learning," says Goksun, who besides Turkish speaks fluent English and some French.

Language is fascinating to her on a personal level, as well. She learned English from both British and American teachers, so she developed a hybrid of the two that she had to modify when she came to the United States. In fact, during her first days at Temple, she recalls going to lunch with her lab group who slung around cultural references so rapidly, "I had no idea what they were talking about!" she says.

But she was determined to connect, and today she easily knows the difference between Lord & Taylor and Ann Taylor. "That's all part of what makes this work really interesting to me," she says.

The importance of verbs

What makes the research by Goksun, Hirsh-Pasek and Golinkoff so interesting is that they are studying verb learning in a field previously dominated by research on noun learning.

"If we don't understand how children learn verbs, we will never understand how they learn sentences or master grammar," because in essence, verbs drive sentences, Goksun says.

To map this area, she and her mentors are drawing on the theories of linguist Leonard Talmy, PhD, of the State University of New York at Buffalo. He suggests that sentences can be broken into conceptual chunks — the figure performing the action, the goal of the action or the ground where the action takes place, for instance. Depending on the language, verbs incorporate these elements in different ways, making the arrangement of these building blocks an intriguing way to study language and verb acquisition, Goksun says.

In her dissertation, Goksun is comparing babies' relationships with ground-related verbs in Japanese and English. The Japanese use different verbs depending on the type of ground where an action occurs. If the ground is bounded, like a road, the Japanese use one type of verb. If it lacks clear borders or barriers, like a field, they use different ones. These distinctions are not present in English, where the verb "crossing," for instance, applies equally to a road or a field.

Goksun's research team collaborated with a Japanese colleague, Mutsumi Imai, PhD, of Keio University, to conduct a series of experiments testing whether English and Japanese babies at 9 months, 14 months and 19 months see the same figure and ground components in events and whether that perception changes over time.

First, the babies watched four quick rounds of a film showing a woman crossing a road. Next, they watched a split screen, with one side showing the original event and the other showing a woman crossing either a bounded space like a road or railroad tracks or an unbounded one like a field or tennis court. A camera recorded the babies' eye movements.

Because babies will look longer at an event they haven't seen before, the researchers could tell that, at 9 months, the infants didn't distinguish between the two events — they spent the same amount of time looking at both sides of the split screen.

At 14 months, all of the babies looked longer at the second screen only when it showed an unbounded ground — suggesting that both Japanese and American babies learned to distinguish between bounded and unbounded space. But, at 19 months, something curious happened: While their eye movements showed that Japanese babies continued to see bounded and unbounded space as fundamentally different, American babies were much less likely to distinguish between them.

This finding suggests that babies lose touch with concepts that are not encoded in their language, says Goksun, though, as other researchers show, they don't lose them entirely. "We and others argue that you just dampen your attention to these constructs because your language doesn't use them," she says.

What's more, the researchers found that these differences weren't related to age per se but to language experience: Same-age American babies with bigger vocabularies lost the distinction more quickly than those with smaller vocabularies, further demonstrating that language learning can change how we see the world.

While Goksun calls these "pretty good results," adviser Hirsh-Pasek is more glowing.

"Tilbe is helping to carve the field both theoretically and empirically about how [babies' perceptions] shift over time as they abut against the language system," she says. "That's headline news."

In addition to Goksun's theoretical contributions, her work could have practical applications, perhaps by pointing to new ways to teach second languages. She also says the findings may apply to understanding why children with autism have trouble learning verbs. That could lead to developing better ways of teaching these youngsters.

Charting new directions

For her next step, Goksun will conduct fMRI research to investigate potential differences in the way the brains of speakers of various languages map the same basic information. She'll have to settle for adult participants, however, because it's tough to conduct brain-imaging studies with babies.

The switch is not entirely unwelcome, chuckles Goksun.

"Working with babies is fun, but you have to be patient," she says. "You have to wait a lot for the data because the discard rate is so high" — the result of babies who are too active to code properly, for instance. Nevertheless, she knows she'll return to infant research in an effort to keep putting the early puzzle pieces into place.

Ultimately, Goksun also wants to return to Istanbul, a hustling, bustling wealth of life surrounded by two seas.

"It's home, and I love it," she says. "You have a feeling of ancient history there, and the food is better!"

"I'm a good cook," she adds, "but I can't get all the ingredients I need here."

She does, though, have the ingredients she needs to make it to the top of her field, adviser Hirsh-Pasek says.

"Tilbe has the drive for discovery, the intellectual force and care, and the creativity that allows us to see things that are otherwise hidden," she says. "Those are vital qualities in baby research."


Tori DeAngelis is a writer in Syracuse, N.Y.