The Learning Curve


The New York Sun
March 22, 2004 Monday
Copyright 2004 The New York Sun, One SL, LLC

Armed with notepads and video cameras, a team of New York University researchers intently monitored 13-month-old Victoria Uberti as she crawled across a platform at their Washington Place lab. Moments earlier, while Victoria was distracted, they had replaced part of the platform with a treacherous piece of soft foam. Now they were luring Victoria into their trap with a colorful xylophone and a handful of Cheerios.

On any given day at NYU's infant and child labs and Columbia University's infant cognition lab at Teachers College, babies, toddlers, and preschoolers like Victoria are falling into foam, watching colorful cartoons, and helping friendly robots, all in the interest of advancing science.

"Every parent that brings in a baby is helping us resolve a mystery," a psychology professor at NYU, Gary Marcus, said. But if a pure love of knowledge isn't incentive enough, he said, parents should consider the practical applications of research like his studies of language acquisition, such as discovering treatments for language disorders or developing computer programs that can more successfully interact with people. "Those are the reasons that we do what we do," he said.

One of Mr. Marcus's ongoing studies examines how children understand verbs, specifically how they comprehend who does what to whom. Preschoolers watch lab assistants use two colorful stuffed animals to act out a pair of made-up verbs, one describing pushing down on someone's shoulders, the other, pulling someone's arms back. Then the children watch the two characters in a series of cartoon snippets while answering questions to help a friendly "robot" (actually, the video monitor) figure out what's going on.

Parents who visit the labs for such experiments can't resist asking researchers to evaluate their children's intelligence. "But we're not looking at individual performance," said a Columbia speech and language pathology professor, Peter Gordon, who oversees the Teachers College lab. "You can't tell anything from individual babies. The idea is to see what human babies are capable of at a given age." And with new technology and more sophisticated experiments, he said, "We can really begin to ask, 'What's going on in a baby's head before language even arrives there?'"

The study of the childhood mind has been hot since the 1990s, when a series a series of magazine cover stories asked, "How smart are babies?" It's now well established that children are more precocious than we'd realized, but much more study is needed to figure out just how they learn. Finding willing participants is "half the work," said an NYU graduate student, Keith Fernandes, who manages Mr. Marcus's lab. The labs seek a diverse mix of families, but none break down their research by race or socioeconomic status. "We're trying to talk about abilities which are human, not specific to race, gender, or environment," Mr. Fernandes said.

For the most part, the labs recruit from purchased call lists, but word of mouth and personal contacts also help, said an NYU psychology professor, Karen Adolph - Mr. Gordon's wife - who runs NYU's Infant Motor Development Lab. Ms. Adolph met Victoria's mother, Maria Uberti, at a children's birthday party, and recruited her for a "longitudinal study," in which toddlers come to the lab once every three weeks from the age of 10 months to 15 months. At Victoria's most recent visit, after some brief playtime with the research team, she stripped down to a onesie and her trials began.

At the center of Ms. Adolph's lab is an ingenious elevated platform made of movable wooden sections, with adjustable ramp and handrail attachments. With the camera recording, Victoria crawled across the walkway several times to reach toys and Cheerios awaiting her at the other end. After these "solid trials," a section of the platform was replaced with an equal-sized piece of foam.

But Victoria had seen this before. Approaching the foam on her first soft trial, she pulled up, tested it with one foot, and then stopped cold, despite the rewards waiting for her at the end. After two more trials with the same result, the pit was replaced with solid walkway and Victoria happily crawled to the end again.

"You're a genius! That was a perfect session!" shouted psychology graduate student Amy Joh, who will use this study in her dissertation. "Doesn't it make you feel good that she learns stuff and remembers it?" she asked Ms. Umberti.

"She's surprised me," Ms. Uberti said. "She sees that [foam], and she says, 'I think I'll go off to the side.'"

That's the idea, Ms. Joh said. "Children are learning about environmental clues and using them to avoid falling." Ms. Joh has put adults on the trick platform as well, and says, "You'd be surprised - they fall the first time, too."

Capturing the cognitive skills of the youngest children requires more sophisticated technology. Children as young as 4 months old are involved in "looking time" studies at NYU and Columbia, in which advanced camera arrays track a baby's pupils to determine how much attention he pays to varying sounds or images. In other words, "We try to get them bored," Mr. Fernandes said. Once children have mastered, or "habituated," a visual or sonic pattern, they pay less attention to it. The cameras record that, and the researchers use frame-by-frame tracking to make their conclusions.

In one current test of infant visual cognition at NYU, babies watch a green circle float across a monitor and disappear behind a blue square at the center of the screen. An adult would perceive the ball as "persistent" and look for it to come out the other side, but will an infant anticipate the ball's return, or imagine it to be gone? (Four-month olds generally do not, although many 6-month olds do.)

Most of the labs' studies require a minimum of 16 to 24 subjects to generate useful results, but several more children need to be recruited for a successful study. Although most experiments take less than 15 minutes, about a quarter of 2-year-olds simply refuse to participate, Mr. Fernandes said, and about one in six infants "fuss out" because they're tired or hungry or just not interested in sitting still. He will ask parents to try to coax a child into taking part, but "We don't push them. We just like them to have a good experience." The labs generally reimburse parents the cost of their subway trip or parking, and will also try to babysit a sibling who has tagged along. After an experiment, the children receive whimsical diplomas, along with a school-logo toy, T-shirt, or photo.

"I've been trying to get some of my friends to come in," Ms. Uberti said, but she's found that many New York parents are leery of having their children studied or videotaped. "But Victoria loves it. She has a good time. And I try to resist thinking she's brilliant, but in my heart, I know."