I Can Hear You Whisper (26 page)

Discourse refers to units of language larger than a sentence. A favorite of the deconstructionists of the 1980s, it's a relatively new field of study that looks at writing, conversation, etc. It is discourse, I realize, that is the aim of a literate, educated person. “The assumption is that you get that for free,” says Poeppel, meaning that if you learn all the more basic parts of language, you will achieve discourse as well. I'm not so sure about that, since one of the emerging areas of concern for cochlear implanted children is how their language develops beyond elementary school, when it needs to take on sophistication and subtlety as they read to learn and begin to write essays. In the brain, discourse would fall into what Poeppel and Hickok call the widely distributed “conceptual network,” meaning we bring to bear much of our neural resources to tackle it.

For all of these language systems, the scientific focus has been on mapping, on figuring out “where stuff is.” But that is not enough, says Poeppel. “We have more blobs, better blobs. But our yearning should be bigger. Namely, what's an explanation for what's going on there? What is the mechanism? A mechanism is some kind of account or explanation for how a set of elements interacts to generate something. That's not what we have.”

In the search for a mechanism, linguists and neuroscientists are asking if there is a hierarchy to these subsystems of language. They are searching for what Poeppel calls the “primitives,” the primary colors of language with which everything else is constructed. For my purposes, thinking about language, and thinking about Alex, in terms of this new list of parts, I see how each of them contributes to the next. Phonology is an obvious problem for a child who doesn't hear well, but morphology figured in many of Alex's progress reports. If he couldn't hear or say parts of words that had meaning, like the “-ed” on a past-tense verb or the “s” of a plural, he wouldn't always understand or be understood. Semantics showed up in the constant effort to expand his vocabulary. Prosody had previously been drowned out for me by the drumbeat of concern over perception and production, but I could see how helpful it is because of how integral it is to the ability to make sense of the stream of language. “One of the absolutely elemental things you have to do for comprehending language is you have to segment it,” says Poeppel. “If you can't segment it, you can't actually look up the words at all. If you can't look up the words, your syntax isn't going to be all that great, either.”

“So how can we use this information to help a child?” I want to know, even though I recognize that Poeppel spends his time in the lab, not the classroom.

“The syllable has practical ramifications,” he says. “From the get-go, what we would want to give a kid are these cues in the signal. The syllable is a segmentation cue that provides rhythmic information and that's easy to remember. That's super-useful. The information comes prepackaged.” And the brain apparently makes use of that fact. Various studies have shown that the brain gives a bit of extra effort to syllables that occur at the beginning of words, something researchers call the “
word onset effect.” Poeppel has also shown that the brain resets itself to track the syllable rate of a speaker.

How can we help children train that ability and lay down those circuits? One way is to let them explore the rhythms of language. Happily, that is something that comes naturally or that was instinctually built in to many children's stories and songs.
Usha Goswami, an educational neuroscientist at Cambridge University, studies whether dyslexia may be related to difficulties in sound processing. Her research suggests that a return to some old-fashioned wordplay would be useful long before children learn to read. “Nursery rhymes are perfect little metrical poems,” she says. “We know children love them, and they enjoy things like singing in time to music.” Engaging in such activities seems to help develop the necessary phonological and prosodic networks. Her laboratory is also studying the benefits of having five-year-olds learn poetry by heart.

Goswami even got me thinking about the neurobiology of
The Cat in the Hat
. “Repetition matters, too,” she says. “That's true of all learning by the brain, the more repetition the better. Dr. Seuss hasn't just got rhyme, he's also got repeating phrases. The child who's good with syllables will be quicker to get onset rhymes and will be quicker to acquire phonemes once they acquire abstract units like letters. That's why developmentally you want to start with amplifying the syllable level by bringing in all this rhyme. That should put the child in a better position to begin reading.”

As soon as I got home, of course, I read to Alex about the cat who came to play on that cold, cold, wet day.

I
n one corner of a classroom at a
Head Start facility in Eugene, Oregon, three preschool children sit at a small table trying to color a frog green without going outside of the lines. It's not easy for them and they have to concentrate. In the opposite corner, another group of children are playing with balloons, batting them back and forth and trying to keep them aloft. These three, with the balloons, have been assigned the role of Dr. Distractor, and their job is to try to get the children at the table to stop paying attention to their work. Over the next eight weeks, every Tuesday night, the two groups will swap roles regularly, and whoever is in the distracting group will move closer and closer to the table until, by the last session, they are practically on top of those doing the drawing. To maintain concentration, the poor four-year-olds at the table have to marshal all their resources, looking only at their papers, thinking hard about their work. At one point, a little girl raises her hand to physically hide the other children and their balloons from her view.

The entire process is an exercise in applied neuroplasticity and it represents a very modern way of thinking about how to help children.

The old way can be summed up in a story that
Mike Merzenich, who made his name studying plasticity, told me. In the 1960s, around the time that David Hubel and Torsten Wiesel were conducting the research on kittens that would establish the concept of the critical period, one of Merzenich's relatives, who taught elementary school in Wisconsin, was honored as the national teacher of the year. The family gathered to celebrate her accomplishment and she gave a small speech. “She stands up and says, ‘My secret was that I figured out, on the basis of testing, what children were really worth my attention and I gave them everything.'” Merzenich was so bothered by that statement that he's never forgotten it. “What do you think the kids that weren't worth her attention got?” he asks. “Nothing.”

The story is proof, for Merzenich, of why it's so important to understand exactly how the brain changes with experience. “[Hubel and Wiesel] did fantastic science, and the description of the critical period was brilliant,” he says, but he argues that the cast on the interpretation—the idea that the window of development slammed shut so firmly—was wrong. “It had very destructive negative consequences. What it meant in American society—and world society—was when you came to the schoolhouse door, what you saw was pretty much what you had. Kids were in a sense doomed to their genetic fate. It was imagined by pediatricians and child psychologists and schoolteachers everywhere that your brain was fixed and it was all about compensation for the circumstances in front of you.”

Merzenich's groundbreaking work with monkeys in the 1970s, at the same time as he was contributing to the invention of the cochlear implant, showed that brains can and do continue to change throughout life. “It's a revolution,” he exclaims, “understanding that in a sense we are continuously remodelable, that in fact, our fate and our ability are under our direction and charge. To understand that we can at any point improve, bodily improve, the things we do in life and that this science can be applied to change the outcomes of children on a large scale, this is revolutionary. It will come into every aspect of human societies—pretty soon it will be everywhere.”

One place the understanding has already arrived is the classroom. The effort to take what we know about brain development and make use of it in schools has been christened “
neuroeducation.” From studies of executive function in toddlers to research on math ability in adolescence, all of the work takes as its foundational principles that learning is driven by brain circuitry and that brain circuitry is wired by experience.

The transition out of the laboratory is not always smooth. Not every basic scientist wants to wade into the classroom or the home and apply what has been learned. Those who do sometimes seem to be tripping over one another in the race to create scientifically based curriculums and video games (Merzenich was one of the first to market such a product, a reading intervention program called Fast ForWord). For their part, educators are not always happy to see the scientists. The two groups don't always speak the same language or share the same immediate goals; a school principal focused on seeing test scores go up will not want to wait for a double-blind study from which some children, by design, won't benefit. Furthermore, educators have grown suspicious of perennial promises of the next big thing in learning as well as defensive over complaints about the job they are doing. Nevertheless, it's undeniable that researchers today do know far more than they used to about how children learn and how their brains develop.

To talk about some of the knowledge underpinning this new trend and see how it might apply to Alex, I turned to Helen Neville at the University of Oregon and the program she and her colleagues have been working on with a local Head Start for nearly a decade. Given that Neville set out to change the world back in the 1960s, it's not surprising that she thinks it's incumbent on scientists to try to use their work to improve children's lives. “Now we're seeing where the rubber meets the road,” she likes to say.

“It all starts with basic research,” she tells me. “You can't jump in and pull an intervention out of the air. You have to know what systems show neuroplasticity, what ones are vulnerable, how fixed they are, how changeable they are, and what are their mechanisms, so you can target them. I've been studying this for thirty years. I've kind of come full circle, I think.”

The connection between preschoolers trying to pay attention and deaf adults exhibiting better vision in the periphery may not be immediately obvious, but there is actually a straightforward plotline to Neville's story of scientific discovery, and her areas of focus have a special resonance for me. She began by studying the brains of deaf and blind people because their experiences were so unlike those of hearing and sighted people. “You had to start with a population that would give you a good chance of finding a difference,” she says. From there, she moved to people who had had slightly different experiences, such as second language learners or those who learned sign language instead of spoken language. And finally, she looked at typically developing children who have different experiences “just by virtue of being different ages or in different stages of cognitive development.”

All of the work is ongoing, but after thirty years Neville has sketched out a nuanced picture of how the brain's systems of vision, audition, language, and attention change with experience. As Mike Merzenich maintains, the window does not necessarily shut firmly all at once. But neither, emphasizes Neville, does it stay open indefinitely for every skill. The details of what she calls “
profiles in plasticity” matter. Some brain systems are so hardwired that they are the same whether a person hears or sees or not. Some change considerably with experience, but only during particular windows in development, which are in turn determined by the specific system in question—hearing has one, language turns out to have several. And then there are certain parts of the brain that continue to be shaped by experience throughout life, “with impunity,” says Neville. Knowing all of this has allowed her to chart “a
road map of plasticity.”

 • • • 

It was a map I thought would be useful to have in hand for what it could tell me about the remaining uncharted territory of language learning. Elissa Newport had given me a glimpse of how the brain approaches learning language. David Poeppel laid out a plan for how language works once we're good at it. Helen Neville was going to connect the dots.

No matter how many languages a child learns before the age of seven, they will all operate in the brain in a similar fashion once they are acquired. Additional languages learned later are processed differently, and Neville tries in part to pinpoint those differences. Work from her laboratory and others has demonstrated that there are different profiles in plasticity for phonology, syntax, and semantics. As a result, scientists have created something like an evolving account of the development of language in the brain, allowing us to follow along as a child's linguistic capabilities mature. The definition of an adult response can vary, depending on the skill—some responses become smaller and more efficient over time, others more widespread. What matters is the change in the way neurons do their work, a sign that they are wiring together with use or being pruned from disuse.

It won't surprise anyone who has mangled the sounds of a second language—that is, most of us who didn't start to learn until high school—to know that the window for phonology, the ability to recognize speech sound contrasts, opens and closes early. Generally, we will have an accent in any language we learn past the age of seven. As babies, we are already working on the underlying skill, learning to distinguish the sounds of our native language—the ones we'll need—from sounds we don't need. This is the process by which English babies learn the difference between “r” and “l,” but before the end of their first year, Japanese babies let that distinction fade, since Japanese doesn't contain those sounds. Furthermore,
the better babies are at responding to the contrasts of their native language at the ripe old age of seven and a half months, the more proficient they are at language as toddlers: Their word production and sentence complexity at twenty-four months are higher and so is the mean length of their utterances at thirty months. In Neville's laboratory, they found that
in thirteen-month-olds the brain response to known words differs from that to unknown words and is broadly distributed over both hemispheres.
By twenty months, the effect is limited to the left hemisphere, more as it is in adults, reflecting increased specialization, vocabulary size, and maturity.

The ability to chop up language into usable chunks—to segment it—is also different in late learners. They can do it, of course, or they wouldn't be able to make any sense of the new language, but they do it much more slowly. The “word onset effect,” part of the N1 that comes early in native speakers, isn't as visible in late learners.

Starting early with language doesn't just get rid of accents, it also makes it much easier to handle grammar and produce the appropriate sentence structure fluently. We acquire those syntactic skills nearly as early as we master phonology. An early study by Elissa Newport shows this starkly. In
English-Korean speakers who came to the United States at varying ages, scores on a test of English grammar drop off steeply between the ages of seven and seventeen. At two, children's brief sentences already reflect the word order of their native language, so those learning English know that “Daddy” comes before “eats” and “pizza” comes after. (This is the very skill that Janet Werker attributed to prosody.) Between the ages of three and five, children are already experts at such grammatical fine points as how to describe something that happened in the past, how to ask a question, how to say that you don't want or like something. They hit these milestones no matter what language they are learning, which, as Neville has pointed out, “supports the proposal that language learning has a significant biological basis.”

Differences don't just show up in second language learners. How proficient you are in your native language is evident in your brain as well. It's a rather infamous fact that much cognitive research is done on college students—if you are a researcher at a university, students are handy and cheap. But are they typical? When he first started reading up on the literature describing how the brain was organized for language, Neville's colleague Eric Pakulak found it fascinating until he looked at the methods section in one particular paper. “It's twenty undergraduates from Harvard,” he says. “That's not really representative.” As it happens, Pakulak plays rugby with a group of men who do represent a wide swath of the socioeconomic spectrum. So he decided to test a group that included some of his teammates and others who weren't university students. He looked at two particular responses in the brain that reflect syntactic processing (grammar): the early anterior negativity, a wave that falls somewhere between a hundred and three hundred milliseconds and is relatively automatic; and the more controlled P600, the late wave thought to reflect repair and reanalysis. In higher-proficiency speakers, Pakulak found that the early response was more efficient and more concentrated. He hypothesizes that this “frees up” mental resources for the later response, which is larger and more widespread.

For those who miss out on developing the early responses, there are alternatives.
Pakulak also studied Germans who spoke English as a second language. Because of the consistency of the German educational system, all had begun learning English around the age of eleven or twelve—late in brain terms—but all were good at it. (This is the university bias at work again: All the subjects were undergraduates, graduate students, or professors.) Pakulak found that the late start on English meant the Germans lost the benefit of the early response, but they made up for it by putting more brain areas to work later in the process. “They're using different resources to achieve a different
level of proficiency,” he explains. “They're using more controlled processes because they don't necessarily have the access to these early and automatic processes that are more constrained by differences in experience.” Those results reminded me of what Greg Hickok told me: that anyone, like Alex, who got less input, was going to have to use more top-down processing. The same apparently was true of German graduate students who wanted to study in the United States.

Here's the good news: We can keep learning new words in any language for as long as we like. “This is a system that continues to change throughout life,” says Neville. “It can be set up in a native-like fashion even if you start learning a language at the age of thirteen or fifteen.” In Chinese-English bilinguals who started learning English either very young, as preteens, or later, the systems in their brains that handle semantics look “completely identical no matter when English is learned.”