The Boy Who Loved Too Much Read online

  People with Williams also proved to be remarkably creative, expressive storytellers. They were intrinsically gifted performers, modulating their vocal pitch, volume, and stress (for example: “OH, my POOOOOOR little wabbit”) to heighten the drama of their stories. Asked to tell a story on the spot, a teenage girl with Williams invented a fairy tale about a chocolate princess in a chocolate world that was in danger of melting, but who saved her kingdom by changing the color of the sun. (The same girl, at fourteen, composed her own music but had the visual-spatial skills of a five-year-old and couldn’t make change for a dollar.)

  One linguist asked kids with either Williams or Down syndrome the question, “What if you were a bird?” Typical answers from kids with Down syndrome were “Fly in the air” and “I not a bird, you have wing.” Kids with Williams were able to offer up more elaborate scenarios, such as “I would fly and if I liked a boy, I would land on his head and start chirping” and “I would fly where my parents could never find me. Birds want to be independent.”

  Bellugi and her colleagues were justifiably impressed. But over time they began to realize that people with Williams were not the linguistic savants they had appeared to be. Their gifts turned out to be limited to certain aspects of language, including the relatively sophisticated ability to grasp metaphors and hypothetical statements, while other elements of language eluded them. They struggled in particular with what is called relational vocabulary and includes concepts such as “under,” “over,” “behind,” and “above,” which parallel the spatial relationships that typically stymie people with Williams. Shown a picture of an apple in a bowl, people with Williams offered a variety of descriptions, including “The bowl is in the apple” and “The apple is around the bowl”—nearly every possible description but the correct one. This discovery posed a direct challenge to the assumption that the brain’s language center was largely insulated from other regions, including the area that processes spatial relationships.

  However, people with Williams also struggled with nonspatial parts of speech, including conjunctions such as “and” and “or,” and disjunctions such as “neither . . . nor.” They had trouble with pronouns. And, despite their chattering, fluent conversational style, they often failed to correctly answer a pointed question, particularly one of the wh- questions: who, what, where, when, and why. One school-age boy, asked when Columbus sailed to America, answered “Niña, Pinta, Santa Maria.” Despite an evident familiarity with the facts of the voyage, he was baffled by a sentence structure most of us have mastered by the age of three. Linguists take failures like these, in contrast to striking abilities with other parts of speech, as a sign that however the Williams brain processes language, it’s very different from the way a normal brain does—and that people with Williams are more impaired than they appear to be in casual conversation.

  In the decades that have passed since Bellugi first encountered a child with Williams, both she and the scientific community at large have come to discount Gardner’s theory of multiple intelligences in the face of compelling counterevidence. Study after study has shown that strength in one area of intelligence does, in fact, predict strength in the others, even in people with Williams. The evidence is unequivocal enough that intelligence researcher Linda Gottfredson could proclaim with confidence, in a 1998 article for Scientific American, “Despite some popular assertions, a single factor for intelligence, called g, can be measured with IQ tests and does predict success in life.”

  Bellugi, now in her eighties, acknowledges that the general intelligence theory appears to have been right all along. The revolution in education and assessment never materialized: old-school teaching and testing styles may be imperfect, but not because of the fundamental assumptions they make about intelligence, as Bellugi now agrees. Contemplating her error in retrospect, she has come to believe that she was so won over by her Williams subjects that she initially overestimated their abilities.

  “It’s my fault. I’m the guilty party,” she confessed cheerfully in an interview. “But in an interesting way. I’m not embarrassed about it. When I started out, a hundred years ago, I was just so impressed with the Williams kids.”

  She’s not the only one to have been sidetracked by the charisma of her research subjects. Just as Eli charmed his evaluators into overestimating his cognitive abilities, people with Williams routinely, if inadvertently, sabotage studies by ignoring the task at hand and focusing all their attention on the researcher instead. A deaf scientist once noted that the people with Williams who came through her lab persisted in talking to her even after she signaled that she couldn’t hear or speak. “They seem to be fascinated, continuing to smile and talk to me, all the time looking right into my face while they try to imitate my signs,” she wrote.

  In one study, people with Williams consistently turned the tables on their interviewer and began asking her questions such as “Do you have a dog?” “What’s your favorite singer?” and “Where were you born?” When asked to describe a significant moment in their lives, several of the interviewees said, “Being here is the best thing that ever happened to me.”

  These studies often reveal less about the nature of Williams than they do about the effects of charm and flattery on a supposedly objective investigator, turning tests of cognition into sociological case studies that demonstrate how easily a warm smile and a kind word can divert our attention and blind us to faults. And they prove that, despite not being perfect speakers, most people with Williams are smooth talkers, abundantly capable of harnessing their strengths to distract from their weaknesses.

  * * *

  THE SECOND BREAKTHROUGH that catapulted Williams out of scientific obscurity was the discovery, in 1993—three years after the official launch of the Human Genome Project—that everyone with the disorder was missing one copy of the elastin gene. Called ELN for short, this gene regulates the body’s production of elastin, the pliable protein that makes skin bounce back to its original position when you pinch it. It’s especially vital in blood vessels, where it provides for stretch and helps facilitate blood flow.

  After pinpointing the elastin gene deletion, scientists raced to study the surrounding DNA. They soon found that people with Williams were missing twenty-six to twenty-eight neighboring genes on one of their two copies of chromosome 7. The missing genes represent a tiny gap in the chain of more than a thousand genes on one half of this chromosome alone. And given that chromosome 7 is just one of the twenty-three paired chromosomes that contain the roughly 20,000 genes that make us human, the gap narrows to what is essentially a paper cut in the human genome, which is why geneticists call it a microdeletion. But its effects are astronomical by contrast.

  A tiny number of people with Williams have “atypical deletions,” meaning they’re missing some but not all of the Williams genes. This subset is so rare that it doesn’t account for the spectrum of abilities among people with Williams that has long perplexed researchers—why some people with Williams have an IQ of 35 and others an IQ of 95, for example, given that all, or mostly all, are missing the same twenty-six genes. Geneticists agree that at least part of the answer has to do with environment and upbringing, and part has to do with the strength of the extant copy of those genes. People with atypical deletions are in high demand as research subjects, though, since they help geneticists zoom in even further on the correlation between missing genes and their effects.

  What makes Williams especially popular with geneticists is that the findings from this rare disorder have far-reaching repercussions for the rest of us. Williams research has so far helped identify genes that play a role in hypertension, low bone density, and high blood sugar, in addition to shedding light on the complicated interplay between gene structure and social behavior.

  “Williams syndrome is a street lamp,” says Dr. Pober, the geneticist who diagnosed Eli and is one of the leaders in Williams research. “It shows us that these characteristics and behaviors are linked to these twenty-six genes. Then the p
roblem is matching up genotype and phenotype: which gene deletion leads to anxiety and phobias, which one leads to musicality. Of course, it’s not always a one-to-one match.”

  As yet, the elastin gene is the only one of the bunch whose function is incontrovertibly accepted by geneticists. Other relationships are still disputed or require further testing. But each of the unique symptoms of the disorder—everything from heart problems and an elevated calcium level to hypersensitive ears and a tendency to develop debilitating anxiety, as well as deficiencies in depth perception, impulse control, planning, and problem solving—could potentially be traced back, at least in part, to the genes that aren’t there. Even the extraordinary abilities that come with Williams, especially in language and music, can paradoxically be traced to those missing genes: people who have an extra copy of the same genes, instead of a missing set, have strong visual-spatial skills but severely delayed speech and expressive language, and they tend to suffer from social phobia. A number of people with a duplicate of the Williams genes, in fact, meet the criteria for autism.

  And while the possible causes of autism are hotly debated, there is compelling evidence of at least some genetic basis. The onetime theory, devastating to many parents, that it was the result of childhood neglect—specifically from uncaring “refrigerator mothers”—has been widely discounted in favor of biological explanations. Research shows that autism is one of the most heritable of all neurological disorders. A study of identical twins found that when one was diagnosed with autism, the other was diagnosed roughly nine times out of ten. Because twins overlap in both nature and nurture, another study examined fraternal twins, who share less DNA than identical twins but, presumably, the same environment. When one fraternal twin had autism, the study found, the other had it between two and three times out of ten. It seemed likely, then, that shared genes more than shared upbringing explained the coincidence. Non-twin siblings, parents, and even distant relatives of people with autism are also more likely to have autism than the general population. So researchers are actively searching for the genes that might play a role, particularly in the region deleted in Williams. But while geneticists have tentatively identified many genes that could be implicated in autism, scientists have yet to agree that any of these plays a definitive role.

  One of the Williams genes is among those correlated, to some degree, with autism. This finding was no great surprise to Williams experts, since people with Williams often demonstrate symptoms more commonly associated with autism: repetitive movements such as rocking, hand flapping, or picking at skin; perseveration, or repeating a word, phrase, or gesture over and over; echolalia, or repeating words that others say (or lines from songs, or fragments of movie dialogue); and a fascination with things that spin—as well as with mechanical objects such as vacuum cleaners. Researchers examining the overlap between the two disorders have concluded that some people with Williams also meet the criteria for autism spectrum disorders, although the proportions have varied widely from study to study, from as few as 5 percent to as many as 50 percent of people with Williams.

  Eli is among the many people with Williams who display some of the classic signs of autism. Even before he could talk, he was so preoccupied with vacuums, floor scrubbers, and spinning things (one of his first words was “twirly”) that he threw volcanic tantrums whenever he encountered one of these beloved objects and was not allowed to stare at it, touch it, or hold it. But he’s never been diagnosed with autism. Dr. Pober considers a double diagnosis both futile and somewhat absurd. Many ailments share symptoms, she argues. Both mononucleosis and depression sap people’s energy, but that doesn’t mean someone suffering from one is also suffering from the other. Likewise, Williams is its own distinct disorder, even if it has some of the attributes of autism. The two disorders even differ anatomically: autopsies have shown that the brains of people with Williams are smaller than normal, while the brains of people with autism are larger than normal.

  A number of genetic disorders, in fact, have symptoms in common with autism. Rett syndrome, which impairs motor function and affects girls exclusively, includes some autistic features, as does fragile X syndrome, which causes intellectual disability. But autistic traits are just part of the larger constellation of symptoms that make each of these syndromes unique. Similarly, a Williams diagnosis trumps an autism diagnosis, Pober says, since—in addition to some shared behavioral issues—Williams also entails health problems that autism doesn’t, which require careful oversight. But she doesn’t argue with the idea that studying the areas where the two disorders intersect, and where they diverge, might hold the keys to a better understanding of both.

  In a 2006 case study of a six-year-old girl who met some of the criteria for autism and some for Williams, DNA testing revealed that she was missing a handful of the Williams genes as well as fourteen other genes “downstream” from those usually deleted in Williams. She was overly friendly with strangers, but she didn’t have the elfin facial features common to Williams. She also lacked the medical problems that usually accompany Williams, and her verbal skills were strikingly worse than usual for Williams. Although affectionate, she had little interest in or aptitude for conversation. Researchers concluded that one or more of her missing genes might have made her susceptible to autism, or exacerbated her autistic symptoms, during the early stages of her development.

  One gene in particular has long been linked to autism, albeit inconclusively: the gene that codes for the protein involved in transporting the neurotransmitter serotonin, which is active in the nerve pathways involved in emotions, sleep, and anxiety, among other things. People with autism commonly have unusually high levels of serotonin in their blood, which has led some researchers to suspect that a variant of this gene might be related to the development of autism (and to a number of psychiatric disorders, including depression, although these links remain controversial). In 2013, geneticists studying the DNA of people with severe autism came across two boys diagnosed with autism who turned out to have the classic Williams syndrome deletion—a surprising discovery, since neither showed any trace of the Williams personality and both were incapable of speech. However, both boys had elevated blood serotonin levels, as well as the same variant of the serotonin transporter gene commonly found in people with autism. The researchers concluded it was possible that this gene variant could have overridden the effects of the missing Williams genes, producing autism symptoms instead.

  In terms of social behavior and language aptitude, of course, people with Williams and people with autism are about as opposite as they could possibly be: hypersocial versus antisocial, loquacious versus laconic. Eye-tracking studies confirm that they look at the world very differently. When shown scenes of people interacting, people with autism spend less time than average watching the faces of the people involved; they pay more attention to inanimate objects. People with Williams, on the other hand, are fixated on the faces of the people in the scenes, which they watch for much longer than normal.

  Studies of language processing in the two groups have revealed dissimilarities that match their social differences. A 2011 study looked at how both groups reacted to sentences ending with a word that fit semantically but was contextually unexpected: for example, “I take my coffee with cream and paper.” Most people, when hooked up to devices that measure electrical waveforms in the brain, show a spike in brain activity when they hear the word “paper,” demonstrating surprise. The researchers expected people with Williams to show a similar pattern, given their language abilities, while they believed people with autism would react less noticeably. They were half-right. People with autism were less surprised than normal, but people with Williams were much more surprised than the control subjects. The experiment showed that both groups process language in unusual ways, but very differently from each other.

  While people with Williams tend to be more proficient speakers, the neuroscientist Carolyn Mervis argues that they have just as much trouble engaging in “successfu
l conversation” as people with autism. In Williams, conversational barriers include overfamiliarity and the persistent repetition of questions or phrases. When they themselves are asked questions, people with Williams have trouble coming up with responses that continue the conversation. Like people with Asperger’s syndrome, people with Williams can’t always tell the difference between relevant and irrelevant information to bring up in conversation. Neither are very good at choosing conversational topics that might interest someone else.

  For all their charms, people with Williams syndrome have as much trouble as people with autism in making meaningful social connections. They just arrive at this trouble from different directions. And, as Gayle learned in the case study she never signed up for, they get there fast.

  Six

  Eli Turns Twelve

  Eli turned twelve in February 2011, but because he wanted an outdoor, pirate-themed party—with a fire in the fire pit—his birthday party was postponed until Memorial Day weekend. Mimi hosted it at the modest ranch house she shared with five other relatives, a twenty-minute drive from Gayle and Eli’s apartment. Two more relatives, Eli’s great-aunt and great-uncle, lived just across the street. Their hilltop cul-de-sac was so quiet that the only sounds before Gayle and Eli’s arrival were the tinkling of wind chimes and the occasional yapping of a neighbor’s Pomeranian. It was safe to assume that when Eli bolted from the car bellowing “WE’RE HERE!” the whole neighborhood heard. It was also likely that they weren’t surprised, since this was how Eli announced himself whenever he came over.

  Eli yanked open the screen door to the house and let it slam behind him. He ran into his aunt Jean—really his eighty-year-old great-great-aunt—on the half flight of stairs to the living room.