The Prodigy's Cousin (6 page)

Bethany and Michael were both extraordinarily talented. Both endured traumatic birth experiences, evidenced a voracious hunger for knowledge, developed at breakneck speed, and spontaneously heard—and maybe even composed—music. But because Bethany had a specialty in which she reached professional status before the age of ten, she qualified as a prodigy, while Michael, who graduated from college at ten (not technically a professional), did not.

It's an imperfect definition. It's based on accomplishments (an external rather than an internal guidepost), and the age cutoff—which was eventually softened—is somewhat arbitrary. Feldman didn't even really mean for it to set the bar for prodigiousness—at least not initially. “
I think it had some value,” he said. “I didn't do it for the purpose of having it be a definition that would work for the field for all time. I did it because of what I was up to.”

After Joanne met Garrett James and his autistic cousin Patrick, she graduated from Case Western and then hopscotched through a series of academic positions.
Along the way, she wrote up her work with Garrett, but she kept quiet about any potential link between prodigy and autism. After all, her idea that the two were somehow connected was little more than a hunch.

That hunch was a highly unorthodox one. In 2005, when Joanne was on the cusp of initiating a small prodigy study, no one searching for the underpinnings of child prodigies' talents had suggested that they might have something to do with autism.

But Joanne was determined to test the waters. Over the years, she had contacted the families of talented children she read about online or in the newspapers. She spoke with the mother of a child who was memorizing books at fourteen months and began adding at eighteen months. She reached out to the parents of a little girl whose paintings were selling for tens of thousands of dollars before she hit double digits. They were fascinating cases, but were they prodigies?

Feldman's prodigy definition was a good starting point. But Joanne thought that any connection between prodigy and autism was likely genetic, so she wanted to use a standard that was more closely focused on factors intrinsic to the child and less oriented around a particular level of achievement.

The thing that seemed most critical, she thought—the dagger that
cut to the heart of what it meant to be a prodigy—was the accelerated development of talent during childhood. That insatiable drive, which the psychologist Ellen Winner has described as a “
rage to master,” seemed less dependent on attributes of the child's field or parents; it seemed more likely to have a distinct biological engine behind it.

Such swift achievement often attracted a fair bit of limelight. A hodgepodge of people from journalists to scientists clamored to applaud children who skyrocketed to the top of a typically adult field. National or international acclaim by adolescence, Joanne reasoned, might serve as a proxy for the runaway-train development pattern she thought characterized prodigy. It was still a behavior-based definition, but by focusing less on whether the child had achieved professional-level success, she tried to create a threshold more closely aligned with attributes intrinsic to the child.

Interestingly, the imprecision of using behavioral criteria to make a diagnosis is something autism researchers know all too well.
Autism, like prodigy, is diagnosed based on external, observable symptoms, not genetic tests, blood samples, or brain scans.

Austism was described in the 1940s by two researchers, Leo Kanner and Hans Asperger, working on two different continents.
Kanner was a psychiatrist who founded the Johns Hopkins Children's Psychiatric Clinic, the first such clinic in the country.
In 1938, Kanner met Donald T., a five-year-old from Mississippi. Donald paid no attention to the people around him, and his mother reported him happiest when left alone. Objects, on the other hand, fascinated him. Donald would spin anything and jump in excitement as he watched it rotate. His days were packed with ritual and repetition. When he spun a block, he started with the same side facing upward; he always threaded buttons in the same sequence. He had an excellent memory for faces and names, and he memorized a large number of pictures from the encyclopedia. His speech consisted almost entirely of parroting phrases he had heard before; his words often seemed unrelated to what was going on around him.

Over the next few years, Kanner saw several other children exhibiting the same “
extreme autistic aloneness.”
Such children were often labeled feebleminded, idiotic, or schizophrenic, but Kanner was convinced that what he was seeing was a unique condition.
He published a paper reporting on this new condition in 1943; the next year, he named the syndrome early infantile autism.

Hans Asperger had the same idea at almost the exact same time. While working
at a children's clinic in Vienna, he encountered a number of children who he believed shared a syndrome marked by social withdrawal and communication difficulties, such as problems with eye contact or an unusual speech pattern, as well as a high degree of creativity. Asperger emphasized that this syndrome could affect individuals of all levels of intellect, including the “
highly original genius.”
He began using the term “autism” to describe what he had seen as early as the 1930s, and in 1944 he published a paper arguing that the collection of behaviors he had observed constituted a new, independent condition.
He called it autistic psychopathy.

The conventional wisdom is that Kanner and Asperger were working independently and that their nearly simultaneous identification of autism—and the nearly identical names they gave the syndrome—was a grand coincidence of history.
This view is shifting, however, in light of new evidence that Kanner worked closely with former colleagues of Asperger's and thus might have been more familiar with Asperger's work than he let on.

But regardless of who identified and named autism first (and who knew what when), both Kanner and Asperger placed behavioral abnormalities at autism's core. Kanner emphasized that the children's “
fundamental disorder” was the “inability to relate themselves in the ordinary way to people and situations from the beginning of life.” Asperger thought a key marker of the condition was “
the shutting-off of relations between self and the outside world.”

A diagnosis based solely on behavioral symptoms is inevitably slippery (just as Feldman discovered while selecting participants for
his prodigy research).
In the early autism studies, scientists used differing criteria for autism. Two decades after Kanner first identified autism, he complained that it had become “
a pseudodiagnostic wastebasket for a variety of unrelated conditions.” And this issue continues to plague autism, as demonstrated by the ever evolving
DSM
diagnostic criteria:
autism has gone from a symptom of schizophrenia to a condition independent of schizophrenia;
the diagnostic criteria have shifted over time such that they encompass an expanding number of people.

When work began on the most recent edition of the
DSM,
there was talk of developing diagnoses based on the underlying neuroscience rather than symptoms. But the new
DSM
offered no drastic rethinking. The drafters slotted some diagnoses into different chapters, combined some separate diagnoses into single conditions, added new diagnoses, eliminated others, and refined diagnostic criteria. But fundamentally, the
DSM-5
maintained its symptoms-based approach to diagnosis rather than focusing on the underlying biochemical mechanisms.

The entry for “Autism Spectrum Disorder” is a combination of what had been four distinct diagnoses (autistic disorder, Asperger's disorder, childhood disintegrative disorder, and pervasive developmental disorder not otherwise specified) in the previous version. According to a fact sheet published by the American Psychiatric Association (the organization that publishes the
DSM
),
this change was made in part because researchers didn't think that these diagnoses were applied consistently across clinics and in part because it was thought that a single diagnosis better reflected the idea that autism symptoms vary in severity. But neither of these reasons has much to do with whether these symptoms actually stem from one distinct underlying abnormality or four.

This emphasis on behavior leaves scientists in a quandary. As the prominent autism researcher Uta Frith once observed, “
Behaviour, however reliably it is measured, is not revealing about its cause. There is a mapping problem. Many different causes can underlie the same
behaviour. On the other hand, behaviour that looks different in different individuals may actually be due to one and the same cause.”

Beyond the challenge of pinning down the definitions of prodigy and autism, Joanne faced another hurdle: her research pockets were empty, so she couldn't travel to the children and their families. She had tried to get NIH funding, but her application had been rejected.

She still wanted to do
something,
though, to at least find out whether her theory had merit. She didn't need ironclad proof. She just needed something quick and dirty that could be done on the cheap to explore whether there was anything to the idea that autism and prodigy were connected.

The answer came from abroad. The Autism Research Centre, an organization nestled within the University of Cambridge's psychiatry department, operates as a hub for scientists studying autism. At its core was Simon Baron-Cohen, one of the most famous and prolific autism researchers in the world.
In 2001, he and his colleagues had published the Autism-Spectrum Quotient, or AQ, a test that measured a handful of traits associated with autism.

It was a snappy little test. With just fifty questions, it was quick and to the point, and it was designed to be self-administered. Most important, it was free.

Joanne thought she could use it as a canary in the mine for her theory.
Autists' family members often demonstrate autistic traits, just not at a level that would merit an actual autism diagnosis. Joanne reasoned that if there was a connection between prodigy and autism, maybe the AQ would show that the prodigies' relatives, too, had some autistic traits, just like the autists' relatives. She decided that if they did, that would be her sign; she would launch a full investigation into the connection between prodigy and autism.

An autism researcher Joanne had known since high school agreed to make the AQ available to her patients and their families. For a
control group, a local school distributed the AQ to families whose children had no documented disabilities. Joanne mailed the AQ to the families of some of the prodigies—those children who showed the accelerated development she was looking for—with whom she had been in contact over the previous few years. She wound up receiving ten completed surveys from each group.

The responses—though relatively few—were intriguing. The autists' relatives and the prodigies' relatives'
AQ results indicated that both groups had elevated levels of some autism-linked traits. Both had more difficulty with “attention switching” (the ability to multitask, switch between activities easily, and embrace spontaneity) and were less drawn to social situations than the control families.

But the headline result, the one that Joanne thought might offer a real clue to the relationship between autism and prodigy, came from the families' scores in attention to detail, another autism-linked trait. The relevant AQ questions measured the degree to which the test takers noticed small changes and absorbed patterns, dates, or other tidbits of information. In this category, the prodigies' families spiked. They demonstrated significantly greater attention to detail than the control families
and
scored higher than the autists' relatives.

The results reassured Joanne that she was onto something. The connection between prodigy and autism, the one she had suspected since seeing Garrett and his cousin Patrick together seven years before, was not a figment of her imagination. But there was also a fascinating twist. The prodigies' families' and the autists' families' results were not identical. The prodigies' relatives seemed to have less trouble with conversation and social niceties than the autists' relatives, at least according to the AQ. It was a curious inconsistency. It seemed that prodigy and autism were connected in some—but not all—respects. A summary of Joanne's findings was published in the fall of 2007 in the academic journal
Behavior Genetics,
revealing the first hints of a connection between prodigy and autism.