Why some siblings with sensory gating issues ended up manifesting the symptoms of schizophrenia and others did not was still a mystery. Freedman’s next step was to try to locate the specific part of the brain responsible for sensory gating. Thanks to DeLisi, he now had access to a family with an unfathomably, overwhelmingly profound manifestation of schizophrenia.
* * *
—
IN FEBRUARY 1986, months after her first visit with the Galvins, DeLisi used data from her families to confirm what Richard Wyatt’s NIMH team had discovered about schizophrenia’s correlation to large brain ventricle size. A year later, she used the data in a study testing a possible link between schizophrenia and human leukocyte antigens, or HLA, a gene complex involved in the regulation of the immune system. No such link was proven. Still, the multiplex family database had begun contributing to the body of knowledge about the disease. As far as DeLisi was concerned, this was only the beginning.
She sent the Galvins’ blood samples to the Coriell Institute for Medical Research, a facility in Camden, New Jersey, that preserves huge collections of cell lines from various diseases. This allowed for the possibility of others using the family’s DNA as a resource in dozens, even hundreds of future studies, conducted in labs around the world. DeLisi held fast to her belief that if she could find a marker for schizophrenia embedded in the genetic data of a family like the Galvins, schizophrenia might one day become like heart disease, an illness with particular benchmarks and risk factors that could be measured. In 1987, DeLisi was recruited away from NIMH by the State University of New York at Stony Brook, which offered her a professorship and a program of her own to run. She kept researching multiplex families there. She had forty already, including the Galvins. With a grant from NIMH, she steadily built on that list, eventually reaching one thousand families—more than anyone else had managed to assemble.
Then came several fallow years. Family studies were yielding amazing results in other diseases, including early onset breast cancer and Alzheimer’s disease, but there was no breakthrough for schizophrenia. In 1995, DeLisi published two studies drawn from her own pool of data on families. The first seemed to confirm that the same genes responsible for schizophrenia are connected to other mental illnesses like depression or schizoaffective disorder. The second failed to find a link between schizophrenia and bipolar illness, at least on one particular chromosome where bipolar illness appeared to be rooted. DeLisi remained confident that someone somewhere could find a genetic fingerprint in this pool—and show that nature, not nurture, determined this condition. “I am not a firm believer in environment having an effect at all,” DeLisi told a reporter in 1999.
DeLisi’s work still had supporters. “It is critical that we avoid premature disillusionment,” Kenneth Kendler, with the Medical College of Virginia, wrote in 1993. “The human brain is very complex and quite difficult to access.” But one of her old colleagues from Richard Wyatt’s lab at NIMH, Daniel Weinberger, started to suspect that researching families was a blind alley. “More than ninety percent of the relatives of schizophrenics do not have schizophrenia according to current diagnostic criteria,” he’d told a reporter in 1987.
Weinberger had a point. The odds of siblings in the same family sharing the condition are indeed low. On the other hand, a sibling of someone with schizophrenia still had about ten times the chance of having the condition as a person in a family without the disease. Compared to the odds of inheriting many other disorders, these odds were extraordinarily high—higher, even, than heart disease or diabetes. Seen that way, it would seem foolish not to keep looking at families.
* * *
—
AT NIMH, THE search for more physical signs of schizophrenia continued, even as the direction of that research seemed almost aimless. Wyatt’s lab had used MRIs to examine identical twins with one sibling with schizophrenia, comparing the size of each twin’s hippocampus. Sure enough, in 1990, they found differences. The hippocampi of the brains of people with schizophrenia were smaller than those without. This finding, like the one about enlarged brain ventricles a decade earlier, seemed to reveal something new about how the disease worked: The hippocampus helps remind you of where you are at any given moment, and it is less developed in the twins that, diagnosed with schizophrenia, have less of a grip on reality.
“We were high as a kite on this stuff,” remembered Daniel Weinberger, who coauthored both of those studies. “But there was a gnawing feeling in the back of my head.” All that this brain research was doing, he thought, was confirming different versions of the same idea: that a schizophrenic brain is physically different from a normal brain. For those who treated schizophrenia patients on a daily basis, this was hardly surprising. “You could talk to these people for five minutes,” Weinberger said, “and you knew their brains couldn’t be functioning the same.”
The MRI studies were seeming less valuable over time—all just pieces of one little corner of a much larger puzzle. Weinberger suspected that the only reason researchers loved them so much was that they had the tools to do them. “One of the things that has characterized psychiatry research forever is the old saying of, ‘Looking for the lost keys where the light is.’ Everything has been, ‘Well, we have this tool. We have a hammer, so we’re going to look for nails.’ And we would find things, because this is the nature of phenomenology—you find things.” Whether they were promising leads or red herrings, no one knew for sure.
In 1987, Weinberger published a theory that went on to change how practically every researcher thought of the illness. Until then, schizophrenia researchers had been fixated on post-adolescence as the moment schizophrenia appears. Brain scans all but confirmed that: The frontal lobe is the last part of the human brain to mature, at the end of adolescence, and MRI studies of the brains of many schizophrenia patients show problems with activity in the frontal lobe. But with his new theory, Weinberger suggested the problems in the brain quietly started much earlier in life. He reframed the conception of schizophrenia as a “developmental disorder,” in which abnormalities that patients possessed at birth, or even in utero, set off a chain of events that, in essence, sent their brains off the rails gradually, over time. All genes did, he said, was establish a blueprint for brain development and function. The rest happened later, in real time, with the help of the environment.
If Weinberger was right, the adolescent phase of brain maturation was simply the final chapter of the story. The brain is having difficulties throughout gestation and birth and childhood, only no one notices anything until the final phase of construction, when the brain is mature. Seen this way, schizophrenia’s onset seemed a little like a bowling ball that veers ever so slightly to the left or right the second it leaves the bowler’s hand and strikes the wood on the lane. For a few feet, the ball seems to be doing well, heading straight. Only closer to the pins does it become clear that the ball has been gradually going off course—so far off-center that it hits just one pin on the side, or falls into the gutter. Back in 1957, Conrad Waddington of the University of Edinburgh had proposed a similar metaphor for explaining the varied directions cells take as they develop and multiply. He envisioned a bunch of marbles rolling down a slope—an obstacle course with an elaborate system of lumps and grooves. Each marble ends up taking a different journey down the slope. That slope is what he called the “epigenetic landscape”—part architecture, part chance.*