Experiments In Mental Illness


In 1972, a psychologist named David Rosenhan convinced some of his friends to fake their way into psychiatric wards across the US.

The pseudopatients were to present themselves and say words along these lines: "I am hearing a voice. It is saying thud." Rosenhan specifically chose this complaint because nowhere in psychiatric literature are there any reports of any person hearing a voice that contains such obvious cartoon angst.

Upon further questioning, the eight pseudopatients were to answer honestly, save for name and occupation. They were to feign no other symptoms. Once on the ward, if admitted, they were immediately to say that the voice had disappeared and that they now felt fine. Rosenhan then gave his confederates a lesson in managing medication, how to avoid swallowing it by slipping it under the tongue, so it could later be blurted back to the toilet bowl.

Once in the admissions unit, Rosenhan was led to a small white room. "What is the problem?" a psychiatrist asked.

"I'm hearing a voice," Rosenhan said, and then he said nothing else.

"And what is the voice saying?" the psychiatrist questioned, falling, unbeknown to him, straight into Rosenhan's rabbit hole.

"Thud," Rosenhan said, smugly, I imagine.

"Thud?" the psychiatrist asked. "Did you say thud?"
"Thud," Rosenhan said again.

Rosenhan was led down a long hallway. Across the country, the eight other pseudopatients were also being admitted. Rosenhan must have been scared, exhilarated. He was a journalist, a scientist at the apex, putting his body on the line for knowledge. He was taken to a room and told to undress.

"When will I get out?" we can imagine Rosenhan asked, his voice perhaps rising now, some panic here - what had he done, my God.

"When you are well," a doctor answered, or something to this effect. But he was well: 110 over 80, a pulse of 72, a temperature that hovered in the mid-zone of moderate, homeostatic, a machine well greased. It didn't matter. He was diagnosed with paranoid schizophrenia and kept for many days.

The strange thing was, the other patients seemed to know that Rosenhan was normal, even while the doctors did not. One young man, coming up to Rosenhan in the dayroom, said "You're not crazy. You're a journalist or a professor." Another said, "You're checking up on the hospital."

And then one day, for a reason as arbitrary as his admission, he was discharged.

Rosenhan's paper describing his findings, On Being Sane In Insane Places, was published in Science, where it burst like a bomb on the world of psychiatry.

The experiment was greeted with outrage, and then, at last, a challenge. "All right," said one hospital, its institutional chest all puffed up. "You think we don't know what we're doing? Here's a dare. In the next three months, send as many pseudopatients as you like to our emergency room and we'll detect them. Go ahead."

Now, Rosenhan liked a fight. So he said, "Sure." He said in the next three months he would send an undisclosed number of pseudopatients to this particular hospital, and the staff were to judge, in a sort of experimental reversal, not who was insane, but who was sane. One month passed. Two months passed. At the end of three months, the hospital staff reported to Rosenhan that they had detected, with a high degree of confidence, 41 of Rosenhan's pseudopatients. Rosenhan had, in fact, sent none. Case closed. Match over. Psychiatry hung its head.

In the 1970s, Spitzer and a group of colleagues completely revised the Diagnostic and Statistical Manual on Mental Disorders, or DSM for short, tightening the diagnostic criteria, taking away from it signs of subjectivity and psychobabble? "No," repeats Spitzer, "that experiment could never be successfully repeated. Not in this day and age."

Psychologist Lauren Slater decided to put Spitzer's words to the test and gave it a try.

I don't feel in control, though. At any moment someone might recognise my gig. As soon as I say, "Thud", any well-read psychiatrist could say, "You're a trickster. I know the experiment." I pray the psychiatrists are not well-read.
I am brought to a small room that has a stretcher with black straps attached to it. "Sit," the ER nurse tells me, and then in walks a man, closing the door behind him - click click.

"I'm hearing a voice," I say.

He writes that down on his intake sheet, nods knowingly. "And the voice is saying?"


The knowing nod stops. "Thud?" he says. This, after all, is not what psychotic voices usually report. They usually send ominous messages about stars and snakes and tiny hidden microphones.

"Thud," I repeat.

"Is that it ?" he says.

"That's it," I say.

"Did the voice start slowly, or did it just come on?"

"Out of the blue," I say, and I picture, for some reason, a plane falling out of the blue, its nose diving downward, someone screaming. I am starting, actually, to feel a little crazy. How hard it is to separate role from reality, a phenomenon social psychologists have long pointed out to us.

"So when did the voice come on?" Mr Graver asks.

"Three weeks ago," I say, just as Rosenhan and his confederates reported.

He asks me whether I am eating and sleeping OK, whether there have been any precipitating life stressors, whether I have a history of trauma. I answer a definitive no to all of these things: my appetite is good, sleep normal, my work proceeds as usual.

"Are you sure?" he says.

"Well," I say, "as far as the trauma goes, I guess when I was in the third grade, a neighbour named Mr Blauer fell into his pool and died. I didn't see it, but it was sort of traumatic to hear about."

Mr Graver chews on his pen. He's thinking hard.

"Thud," Mr Graver says. "Your neighbour went thud into his pool. You're hearing 'thud'. We might be looking at post-traumatic stress disorder. The hallucination could be your memory trying to process the trauma."

"But it really wasn't a big deal," I say. "It was just ..."

"I would say," he says, and his voice is gaining confidence now, "that having a neighbour drown constitutes a traumatic loss. I'm going to get the psychiatrist to evaluate you, but I really suspect that we're looking at post-traumatic stress disorder with a rule out of organic brain damage, but the brain damage is way far down the line. I wouldn't worry about that."

He disappears. He is going to get the psychiatrist. The psychiatrist enters the little locked room. The psychiatrist looks sad, and baffled, and then says, "But the voice is bothering you."

"Sort of, yeah."

"I'm going to give you an antipsychotic," he says, and as soon as he says this the sadness goes away. His voice assumes an authoritative tone; there is something he can do. "I'm going to give you Risperdal," he says. "That should quiet the auditory centres in your brain."

"So you think I'm psychotic?" I ask.

"I think you have a touch of psychosis," he says, but I get the feeling he has to say this, now that he's prescribing Risperdal. It becomes fairly clear to me that medication drives the decisions, and not the other way around. In Rosenhan's day, it was pre-existing psychoanalytic schema that determined what was wrong; in our days, it's the pre-existing pharmacological schema, the pill. Either way, Rosenhan's point that diagnosis does not reside in the person seems to stand.

"But do I appear psychotic?" I ask.

He looks at me. He looks for a long, long time. "A little," he finally says.

"You're kidding me," I say, reaching up to adjust my hat.

"You look," he says, "a little psychotic and quite depressed. And depression can have psychotic features, so I'm going to prescribe you an antidepressant as well."

Was this a freak accident? Surely most doctors do not prescribe antipsychotics and antidepressants at the drop of a hat? Or do they?

It's a little fun, going into ERs and playing this game, so over the next eight days I do it eight more times, nearly the number of admissions Rosenhan arranged.

Each time, I am denied admission, but, strangely enough, most times I am given a diagnosis of depression with psychotic features, even though, I am now sure, after a thorough self-inventory and the solicited opinions of my friends and my physician brother, I am really not depressed. (As an aside, but an important one, a psychotic depression is never mild; in the DSM, it is listed in the severe category, accompanied by gross and unmistakable motor and intellectual impairments.)

I am prescribed a total of 25 antipsychotics and 60 antidepressants. At no point does an interview last longer than 12 and a half minutes, although at most places I needed to wait an average of two and a half hours in the waiting room. No one ever asks me, beyond a cursory religious-orientation question, about my cultural background; no one asks me if the voice is of the same gender as I; no one gives me a full mental status exam, which includes more detailed and easily administered tests to indicate the gross disorganisation of thinking that almost always accompanies psychosis. Everyone, however, takes my pulse.

I call back Robert Spitzer at Columbia's Institute for Biometrics.

"So what do you predict would happen if a researcher were to repeat the Rosenhan experiment in this day and age?" I ask him.

"The researcher would not be admitted," Spitzer replies.

"But would they be diagnosed? What would the doctors do about that?"

"If they only said what Rosenhan and his confederates said?" he asks.

"Yeah," I say.

"They would be given a diagnosis of deferred."

"OK," I say. "Let me tell you, I tried this experiment. I actually did it."

"You?" he says, and pauses. "You're kidding me." I wonder if I hear defensiveness edging into his voice. "And what happened?" he says.

"I went in," I say, "with a thud, and from that one word a whole schema was woven and pills were given, despite the fact that no one really knows how or why the pills work or really what their safety is."

Spitzer clears his throat. "I'm disappointed," he says, and I think I hear real defeat, the slumping of shoulders, the pen put down. "I think," he says slowly, and there is a raw honesty in his voice now, "I think doctors just don't like to say, 'I don't know'."

"That's true," I say, "and I also think the zeal to prescribe drives diagnosis in our day, much like the zeal to pathologise drove diagnosis in Rosenhan's day, but, either way, it does seem to be more a product of fashion, or fad."

I am thinking this: in the 1970s, American doctors diagnosed schizophrenia in their patients many times more than British doctors did. And now, in the 21st century, diagnoses of depression have risen dramatically, as have those of post-traumatic stress disorder and attention deficit hyperactivity disorder. It appears, therefore, that not only do the incidences of certain diagnoses rise and fall depending on public perception, but also the doctors who are giving these labels are still doing so with perhaps too little regard for the DSM criteria the field dictates.

Despite being appalled at how easily one can obtain drugs that strongly manipulate brain chemistry (antipsychotic drugs are scary; they do good in some people that cannot otherwise be helped, but they can also irrevocable harm some patients), there are some bright spots to this story. First of all, Lauren Slater was not locked up. Secondly, she was treated with kindness. This is a long way from the original experiment where Rosenhan and his accomplices were locked up and treated with disdain. Psychiatry is still a very soft science (or still an art, however you prefer), but at least they now treat those with mental illness as people.
Second, it is remarkably easy to convince a psychiatrist that one is insane. This is not to cast fault at the psychiatrists, they must assume their patients are telling the truth, as they see it, in order to help. However, it is a strong argument for disallowing the "not guilty by insanity" plea that was once popular in the courts.

If you would like to read the entire description yourself, Lauren Slater summarizes the experiment in a lengthy weekend article at the Guardian. The links no longer work, but I was able to find cached versions at Google for Part I and Part II.


Share this with your friends
Mental llness and Nutrition.pdf286.18 KB