Autism Is Not Psychosis

People with autism are less likely to commit crimes. Identifying individuals at risk for psychotic breaks, though, may be the one appropriate focus of the mental health discussion raised around Adam Lanza.  

batman 615 police.jpg
A Connecticut state trooper mans a checkpoint on the road leading towards Sandy Hook Elementary School [AP Photo/Jason DeCrow]

Given the unspeakable events last week in Newtown, everyone is searching for answers about what might have driven Adam Lanza to kill. Attention has focused on his possible autism spectrum diagnosis, but as researchers and advocates have pointed out, autistic people are actually less likely to commit crimes compared to those without the condition.

There is, however, another disorder that can both be linked with planned violence and mistaken for autism, which may account for part of what went wrong, in this instance or others. That's psychosis, which can occur as part of schizophrenia or in some cases of severe depression or drug misuse. While it's important to understand the differences between autism and psychosis, it's also critical to know that stigmatizing, bullying, rejecting, and isolating people who are different exacerbates every mental illness and developmental disability ever studied.

One of my colleagues, Dr. Bruce Perry, is senior fellow at the Child Trauma Academy and a child psychiatrist who has worked with or consulted on cases involving both children exposed to extreme violence and young perpetrators. He consulted with authorities guiding the response to Columbine.

To be clear, no one can diagnose a patient from afar, and there are too many unknowns in this case to diagnose Lanza. Still, for academic purposes and to the point of accurate mental health discussion and addressing ideas about identifying at-risk individuals, Perry says of the shooter: "My first thought was that he might have had a psychotic break -- and that his odd, disengaged behaviors earlier in life -- reasonably labeled as something like an ASD by some -- might actually have been prodromal psychotic disorder."

Schizophrenia is often not diagnosed until a first psychotic episode occurs, typically in the late teens or early 20s. These episodes are marked by delusions, hallucinations (mainly auditory), and often extreme paranoia. But long before these occur, some children who ultimately develop the disorder seem "off" and retreat from social contact. This is part of a set of symptoms known as a "prodrome" because it precedes those that characterize the later illness.

Like autism, schizophrenia is now believed to be a condition involving differences in early brain development, which probably begin in the womb, even though symptoms may not appear until years or even decades later. Lanza was 20, which is right in the typical range for a first psychotic episode to occur. But that is the most we can say.

Early symptoms of schizophrenia can sometimes be indistinguishable from those of autism: they include social withdrawal; communication problems and restricted speech; odd, repetitive behavior; an apparent lack of emotion or emotional expression and often, lack of conformity in terms of hygiene and dress. In both cases, the symptoms can start in early childhood, though with schizophrenia, there is often a noticeable exacerbation of prodromal symptoms in the year or two before the first psychotic episode.

Jorge Dan Lopez/Reuters

Both conditions are also sometimes linked with extremely high intelligence, which makes our failure to provide effective help even more wasteful. Lanza was an honors student, whom some of his classmates labeled a "genius." He was known to dress formally and carry a briefcase, not the backpacks favored by most schoolchildren. In some instances then, because we don't have biological markers to provide a definitive diagnosis, it may be impossible to tell if someone has an ASD or is going to develop schizophrenia until psychosis itself occurs.

Indeed, some of the same genes that have been linked to autism risk are also linked to schizophrenia risk -- and some of the same environmental triggers also raise risk for both. For example, having an older father or being in the womb while your mother suffers from certain infections elevate both the risk of autism and schizophrenia. It is not known why the predisposition becomes one disorder rather than the other.

Says Perry, "Certainly from what little we know about these kinds of spree events involving children, there are often intrusive, recurring psychotic ideations [in the perpetrators], such as seen in psychotic breaks such as the Aurora shooter or in a severe depression with psychotic features such as with Andrea Yates. [These] help drive the behavior."

Acute psychosis is linked with an increased risk of violence, which is at least doubled in schizophrenia compared to that seen in normal people and arises in other disorders involving paranoid fears and delusions. This risk is even higher for people with schizophrenia who have alcoholism or addiction: their risk of violence is nine times higher. A 2011 review of the research found that 35 percent of people with a first episode of psychosis have committed some type violence, but only 1 percent of this group engaged in violence serious enough to result in hospitalization or death.

That said, people with schizophrenia in treatment taking appropriate medication are no more likely to be violent than anyone else and only 5 to 10 percent of all murders are committed by people with any type of mental illness.

It's important here, too, to distinguish between psychosis -- which involves delusions, hallucinations, and loss of contact with reality -- and psychopathy, which involves predatory, self-centered and violent behavior and complete lack of concern for the feelings of others. Psychopathy is the extreme form of antisocial personality disorder and is, not surprisingly, the diagnosis linked to the most extreme crimes, like serial killing. Neither autism nor schizophrenia is linked with psychopathy.

And as for autism -- if it ultimately turns out to be what Lanza had -- that diagnosis is actually linked with reduced crime risk. People with autism tend to be extremely conscientious, often strictly following moral rules in ways that cause social problems; for example, by avoiding white lies. "From what we do know about ASD, [the shootings] would be pretty atypical," says Perry. "There's no risk in the diagnosis itself," agrees Dr. Harold Bursztajn, co-founder of the Program in Psychiatry and the Law at Harvard Medical School, who has reviewed the research.

In some cases, however, Bursztajn notes, "There is a risk that people who are impaired in social functioning -- who have a difficult time communicating, who become isolated-- can at times respond impulsively and at other times hold grudges forever and be hypersensitive and never able to get over a slight. They can become depressed, psychotic and delusional. So what we're looking at is not a risk associated with a diagnosis, but a risk associated with an impairment." That impairment is not inherent to autism, but can come from the isolation that stigma and lack of appropriate services creates.

Both the major autism charity, Autism Speaks and the Autistic Self Advocacy Network (ASAN) have issued press releases trying to prevent the Connecticut attacks from leading to increased stigmatization and fear of autistic people, noting the data showing no connection between the diagnosis itself and violence. Fears related to autism, however, remain. "Some popular perceptions are based in our own social anxieties in the face of such tragedy," says Bursztajn, "The quick fix becomes diagnosis instead of analysis."

Another reason many believe that autistic people are at higher risk of committing violent crimes may be a misunderstanding of what is meant by deficits in empathy, which are associated with autism in the public mind. Says Ari Ne'eman, who is autistic and is president of ASAN, "One of the most damaging stereotypes surrounding autistic people is the myth that we lack empathy. In fact, the research literature has confirmed what we've been saying for the longest time: we experience empathy at the same levels as the neurotypical population, even if the outside world doesn't always understand our ways of communicating it,"

While autistic people sometimes have difficulty understanding the thoughts and intentions of others -- an ability sometimes known as cognitive empathy or theory of mind-- research finds that they are not impaired in "emotional empathy" or the ability to share the pain and pleasure of others. In fact, some autistic people have such high levels of emotional empathy that their distress about other people's pain prevents them from actually being able to reach out and help.

Problems with cognitive empathy, however, can lead autistic people to behave insensitively at times. They may not automatically understand, for example, that telling someone "you look fat in that," is not the socially appropriate response to the question, but this does not mean that they don't care if they hurt people. Unlike in the case of psychopaths -- who are high on cognitive empathy but lack emotional empathy -- when autistic people hurt others, it's usually unintentional.

Autism is also accompanied by sensory problems, typically various oversensitivities to bright lights, loud noises, scratchy clothing and certain tastes. Being overwhelmed by these experiences may sometimes lead them to melt down and lash out -- but this is impulsive, not planned, and is more likely to harm the person themselves than anyone else.

A memorial for the 27 victims [Mike Seegar/Reuters]

Reduced sensitivity to some sensations, particularly pain, is also reported in some autism cases. Some who knew Lanza have said that he had an unusually high tolerance for physical pain and was seemingly unaffected by it. It is possible that this symptom might reduce empathy for the pain of others: if you don't know what pain is like yourself or feel it less intensely, you may wrongly believe others are similarly numb. However, there are people who are entirely without the sensation of pain who are not impaired in empathy. It's complicated.

Schizophrenia is linked with a different sort of impairment in cognitive empathy or "theory of mind," compared to autism. In autism, there can be difficulty recognizing other people's thoughts or intentions, while in schizophrenia, there's a tendency to see intentional action where it doesn't exist. For example, an autistic person might ignore someone who is crying because he doesn't realize that means that the person is upset-- while a person with schizophrenia might see those tears and that emotional experience as part of an elaborate plot.

You might say autistic people have too little theory of mind -- while schizophrenic people have too much, seeing conspiracies, connected behavior and intent in what are actually random coincidences. When this paranoia attaches to a particular person or institution, a real risk of violence can ensue.

As advocates point out, however, the vast majority of people with mental illness are no threat to anyone, except sometimes themselves. The mentally ill and developmentally disabled are more likely to be victims of violent crimes than perpetrators of it.

So what about the psychiatric medications Lanza was said to have been taking: could they have spurred the killing spree? If he was on antidepressants in the same class as Prozac, known as SSRI's, they could have been part of the problem, particularly if he had just started medication. "We do know that there have been reports of idiosyncratic responses to SSRI's in adolescents and young adults suggesting increased suicidal and homicidal ideations," Perry says.

David Healy, a British psychiatrist who has studied the connections between medications and violence and testified in court cases related to the links, says, "There is a general discounting of the possibility that psychotropic drugs could cause the kind of violence seen in Connecticuit or Aurora."

He adds, "Doctors and others stress that it is the illness not the drugs that causes violence and that we are grossly undertreating patients. But if we are undertreating patients at least half of these school shootings should be perpetrated by individuals not on treatment when in fact comfortably over 90% involve individuals on drugs."

Concludes Perry, "The bottom line is that we need to do a psychological autopsy on the shooter -- and other similar cases -- to see if there are reasonable places where we can be better at identification of at risk individuals. And finally, the lethality of this kind of (almost incomprehensible) spree behavior is magnified by the weapons used. 22 slashed Chinese school children is terrible -- but not as bad as 26 dead in Connecticut."

Moreover, a tragic irony of these cases is that the stigma they associate with mental illness actually contributes to the disorders themselves by deterring the sick from seeking treatment and creating fear in others that adds to their isolation and depression. Says Bursztajn, "Stigmatizing people who are already isolated makes it much more difficult for them to use their best social judgment and be able to distinguish between fantasy and delusion. The highest risk factor for violence is social stigmatization." That's something we all need to keep in mind in coping with this complex, tangled tragedy in light of our own emotions and the debate that will appropriately follow.