« SCOTUS grants cert in significant habeas case and resolves one ACCA issue | Main | Notable skepticism about making a federal criminal case against John Edwards »

June 6, 2011

"Can A Test Really Tell Who's A Psychopath?"

The title of this piece is the headline of this fascinating recent NPR segment, which examines the creation and use of a test for psychopaths.  Here are some extended excerpts from the "science" part of the segment:

Canadian psychologist Robert Hare began studying psychopaths in the 1960s, and it's easy to forget now — in part because Hare's work has made the concept of the psychopath so commonplace — but a half-century ago, research on psychopaths was considered both obscure and largely irrelevant to understanding crime.

Back then, Hare says, there was a very clear consensus about where crime came from: Criminals were made, not born.  "In those days, social factors, environmental factors were the explanation for all crime," Hare says....

Hare, for one, didn't fully buy this.  He thought inborn personality was important . He says that as a psychologist, when he looked at people, he just saw incredible differences in temperament: differences in impulsivity, differences in the capacity for empathy, for feeling guilt....  Ultimately, [his research] led Hare to theorize that people with psychopathic personalities were essentially emotionally deaf.  They simply did not have the capacity to feel, in a firsthand way, emotions like empathy and love and remorse....

Hare sat down with his research assistant and together they wrote down all the personality traits they'd consistently seen in the psychopaths they'd studied.  Things like lack of empathy, lack of remorse, manipulation, egocentricity, impulsivity, superficial charm, psychological lying.  For each of these qualities, Hare wrote up a description so it would be clear what he meant by, say, lack of empathy....

The test listed 20 traits to check, and so Hare called it the Psychopath Checklist.  Scores were totaled at the end — 40 was the highest score, but anything over 30 certified the test taker as a psychopath.  Hare next tested his test to make sure that it was "scientifically reliable" — that two people using the test on the same person would reach the same conclusion about whether that person was a psychopath.  In research settings, the PCL-R's reliability appeared astonishingly good.....

For about five years, Hare's test did exactly what he wanted it to do: make the science of psychopathy better.  Psychopathy researchers from around the world bombarded Hare's lab with requests to use the PCL-R.  They published study after study on their findings.

Then, in the mid-'80s, one of Hare's students, an undergraduate named Randy Kropp, decided to conduct a different kind of study using the PCL-R. Kropp selected a group of prisoners with high, low and moderate scores on the PCL-R, then followed them after their release from prison.  He wanted to see whether prisoners with high scores were more likely to commit crimes than those with low scores once they were out on parole.  About a year later, he published his findings.

"Those who had low scores on the PCL-R, about 20 to 25 percent would be re-convicted within four or five years," says Hare.  "In the high group, it was about 80 percent."  So a parolee who scored high had an 80 percent chance of committing another offense within the next five years.  Low scorers had just a 20 percent chance of recidivism....

Suddenly, the PCL-R — a personality test used only in marginal academic research — appeared to identify the world's most serious chronic criminals.  The research community was stunned, says Stephen Hart, a former student of Hare's who is now a leader in the field of psychopathy research....

Its predictive ability made the test potentially useful outside the lab. Shortly after Kropp's finding went public, Hart recalls, Hare's lab got a visit from Canada's National Parole Board. It wanted the test: "They said quite literally, 'What we want to do is give everybody this test, and then have the test score written in big red numbers on the front of the file. No parole board should be able to make a decision without having some knowledge of whether or not somebody is psychopathic!' "

[A]t least initially, Hare was deeply concerned about letting people in the criminal justice system use the PCL-R.  He feared that the test, created purely for research purposes, might be used incorrectly in the real world and could hurt people.  Hare was particularly worried, he says, because by that point, the test had become widely respected as a scientifically reliable instrument.... For years, Hare made it clear to his students that he would not give the test out to anyone working in the criminal justice system....

While Hare remains a strong believer that his test works well for the kind of basic scientific research that it was originally designed for, he and others have begun to wonder if it does as good a job outside the lab.  "Once you get into the real world, there does seem to be some lessening of reliability," says Daniel Murrie, a professor at the University of Virginia who has studied what happens when psychological tests are taken from a rarefied research environment and transferred to the rough-and-tumble world of criminal justice.

About four years ago, Murrie decided to study the PCL-R to look at what happened when a psychologist hired by the prosecution gave Hare's test to the same prisoner as a psychologist hired by the defense.  Did those two psychologists give the same score to the same person?  The answer, says Murrie, was no. "Ten, 15, even 20-point score differences we found," he says, " And overall there was about an 8-point difference in scores."

The question is why.  One possibility, Murrie argues, is that the psychologists using the test in prisons and courts might not be well-trained.  "We don't know if the people giving the test in the field have gotten formal, rigorous training, or if they've just sort of bought the manual and maybe read a couple of papers and just decided to start using it," Murrie says.

But Murrie thinks it's also something else.  He says that in his study, psychologists hired by the prosecution consistently gave higher scores than psychologists employed by the defense.  Probably, Murrie says, because they're being paid for those opinions, and that money influences them.

The idea that criminal behavior is primarily a product of poor environments has much less power today, in part because Hare's work seemed to teach us that crime resides inside the person. Inborn personality traits, like empathy, can influence whether people participate in crime.

When you think about criminals this way — as people who are almost genetically predisposed to crime — you are much less likely to invest in their rehabilitation than if you saw their acts as the product of unfortunate environmental circumstances.

This is why it's so important to figure out if bias and bad training are affecting Hare's test to the point that it is potentially mislabeling people. After all, once someone is labeled as a psychopath, what can you do with him? Nothing but lock him away.

Along with this segment, the NPR website has this companion page titled "Expert Panel: Weighing The Value Of A Test For Psychopaths."  This page sets out these views on the PCL-R's role in the criminal justice system:

June 6, 2011 at 11:50 AM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference "Can A Test Really Tell Who's A Psychopath?":


"Things like lack of empathy, lack of remorse, manipulation, egocentricity, impulsivity, superficial charm, psychological lying."

Think Nancy Grace.

Posted by: George | Jun 6, 2011 12:53:37 PM

I was thinking prosecutors in general.

Posted by: anon2 | Jun 6, 2011 1:27:39 PM

I work daily with offenders (state parolees, feds, county) and addressed this topic obliquely at a conference.

This NPR report, which I heard as well, MISSES this monumental truism, i.e. "Inborn personality traits, like empathy" are not necessarily innate at all! Early maternal influence, parental modeling and correction, the coarseness of society, and even punishment and Cognitive Behavioral Therapy are extremely potent.

Perhaps by the time one tends to encounter a wildly violent offender, as it were, the criminal appears to have been 'born this way'.

However, just as gender identity largely develops before the age of 4, could not observed sociopathic tendencies, as recorded by this PCL-R test, have been birthed and nurtured in and by a twisted environment? At least partly so?

In brief, although perchance "crime resides inside the person", the claim that wickedness is congenital is unsupported. Inborn or no, and regardless of rehabilitation, the perpetrator of a deplorable act ought be swiftly and firmly deterred by punishment.

Posted by: adamakis | Jun 6, 2011 2:00:18 PM

"He says that in his study, psychologists hired by the prosecution consistently gave higher scores than psychologists employed by the defense. Probably, Murrie says, because they're being paid for those opinions, and that money influences them."

Obviously. In addition, there is an obvious selection bias. If an expert consistently finds the "wrong" result as one side sees it, that side will stop hiring him and the other side will hire him.

Posted by: Kent Scheidegger | Jun 6, 2011 6:57:05 PM

Oh fantastic, another pretense to justify the next national offenders registry.... the real equalizer will be when they figure out how to read minds.

Posted by: james | Jun 6, 2011 7:36:40 PM

anon2 --

"I was thinking prosecutors in general."

The willingness of the fruitcake Left to spit at random never ceases to amaze me.

Correction: It actually ceased to amaze me years ago.

Posted by: Bill Otis | Jun 6, 2011 10:59:51 PM

in that case james the politicians and govt officials of the world will be SAFE


Posted by: rodsmith | Jun 6, 2011 11:19:19 PM

An argument can be made just as easily that the defects of Antisocial Personality Disorder (ASPD) are a disability. Say, a wheel chair is a remedy for the paraplegic. What are the remedies for the disability of ASPD?

1) Severe, certain punishment, and continual external supervision and control are the remedy for ASPD. Sentencing guidelines dropped crime 40% by increasing the prison population. The crimes may have shifted to the inside of the prison. So, in prison severe, cheap, swift corporal punishment should be allowed at the administrative level, without a hearing, without benefit of counsel. The inmates, even the kingpins, should be physically terrified of the guards, and prisons should be 100% crime free, as a standard of due care. Those who fail to respond to punishment should be executed, summarily, upon counts of violent offenses.

2) We know one strong environmental factor in crime. Alcohol.

3) Another is bastardy.

4) Lawyer density is a factor. Nations in South America have even more lawyers per population then we do. Their crime rates are higher then ours. Japan and Switzerland have low lawyer densities. Low crime rates. Poor countries such as Egypt have great poverty, ineffective police and justice systems, yet low crime rates. Low lawyer density in Egypt. High rate of self-help. Crowds beat the ass of the street criminal at the scene. Crime rates are virtually nil in lawyer residential neighborhoods. Police arrives in 2 minutes, blasting. The death penalty for any violent criminal is at the scene in the lawyer neighborhood. No excessive force litigation in lawyer residential neighborhood. Infinitesimally small crime rates just a few miles from Fallujah like ghettos in the American inner city.

Think of ASPD as a talent, as Mozart could play the piano as a child. That may be innate. If Mozart had lived in times before the invention of musical instruments, no musical prodigy. So innate ability or disability combines with real world praxis (being modeled violence in the home, alcohol, getting away with 90% of crimes, long delays in punishment, etc.) to result in a crime.

Most of the environmental inputs into crime are under the total control of the lawyer profession, itself a criminal cult enterprise (CCE), intentionally maintaining high crime rates for rent seeking purposes, for lawyer employment, and the growth of government. Government is a wholly owned subsidiary of the CCE, and where staff lawyers make 99% of the policy decisions.

Posted by: Supremacy Claus | Jun 7, 2011 9:03:07 AM


They keep working on that. No luck yet though.

I do however have to question the idea behind the guy's early claiming he would never allow his test to be used by the criminal justice system. Presumably his papers had to identity the questions and the manner of scroing those questions. And especially if he took any grant money for the studies that led to his questions. It just seems like his cooperation would not really be needed, helpful perhaps but not needed.

Posted by: Soronel Haetir | Jun 7, 2011 10:25:15 AM

The research indicating that psychopathy is a real psychological phenomena is pretty solid, and there is mounting evidence that is is, if not hereditary (i.e. inherited from parents), probably either congenital (i.e. present at birth), or present by very early childhood and largely a permanent part of a person's personality. It does not have a simple single gene inheritance pattern (suggesting that a mishmash of genes must combine just so for it to manifest), but is typically found in individuals where there is a high prevalence of a variety of mental health conditions in the immediate and extended family.

The committees preparing the DSM-V (the bible of psychiatric diagnosis) are considering seriously breaking Oppositional Defiant Disorder into subtypes, one designed to capture psychopathic traits and the other designed to capture other kinds of conduct captured by the current diagnosis.

Among other things, the research has found that in children, parenting and discipline strategies based on future rewards as incentives, rather than punishment for past wrongs or strategies based on appealling to guilt seem to be most effective with these children. Efforts to change their psychological makeup seem to have successes that are few and far between and involve immense effort.

Also, while all psychopathy are weak on empathy, guilt and conscience, a large share are not violent, even though they may more freely engaged in deceipt and calculating manipulation of people. There appears to be a "plus factor" perhaps associated with novelty seeking or impulsivity that psychologically distinguishes the ordinary psychopath from the violent criminal one.

Many gradations in sentencing law and offense grading are basically designed to punish psychopaths more harshly than non-psychopaths, a policy that makes sense because while most crimes are not committed by psychopaths, they make up an extremely large share of the people committing pre-meditated first degree murders and other pre-mediated serious crimes. Even within the context of criminal gangs, individuals who show psychopathic traits commit a very disproportionate share of the violent crimes carried out by the gang.

But, the standard instrument used to measure psychopathy, like a lot of tests in pscyhology, is more useful in evaluating someone about whom you already have a lot of truthful information about, but it requires both truthful information that a psychopath may not be inclined to share and some subjective judgment calls about what constitutes a real world fact that fits an item on the list that is difficult to make very stable without a cohesive group of clinically experienced people who communicate regarding what fits each item on the list. We don't have paper and pencil tests that are not easily spoofed (like a classic test that involved doing a children's maze-like drawing with instructions never to lift the pencil and grading it by how many times the pencil is lifted, which worked to predict recidivism by acting as a proxy for self-control).

Another key thing that psychopathy related research has revealed is that virtually all people who are violent as adults, pscyhopathic or not, were violent as children. If a kid isn't violent at twelve, he's not going to be violent at twenty barring extremely rare and exceptional events like brain damage. Lots of people are violent as young children, but most people grow out of it. This suggests that excluding character evidence from criminal trials (at least to prove lack of violent tendencies) and closing juvenile records for people facing adult sentencing, may be quite poor policy.

Posted by: ohwilleke | Jun 7, 2011 2:15:32 PM

This test should be given to every police officer and prosecutor. I think the results would be interesting.

Posted by: mls | Jun 7, 2011 5:35:41 PM

Is Dexter a Successful Psychopath?
What does it mean to be a "successful" psychopath?
Published on September 13, 2010 by Melissa Burkley, Ph.D. in The Social Thinker

Posted by: George | Jun 7, 2011 8:25:38 PM

Ohwilleke has done a nice review of the state of understanding of criminality.

I would only add that the psychopath also has a high rate of substance abuse, compounding their dangerousness by 1) need for money for drugs; 2) intoxicated out of control violent behavior; 3) have a disproportionate number of social pathologies, including drunk driving offenses with car crashes.

Kill one as early as possible, and you prevent $100's of millions in damages to the nation. Half already die by unnatural causes, including murder, suicide, accidents before age 30. We just need to get rid of the other half.

These have the absolute protection of the lawyer from the earliest ages. You will have to kill the lawyer hierarchy before reaching the remainder of psychopaths devastating our nation. I encourage a direct action movement of crime victim families starting with an absolute boycott of the lawyer cult criminal hierarchy by all product and service providers. Hunt them, list them, shun them to death. No food, no heat, no shelter, no medical care for these internal traitors. Run them out of every neighborhood. Let them live in the wild. These enemies are pitiless, just like their clients. They deserve no human consideration.

Posted by: Supremacy Claus | Jun 7, 2011 10:31:06 PM

From the Dexter article, coming to the flip side the the conclusion that impulsivity is what separates the violent criminal psychopath from the non-violent one:

"[W]hat makes a successful psychopath different than an unsuccessful or "prototypic" psychopath? My colleague . . . asked experts in the areas of psychology and law to describe an individual they knew personally who matched the description I gave above regarding a successful psychopath. These experts were then asked to rate this individual on a variety of personality characteristics. From these responses, a clear, consistent description emerged that matched the typical characteristics of a prototypic psychopath in all ways but one: [Big Five Personality trait] Conscientiousness."

Posted by: ohwilleke | Jun 8, 2011 5:08:38 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB