NPR

Brain Scans Predict Who's Likely To Be A Repeat Offender

In research published in the Proceedings of the National Academy of Sciences, scientists found brain scans can predict with startling accuracy the likelihood that criminals will run afoul of the law again. The results require serious legal and ethical debate before being introduced into the criminal justice system. David Greene talks to Kent Kiehl, a professor of psychology at the University of New Mexico and lead author of this mind research study.

Transcript

DAVID GREENE, HOST:

Imagine using mind technology to arrest and convict people before a crime actually happens. Sounds like something out of the movie "Minority Report."

(SOUNDBITE OF MOVIE, "MINORITY REPORT")

TOM CRUISE: (as Chief John Anderton) I'm placing you under arrest for the future murder of Sarah Marks and Donald Dubin that was to take place today, April 22nd, at 0800 hours and four minutes.

UNIDENFIED ACTOR: (As character) No!

GREENE: OK, the movie is obviously fantasy. But there is real research that brings us a little closer to predicting if someone will commit a crime. We talked to Kent Kiehl, a professor of psychology and lead author of a new mind research study. Professor, welcome to the program.

KENT KIEHL: Thank you very much for having me.

GREENE: So can you explain your research a bit more? As I understand it, you're saying you may have found a way to use brain scans to predict whether criminals will break the law again.

KIEHL: Yes. So we know that, you know, important decisions are made every day in the criminal justice system about, for example, whether to release an inmate; under what conditions to release an inmate, parole or probation. And one of the things that parole boards typically do is some form of what's called the risk assessment - to try to identify the factors that are likely to promote desistance to future crime, or persistence to future crime.

And one those variables is impulsivity. So if an offender scores high on impulsivity, for example, let's say you might use a pen and paper test to have them fill out and say that they buy things impulsively, they quit jobs frequently...

GREENE: That might mean you would commit a crime impulsively...

KIEHL: Yeah. Exactly. And it turns out, it does. So it turns out that those scales can predict that an individual is likely to come back to prison quickly. But you also might study nolo (ph) games or neuro psych tests, and those are also going to also give you a picture of what's happening.

KIEHL: What we did, though, that was different, was we said well, those are trying to measure brain function by proxy. And so what if we directly measure brain function using an MRI while they're doing those games and tasks? And what we did is, we found that the brain activity associated with impulsivity was incrementally predictive above those other measures. And it was really predictive. Inmates who had low activity were up to four and a half times more likely to reoffend - looking out four years, than inmates that had higher levels of activity.

GREENE: Break this down for me, in layman's terms. If you were sitting me down to do this test right now, what would you tell me?

KIEHL: I'd say that you are going to see a series of letters presented on the screen to you, one at a time; and they're going to come fast and - quickly. And every time you see this letter X, I want you to press a button here.

GREENE: Uh-huh.

KIEHL: But occasionally, there's going to be the letter K, and when you see the letter K, you're not supposed to press the button. And it's fast and quick and you're going to make mistakes, but you need to keep going as fast and accurately as you can. And here we go.

GREENE: Someone who sees the K, they're not supposed to put - the button but they can - stay in control, means they're less impulsive and probably less likely to be a criminal; and you're watching this whole thing in the brain, and what it looks like.

KIEHL: Exactly. Exactly.

GREENE: Well, aren't there other factors, when it comes to criminal behavior? I mean, I'm thinking drug, alcohol abuse, mental health. I mean, how predictive is impulsivity, when it comes to whether or not someone will commit a crime in the future?

KIEHL: So you're absolutely right. There are many actors that go into identifying the factors associated with risk for re-offending. And in this study, we did control for the effect of alcohol and substance abuse, age, education, socioeconomic background. And what was interesting was that the measures of brain function were still incrementally predictive above and beyond those other measures.

GREENE: I can imagine a whole slew of ethical concerns and questions coming up, as your research goes forward. What sorts of ethical questions do you think you'd have to kind of get through before this might actually be part of the criminal justice system?

KIEHL: I completely agree with you. In fact, we have a whole team that works to study these ethical issues, and the potential abuses or uses of the science that we do. You know, the brain is a very plastic thing. So even if we do find areas of the brain that are related to future recidivism, that doesn't mean those things can't be changed. I think that's the number one message here - is that what we've figured out, we think, is an area of the brain that we believe we can actually train or improve the function. So if I was able to increase activity with some sort of treatment - like a cognitive behavioral treatment, or a mindfulness meditation treatment; or maybe a pill that helps to increase activity there, and that reduces recidivism - pretty much all the inmates that we work with will say, I'll take that treatment because I don't want to be back here, either. And if I can be a little less impulsive and a little bit more likely not to come back to prison, I'll do that.

GREENE: Professor, let me ask you if there's a potential slippery slope here; I mean, that your research might be used in ways that people might not want. You start testing kids, testing people, using these brain scans to see if they might be first-time offenders in the future.

KIEHL: It's absolutely a potential concern. And we don't have any data yet that's really sound, that would suggest that we could predict before someone ever did anything wrong. So we're really working with, you know, trying to treat those individuals who are already on a trajectory towards getting themselves in a lot of trouble. The science is not there, accurately enough, to say for certain whether or not any of these offenders are going to re-offend. It's simply an increased risk.

But you also - you just need to be very careful making sure that you're not stigmatizing, especially children or adolescents. And, you know, we're very sensitive to these issues and we want to, you know, educate the public about these issues so that we don't have these types of concerns coming up.

GREENE: Interesting stuff. Professor Kiehl, thanks so much for talking to us.

KIEHL: Well, thank you very much for having me .

GREENE: He's a professor at the University of New Mexico. And his research is published in the "Proceedings of the National Academy of Sciences." Transcript provided by NPR, Copyright NPR.

Most Popular