Therapy Helps Troubled Teens Rethink Crime

Student Kahlil Quato fights tears as he speaks at a University of Chicago candlelight vigil in 2007. The service was held in remembrance of Amadou Cisse, a graduate student, who was shot to death at point-blank range. (AP)

Late one night in November 2007, a student at the University of Chicago named Amadou Cisse was accosted by a young man named Demetrius Warren.

Warren demanded Cisse's backpack and water bottle — at the point of a .22-caliber gun. When the bag and bottle were not forthcoming — or not forthcoming quickly enough — Warren shot Cisse at point-blank range, killing him.

The 29-year-old Cisse was a month shy of completing his Ph.D. in chemistry. In 2011, Warren was sentenced to 120 years in prison.

The case spurred the University of Chicago to establish a center to study how to prevent homicide, using the tools of empirical science. Now, in a study issued by the National Bureau of Economic Research, investigators from the university's Crime Lab say they have identified an intervention that could help reduce violent crime.

To understand the technique, first consider what happened that night in 2007. Jens Ludwig, director of the Crime Lab, says that the conventional ways we think about crime don't make much sense in this incident: Unlike the scenarios on TV crime shows such as The Sopranos and The Wire, where violence is usually premeditated and aimed at obtaining specific goals, Warren had little to gain by shooting Cisse, Ludwig says.

"If they thought about it for even one second," he says, "it's very hard to imagine that anyone would think it was a good idea to shoot someone at point-blank range in exchange for a book bag and a water bottle that would surely have a resale value of not more than a couple of dollars at best."

Another popular crime theory suggests that increasing the severity of prison sentences will deter crime. But Ludwig says the Warren case calls that theory into question, too. "Demetrius Warren was sentenced to 120 years in prison," Ludwig says. "It is very hard to imagine that at the time Demetrius Warren has this .22 stuck in the chest of Amadou Cisse, that Demetrius Warren is going to be more deterred by a 200-year prison sentence [than] a 120-year prison sentence."

Ludwig says the case prompted one of his colleagues, Harold Pollack, to look into every youth homicide in Chicago in a single year. The crime researcher conducted a "social autopsy" on every case, to understand how and why the murder occurred.

He found that homicide among young people in Chicago tends to happen much as it did with Cisse and Warren: A sudden altercation, hastily planned and poorly considered, has terrible consequences because one of the parties has a gun and decides to use it.

In response to such killings, some people call for tighter restrictions on gun availability. Others have called for more guns — on the grounds that armed victims might deter attackers. But the Chicago researchers decided to focus on another element of violent crime: the perpetrator's frame of mind.

Ludwig points again to the case involving Warren: The shooting occurred, Ludwig says, after a rapid escalation of a dispute.

"Demetrius Warren presumably just didn't even think at all," Ludwig said. "He has this automatic response. When the backpack and water bottle are not forthcoming quickly enough, [he thinks,] 'I am not going to be disrespected.' And then he pulls the trigger."

Ludwig says the Warren case matches what Pollack found in his study of other youth homicides in Chicago: "Most serious violent events are almost Seinfeldian in their origin — someone saying something stupid to someone else, and that escalating and basically turning into a tragedy because someone had a handgun in their waistband at the time."

The solution to the problem, Ludwig, Pollack and their colleagues surmised, might lie in getting kids to slow down and think about their actions. The researchers conducted a randomized controlled experiment to test their hypothesis. They had about 1,400 school kids in grades seven to 10, drawn from high-crime areas of Chicago, undergo a 30-week training course called Becoming a Man. A similar group of students, also chosen at random, was tracked, but did not go through the course. At the end of the year, Ludwig said, researchers found 44 percent fewer arrests among the students who had been through the course.

Ludwig says the course was based on a kind of training called Cognitive Behavior Therapy — a mainstay in modern psychotherapy. The technique aims to get people to think about the way they think, and to recognize unconscious patterns of thought that produce unhappy life outcomes.

In one exercise, Ludwig says, the students were grouped into pairs, and one member of each pair was given a ball. The other was told to get the ball out of his partner's hand. This invariably led to a fight, Ludwig says, as the kids brawled over the ball. After watching the fight, the program leader would ask the student who was trying to get the ball a question: "Why didn't you ask the other kid to give you the rubber ball?"

None of the adolescents, Ludwig says, ever thought to ask their partners for the ball.

"The kids will say things like, 'Oh, if I would have asked, he would have thought I was a punk,' " Ludwig says. "Then the group leader will turn to the partner and ask, 'What would you have done had this other kid asked you to give him the rubber ball?' And usually this other kid will say, 'I would have just given him the rubber ball. What do I care?' "

The goal of such exercises, Ludwig explains, is to help the teens understand that their strong, negative reactions during confrontations are often based on what they falsely imagine is happening in other people's minds. Getting the students to put a more benign spin on what they imagined the other guy was thinking was key to helping them control their own impulsive behavior, Ludwig says. And the approach worked — at least for a while.

Unfortunately, within a year after the program ended, its effect seemed to fade. Teens in the group who had gone through the training went back to having the same arrest rates as kids who hadn't gone through the program. Ludwig says the researchers are still exploring how to help young people retain the powerful benefits of this sort of psychological training, as part of a range of efforts in Chicago to stem homicide.

Copyright 2015 NPR. To see more, visit

More Photos


Police departments and community groups are always looking for ways to reduce crime. Much of the time though their success is not measured scientifically, so we don't really know what works and what doesn't. Well, at the University of Chicago, researchers actually tested a crime intervention and they found something pretty interesting. To tell us about it, we're joined by NPR's Shankar Vedantam who comes by regularly to tell us about new social science research. Shankar, welcome back.


GREENE: OK. So tell us, this is the city of Chicago. I mean, it's a city that has really struggled recently to reduce crime. What are these researchers up to?

VEDANTAM: Well, the researchers first decided to explore how homicide actually takes place and to see if the way it takes place lines up with our intuitions. So Harold Pollack at the University of Chicago, he conducted what he calls social autopsies on every youth homicide that took place in the city of Chicago in a year. And what he found was that murder doesn't happen the way we think it happens or they way it happens on the "The Sopranos" or "The Wire."

You know, it usually isn't premeditated or aimed at getting something or settling a score. I spoke with Jens Ludwig. He's the director of the University of Chicago Crime Lab. Here's how he described it to me.

JENS LUDWIG: Most serious violent events are almost Seinfeld-ian in their origins. Someone saying something stupid to someone else and that escalating and basically turning into a tragedy because someone happened to have a handgun in their waistband at the time.

GREENE: I guess one unresolved debate in the country is whether or not taking the gun out of the hand of this person would prevent this tragedy from happening.

VEDANTAM: So, sure, David. I mean, the whole debate about gun control is a big debate, but Pollack and Ludwig and their colleagues decided to try a different intervention that was aimed at how these kids think. So they ran an experiment with about 2,800 kids from grades 7 to 10, all drawn from high-crime areas. Half of them went through a program that was designed to help the kids to stop and think, and half did not.

And when they finished the program they found that the arrest rates for the kids who went through the program dropped by 44 percent.

GREENE: What exactly were they being taught? What's the intervention?

VEDANTAM: So it really is a form of psychotherapy that people call cognitive behavior therapy. And the idea is you need to stop and think about the way you think. And I'm actually going to give you an example, David. This actually comes from one of the interventions that they did with these kids. I brought in a granola bar into the studio right now.

GREENE: Thank you for that.

VEDANTAM: And I have it in my fist. And let's say you want this granola bar. How would you get it out of me?

GREENE: I would say, Shankar, hi, I'm kind of hungry right now. I wouldn't mind having that granola bar if you're not going to eat it yourself.

VEDANTAM: Here you go, David.

GREENE: Hey, thank you. I appreciate that.

VEDANTAM: Here's the thing. Not one of the kids who went through this program ever thought of doing what you just did.


VEDANTAM: So they way they played this game, the kids were trying to get a ball out of a partner's hand and when they were told that the partner had it in his fist, they went after the partner and they fought for the ball and they brawled over it. And after about five minutes of this, the program leader stopped them and asked them a question. And here's how Jens Ludwig explained it to me.

LUDWIG: Why didn't you ask the other kid to give you the rubber ball? And then the kids will say things like, oh, if I would've asked, he would have thought I was a punk. Then, the group leader will turn to the partner and say what would you have done had this other kid asked you to give him the rubber ball? And usually the kid will say I would have just given him the rubber ball. What do I care?

GREENE: So it's a way of helping these kids understand that they're not going to be perceived necessarily the way they think they will be.

VEDANTAM: Exactly. I mean, so the central idea of cognitive behavior therapy is that there are these unconscious patterns to our thinking that when we expect an interaction to be hostile, we're going to be very aggressive coming into the interaction. So the goal is to train kids to slow down, not to leap to conclusions, to question your own assumptions.

You know, and I have to give you credit, David, because you didn't leap to the assumption that I was going to be resistant in giving you the granola bar.

GREENE: We didn't brawl in the studio. That's a good thing.


VEDANTAM: But I have to tell you there's also good news and bad news here, David. The good news is the intervention was very successful, but the bad news is one year after the program ended, the arrest rates of the kids who went through the program was no different than the arrest rates of the kids who didn't go through the program.

GREENE: Oh, that's sad. So is there some way to basically extend the success here?

VEDANTAM: So I think they're going to test different interventions, David. But Ludwig told me that even in just the year of the program, the benefits of the program far outweighed the costs. And not just in financial terms, but in long-term human terms. You know, someone at the Cook County Juvenile Detention Center told him, look, 20 percent of the kids who are here probably need to be here because they're dangerous.

For 80 percent of the other kids, they're here because of a single incident. If I could give them back 10 minutes of their lives, they wouldn't be here because they could make a different choice in that 10 minutes. And that's what this kind of program's designed to achieve.

GREENE: Shankar, really interesting. Thanks for coming in.

VEDANTAM: Thanks, David.

GREENE: That's Shankar Vedantam. He regularly joins us to talk about interesting social science research. You can follow him on Twitter @hiddenbrain. And while you're doing that you can also follow this program @morningedition and @nprgreene.


GREENE: This is NPR News. Transcript provided by NPR, Copyright NPR.

Most Popular