Saul Kassin reveals underbelly of the police interrogation business

(Editor’s Note: Recently, the Record had the opportunity to talk with Saul Kassin, professor of psychology and chair of the legal studies department. Kassin has published numerous books and articles on the psychology of police interrogation techniques, criminal confessions and eye witness behavior. In addition to this scholarly work, he selectively consults in criminal cases.

You seem to be involved in so many different fields. How would you characterize your most important work?

In a nutshell, I look at the psychology of evidence and jury decision making. Specifically, the type of evidence I’m interested in is police interrogations and how they get confessions. All of that brings into play social psychology, effects on behavior, effects on memory and effects on decision making.

How did you become interested in the psychology of evidence?

Years ago, when I came out of grad school, I went to the University of Kansas to do research on jury decision making – that’s just a fantastic application of basic theoretical psychology. The more interested I got in juries and how they make their decisions, the more interested I got in evidence, the stuff on which they make their decisions. It turns out the evidence is every bit as psychological in the way it’s taken as the decision-making process is for juries. Imagine a suspect in an interrogation room: the police are trying to get a confession, and the suspect might be guilty, might be innocent. The suspect has to deal with a host of social pressures and has to make a decision – “should I confess or not, should I confess to this crime?” What’s most interesting is that people often choose to confess even when innocent.

What would make a person confess to a crime they absolutely didn’t commit?

It comes down to so many factors. Sometimes the pressure involves being isolated from friends and family for a lengthy period of time in an unfamiliar place. Sometimes the pressure involves the sense that police have independent evidence of guilt. Some of the pressure then involves the feeling that it’s in their best interests to confess. So, if they think they’re trapped, if they think the police have evidence, even if it’s false and they think that confessing will somehow bring leniency, all of that might bring an innocent person to confess to something he didn’t do.

So, where can you draw the line between an appropriate interrogation and one that has a real chance at coercing a false confession?

[Laughs] That’s a great question. Ideally, if we were to construct from scratch an interrogation technique, in an ideal world that interrogation technique would be surgically precise: it would be compelling enough to draw confessions from the guilty, but not so compelling as to draw them from the innocent. The problem is that American police, American interrogations, are quite aggressive and very powerful. Many detectives are specially trained in the use of certain techniques – these are powerful techniques, but not surgically precise. They draw a sufficient number of confessions from the guilty but they also draw confessions from the innocent. I don’t think anyone has found that line.

So in your research, how do you go about studying the psychology of these confessions?

There are lots of different methods we’ve employed, ranging from laboratory simulations to field studies to surveys. For example, in a laboratory simulation I published with a student in 1996, we were trying to test the hypothesis that certain conditions increased the likelihood that people would not only confess to something they didn’t do, but actually think they did it. We did that using a simulation where we led people to think that they broke a computer. We tried to get them to confess to that breaking of a computer. So that’s one method, and that study’s been published and since been replicated by people in several countries. It’s an easy demonstration of a false confession.

I had a student a couple of years ago and we went to a prison outside of Boston. We had prisoners give us confessions on videotape to the things they did that had them incarcerated. And then we gave them a couple of minutes and asked them to come up with a completely bogus confession. We gave them two sentences of the basic details of the bogus crime they didn’t commit and asked them to make up a false confession, and they did that. And we took those tapes of true and false confessions and we showed them to people and asked them if they could tell the difference. Williams students could barely tell the difference at about a 60 percent accuracy rate, which is not unusual; people are not very good at judging truth or deception. Then we showed those tapes to police detectives and found that they weren’t any better – in fact they were slightly worse.

That’s a little disturbing.

It is disturbing. They come at the task with a guilty response bias; they tend to see guilt. We neutralized that bias and found that they’re still not very accurate. And again that’s a finding that I’ve since seen replicated in a number of labs.

In addition to all of these scholarly exercises, you also spend time on the front lines in the courtroom. How are you involved there?

I get involved in a number of cases as a consultant and those cases often provide me a wealth of ideas. My role in general would be to assist the district attorney or defense lawyer, who is faced with questionable, disputed evidence – particularly a confession or an eyewitness identification. It can be on a court-appointed basis, but it isn’t usually. I might assist in providing recommendations about what to look for in how to analyze that confession.

Do you ever go on record appraising the validity of a confession?

I won’t do that. I get involved in very few cases, certainly relative to the number of telephone calls I get. What I will do is, if the conditions under which a confession was taken are problematic, and if there’s no external corroboration for the confession, I would agree, sometimes, to testify.I can explain to a jury what the conditions were under which this confession was taken and what we know about those conditions in terms of putting people at risk to give false information. I won’t conclude one way or another, because I don’t know any better than the jury would, whether the confession was true or false. The only way you’d know that is to not only look at the confession, but to consider everything else that’s in evidence.

Are the conditions involved in each case unique, or are there a lot of commonalities?

The conditions are very case-specific. What I’m happy to do in most cases I work on is basically inform a judge at a suppression hearing or a jury at trial about the general principles. These are the principles of interrogation, these are the principles of social influence, this is what we know about the risks, and then I would simply inform them and educate them about that so they can better evaluate the statement. Typically, those cases come as a result of a phone call from a lawyer, but sometimes as a result of mail I receive from prisoners. I’ve got tons of mail from people who are in jail, who claim they’re innocent, who’ve claimed it for years. . . and these are typically cases involving confessions. The mail just piles up – I just don’t have time to deal with it all – but sometimes I get involved in cases as the result of a letter.

What cases that you’ve worked on or heard about stand out in your mind as the most egregious examples of false confessions?

Oh, gosh. . . I’ve seen such strange cases. Just recently there was a case in Australia, a murder trial where a guy confessed to killing a girl who just showed up alive. Now they have to deal with the fact that he confessed under probably coercive circumstances.

But to me what stands out the most right now is the recent Central Park jogger case [Editor’s note: In 1989, a female jogging in New York’s Central Park was beaten, raped and left for dead. Five teenage males confessed to the crime and were subsequently convicted. In 2002, a man convicted on several other rape and murder charges confessed to raping the jogger. His confession was corroborated by DNA evidence found at the scene.] – in part because there were five false confessions there and in part because it certainly fooled me.

I’m a native New Yorker and I followed the case closely at the time, and I never suspected that there was anything wrong with those confessions. The confessions fooled a lot of people: They fooled a judge, fooled two juries, fooled a whole bunch of lawyers, fooled a whole city and fooled me. And I now firmly believe that they were false. Somebody new stepped forward with a confession, that person’s DNA was at the crime scene, that person didn’t know these kids, that person knew all about the attack and gave the police details they didn’t already know, which none of those five confessions did.

What have you learned from looking at the conditions present in that case, in retrospect?

I’ve looked at them, and so has the Manhattan District Attorney’s office. The problem in doing that in this case, and in an awful lot of cases, is that only the confessions were videotaped, and that’s what the judge and juries got to see. But they didn’t get to see any videotape of the circumstances that produced those confessions, because the interrogations themselves were not taped. The problem with that policy, which is ever-present, is that the jury gets to see the confession, but not the eliciting circumstances. And as far as those eliciting circumstances are concerned, it’s very common for the detectives who took the statement, and the defendants who gave the statement, to completely disagree about what happened. Some of that reflects memory error, some reflects bias, but it creates a problem. You can’t adequately analyze an interrogation if you’ve got disputes about what the simple facts of that interrogation are.

Let’s switch gears for a second. America after Sept. 11: Have you seen any changes in the way police conduct interrogations? Are cops more aggressive, more coercive now than before?

I don’t think that’s necessarily the case in the criminal domain. I do think that law enforcement reasonably is starting to make distinctions among their suspects. I think we’ve certainly seen a higher level of acceptance of aggressive tactics when it comes to interrogating alleged terrorists than we would ever consider appropriate for the common criminal.

Does that include torture?

I don’t know what it includes; they’re not being very forthcoming about what’s happening in these cases. Again, as an objective matter, I know that some of the guys in Guantanamo Bay have been there for a long period of time. You could never hold a criminal suspect for that period of time. As to what specifically is going on, I kind of lost touch with that a few months ago. Again, the FBI was not very forthcoming.

With so much secrecy, can we be sure the interrogations are producing honest information that justifies the harsher methods?

Well, again, the goal is different. Interrogators have to get information that somehow has to get corroborated; without external corroboration of the information produced, the information itself is worthless. In contrast, unfortunately, in many criminal cases a confession standing alone without any corroboration is sufficient for a conviction. So whatever coercive elements may be present at Guantanamo Bay, I’m sure the interrogators are mindful of the fact that the more coercive the tactic, the more at-risk you are to obtain false information.