http://online.wsj.com/article_email/0,,SB111351995072807538-IdjfYNllad4mpyubHuHb62Am4,00.html

SCIENCE JOURNAL

By SHARON BEGLEY       
Interrogation Methods Can Elicit Confessions From Innocent People
April 15, 2005; Page B1

For cops, this was as good as it gets: The 14-year-old boy they arrested in the February murder of a man who found an intruder in his parked car in Rockford, Ill., didn't just confess. After the police took him from his home around midnight and isolated and interrogated him until dawn, he also re-enacted the crime for them, describing the inside of the car and relating how he had broken into it, struggled with the victim and shot him in the chest.

There was only one problem. After the boy had spent two weeks in detention, police, acting on a tip, discovered the real shooter was a 17-year-old.

Scientists who study false confessions aren't surprised. During the hours-long interrogation, says Shelton Green, the boy's public defender, detectives called the boy a liar, told him he would go to prison for 10 to 15 years if he didn't admit his role, suggested he shot the man in self-defense and promised to help him if he would own up.

"This was almost a perfect storm of criminal injustice," says Rockford prosecutor Paul Logli, president-elect of the National District Attorneys Association.

Suspects confess for a number of reasons. "But the most important," says Saul M. Kassin, professor of psychology at Williams College, Williamstown, Mass., "is that standard interrogation techniques are masterfully designed to leave people with almost no rational choice but to confess."

Typically, detectives isolate the suspect, heighten his stress and let him know that denial is futile. Crucially, says Prof. Kassin, they insist "we know you did it," make him think he can go home if he confesses, and lead him to think the evidence against him is strong. "If he thinks this is what he'll face at trial, a young suspect in particular may think it's better to confess" and hope for leniency, says Prof. Kassin, who testifies for defendants "two or three times a year, in false-confession cases so egregious they break my heart."

In a review of 50 years of studies, he and Gisli H. Gudjonsson of King's College London analyze why an innocent person would confess to a heinous crime. Isolation, confrontation, offering (false) incriminating evidence and implying the crime was justified can elicit confessions from the guilty and are recommended in police manuals. The U.S. Supreme Court has upheld the use of manufactured evidence in interrogations.

"Interrogators are trained to suggest to suspects that their actions were spontaneous, accidental, provoked, peer-pressured, drug-induced or otherwise justifiable by external factors," Profs. Kassin and Gudjonsson write in the journal Psychological Science in the Public Interest.

But what Prof. Kassin calls the "social-psychological weapons" of interrogators are so powerful they also can extract confessions from the innocent. Making the suspect anxious about his denials, challenging inconsistencies (a taste of what he would face at trial) and justifying the offense all induce confessions.

Those most likely to confess to a crime they didn't commit are compliant, suggestible, young, mentally retarded, mentally ill, or afraid of confrontation and conflict.

These folks aren't confessing to jaywalking. Of 125 proven false confessions from 1971 to 2002, 81% were for murder and 8% for rape. Although it is impossible to know how many confessions are false, of the first 130 exonerations that the New York-based Innocence Project obtained via DNA evidence, 35 involved people convicted after false confessions. People have confessed to murdering someone who is still alive, and to crimes committed when they were demonstrably somewhere else.

Some innocent people even come to believe they are guilty. In one infamous case, Michael Crowe, 14, was suspected in the 1998 stabbing death of his sister in Escondido, Calif. Through hours of questioning (with neither a lawyer nor parent present), he denied any involvement.

But after detectives told Michael (falsely) that his hair was found in his dead sister's hand, that her blood was in his bedroom and that he had failed a polygraph, he came to believe he had a split personality and confessed. Last year, a drifter who was seen in the neighborhood on the night of the murder and had the girl's blood on his clothing was convicted in the killing.

Police and prosecutors are starting to express concern about false confessions. "There are interrogation techniques that can lead to this," says Mr. Logli, the Rockford prosecutor. Minnesota, Alaska, Illinois and Maine mandate videotaping interrogations so prosecutors and juries can judge whether cops used methods likely to elicit false confessions. A report from Canadian prosecutors notes "hundreds of cases where confessions have been proven false" and recommends that investigators and prosecutors receive training about "the existence, causes and psychology" of false confessions. Earlier this year, a Chicago firm that trains detectives offered a course about permissible "trickery and deceit during an interrogation."

I have written in the past about the lack of a rigorous scientific foundation for fingerprints, eyewitness testimony, standard lineups and other forensic techniques. Add to that list the assumption that only the guilty confess.

* You can e-mail me at [log in to unmask]