Criminologists call it mind-boggling but true: A third of people who confess to crimes didn't do it and incriminated themselves without physical coercion by police.
Social scientists say years of research have let them identify types of people especially vulnerable to confessing to something they did not do, and that expertise is now being used in criminal trials.
But not in Pennsylvania, where a divided state Supreme Court has said no to using expert testimony to help jurors understand the puzzling phenomenon of false confessions.
In a 4-2 decision filed late last month, the justices reversed a Superior Court decision allowing testimony of a nationally known expert on false confessions in the murder trial of Jose Alicea, an Olney man accused in a 2005 shooting during a brawl at a neighborhood cafe.
Writing for the majority, Justice Seamus P. McCaffery concluded allowing such testimony would turn into a battle of hired experts opining "generalities" that would not help the jury understand why people offer false confessions.
"Ultimately, we believe that the matter of whether [a] confession is false is best left to the jury's common sense and life experience, after proper development of relevant issues related to . . . the particular circumstances surrounding the elicitation of his confession," McCaffery wrote.
Justice Thomas G. Saylor, joined by Justice Debra McCloskey Todd, dissented, saying whether to allow expert testimony on false confessions should be left to trial judges to decide on a case-by-case basis.
Saylor criticized what he called the "blanket exclusion" of social science research "based upon unanalyzed assumptions about juror capabilities, even as these assumptions are challenged by demonstrations of wrongful convictions and developing behavioral science."
The high court's decision means that, almost nine years after his arrest, Alicea, now 28, will stand trial for the Oct. 30, 2005, shooting of 21-year-old Esroy George Rowe.
Alicea's lawyers have argued it is crucial for jurors to understand why he confessed to Rowe's slaying.
According to court documents, Alicea, then 19, had no record or previous contact with police - and an IQ of 64, six points below the traditional threshold for mental retardation - when he was brought in for questioning at 2 a.m. on Nov. 1, 2005.
Five hours later, nervous and shaking, Alicea began a statement implicating himself.
Except for Alicea's confession, the defense argues, the evidence against him is equivocal: eyewitnesses who identified two others as the shooter. Alicea's lawyer, Lawrence S. Krasner, called the Supreme Court's ruling "a wrong decision by a court staking out an unscientific position that will continue to convict innocent people, encourage improper interrogations by police, and cost citizens a fortune in lost lives and lost taxpayer dollars."
Krasner cited Innocence Project data showing 30 percent of convicted people later exonerated by DNA originally confessed to the crime.
"The Supreme Court can't explain that away," Krasner said.
In his pretrial motion, Krasner asked to be allowed to call as a witness Richard Leo, a nationally known expert on police interrogation and false confessions, to explain why some people are vulnerable to pressure and more likely to falsely confess.
In August 2008, veteran Philadelphia Common Pleas Court Judge Benjamin Lerner said yes - as long as Leo's opinions were about the general scientific findings regarding false confessions and not about Alicea.
The Philadelphia District Attorney's Office appealed, contending Leo's opinions would infringe on the jury's role to determine the facts and assess witness credibility.
In March 2011, a Superior Court panel affirmed Lerner by a vote of 2-1, writing that even jurors "who are aware of police interrogation techniques, or believe that they are aware by watching media and television, are unlikely to understand how these methods can lead to an innocent individual confessing."
The District Attorney's Office appealed again, to the Supreme Court, resulting in last month's decision.
Hugh J. Burns Jr., chief of appeals for the District Attorney's Office, said he was pleased by the high court's decision but puzzled by another decision the same day, in Commonwealth v. Walker, to allow expert testimony on the reliability of eyewitness testimony.
"Frankly, I don't understand the reasoning and the difference between the two cases," Burns said. "Why do you allow expert testimony about the reliability of the testimony of the eyewitness to a gunpoint robbery? But there you have it."
Marissa B. Bluestine, legal director of the Pennsylvania Innocence Project, who submitted a friend-of-the-court brief in the Alicea appeal, said reconciling the two decisions is a "real head-scratcher."
Bluestine was involved in a November 2012 conference on the subject of false confessions sponsored by her organization and Temple Law Review.
Bluestine said the Alicea decision would make it harder for juries to assess credibility, "to the detriment of all of us."