Confessing to a crime usually is not enough to throw you behind bars. Many states require independent evidence to corroborate a confession. But if a suspect confesses and forensic investigators know, it can cause them to favor evidence in support of a guilty verdict—even if the confession is coerced or false. So says a study in the journal Psychological Science. [Saul Kassin, Daniel Bogart and Jacqueline Kerner, Confessions that Corrupt: Evidence from the DNA Exoneration Case Files, January 2012 Psychological Science (no link yet)]

Researchers analyzed 241 cases from the Innocence Project, which uses DNA tests to try to exonerate prisoners who are in fact not guilty. Most of the wrongful convictions were based on eyewitness mistakes. But a quarter of the bad verdicts involved false confessions.

And such cases were much more likely to involve botched forensic evidence—which tended to pile up after the confessions were made. That sequence suggests that investigators’ scientific conclusions were corrupted by belief in the defendant’s guilt.

Even more troubling, the authors say, is that tainted evidence can influence a trial long after a confession has been thrown out—as with the Amanda Knox trial in Italy. For judges and juries, the message is clear: even evidence that appears to be a smoking gun may be smoke and mirrors.

—Christopher Intagliata

[The above text is a transcript of this podcast.]