In response to a growing body of scientific evidence suggesting that witnesses to crimes often misidentify suspects shown to them in police photo lineups, the Baltimore City Police Department announced recently that is changing the way it conducts the procedure to make it less prone to error. That's a long overdue change that not only brings the department more in line with modern best practices but also makes it less likely that innocent people will be sent to prison for crimes they didn't commit based on faulty witness identifications.
Traditionally, the decisions of police, judges and juries tended to be strongly influenced by witnesses who claimed to recognize criminal suspects when they saw them again, either in person during a line-up at the jail or in photo arrays that investigators asked them to study as they were being questioned. When a witness or victim could point to a particular individual and say with assurance "Yes, that's the guy who did it!" detectives usually figured they had the case wrapped up. Or at least that's how it happened in the movies.
Direct eyewitness testimony of this sort was for years considered one of the strongest types of evidence that prosecutors could bring against a criminal defendant, and many a convict ended up spending decades behind bars as a result. The only problem was that a significant number of people convicted on the basis of such evidence actually turned out to be innocent of the crimes they were accused of.
That became apparent after the development of DNA testing, which for the first time allowed investigators to check the genetic markers found at a crime scene against the testimony of witnesses who claimed to have seen a suspect there. Of the first 250 cases of people freed from prison after being cleared by DNA testing, 190 had been convicted on the basis of false identification. Today, researchers estimate that up to a third of the 75,000 identifications made annually in this country are in error.
More recent research has shown that human recall rarely possesses the definitive certainty of a photograph. Instead, it is malleable and easily influenced by context. People don't so much consciously "make up" or "forget" things as unconsciously change them around to fit the circumstances in which they find themselves. Researchers suspect one way they do that is by responding to subtle cues given off by police that unintentionally communicate to them which suspect investigators believe most likely committed the crime. Whether they simply want to help the police — or just want to end the questioning — a witness may falsely identify a suspect based on such intuitions.
Baltimore police have been following guidelines issued by the International Association of Chiefs of Police and the National Institute of Justice that require witnesses to view photographs of suspects sequentially rather than in groups, with the entire process recorded so that the witness' level of certainty about an ID can later be reviewed. Research has shown that when witnesses are shown several pictures at once they tend to compare them and pick out the one that looks most like their memory of the suspect, even if it is the wrong choice.
What city Police Commissioner Anthony Batts is proposing in addition to those safeguards is a "double-blind" system, under which witnesses are shown pictures sequentially by officers who themselves do not know which images are of suspects. That prevents officers from unconsciously influencing the witness' choice and further reduces the chances for misidentification. The "double-blind" system has long been used in pharmaceutical research as a way of ensuring that subjects aren't simply reflecting what they think researchers conducting the study expect them to.
Whether the change in ID procedures will result in more criminal convictions is still an open question, of course. But at the very least it will give police, prosecutors, judges and juries greater confidence that the people they send to prison actually committed the crimes they were accused of — and that the police don't close cases while the real perpetrators are still walking the streets. That's still the greatest risk of putting the wrong people behind bars, and Baltimore's reforms ought to serve as an example for police departments across the state.