To ID the perp, one photo at a time

IOWA STATE U. (US) — Sequential photo lineups—those in which crime witnesses view one suspect photograph at a time—produce fewer mistaken  identifications than simultaneous photo lineups.

Those are the findings of a report issued this week by both the American Judicature Society (AJS) and the Police Foundation.

The report comes on the heels of a landmark decision last month by the New Jersey Supreme Court mandating major changes in the way courts evaluate eyewitness identification evidence.

Psychology professor Gary Wells was principal investigator on a study that gathered data from police criminal investigations and found that sequential photo lineups produce fewer mistaken eyewitness identifications than simultaneous photo lineups. (Credit: Bob Elbert, ISU News Service)


The study has implications for reducing wrongful convictions in the United States criminal justice system. Of the 273 exonerations granted since 1989, more than 75 percent can be attributed to mistaken eyewitness identifications according to the Innocence Project—a non-profit national litigation and public policy organization dedicated to exonerating wrongfully convicted people through DNA testing and reforming the criminal justice system to prevent future injustice.

“It’s a good study for purposes of reassuring police departments that they’re not really going to lose suspect picks from it, but they will get fewer mistakes,” says Gary Wells, a psychology professor at Iowa State University, who was principal investigator on the study.

“That’s what this is all about, and that’s what police departments have always been interested—they don’t want to get the wrong person [perp]. They want to get the right person. And this is a tool that’s going to help move them in that direction.”

AJS, in collaboration with the Police Foundation, the Innocence Project and the Center for Problem-Oriented Policing, implemented a national field study at four law enforcement agencies to determine which lineup method—sequential or simultaneous—is more accurate. Police departments in Texas, Arizona, California, and North Carolina took part in the study.

All of the lineups in this study were conducted double-blind—meaning the officer administering the lineup did not know the suspect’s identity and the witness was told that the officer didn’t know.

The AJS National EWID Field Studies were conducted between 2008 and 2011 using a specially designed software application created by SunGard Public Sector Inc. The software application was uploaded onto laptop computers and ensured that instructions were consistently administered at the start of each lineup; the lineup condition (sequential or simultaneous) was randomly applied; and the order of mug shot photographs was randomized prior to the witness beginning the procedure.

The computer-based photographic lineups included one suspect photo and five known-innocent filler photos.

“The way it’s done in this field work is that one person is the suspect—who may or may not be the perpetrator—but the other people who are being shown to the witness are known innocent fillers,” Wells says.

“And so whenever the witness picks one of them, we know it’s a mistake. And so if the sequential procedure is doing the job that we had said it would all along, it would suppress those filler picks without suppressing picks of the suspect. And that’s exactly what was found in this work.”

Eyewitnesses shown photos in sequence, who made an identification, chose the suspect 69.1 percent of the time and chose known innocent fillers 30.9 percent of the time. By comparison, eyewitnesses shown a simultaneous array of six photos, who made an identification, picked the suspect at a 58.4 percent rate and the fillers 41.6 percent of the time.

Although hundreds of controlled laboratory studies have consistently found that sequential lineups substantially reduce mistaken identifications, Wells says many police departments have been hesitant to change their procedures based on laboratory findings alone.

“The big question that was driving this research was that while many jurisdictions have switched over to what we call the sequential procedure—which I developed back in the 1980s—there still are jurisdictions that are reluctant,” Wells says. “And the reason they’re reluctant is because they’d say, ‘That’s what your experiments show, but what about real eyewitnesses?’ So, it’s difficult to do this work with real eyewitnesses, but we found a way to do it with laptop computers.”

“Drawing on an unusually rich body of research from experimental psychologists, we have the tools to reduce eyewitness error, to protect the innocent and help law enforcement apprehend the guilty,” says Barry Scheck, co-director of the Innocence Project. “This scientifically rigorous field study is a model for how that can be done practically, systematically and transparently.”

The Police Foundation is conducting a second phase of the AJS National EWID Field Studies to evaluate the cases captured during data collection and to assess whether lineup presentation methods have an impact on witnesses’ accuracy and ability to identify the actual perpetrators. The Police Foundation is expected to release a report on the second phase of the AJS National EWID Field Studies in 2012.

In the coming months, lead scientists Wells, Nancy K. Steblay (Augsburg College, Minn.), and Jennifer E. Dysart (John Jay College of Criminal Justice, N.Y.) plan to continue their analysis of data gathered during the field studies, such as the witnesses’ certainty, cross-race comparisons, lighting conditions, and the presence or absence of a weapon at the time of the crime.

The AJS National Eyewitness Identification Field Studies are supported by grants from the Open Society Foundations, the Laura and John Arnold Foundation, and the JEHT Foundation.

More news from Iowa State University: