Eyewitness identification reform
Landmark study gives law enforcement better tools to reduce errors
The September execution of Georgia inmate Troy Davis compels us to consider the unthinkable: whether a state’s criminal justice system has put to death a person innocent of the crimes for which they were convicted. The execution, based almost exclusively on eyewitness identification evidence, came the same week the American Judicature Society (AJS) released a national field study on eyewitness identification that has major implications for reducing wrongful convictions in the United States.
The two events reinforce the importance of employing the most reliable tools available to us in the administration of justice. A Test of the Simultaneous vs. Sequential Lineup Methods: An Initial Report of the AJS National Eyewitness Identification Field Studies, describes eyewitness identification procedures that can counter a leading cause of wrongful convictions nationwide: the misidentification of suspects in police lineups.
Lineups are an age-old tool used by law enforcement to convict the guilty and exonerate the innocent. The technique is simple and straightforward; the eyewitness views a lineup in which the suspect is embedded among fillers. The presumption is that if the suspect is guilty, the witness will identify the suspect—whereas if the suspect is innocent the witness will identify no one.
However, eyewitness identification evidence is not always reliable. DNA-based exonerations of the innocent show that 75% are cases involving mistaken eyewitness identifications. Psychologists have been conducting laboratory studies on this problem for more than 30 years, and have proposed a number of possible reforms to the procedures used in conducting lineups.
AJS, in collaboration with the Police Foundation, the Innocence Project, and the Center for Problem-Oriented Policing, implemented this national field study at four law enforcement agencies to determine which lineup method—sequential (witness views lineup members one at a time) or simultaneous (all lineup members are presented as a group)—is more accurate. The study was designed to correct some of the problems in a 2006 Illinois study that later was found to have used flawed methodology. Unlike that study, which attempted to compare double-blind sequential lineups to non-blind simultaneous lineups, all of the lineups in the AJS 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 did not know. Additionally, the witnesses or victims in the AJS study did not know they were part of an actual study and were randomly assigned to either a simultaneous or sequential procedure.
Iowa State University Professor of Psychology Dr. Gary L. Wells, the study’s lead scientist and director of social sciences for the (AJS) Center for Forensic Science and Public Policy, established long ago that eyewitness identification of suspects in police lineups is much more prone to error than people think. Part of the problem stems from the standard procedure in which lineup members are presented as a group. When lineup members are presented as a group, witnesses tend to compare the members to each other, to figure out who looks most like the perpetrator and then identify that person. But, if the actual perpetrator is not in the lineup, there remains someone who looks more like the perpetrator than the others, and that person is at risk of being mistakenly identified. The sequential lineup method, which presents lineup members one at a time to the witness, tends to prevent the mere comparison of one lineup member to another and instead forces witnesses to compare each lineup member to their memory of the perpetrator.
Controlled laboratory studies, which are based on simulated crimes, have consistently found that sequential lineups produce a better ratio of accurate identifications to mistaken identifications than the double-blind simultaneous procedure. Nevertheless, many police departments have been hesitant to change their procedures based on laboratory findings alone. An important difference between laboratory studies and field studies that examine actual criminal cases is that the identity of the perpetrator is not necessarily known in the actual criminal cases. Although the identification of a known innocent filler is clearly an error, it cannot be presumed that the identification of the suspect is necessarily correct. Therefore, AJS and its collaborators saw an opportunity to coordinate a test of the simultaneous versus sequential lineup question in the field. The AJS Field Studies, which are based on actual criminal cases, represent the first time social scientists, prosecutors, criminal defense attorneys and the law enforcement community have come together to test new ways to improve the reliability and credibility of eyewitness evidence.
The results obtained during the AJS Field Studies found no statistically significant difference in the identification of suspects using simultaneous versus sequential procedures. Both methods produced suspect identifications in approximately one-fourth of the lineups conducted. However, the mistaken identification of filler subjects was significantly lower using sequential identification processes; fillers were picked in 18% of simultaneous lineups, versus 12% of sequential lineups.
AJS is beginning a follow-up analysis on many other aspects of the data, including the certainty of statements by the eyewitnesses. This could help determine, for example, whether mistaken identifications of fillers using one procedure versus the other are associated with more certainty. Analyses will also be conducted on whether the witness was a victim-witness or a bystander-witness, whether the identification was same-race versus cross-race, and numerous other measured variables to determine whether the sequential procedure is more accurate in certain circumstances. These analyses will be particularly important in light of recent oral arguments before the U.S. Supreme Court in Perry v. New Hampshire and in a recent ruling from New Jersey’s Supreme Court in State v. Henderson—both of which concern the reliability of eyewitness identification evidence.
It is understood that eyewitness identification will always be subject to human error. It must also be understood that only subsequent rigorous research can lead to the development of additional techniques to increase the accuracy of eyewitness identification. This landmark field study serves as a model for how those improvements can be made practically, systematically and transparently—and clearly demonstrates a promising path for further research to take toward convicting the guilty through accurate eyewitness identification, and reducing or even eliminating the incarceration of the innocent through inadvertent misidentification.