The Innocence Project has released a new report pointing to the problems with eyewitness identifications in criminal cases and offering recommendations for making the system more reliable. The report, “Reevaluating Lineups: Why Witnesses Make Mistakes and How to Reduce the Chance of a Misidentification,” states that over 175 people (including some who were sentenced to death) have been wrongfully convicted based, in part, on eyewitness misidentification and later proven innocent through DNA testing. But DNA testing is not a solution to the problem since it is only available in 5-10% of all criminal cases, according to the report. The findings in the report included:

• In 38% of the misidentification cases, multiple eyewitnesses misidentified the same innocent person.
• Fifty-three percent of the misidentification cases, where race is known, involved crossracial misidentifications.

• In 50% of the misidentification cases, eyewitness testimony was the central evidence used against the defendant (without other corroborating evidence like confessions, forensic science or informant testimony).

The report offered a number of recommendations for reform, including:
• Double-blind presentation: photos or lineup members should be presented by an administrator who does not know who the suspect is.

• Lineup composition: “Fillers” (the non-suspects included in a lineup) should resemble the eyewitness’s description of the perpetrator and the suspect should not stand out. Also, a lineup should not contain more than one suspect.

• Witness instructions: The person viewing a lineup should be told that the perpetrator may not be in the lineup and that the investigation will continue regardless of whether an identification is made.
• Confidence statements: At the time of the identification, the eyewitness should provide a statement in her own words indicating her level of confidence in the identification.
• Recording: Identification procedures should be videotaped.
• Sequential presentation (optional): Lineup members are presented one-by-one (by a “blind” administrator) instead of side by side.

(“Reevaluating Lineups: Why Witnesses Make Mistakes and How to Reduce the Chance of a Misidentification,” The Innocence Project, Benjamin Cardozo School of Law, July 2009). See Studies and Innocence.