Police officers in New York and around the country have long relied on lineups to provide prosecutors with evidence linking suspects to the crimes they are accused of committing, but a growing number of cases involving individuals who were wrongly incarcerated for years or even decades has raised questions about the reliability of witness identifications. A Louisiana man convicted after a witness pointed him out in court was released in January after 38 years behind bars, and lawmakers in the state responded by introducing rules designed to make the identification process more transparent and fair.

While few witnesses intentionally misidentify suspects, studies have found that they often make mistakes because they want to do their civic duty and please the police officers involved. To address this problem, lawmakers in Louisiana and several other states now require what are known as double-blind lineups. In these lineups, the police officer assisting the witness does not know who the suspect is.

Researchers began to question using lineups to identify suspects as early as the 1930s, but studies revealing the unreliability of eyewitness identification were largely ignored until DNA evidence was introduced in the 1980s. The Innocence Project has used DNA evidence to overturn more than 350 wrongful convictions, and eyewitness misidentifications played a role in 71 percent of those cases. Statistics like this have prompted police departments in several states to revise their witness identification procedures.

Experienced criminal defense attorneys may urge prosecutors to reduce or dismiss charges when eyewitness identifications are not supported by compelling physical and forensic evidence, and these arguments could be particularly dogged when the witness or witnesses involved are a different race or ethnicity from their client. This is because a number of studies have revealed that many people find it hard to identify individuals of a different race.