Should We Trust That Police Lineup ID?

Common Methods of Identifying a Suspect
Most law enforcement agencies continue to use “line-ups” for witness identification. Some consider this the most reliable method of suspect identification available to them, and reserve these procedures for their most serious cases. The process takes time and effort, after all. Police officers must gather “fillers”, who are the additional people standing in the line-up when a witness attempts to identify a suspect. Often, and surprisingly even in very serious cases, police will simply print a few photos including one of their suspect, and present the photos to a witness all at once, asking for an identification. Worse still, sometimes police will simply provide one photo and ask “is that the guy”, or have a witness identify a suspect in person who has been detained or arrested near the crime scene. Of course, all these methods have varying success rates when it comes to whether a witness is able to accurately identify a suspect.

Why So Many?
14 years have passed since the American Bar Association adopted a resolution calling on law enforcement agencies, including prosecutors, to use the most reliable and accurate means available for suspect identification by eyewitnesses. If that had been heeded, and subsequently required by prosecutors, law enforcement would no longer rely on such a wide range of identification procedures. They would research this issue and work to reach a consensus on the most reliable method. Accounting for some disagreement over a couple of the most reliable methods, we would see a narrowing of identification methods to just a couple of approaches. Instead, law enforcement agencies continue to use a wide variety of identification procedures, often within the same agency. The problem with this slap-dash approach to such a central and important piece of evidence in any criminal case is that failure to use the most reliable method is often a result of a choice based upon convenience. Investigating officers will claim they either did not have the time or resources available to conduct a proper and reliable identification. Investigating officers, those within their agencies responsible for establishing these procedures, and the lawyers who prosecute their cases in court are willing to settle for less reliable identifications in the interests of convenience and budget.

The Trouble With Eyewitness Testimony
Eye witness testimony is often unreliable. People tend to believe that if a person sees something they will accurately remember what they saw. Unfortunately that is often not at all the case. Especially in stressful or traumatic circumstances, a witness can be focused on, for example, the size or color of a gun, and completely overlook someone’s clothing or facial features. Sometimes memories can be “filled in” after the fact. For example, a victim of a robbery may be completely focused on the threat of a weapon, miss the face of the culprit, then see someone running nearby shortly afterward and attribute details of that runner to the culprit. Some approaches to suspect identification make these kinds of mistakes much more likely to happen. The best example would be the police hauling a suspect in front of a witness just after a crime occurs. The suspect is in handcuffs, maybe even in the back of a police car. That is suggestive to the witness on some level that this person is the person who committed the crime. Almost anything can be subtly suggestive in an identification, including placement of a suspect in a series of photos, selection of filler stand-ins, clothing and heritage, features, and even statements made by those in the line up or officers conducting the line up.

A study conducted by the American Judicature Society, a non-partisan group focused on researching and reporting on best practices within the judicial system, conducted field studies on simultaneous vs sequential lineups and published a report in 2011. Researchers in this group consisted primarily of professors from a variety of independent universities. In that study, the AJS researchers studied the method most law enforcement agencies believe to be the best method of eyewitness identification, which is a simultaneous line up in which a witness attempt to identify a suspect from a group presented in person, together, and all at once. Those researchers found that this method resulted in witnesses correctly identifying a suspect in only 25.5% of line ups. Perhaps more importantly, those witnesses identified the wrong person 18% of the time. The rest were apparently some form of “I’m not sure” or “I don’t see him/her in this group.” So nearly 3 out of 4 line ups failed to accurately identify a suspect, and this is law enforcement’s best option.

What Would Be Better?
The primary concerns about the accuracy and reliability of identification procedures are false accusations and conviction of the innocent, as well as failure to prosecute those actually responsible for a crime. Law enforcement really has just one job. That is to enforce the law. One would think law enforcement, being focused on that one job, would consistently use the most reliable means of correctly identifying perpetrators, even if that most reliable mean still is not great. Perhaps they should use their most reliable method with caution, keeping in mind that eyewitness testimony will always have its limitations. However, it is possible to reduce that very troubling number of false/wrong identifications. The same study cited above also found that the rate of false identifications by witnesses went down from 18% to 12% when potential suspects were presented sequentially, meaning one at a time, to witnesses . So if the witness had to make a decision by looking at each potential suspect in turn, they were less likely to incorrectly identify a filler as a suspect. In this study, accurate identifications of suspects where witnesses were presented in turn rather than all at once were also slightly higher.

Why Aren’t They Using The Better Method?
One explanation could be that law enforcement, and by extension their prosecuting attorneys offices, are more interested in the total number of accurate results than reducing the number of those falsely accused. Where neither method has significantly higher, or even very good overall, rates of accurate identifications, maybe law enforcement just doesn’t care that much that the method they assume is most reliable also has a much higher percentage of false positives. It could also be that police are bad at science. Scientists are typically motivated by a search for truth. Law enforcement agencies are very often motivated by a search for closure. “Scientists” who work in crime labs are biased. Failure to acknowledge that fact is naive. Convictions are both the job and the goal. Questioning the conviction is extremely unpopular in that setting.

The same bias holds true for researchers funded by law enforcement agencies. One such study of simultaneous vs sequential lineups conducted in Illinois in 2004 was subsequently rejected by the scientific community for failure to follow basic scientific principles for conducting research and experimentation. For example, older and more difficult cases were more often assigned to sequential lineups, which resulted in a biased result against that procedure.

Accuracy or Convenience?
Imagine you were suspected of a crime. Jail time is possible. For this example, let’s assume you didn’t actually commit any crime. But you could lose your freedom if you look like the person who did. If you have children, you may not get to see their next five birthdays. You’re not happy about being in any form of line up. After all, you know eyewitness testimony gets it wrong almost as often as it gets it right. Most of the time, it’s not helpful in identifying a suspect at all. But you’re given two options of line up format. Which would you choose?