Eyewitnesses and the Dangers of the Photo Lineup


Yesterday I posted a link on my Twitter and McCathern’s Facebook page of a news story in Dallas that highlights the latest example of police misconduct in a homicide investigation.  The story centers around a Dallas homicide detective who unduly influenced a mentally-challenged suspect (“Miller”) into picking another suspect (“Jones”) out of a photo lineup.

The story reports that Miller was unable to pick anyone out of a proper photo lineup, but when other officers left the room, Detective Elena Perez showed the young man a single picture of Jones and had him ID the picture as the man who shot and killed a Subway store employee back in June.  Detective Perez then lied in her arrest affidavit, claiming that Miller had properly identified the suspect.  When information of Perez’ misconduct surfaced, both Miller and Jones were released from custody.  You can read the full story here: http://www.wfaa.com/story/news/local/dallas-county/2015/08/18/dallas-police-arrests-improper-influence/31952819/

Unfortunately, examples of false arrests resulting from unreliable eye witness identifications are all too common in the legal community.  The Innocence Project reports that eye witness misidentification is a contributing factor in 75% of wrongful convictions.  Locally, members of the Dallas metroplex are familiar with the surge of DNA exonerations of persons wrongfully convicted by eye witness misidentifications that surfaced through ex-Dallas District Attorney Craig Watkins’ Conviction Integrity Unit.

Still, many people outside of the criminal justice community are left scratching their heads when they hear about the unreliability of eyewitness identifications and police lineups.  What is the danger?  How can someone who witnessed a crime or was even the victim of a crime pick the wrong person out of a lineup?  The inherent problems with suspect lineups aren’t necessarily obvious problems, and perhaps the biggest problem of all is how trusting juries are of eyewitness testimony.


I will start by saying that I could write, and many have written a whole book on the problems and inherent dangers with photo lineups and eyewitness identifications in general.  However, here is the Cliff’s Notes version.

First, we’ve all heard or seen news stories about the problem with cross-racial identification.  Countless studies have show that everyone, regardless of their race, has difficulty specifically discriminating the facial differences between people of a different race.  That is, someone who is Hispanic naturally will not be as discerning among Caucasians as Caucasians themselves are, someone who is Caucasian naturally will not be as discerning among African Americans as African Americans themselves are, etc. The cross-racial aspect is a huge factor when talking about eyewitness identifications in general.  You can learn more about cross-racial identifications here.

Second, there is a huge danger with suspect lineups that the persons represented in the lineup simply do not look similar enough.  The extreme example would be a 6 person photo lineup where two men are bald, three have glasses, and one has a mustache.  In that example, if the witness remembers the suspect had a mustache, then obviously the one person in the lineup with a mustache will be the one ID’ed.  This might sound absurd, but similar examples are unfortunately littered throughout the history of suspect identifications.

Next, much research has been done to illustrate the natural desire that witnesses, and especially victims have to “get it right,” or “find the culprit.”  This deep desire naturally makes eyewitnesses more suggestible and even more likely to pick someone out of a lineup even when the actual perpetrator isn’t featured.  There’s an overwhelming amount of pressure on an eyewitness and studies have shown that most simply cannot grasp or accept the possibility that the actual suspect might not be in the photo lineup.

Finally, over the years studies have shown that police officers, despite their best efforts, often give subtle cues or indications as to which photo in a lineup belongs to the person they are investigating.  Even when an officer makes every attempt to appear neutral, eyewitnesses are able to pick up on cues that suggest which person the detective wants them to pick out of the lineup.  Coupled with a “desire to get it right,” eyewitnesses are often unconsciously swayed by this influence.


So the above heading is a little misleading.  As things stand today in Texas, there is not a specific state-wide procedure for administering photo lineups.  Rather, Article 38.20 of the Texas Code of Criminal Procedure requires law enforcement agencies to have a department-wide policy, and lists several factors that should be included or considered in that policy.  However, there has been a lot of training, a lot of writing, and a lot of proposals across the state and nation when it comes to the ideal method of administering a photo lineup.

As a former Collin County prosecutor, I had the opportunity to work with some of the top agencies in the state (holler Plano PD!) whose photo lineup policies mirror the “ideal” requisites to minimize any chance of unintentional witness influence.  Here’s how the best and most up-to-date agencies are handling their photo ID’s.

First, whenever possible, the suspect’s drivers license picture is used, and the other “filler” photos are taken from a drivers license database so that the background, form, and even angle of the pictures all match.  Every detail of the picture is also matched among the different individuals, meaning, if the suspect has short red hair, round glasses, a long nose and no facial hair, then each of the other pictures will also have short red hair, round glasses, a long nose and no facial hair.  This aspect is absolutely key and the best photo lineups do not have even one distinguishing detail among the photos.

Next, all six photographs in the lineup are handed to an officer or a detective who knows absolutely nothing about the case.  This detective doesn’t know what the case is about, which one of the photographs belongs to the suspect, what kind of witnesses is coming in to look at them (third party witness, victim, family member, etc.), nothing.  That way, the person administering the photo lineup is immune from subconsciously giving those unintended signals or cues as to who the suspect might be.  When possible, the photo lineup is double-blind, meaning both the person administering the lineup and the person who collected the photos are wholly uninvolved in the case.

Then, the witness is taken into a room away from everyone else, where they don’t see, speak to, or hear any of the officers who are associated with the case.  The detective administering the lineup then reads or recites a scripted intro explaining the process.  This intro is included in the department’s policy and is read exactly the same every time, so that if the lineup is ever questioned, a judge, attorney or jury all know what was said to the witness and what wasn’t said.  That intro will include instructions about how the process is going to work, how to indicate if they recognize an individual, and should always include a warning that “the perpetrator may or may not be in one of the photos.”

This is where the gold standard photo lineups differ from what you see on TV.  Rather than showing a witness one sheet of paper with six or eight different photos all on a grid, the detective should have each photo on its own individual sheet of paper, and show each photo to the witness one at a time.  The witness is put in a position of having to actually recognize the photo solely based on memory rather than by comparing one photo against five others.  If the witness recognizes a photo as the suspect, they are to initial and date the photo and are asked by the detective “how sure are you?” to which they usually respond with a percentage (“80%” or “100%”).

The administering officer then leaves the room without giving the witness any indication of whether they picked the suspect or not (remember, since it’s a blind procedure, the officer would have no way of knowing whether the person the witness picked was the suspect anyway).  In the paperwork the officer fills out, he indicates the number of the photo selected (if any), how sure the witness said they were, and how long it took for the witness to make their decision.  All of that information is eventually turned over to both the prosecutor and the defense attorney so they can consider those factors when determining the reliability of the witness’ ID.


As is made apparent in the Dallas story this post began with, perhaps the most dangerous part of eye witness identifications are the occasional “bad apple” officers who taints them.  Now, this isn’t Hollywood, and I do not for a minute think that officers are sitting around intending to “frame” suspects or “pin a case” on someone they know is innocent (I’m just not that jaded yet).  However, you do not have to google long to find a story of officers who bent the rules, or even just ignored them all together, in order to make a case they thought they had fit.  There are countless examples of officers committing misconduct to make a suspect identification fit their narrative of a case, just like Detective Elena Perez did in the Subway shooting.

In the case of defendants Michael Evans and Paul Terry, police arrested the two 17 year old boys for murdering a 9 year old girl based on the identification of supposed eyewitness, Judith Januszewski.  Ms. Januszewski’s credibility was mangled from day one, as she called in to claim a reward that was being offered for the case, and claimed to have witnessed Evans and Terry struggling with the victim at a time when Januszewski’s employer says she was still at work.  But, police pushed forward anyway urging the victim’s family to change their timeline about when their daughter left to fit the narrative of the “eyewitness'” testimony.  27 years later, DNA evidence exonerated both Evans and Terry, and you can read their full story here.

The case of Lesly Jean is equally disturbing.  In 1982 Jean was arrested for sexually assaulting a woman in her home after he was seen in a donut shop matching the perpetrator’s description.  When first shown a lineup, the victim pointed out two different men that each gave her a bad feeling, but never identified someone as the assailant.  Police then hypnotized the victim to “improve her memory” (witness hypnotizing is a whole other discussion we won’t delve into in this post) and she then confidently selected Jean.  Weeks later, police showed the victim yet another photo lineup, this time with only three pictures: one of Jean, one of a man much taller and one of a man much shorter.  Not surprisingly, the victim again identified Jean who was convicted and sentenced to life in prison.  In 2001, DNA evidence exonerated Jean, and you can read the details of his story here.

Finally, one of the most infuriating stories is that of Marvin Anderson.  When a young woman was raped by an African American man who told her he “had a white woman,” the officer investigating the case slated Marvin Anderson as a suspect because he was the only African American in town the officer knew who lived with a white woman.  The officer showed the victim a photo lineup that included a color photo of Anderson next to six black and white photos, and naturally, the victim ID’ed Anderson’s color photo.  Anderson served 15 years in prison before being released on parole and it was still another four years before he was finally exonerated by DNA evidence.  You can learn more about Anderson’s story here.

In closing, I do want to again emphasize that photo lineup procedures have come a long way, and as I mentioned earlier, I have had the privilege of working with some of the best law enforcement agencies who truly set the gold standard when administering these lineups fairly and absent any bias.  However, as the cases outlined above illustrate, the police officer behind the lineup is as influential in a misidentification as the process of the lineup itself.

Finally, the whole inspiration for this post was not actually the story that I posted yesterday, but this follow-up story I read this morning was actually what spurred me to write about eyewitnesses and police influence.  The above link redirects to the Subway victim’s family and their reaction to the news that Jones and Miller were released from custody.  To read about the family feeling such relief and joy when they thought the murderers had been caught to the devastation when they learned the killer was still free is absolutely heartbreaking.  In fact, of the four cases featured in this post, all of them except for Marvin Anderson’s still remains an open case where the true perpetrator has not yet been caught.  When corners are cut, and rights are violated, it does not just harm the individuals charged with a crime, it harms the victims and their families as well.

Related posts

Second Chances

That One Dumb Mistake As a criminal attorney the phone call I receive the most, the online post I see the most, the question I get asked when I...

Leave a Reply