Current Eyewitness Identification Procedure Reinforce False Memories and Lead to Wrongful Convictions

By: Houston Criminal Lawyer John Floyd and Paralegal Billy Sinclair

There have been 258 DNA exonerations in this country over the last two decades, according to the New York-based Innocence Project. In approximately 75 percent of those cases, eye misidentification played a significant role. It is an issue we have thus far blogged about four times this year (here, here, here, and here) and four times last year (here, here, here, and here)—the latter two 2009 posts dealing with the wrongful conviction of Timothy Cole.


The wrongful conviction of Cole is a tragic affair no matter how it is viewed. It has had much the same impact on the Texas criminal justice system as the 1999 wrongful conviction of Clarence Elkins, Sr., had on the Ohio criminal justice system. The same year Elkins was wrongfully convicted Cole died in a Texas prison from asthma complications. Fourteen years earlier he had been a 26-year-old student at Texas Tech University. The university and the entire Lubbock community were under siege from a serial rapist who had sexually assaulted five women between December 1984 and April 1985. The fifth woman attacked was 20-year-old Tech student Michele Mallin.


In an effort to apprehend the serial rapist, the local police assigned an undercover female officer to hang around the university campus. One evening the officer walked into a popular pizzeria frequented by students. Timothy Cole happened to be in the pizzeria. After having a coke, the officer got up and walked out of the restaurant. Cole followed. He walked directly to his car, but before driving off, he pulled up alongside the undercover officer who was strolling down the street. Cole struck up a conversation with officer and they traded names, although the officer refused to give him her telephone number. Cole drove off.


Since Cole was the only man who approached the officer that night, she turned his name into investigators working the serial rapist case. They ran Cole’s name and discovered he had reported being robbed at a local pool hall several weeks earlier. When the police went to investigate the robbery report, they noticed he had a weapon that appeared to have been fired. This led to a search of Cole—a search that revealed he had a small amount of marijuana in his possession. He was arrested on misdemeanor drug and weapon charges. One of the arresting officers in this case was the same “undercover” officer who turned his name into the serial rapist investigators.


These two totally unrelated and dissimilar events, somehow in the minds of the Lubbock Police Department, made him a “suspect” in the serial rape case. Investigators took a Polaroid photo from Cole’s misdemeanor file, put five much smaller mug-shots with it, and presented this highly suggestive photo lineup to Michele Mallin who immediately identified Cole as her assailant. The lid on his wrongful conviction was sealed from that point. It would take the Texas justice system more than two decades to fully and completely own up to Timothy Cole’s wrongful conviction, even after the real rapist had been identified and found in the state’s prison system serving time for a series of other rapes.


But, finally, the system did respond to this horrific injustice—a man wrongfully convicted who spent 14 years and eventually died in prison without having his name cleared. In 2009 the Texas Legislature created the Timothy Cole Advisory Panel on Wrongful Convictions (“Panel”) and charged it with the responsibility of providing assistance and advice to the Task Force on Indigent Defense (“Task Force”) which has been charged with the responsibility of conducting a study designed to cover a broad array of issues relating to wrongful convictions. The study will include the following areas:


  • The causes of wrongful convictions;
  • The procedures and programs that may be implemented to prevent future wrongful convictions;
  • The effects of state law on wrongful convictions, as determined based on state statues regarding eyewitness identification procedures, the recording of custodial interrogations, post-conviction DNA testing, and writs of habeas corpus based on relevant scientific evidence; and
  • Whether the creation of an innocence commission to investigate wrongful convictions would be appropriate.


On August 26, 2010 the Panel issued its “report” putting forth a comprehensive package of recommendations which, if implemented, would be the first steps toward improving the quality and integrity of the Texas criminal justice system. Perhaps the most significant of these recommendations focuses on methods for improving the way more than 1,000 Texas law enforcement agencies prepare and administer eyewitness identification procedures. These methods include the following five proposals:

  • The State of Texas should require Bill Blackwood Law Enforcement Management Institute of Texas (LEMIT) to work with scientific experts in eyewitness memory research and law enforcement agencies to develop, adopt, disseminate to all law enforcement agencies, and annually review a model policy and training materials regarding administration of photo and live lineups. That model policy should comport with science in the areas of cautionary instructions, filler selection, double-blind administration, documentation of identification procedures, and other procedures or best practices supported by credible research.
  • The State of Texas should require all law enforcement agencies to adopt eyewitness identification procedures that comply with a model policy promulgated by the Bill Blackwood’s LEMIT.
  • The State of Texas should integrate training on eyewitness identification procedures into required curricula of the Bill Blackwood’s LEMIT and the Texas Commission on Law Enforcement Standards and Education (TCLESE).
  • The State of Texas should permit evidence of compliance or noncompliance with the model policy to be admissible in court.
  • The State of Texas should allow law enforcement agencies discretion on the adoption of sequential procedures.


The first proposal contains the “red meat” of the changes that need to be made in law enforcement’s witness identification procedures. We will discuss each of these proposed changes.

Memory Research


In 1999, Stanford Law School Professor Laura Engelhardt, writing in the Stanford Journal of Legal Studies, discussed a presentation given to the law school by memory researchers Barbara Tversky, Professor of Psychology, and George Fisher, Professor of Law, about the problems associated with eyewitness testimony. Professor Engelhardt succinctly opened the discussion with this preliminary observation: “The bedrock of the American judicial process is the honesty of witnesses in trial. Eyewitness testimony can make a deep impression on a jury, which is often exclusively assigned the role of sorting out credibility issues and making judgments about the truth of witness statements. Perjury is a crime, because lying under oath can subvert the integrity of a trial and the legitimacy of the judicial system. However, perjury is defined as knowingly making a false statement—merely misremembering is not a crime. Moreover, the jury makes its determinations of witness credibility and veracity in secret, without revealing the reason for its final judgment. Recognizing the fallibility of witness memories, then, is especially important to participants in the judicial process, since many trials revolve around factual determinations of whom to believe. Rarely will a factual question result in a successful appeal—effectively giving many parties only one chance at justice. Arriving at a just result and a correct determination of truth is difficult enough without the added possibility that witnesses themselves may not be aware of inaccuracies in their testimony.”


Studies on eyewitness testimony are replete with examples of groups of subjects simultaneously witnessing a staged crime only to present widely divergent accounts short after the episode. For example, Professor Engelhardt wrote about studies conducted in the 1970s by University of California psychologist Elizabeth Lofus who conducted experiments dealing with the effects of third parties injecting false facts into memory. In one experiment the subjects were shown a “yield sign” but asked about a “stop sign” while other subjects were shown “stop signs” but asked about “yield signs.” The results revealed the subjects remembered the false image. “In the initial part of the experiment,” Englehardt wrote, “subjects also viewed a slide showing a car accident. Some subjects were later asked how fast the cars were traveling when they ‘hit’ each other, others were asked how fast the cars were traveling when they ‘smashed’ into each other. Those subjects questioned using the word ‘smashed’ were more likely to report having seen broken glass in the original slide. The introduction of false cues altered participants’memories.”


Earlier this year we posted a blog about a Scientific American (Jan. 6, 2010) article titled “Why Science Tells Us Not to Rely on Eyewitness Accounts” written by Hal Arkowitz and Scott Lilienfeld. The researchers informed us that: “ … many people believe that human memory works like a video recorder: the mind records the event and then, on cue, plays back an exact replica of them. On the contrary, psychologists have found that memories are reconstructed rather than played back each time we recall them.” The two authors also called upon Elizabeth Loftus who explained the act of remembering is “more skin to putting puzzle pieces together than retrieving a video recording.”


This “reconstruction” process produces a slew of false memories. “Many researchers have [even] created false memories in normal individuals,” Arkowitz and Lilienfeld wrote. “What is more, many of these subjects are certain that the memories are real. In one well-known study, Loftus and her colleague Jacqueline Pickrell gave subjects written accounts of four events, three of which they had actually experienced. The fourth story was fiction; it centered on the subject being lost in a mall or another public place when he or she was between four and six years old. A relative provided realistic details for the false story, such as a description of the mall at which the subject’s parents shopped. After reading each story, subjects were asked to write down what they remembered about the incident or indicate that they did not remember it all. Remarkably about one third of the subjects reported partially or fully remembering the false event. In two follow-up interviews, 25 percent still claimed that they remembered the untrue story, a figure consistent with the findings of similar studies.”


We’ve all heard the great “fish story” where the fish grows larger with each retelling of the story. Like most repeated exaggerations, the fisherman soon believes the embellished stories. And that’s exactly what happens with most eyewitness identifications, as the one Ms. Mallin made of Timothy Cole. “Once witnesses state facts in a particular way or identify a particular person as the perpetrator, they are unwilling or even unable—due to the reconstruction of their memory—to reconsider their initial understanding. When a witness identifies a person in a lineup, he is likely to identify that same person in later line-ups, even when the person identified is not the perpetrator. Although juries and decision-makers place great reliance on eyewitness identification, they are often unaware of false memories.”


Professor Engelhardt further explained that it is the actual “retelling” of story that affects memory because stories are rarely told in a “neutral fashion.” Stories are tailored for the listener, particularly by our own biases or a bias recognized in the person listening to the story. And even when the police and prosecutors do not assist or manipulate a witness into an identification, “the mere act of telling prosecutors what happened [by the witness] may bias and distort the witness’s memory,” said Engelhardt. “Eyewitness testimony, then, is innately suspect.”

That is precisely why law enforcement authorities must understand the dynamics of memory, particularly its “reconstruction” process and its innate tendency to distort or misremember. Because a witness’ bias inevitably influences the accuracy of any identification, good or bad, it is crucial that law enforcement not give, no matter how inadvertent, misleading or suggestive information to a witness. As Engelhardt pointed out: “ … the very act of forming a memory creates distortion.” Crime eyewitnesses, therefore, do not need any corrupting influences from either the police or prosecutors—like the highly suggestive photo lineup employed in the Timothy Cole case.




A sequential lineup is one where an eyewitness is shown each photo individually and asked specifically if that photo is the perpetrator. Many researchers have concluded based on preliminary data that (here) this is a superior method over the simultaneous method (displaying all the photos in one array). But there is a drawback as pointed out by the Panel: while the sequential method may reduce the number of false identifications, it may also reduce the number of correct identifications. At least one study has shown that witnesses who view a simultaneous lineup identify the suspect more often than those who view a sequential lineup, 59.9 to 45 percent; choose a filler less often than those who view a sequential lineup, 2.8 to 9.2 percent; and are less likely to choose no one than those who view a sequential lineup, 37.6 to 47.2 percent.


The latter drawback must be acceptable. The simultaneous method simply does not lend itself to independent recollection. The eyewitness knows that one of the photos is the perpetrator. He/she will then look for a photo that matches the witness’ biased recollection, thus, any photo which remotely resembles the initial biased recollection will become the perpetrator. In the sequential method, the witness must make a decision on the single photo before him/her—there is no opportunity to “shop” among the array to find the photo that matches the initial biased recollection. In a sequential photo lineup, the identification process is narrowed: it either is or isn’t. There is no “wait and let me see all of them first.” The sequential method, therefore, forces the very process of memory to focus, to sharpen its recollection of details that may assist in a reliable identification.




The administrator of the lineup does not know who the suspect is. Prior to the actual lineup, the administrator collects a “confidence statement” from the witness. The Panel reported the double blind method is important because the witness’ identification will not be contaminated by any feedback from the administrator. If the administrator knows who the suspect is, he/she can influence the witness’ identification, perhaps unintentional, with both verbal and nonverbal cues. To make sure the administrator does not attempt to influence the lineup deliberately or inadvertently, he/she should know that one of the “fillers” in the lineup is a police officer but the administrator doesn’t know which one is the actual officer. The Panel reported that “researchers have found that these measures all but eliminate administrator influence from the procedures.”




We warmly embrace the recommendations made by the Panel. Having said that, let us say that we strongly urge the Task Force to follow the lead of the excellent “concurring report” prepared by University of Houston law professor Sandra Guerra Thompson who recommends that “single suspect” show-ups be completely eliminated in the State of Texas through legislation.  A single suspect show-up is the procedure used when a law enforcement officer brings a potential suspect to the witness and asks “is this the person.”  Thompson’s recommendation makes the most sense because, as the Panel itself pointed out, 20 percent of the nation’s DNA exonerations involved the highly suggestive “single suspect” show-ups. The panel’s recommendation that the legislature enact a “training bill” is not enough to cure this tragic problem.


But if the Task Force elects not embrace Thompson’s recommendation that there be a statutory elimination of “single suspect” show-ups, then it should at the very least strongly recommend that law enforcement agencies follow the lead of a 1999 U.S. Justice Department Report titled “Eyewitness Evidence: A Guide for Law Enforcement” governing such show-ups, and cited by Professor Thompson, which require lineup administrators to “(1) document a witness’ description of the suspect prior to a show-up and (2) separate witnesses during a show-up. It recommends that if a witness makes a positive identification [sic], investigators should consider using other types of identification procedures for all subsequent confirmatory identifications, rather than this highly suggestive method. As with lineups and photo arrays, the DOJ report also requires investigators to give cautionary instructions to the witness that the person in the show-up may or may not be the perpetrator, and it urges investigators to obtain a statement of the witness’s certainty following a positive identification and maintain written documentation of that statement. Specifically, the DOJ report requires written documentation of the time, place, and result of the show-up. The Innocence Project further recommends that the show-ups occur in a neutral, non-law enforcement location, without handcuffs (when practicable), and with the suspect removed from the squad car. It also recommends that show-ups be videotaped when practicable.”


We understand that the Panel’s report was not warmly embraced by everyone. Attorney Jeff Blackburn, who has worked closely with the Innocence Project of Texas, had pointed criticism of report in a guest column for the Grits for Breakfast website. “Intention and effort aside,” Blackburn wrote, “what the panel finally came up with was a largely watered-down version of what has already been in play in prior legislative sessions. The panel could have gone much further. Instead, it chose to take a predictable path of limited resistance and avoid controversy.”


While we share in the sentiment, we must also consider the report in the political arena in which it evolved. If what the Panel has recommended did not make it through “prior legislative sessions,” it is unlikely that going “much further,” with more comprehensive recommendations, would anything other than diminish the chances of success. At least these recommendations, which concededly have failed in prior legislative sessions, are now coming from an “advisory panel” set up by the Legislature itself. Even a Republican-dominated legislature will have some serious explaining to do if it rejects, in effect, its own recommendations. Even conservative, law-and-order minded Texans have grown weary and disgusted with repeated, highly publicized cases of innocent people being wrongfully convicted and who have spent decades in prison.


We are grateful that the Panel took its recommendations as far as it politically could and we truly admire the courageous “concurring report” by Professor Thompson. The issue of wrongful convictions is not easy bronc to saddle and ride in the State of Texas.  The state leads the nation in such convictions. It is both a dubious and highly sensitive political distinction. We were buoyed when the Legislature created the Panel and we are encouraged that lawmakers may now be willing to put forth a legal, moral and financial effort to reduce the state’s number of wrongful convictions through the Panel’s efforts. And if, at the end of the day, the only thing we get on the identification issue is a “training bill” as suggested by attorney Blackburn, then we will consider that a first step, albeit a minor one, in the right direction.


And, fittingly, it may be the lasting legacy of Timothy Cole, who this past March was posthumously pardoned by Texas Gov. Rick Perry, to have these “baby steps” taken in his name. Our prayer and hope is that no future governor has to issue such a pardon—saying “we forgive” over the grave of a person who did no wrong has a hollow ring. A pardon is an official act of forgiveness. We can only hope that the soul of Timothy Cole has forgiven us.


By: Houston Criminal Attorney John Floyd and Paralegal Billy Sinclair