Are Eyewitness Accounts Wrong A Quarter of the Time?

Share:

Franky_CarrilloIn a few weeks, Franky Carrillo will speak at the Vanguard‘s annual Dinner and Awards ceremony.  His wrongful conviction, that cost him 20 years in custody, was based not just on the faulty memory of the witnesses but also intentional manipulation by law enforcement.

Carrillo was sentenced in 1992 to one life term and 30 years to life in prison after being convicted of one count of first-degree murder and six counts of attempted murder in a fatal drive-by shooting.

Mr. Carrillo had been in prison for nearly 14 years when a Sacramento-based attorney with the California Public Defender’s office, Ellen Eggers, learned of his case and met with Mr. Carrillo in prison in Folsom.

Critical to Mr. Carrillo’s case were six eyewitnesses who testified to seeing him pull the trigger.  It was the victim’s son who finally admitted to Ms. Eggers, “He hadn’t seen anything and he had relied entirely on one of the other witnesses because [he] had assured [the son] it was Franky.  But we already knew [from the second witness] that the Sheriff’s Department had told [him] who to pick.”

“So it all led back to a corrupt Sheriff’s deputy that made a mistake and thought he had the right guy but he didn’t,” Ms. Eggers told us.

Ms. Eggers convinced the defense team to hire an eyewitness identification expert, Scott Frasier, to do a complete reenactment of the crime.  Eventually she convinced the judge to go to the crime scene.

“[The reenactment] established that none of those witnesses could have seen what they claimed to have seen,” she said.  This meant it was not merely adult witnesses recanting years later – there was actually evidence to support that recantation.

It was the revelation that one could not make out the facial features in the lighting conditions that led even the prosecution to believe that Mr. Carrillo was not the shooter that night.

Even under the best of conditions, eyewitness identification is simply not reliable.  Researchers have noted that our minds do not work like tape recorders.

Writes the Innocence Project, “The human mind is not like a tape recorder; we neither record events exactly as we see them, nor recall them like a tape that has been rewound. Instead, witness memory is like any other evidence at a crime scene; it must be preserved carefully and retrieved methodically, or it can be contaminated.”

Moreover, memory is subject to contamination that is undetectable to the eyewitness.  Dr. Geoffrey Loftus, a Psychology Professor at the University of Washington, is one of the foremost authorities on memory and human perception.

In his testimony as an expert witness in a recent trial, he spoke of post-event information.  This is information, as its name implies, creates a coherent story rather than the fragmented information that is initially processed.  At the time, making this more coherent does not make it more accurate.

While post-event information makes it seem more “real,” the memory could actually be based on a false premise.  For instance, an eyewitness may hear from someone else that the attacker wore a certain color, and subconsciously accommodate that information to his or her memory of the event.

“A fundamental principal in American criminal justice is that one is innocent until proven guilty beyond a reasonable doubt,” writes Nancy Petro, a contributor to the Wrongful Conviction blog. “In the past two decades, DNA-proven wrongful convictions have revealed that we’ve routinely met the standard of ‘beyond a reasonable doubt’ with evidence that is quantifiably incorrect one-fourth of the time.”

“A 25 percent error rate in school has historically earned the very lackluster grade of D. A 25 percent margin of error would shutter any hospital and ground any airline,” she noted.  “But, in the criminal justice system, most Americans, blinded by trust in the system and a popular allegiance to ‘tough on crime’ policies, have yet to demand best practices in securing the most accurate evidence possible from those who have witnessed a crime.”

Studies show we lack an appreciation of just how delicate memory is and how much it is subject to contamination if not manipulation.

The Associated Press just ran an article that the state of Virginia, plagued like so many states by problems of wrongful conviction, is attempting to address the issue of false identifications during police lineups by overhauling its model policy on police lineups and eyewitness-identification procedures.

AP reports, “The guidelines call for double-blind identification lineups, meaning an investigator doesn’t know the suspect’s identity so he can’t give intentional or inadvertent clues to the eyewitness. The agency is also developing training sessions for police officers on best practices and the importance of adhering to such standards.”

“No one benefits from a misidentification. For the police, it hinders investigations, for prosecutors it burns witnesses,” said Rebecca Brown, the Innocence Project’s director of state policy reform.

University of Virginia Law School Professor Brandon Garrett, author of Convicting the Innocent: Where Criminal Prosecutions Go Wrong, and a key consultant on Virginia’s model eyewitness-identification policy, said, “It’s a very sensitive task.  If you do the lineup wrong, you can actually change the memory of the face they actually saw.”

That problem was made worse in the case of Franky Carillo, who believes that official misconduct played a crucial role in his conviction.  Five of the six witnesses, all young men, testified to seeing him shoot and kill the victim.

The young men were coaxed by sheriff’s deputies to pick Mr. Carrillo’s picture out of a lineup.  Each of those individuals have since recanted their original trial testimony.

However, the rate of error even under the best of conditions should alarm us.

“Studies have revealed that, among police lineups in which an eyewitness made a selection, the eyewitness chose a non-suspect filler 30 percent of the time,” Ms. Petro writes.

“If a witness fingers a police-selected colleague in the lineup, no harm is done. The police will not arrest him. However, if the witness selects a suspect who happens to be innocent, the stage is set for a wrongful conviction,” she added.

Barry Scheck and Peter Neufeld, in the landmark 1996 National Institute of Justice study, Convicted by Juries, Exonerated by Science: Case Studies in the Use of DNA Evidence to Establish Innocence After Trial found: “Every year since 1989, in about 25 percent of the sexual assault cases referred to the FBI where results could be obtained (primarily by state and local law enforcement), the primary suspect has been excluded by forensic DNA testing.”

Nancy Petro argues even when multiple people point out the same individual, misidentification can occur.

Professor Garrett found in a study of 190 DNA-proven wrongful convictions that multiple witnesses misidentified the same innocent person in 36 percent of the cases.

Ms. Petro cites the case of Dean Gillispie who was misidentified by three victims and whose clothing containing DNA evidence was discarded before it could be tested.

In an article for the Ohio Innocence Project, she found that the police in that case “violated virtually every best practice in lineup procedures.”  It took Mr. Gillispie 20 years to be freed, in December 2011.

A number of states following this wave of research are requiring improved procedures.

“Responsible citizens must urge our elected officials to require best practices in the criminal justice system,” Ms. Petro writes.  “Any case based solely upon eyewitness identification should raise red flags. It’s becoming increasingly obvious that eyewitness identification alone, with its 25-percent error rate, cannot prove guilt beyond a reasonable doubt.”

—David M. Greenwald reporting

Share:

About The Author

David Greenwald is the founder, editor, and executive director of the Davis Vanguard. He founded the Vanguard in 2006. David Greenwald moved to Davis in 1996 to attend Graduate School at UC Davis in Political Science. He lives in South Davis with his wife Cecilia Escamilla Greenwald and three children.

Related posts

12 thoughts on “Are Eyewitness Accounts Wrong A Quarter of the Time?”

  1. jimt

    “Any case based solely upon eyewitness identification should raise red flags. It’s becoming increasingly obvious that eyewitness identification alone, with its 25-percent error rate, cannot prove guilt beyond a reasonable doubt.”

    In the real world, is is not true that there are cases where the eyewitness has very clear viewing conditions and is absolutely confident of perp identification; just as there are cases where viewing conditions are poor and the eyewitness thinks he may be able to identify the perp (or someone who resembles the perp); but is unsure of absolute identification?

    So what about asking the witness his degree of confidence in identifying the suspect?
    For example, if the eyewitness is 100% sure the perp was tall and white, but is unsure of the facial features, at least short or dark-skinned suspects can be ruled out.

    Is an eyewitness allowed to testify that the suspect resembles the perp he saw; but that he can’t be absolutely sure?
    Maybe eyewitnesses are pressured too much to make absolute statements (yes or no) on identification of the suspect; when they perhaps should be allowed to make more qualified statements.
    For example if there are 5 witnesses who testify the suspect strongly resembles the perp they saw (but they can’t be 100% sure); is this admissible as evidence? By itself it would not be enough for a conviction; but perhaps together with other evidence it may tip the scales for (or against) conviction.

  2. medwoman

    jimt

    I understand your rationale. However, if the standard for conviction is “beyond a reasonable doubt” where would one draw the line ? 75% sure? 90% ? I would have a real problem with this. Of course, our system will almost always exclude me from jury duty as my past experience has shown, so it really doesn’t matter.

  3. E Roberts Musser

    [quote]I understand your rationale. However, if the standard for conviction is “beyond a reasonable doubt” where would one draw the line ? 75% sure? 90% ?[/quote]

    I think you are missing the point. Suppose the witness could say for an absolute certainty the perp wore a red shirt. He cannot identify the perp beyond that, but he can say what the color of the perp’s shirt was. Shouldn’t the witness be able to testify that no, he could not positively identify the perp, other than he wore a red shirt? Suppose the witness could identify the perp’s race. Shouldn’t that be admissible evidence? Eyewitness identification is not necessarily black and white (pardon the pun); or 100% or not 100% certain. The witness should be asked what things can s/he be sure of…

  4. David M. Greenwald

    The question is whether their recollection is accurate and one thing the research has discovered is that if there is a witness, they overhear the police and hear something about a red shirt. It is possible that they will absorb that post event information and it will become part of their memory – accurate or not.

  5. medwoman

    Elaine

    [quote]I think you are missing the point. Suppose the witness could say for an absolute certainty the perp wore a red shirt. He cannot identify the perp beyond that, but he can say what the color of the perp’s shirt was. Shouldn’t the witness be able to testify that no, he could not positively identify the perp, other than he wore a red shirt? Suppose the witness could identify the perp’s race. Shouldn’t that be admissible evidence? Eyewitness identification is not necessarily black and white (pardon the pun); or 100% or not 100% certain. The witness should be asked what things can s/he be sure of…[/quote]

    Perhaps I misinterpreted jimt’s proposal, or perhaps I did not express my concern well. I understood the proposal to be that witnesses estimate their degree of certainty, not that they state clearly of what they were certain, and of what they were uncertain. The problem with asserting certainty is that this can be seen as much as a characteristic of the witness as the accuracy of what they are reporting. Some people are simply, because of their nature, much more certain that they are “right” once an opinion is formed and much more likely to be convincing in relating their certainty. Others, by their nature are much more likely to be more tentative in their convictions. This does not mean that they are less accurate observers, just that their observations are less likely to carry weight with a jury. This is not about finding the truth,
    it is about how to sell your case to a jury.

  6. Superfluous Man

    ERM,

    [i]The witness should be asked what things can s/he be sure of…[/i]

    Isn’t the contention that eyewitnesses may believe they are sure of what they observed, but the degree to which their recollection is accurate varies or is completely inaccurate?

  7. E Roberts Musser

    [quote]The question is whether their recollection is accurate and one thing the research has discovered is that if there is a witness, they overhear the police and hear something about a red shirt. It is possible that they will absorb that post event information and it will become part of their memory – accurate or not.[/quote]

    Then I would advise that the identification procedures be cleaned up so that bias of the police does not enter into the picture. When my son had to identify the guy who beat him up, he was given six photos by the police. The police said nothing, just handed him the photos. My son had no problem identifying his attacker. As it turned out, the perp who attacked my son and was identified in the photo line-up had just attacked and severely beaten a helpless inebriated college student. Eyewitness identification is a useful tool. It just has to have certain limiting parameters to make sure bias does not creep in.

  8. E Roberts Musser

    [quote]Isn’t the contention that eyewitnesses may believe they are sure of what they observed, but the degree to which their recollection is accurate varies or is completely inaccurate?[/quote]

    No evidence is 100% perfect. That does not mean one doesn’t use evidence, even if imperfect. For instance, some DNA evidence is used, even if flawed. In fact, DNA is given as a probability as to its degree of accuracy. Evidence has to be taken in the aggregate. It is the sum total of the evidence that will determine guilt, not any one piece. The defense is always free to attack eyewitness testimony during trial as imperfect or flawed because of the way it was done. But because eyewitness testimony can be inaccurate does not, IMO, stand for the proposition it should never be used/allowed/depended upon. I’m all for improving identification procedures, but not for eliminating eyewitness testimony as admissible evidence…

  9. David M. Greenwald

    “Then I would advise that the identification procedures be cleaned up so that bias of the police does not enter into the picture.”

    We did a big story on how that is happening.

    “Eyewitness identification is a useful tool. It just has to have certain limiting parameters to make sure bias does not creep in. “

    It can be, the problem is that like any tool not only can it be misused, but we might not be able to detect when it has been.

  10. David M. Greenwald

    “No evidence is 100% perfect.”

    The question is whether a 75% success/ 25% failure rate is good enough. I would argue that it’s not.

  11. jimt

    From what I can discern in the article above by David, all the eyewitnesses are lumped into one group, and the resultant failure rate is 25%.

    How about a more thorough study that details the responses of each group; the failure group vs. the success group? I would contend that nearly all eyewitnesses with good clear viewing conditions and reliable memories can be >99% sure of a suspect identity (perhaps witnesses can be tested for visual acuity and for facial memory; both simple tests). I strongly suspect most of those in the 25% failure group were pressured to make a definitive statement about the suspect identity; and even though they may have been uncertain; gave a statement of certainty due to this pressure. It seems to me that this is the problem that needs to be addressed, and also to allow qualified statements of suspect ID into testimony–I think juries are intelligent enough to deal with uncertainty in identification; after all facial recognition is an everyday type of situation for everyone; and often there is some uncertainty (Is that really Joe over there?–oh, what a terrible new haircut).

    David, I’m not sure what you are proposing? That eyewitness testimony should be thrown out altogether, or discounted? I would certainly agree there is room for improvement, and that juries should be advised of potential problems with eyewitness testimony, etc.

  12. David M. Greenwald

    I would mainly suggest that types of measures already being put into place as safeguards for how and under what conditions people are asked to identify the subject. Second, I think your point about conditions is a very good one, we need a full understanding about the conditions under which a person made the identification. Third, I think we also need to understand even under good conditions, the chance that stress and contamination can occur. My real concern is our ability to control the environment because I fear that sometimes contamination can be both inadvertent and undetectable. I actually believe that the failure rate of identification is far higher than 25%, what keeps the number that low is that other police investigation work bolsters some identifications while throwing out others and that would be another “check” I would make sure to add – there needs to be more than just an identification to link someone to a crime and convict them beyond a reasonable doubt.

Leave a Reply

X Close

Newsletter Sign-Up

X Close

Monthly Subscriber Sign-Up

Enter the maximum amount you want to pay each month
$ USD
Sign up for