National Academy of Sciences Issues Landmark Report on Memory and Eyewitness Identification


Innocence Project Urges Adoption of Its Recommendations for Improving Police Identification Procedures


Should Hasten Trend Started by New Jersey and Oregon Supreme Courts in Overhauling the Way Courts Handle Identification Evidence


Contact: Paul Cates,

Nick Moroni,


(Washington, D.C. – October 2, 2014) Today the National Academy of Sciences issued a landmark report evaluating the scientific research on memory and eyewitness identification. Researchers conducted an in-depth review of three decades of basic and applied scientific research on eyewitness identification and provided recommendations for improving police identification procedures and for how courts handle eyewitness evidence. The Innocence Project, which has long advocated for many of the reforms recommended in the report, urges states and courts across the nation to enact the recommendation in order to prevent wrongful convictions.


“This report should serve as a powerful incentive for states and courts around the nation to enact reforms that will prevent eyewitness misidentifications. We’ve known for quite some time that eyewitness testimony is simply not as accurate as juries often believe, but we now have a definitive report that has analyzed three decades of science and makes proven recommendations for how law enforcement and the courts can prevent innocent people from being wrongly arrested and convicted,” said Barry Scheck, Co-Director of the Innocence Project, which is affiliated with Cardozo School of Law.


According to Innocence Project, eyewitness misidentifications contributed to 72% (229) of the 318 wrongful convictions that were later overturned by DNA evidence. The real perpetrators were eventually identified in 90 (39%) of these cases. While the innocent where languishing behind bars in these cases, the real perpetrators committed an additional 98 additional violent crimes (63 rapes, 17 murders, and 18 other violent crimes).


Recognizing that police eyewitness identification procedures can have a big effect on the accuracy of identifications, the report endorsed the following best practices, which have long been supported by the Innocence Project as a means to reduce the likelihood of wrongful convictions:

  • Blind Administration — Research shows that the risk of misidentification is sharply reduced if the police officer administering a photo or live lineup is not aware of who the suspect is. This prevents the witness from picking up intentional or unintentional clues from the officer conducting the lineup.
  • Confidence Statements — Immediately following a lineup, the eyewitness should be asked to describe in his or her own words how confident he or she is in the identification. As the report notes, the level of confidence a witness expresses at the time of trial is not a reliable predictor of accuracy. Having the witness describe their level of confidence at the time an identification is made will provide juries with a useful tool for judging the accuracy of the identification.
  • Instructions — The person viewing the lineup should be told that the perpetrator may not be in the lineup and that the investigation will continue regardless of whether the witness identifies a suspect.
  • Videotape the procedure — The report recommends that police electronically record the identification procedure to preserve a permanent record of the procedure.

Ten states have already uniformly adopted these best practices through law, policy or court action, and many jurisdictions around the country have voluntarily adopted policies embracing these important best practices. Just last year, the International Association of the Chiefs of Police (IACP), the world’s oldest and largest organization of police executives, came out in support of these reforms.


The report notes that the legal standard that most courts use regarding the admissibility of eyewitness testimony was established before most of the scientific research was conducted. However, landmark decisions by the New Jersey and Oregon Supreme Courts have already taken note of the robust research on memory and identification in overhauling the way courts in those states deal with identification evidence. Today’s report should help to accelerate this trend by making the following recommendations for courts:

  • Conduct pre-trail judicial inquiry — Judges should inquire about the eyewitness evidence being offered. If there are indicators of unreliable identifications, judges could limit portion of the eyewitness’ testimony or instruct the jury on how to properly evaluate the reliability of the identification based on the scientific research.
  • Make juries aware of prior identifications — Because in court identifications can unduly influence the jury, juries should hear detailed information about any earlier identification, including the confidence the witness expressed at the time of the identification.
  • Permit expert testimony — The report recognizes that expert witnesses who are capable of explaining the nuances of memory and identification are helpful in assisting juries in how to evaluate eyewitness testimony and should be permitted. The report also encourages local jurisdictions to provide funding to defendants to engage qualified experts.
  • Better instruct juries — Jury instructions can be used to educate jurors on how to properly evaluate the factors affecting eyewitness identifications and should be tailored to the relevant facts in a particular case.

Additional information about eyewitness misidentification is available at

. A copy of today’s report is available at



Leave a Reply

Thank you for visiting us. You can learn more about how we consider cases here. Please avoid sharing any personal information in the comments below and join us in making this a hate-speech free and safe space for everyone.

This field is required.
This field is required.
This field is required.

We've helped free more than 240 innocent people from prison. Support our work to strengthen and advance the innocence movement.