Scholars, lawyers, and courts around the country are pushing for reforms that will challenge the long-held assumptions on the reliability of eyewitness testimony. For a quick primer on the problems with eyewitness identifications, view the 60 Minutes piece Eyewitness: How Accurate is Visual Memory?” Featured below, the Exoneration Project has outlined the suggested reforms to current eyewitness identification practices utilized in criminal investigations, including a critique of the the policies of the Chicago Police Department. You may also view the recently released (2014) report by the National Research Council detailing best-practices to increase likelihood of accurate eyewitness identifications used in criminal cases.
Best-Practices for Eyewitness Identification
In the last several years, numerous state and city jurisdictions have overhauled the eyewitness identification procedures used by their officers when investigating a crime. These jurisdictions have implemented methods of conducting photo and physical lineups that have been scientifically proven to reduce the frequency of false-identifications. The implementation of these new procedures by progressive police departments around the country represents a cutting-edge transformation in the field of law enforcement, namely, a shift away from procedures conducive to suggestive identifications that easily corrupt the memories of victims and witnesses alike, towards a scientifically-proven assortment of “best practices” that promise to reduce the tragic injustice caused when the innocent are wrongfully identified as perpetrators of crime. Unfortunately, the City of Chicago has failed to keep pace with the innovations in the field of eyewitness identification, and currently employs anachronistic practices that continue to generate false identifications unless major reforms are enacted.
The scientific literature indicates that two factors, in particular, are essential to reducing the incidence of false-identifications when conducting lineups. First, lineups must be double-blind; the administrator must not know who is the suspect, and who are merely fillers. Double-blind administration has been called “the most important single reform that can be implemented to enhance the integrity of eyewitness identification evidence.” Double-blind administration eliminates the possibility for witnesses to be influenced by administrator’s conscious or unconscious reactions to suspects presented in a live or photo lineup. The states of New Jersey and North Carolina, as well as the cities of Denver, Boston, and Dallas, have already adopted double-blind lineup procedures. Several studies have confirmed the importance of blind administration. For instance, a meta-analysis by Nancy Steblay, Gary Wells, and Amy Douglass found that witnesses who were given confirming feedback by lineup administrators reported a “stronger basis for their identification,” “greater trust in eyewitnesses who have a similar experience,” “a better memory of faces,” and “increased willingness to testify about their identification.” All of these results were statistically significant. Double-blind administration prevents these insidious effects of officer bias.
Second, lineups must be sequential, not simultaneous. A meta-analysis of 497 lineups, published in 2011 by Gary L. Wells, Nancy K. Steblay, and Jennifer E. Dysart, found that the use of sequential lineups yielded 5.9% fewer false-identifications than the simultaneous method. Moreover, the study found that simultaneous and sequential procedures yield rates of correct suspect identification that are not different to a degree that is statistically significant. The researchers concluded, “The results are consistent with decades of laboratory research showing that the sequential procedure reduces mistaken identifications with little or no reduction in accurate identifications.”
The following list of procedures has been complied from the Best Practice Guidelines published by several jurisdictions that have recently implemented reforms, including New York State, Santa Clara County (California), the Commonwealth of Virginia, the State of New Jersey, and the City of Boston.
Inviting Witnesses to View a Lineup:
The administrator or officer who schedules the line-up should be neutral and non-suggestive when communicating with the witness, and should not inform the witness than arrest has been made in connection to the crime.
Instructions to Witnesses Before Viewing a Lineup:
- Witnesses should be told that the person who committed the crime may, or may not, be in the lineup.
- Witnesses should be told that the administrator does not know who the suspect is.
- Witnesses should be told that the lineup will be shown in random order.
- Witnesses should be told that all persons/photos selected for the lineup will be shown, even if/after an identification is made.
- Witnesses should be told that they will have as much time as they need when viewing the lineup. Moreover, witnesses should be shown the photos/persons a second time at their request. However, if a witness requests to see the lineup again, the entire lineup should be shown again sequentially, not just one photo or one person.
- Witnesses should be told that they will be required to state the level of certainty they have that the person(s) they identify is the perpetrator.
Lineups Should Be Sequential, Not Simultaneous:
In the case of a physical lineup, the witness should be shown the individuals one at a time. In the case of a photo lineup, the witness should be shown the individuals one at a time.
Physical and Photo Lineups Must Be Double-Blind:
The administrator of the lineup must not know who the suspect is in a given lineup or photo array.
- Persons should be selected who generally fit the witness’ description of the offender. The filers should resemble the suspect so that a person who knew the suspect would have a difficult time distinguishing him/her from the fillers.
- Photos used should be of the same size and composition. Black and white photos should not be mixed with color photos. Mugshots should not be mixed with snapshots, etc. (The Rodel Sanders Rule).
After the Lineup:
Officers should submit a report on every identification procedure, whether or not a subject is selected, the instructions given to the witness, the exact words spoken by the eyewitness pertaining to any identification made, and the witness’ statement of certainty that his/her identifications were accurate.
CPD Lineup Procedures
The Chicago Police Department has failed to keep pace with the recent reforms in police practices involving eyewitness identification, and continues to employ outdated methods that are have been scientifically proven to elicit false identifications with astounding frequency. [ 13 ] The Chicago Police Department’s most recent directive for lineup procedures [ 14 ] underlines a need for procedural re-evaluation.
The Chicago Police Department’s directive does not suggest, much less require, that photo or physical lineups be double-blind or sequential. The directive makes no mention of how important it is for officers and administrators to remain impartial when inviting witnesses to view a lineup or photo array. Moreover, the directive neither suggests, nor requires, that officers remain neutral while administering a lineup.
This is problematic given that fact that the person administering the lineup or photo array is often an investigator on the case, a fact that disallows for any semblance of administrator neutrality. Additionally, lineups are done simultaneously, a practice that willfully ignores the growing body of scientific research discussed earlier in this report. Finally, records are rarely kept of positive “filler” identifications— any identification made by witnesses must be recorded, even when incorrect suspect identifications are made.
Furthermore, while the directive requires that the officer submit a Supplemental Report after a physical or photo lineup is conducted, absent is any requirement that this report contain an account of the words spoken by the witness during the administration of the lineup, information that could speak to how confident the witnesses was of any identification made. Additionally, there is no requirement, much less a suggestion, that the witness be asked to state his/her level of certainty regarding any identification he/she makes, nor that this information be included in the Supplemental Report.