§ 10.03. Threshold for Conducting Eyewitness Identifications

Policing agencies should not conduct eyewitness identifications unless they have:

  • (a) a substantial basis to believe that the suspect committed the crime and should therefore be presented to the eyewitness, and
  • (b) a substantial basis to believe that the eyewitness can reliably make an identification.


a. Sufficient suspicion. Police should not place a suspect in an eyewitness identification procedure without a strong basis for doing so, including reasonable cause or suspicion that the suspect actually is responsible for the crime. Preferably, the officers should have evidence of guilt independent of the eyewitness’s belief that he or she can make an identification. In addition, officers, consistent with § 10.05, should not convey to the eyewitness any of that basis for suspecting a person, because that would constitute highly suggestive conduct. Live identification procedures must be conducted within the limits of any applicable rules on seizing persons.

b. Basis to conduct identification procedure. The decision to conduct an eyewitness identification procedure should not be undertaken lightly, or without adequate cause and evidentiary support. The strong basis to conduct such a procedure should include a basis to believe that the eyewitness can make an accurate identification. Officers should inquire into the circumstances concerning the eyewitness’s initial viewing of the suspect. Officers should not ask an eyewitness who lacks the ability, or who expresses an inability, to recall the appearance of the culprit to make an identification. In addition, officers, consistent with § 10.05, should not make any suggestions to the eyewitness that the eyewitness can make a successful identification.

c. No trawling. Officers normally should not conduct eyewitness identification procedures if they do not have a suspect. Officers should not engage in forms of “trawling,” which is the use of mugshot presentations of large sets of images of individuals for whom there is no cause for suspicion related to the incident in question. The risks of eyewitness error are too great to justify placing large numbers of innocent individuals at risk of having their images erroneously identified.

Reporters’ Notes

We do not know how often eyewitness identifications are conducted, but according to one estimate, they may be conducted in many tens of thousands of cases a year. Alvin G. Goldstein, June E. Chance & Gregory R. Schneller, Frequency of Eyewitness Identification in Criminal Cases: A Survey of Prosecutors, 27 Bull. Psychonomic Soc’y 71, 73 (January 1989). Yet, human facial recognition poses real challenges for individuals. Scientific research has documented how even under optimal viewing conditions, eyewitnesses can have great difficulty identifying strangers and even non-strangers.

Constitutional rulings do little to address the preliminary question that agencies face: whether to conduct an eyewitness identification procedure at all. The few lower courts to have considered the issue are divided on whether police must have probable cause under the Fourth Amendment to place an individual in a live (but not a photo-array) eyewitness identification procedure. Biehunik v. Felicetta, 441 F.2d 228, 230 (2d Cir. 1971); but see, e.g., Wise v. Murphy, 275 A.2d 205, 212-215 (D.C. 1971); State v. Hall, 461 A.2d 1155 (N.J. 1983). Mug-shot arrays or composite images, or photo arrays, are not regulated under the Fourth Amendment at all, since they do not involve a “seizure” of a person, but rather the person’s image.

The U.S. Supreme Court held that when officers do not engage in intentional conduct during an eyewitness identification, officers are not regulated under the Due Process Clause at all. Perry v. New Hampshire, 132 S. Ct. 716 (2012). Some state courts have adopted different rules, stating that reliability review does apply regardless of whether there was police action. See, e.g., State v. Chen, 27 A.3d 930, 937 (N.J. 2011).

In United States v. Wade, the Supreme Court held that, once indicted, a person has a right under the Sixth Amendment to have a lawyer present at a lineup. 388 U.S. 218, 235-236 (1967). However, that right does not extend to photo-array procedures, which are far more commonly used today than live or in-person lineups. U.S. v. Ash, 413 U.S. 300, 321 (1973); Gary L. Wells & Deah S. Quinlivan, Suggestive Eyewitness Identification Procedures and the Supreme Court’s Reliability Test in Light of Eyewitness Science: 30 Years Later, 33 Law & Hum. Behav. 1, 16 (2009) (a “large percentage of jurisdictions in the U.S. use only photographs and never use live lineups”).

It is essential, though, for agencies to determine whether an eyewitness and an identification procedure using that eyewitness are likely to be reliable. Unfortunately, many crimes occur under suboptimal viewing conditions. For example, research suggests that the presence of a weapon at a crime scene and heightened stress both can make it more difficult later to recall accurately the face of a suspect. Nat’l Research Council of the Nat’l Acads., Identifying the Culprit: Assessing Eyewitness Identification 93-94 (2014). In addition, eyewitness identification procedures often occur after the passage of time.

It is particularly important to resolve whether a witness is capable of making an eyewitness identification before proceeding with an identification procedure, because once an eyewitness is asked to make an identification and does so, confidence in the identification will predictably increase over time. An eyewitness may appear highly confident and reliable in court, even if the eyewitness was highly uncertain and tentative during an eyewitness identification procedure at a police station. This phenomenon is termed “confidence inflation.” Gary L. Wells et al., Eyewitness Identification Procedures: Recommendations for Lineups and Photospreads, 22 Law & Hum. Behav. 603-647 (1998).

Officers thus should carefully inquire into the viewing conditions under which an eyewitness saw the suspect, as well as the passage of time since the viewing occurred. Id. at 98. Officers should be trained to interview possible eyewitnesses in order to elicit as much information as possible and without asking leading questions that might suggest information to the witness. Research on cognitive interviewing can inform such training. Ronald P. Fisher & R. Edward Geiselman, The Cognitive Interview Method of Conducting Police Interviews: Eliciting Extensive Information and Promoting Therapeutic Jurisprudence, 33 Int’l. J. L. & Psychiatry 321, 321 (2010); Ronald Fisher, Interviewing Victims and Witnesses of Crime, 1 Psychol., Pub. Pol’y & L. 732, 735 (1995). Such interviews not only can produce descriptions of suspects, but they can inform an understanding of what factors may have affected the memory of an eyewitness, such as whether a weapon was present, or whether the event was highly stressful. See Henderson, 27 A.3d at 904-905 (“When a visible weapon is used during a crime, it can distract a witness and draw his or her attention away from the culprit.”).

Similarly, one safeguard before proceeding with an identification procedure is to test the face-memory ability of an eyewitness. Different people have differing abilities to remember the faces of strangers. Agencies can, as a matter of policy and practice, assess the face-memory ability of an eyewitness prior to deciding whether to conduct an identification procedure. One such test is the Cambridge Face Memory Test. See Cambridge Face Memory Test, at http://www.bbk.ac.uk/‌psychology/psychologyexperiments/experiments/facememorytest/startup.php.

Finally, agencies should be cognizant that eyewitnesses may seek to identify offenders on their own. The U.S. Supreme Court has held that when unreliability in eyewitness identifications is not due to intentional police action, it is not regulated under the Due Process Clause. However, agencies should seek to prevent situations in which eyewitnesses themselves, without the supervision of officers, search online and on social media, or in physical locations, in order to try to locate suspects. In doing so, they may be affected by suggestive circumstances, and police cannot control the viewing conditions or aim to prevent misidentifications. For those reasons, agencies should discourage such trawling activities and question eyewitnesses to ascertain and to document whether or how they have engaged in any such trawling. Agencies have not always been cognizant of the widespread and online dissemination of mugshot images. Eumi Lee, Monetizing Shame: Mugshots, Privacy, and the Right to Access, 70 Rutgers L. Rev 557 (2011). In addition to the privacy concerns raised by such dissemination, it also has the potential to contaminate eyewitness memory, if such mugshot repositories make it easier for eyewitness to try to locate suspects online.

Table of Contents