The Trouble with Eyewitness Testimony: Toward a model jury instruction on witness accounts


By Julie Jonas, Jevon Bindman, and David Herr

One of the most gripping moments in any trial occurs when an eyewitness identifies the defendant as the perpetrator of a crime in open court. But what the jury doesn’t know is this is primarily for show; the “real” identification occurred months earlier in a police lineup, photo array, or other identification procedure. Dozens of factors affect what the eyewitness observed, the witness’s ability later to recall that information to make the identification, and the witness’s confidence in the identification. And even the best-intentioned law enforcement officer or investigator can unintentionally bring bias to the procedure.

Recent scientific advances have led to a greater understanding of how the circumstances under which witnesses observe something affect their ability to recall that information later. Events that occur between the initial observation and the identification can also affect a witness’s memory. As a result, researchers have advanced numerous recommendations to educate all stakeholders on methods to ensure greater certainty in eyewitness identifications and greater understanding of their sometimes-counterintuitive limitations. 

These recommendations are essential to the justice system’s efforts to ensure that the correct person is held responsible for the wrong committed. But the justice system has been slow to incorporate these recommendations, and jurors are often uninformed about critical factors that affect the reliability of eyewitness identification. It is important to understand that these issues do not arise only on “one side of the v” in criminal cases—nor, for that matter, only in criminal cases. Any party proffering or challenging eyewitness (or “earwitness”) testimony may need to have the testimony put in proper perspective so that the factfinder can assess it fairly.

One solution, consistent with how we handle other matters of guidance to juries, is a more detailed jury instruction that provides useful information regarding the science behind identification and recall. A well-crafted jury instruction is a cost-effective way to educate jurors about best practices for obtaining a credible identification, particularly since the pertinent research is so broadly accepted. 

This article describes the current jury instruction used in Minnesota, its origins, and recent attempts by Minnesota courts to address this issue. It identifies factors not included in Minnesota’s jury instruction that may be helpful in weighing the credibility of a testifying eyewitness, and offers a sample jury instruction that addresses those factors.

Brathwaite: The unintended jury instruction

Like most states, Minnesota follows the constitutional standard for admission of eyewitness testimony set by the U.S. Supreme Court in Manson v. Brathwaite.1 Under that standard, if an identification procedure is unnecessarily suggestive, the evidence should be excluded unless the prosecutor can show through the totality of the circumstances that the identification is still reliable. In determining whether the evidence is admissible, a court may consider the opportunity of the witness to view the perpetrator, the witness’s degree of attention, the accuracy of the witness’s description, the witness’s degree of certainty, and the time elapsed between the crime and the identification. In the decades since 1977, the Supreme Court has provided little additional guidance regarding factors that make a lineup unnecessarily suggestive.

 Brathwaite provides a standard for judges to exclude unreliable eyewitness identification evidence. But it does not inform jurors how to weigh such evidence when it is admitted, and was never intended to do so. Nevertheless, Minnesota and many other states use Brathwaite as the foundation for their standard criminal jury instruction on eyewitness testimony. For example, Minnesota’s instruction (CRIMJIG 3.19) states: 

Testimony has been introduced tending to identify the defendant as the person observed at the time of the alleged offense. You should carefully evaluate this testimony. In doing so, you should consider such factors as the opportunity of the witness to see the person at the time of the alleged offense, the length of time the person was in the witness’s view, the circumstances of that view, including light conditions and the distance involved, the stress the witness was under at the time, and the lapse of time between the alleged offense and the identification. (If the witness has seen and identified the person before trial and after the alleged offense, you should also consider the circumstances of that earlier identification, and you should consider whether in this trial the witness’s memory is affected by that earlier identification.)2

The jury instruction identifies some of the factors considered under the Brathwaite test, such as the witness’s stress level, opportunity to observe, and lapse in time, but not others. And even for those factors that are mentioned, CRIMJIG 3.19 does not explain how those factors affect the reliability of a witness’s identification. For example, a juror might plausibly believe that increased stress sharpens a witness’s ability to form a clear memory, when in fact the opposite is true. 

More concerning, CRIMJIG 3.19 ignores many additional factors not included in Brathwaite that we now know affect a witness’s ability to make a reliable identification, including the following:

1. Weapons focus: The presence of a weapon during the event may focus the witness’s attention toward the weapon and away from the actor’s face. This may make the witness prone to a false identification or more susceptible to suggestion from other witnesses or interested parties.

2. Cross-racial effect: Differences in race or ethnicity between the witness and the person being identified affect the reliability of an identification. People are worse at identifying individuals with an ethnic or racial background different from their own.

3. Intoxication: Intoxication can inhibit a witness’s ability to focus on an event as it occurs. It can also prevent the formation of a clear and detailed memory.

4. Complexity: The presence of multiple actors can limit the witness’s ability to focus on specific people. 

5. Confidence statements: A confidence statement made at the time of the initial identification is more reliable than one made at a subsequent identification or at trial. Additionally, the very act of telling a witness that the suspect they chose was the “correct” one can make the witness feel more certain about the identification even when it is incorrect.

6. Exposure to extraneous information: Opinions, descriptions, or other identifications provided to a witness may affect the independence of the witness’s identification or the witness’s confidence in the identification.

7. Procedure for out-of-court identification: The procedure for the out-of-court identification also may affect the witness’s memory of the encounter. Factors include (1) whether the lineup “fillers” match the description of the suspect; (2) whether the lineup is “double blind” (i.e., did the administrator know which person in the line-up was the suspect and thereby intentionally or unintentionally bias the procedure?); (3) the quality and specificity of instructions given to the witness, including whether the witness is told that the suspect may be absent from the lineup; (4) whether the witness sees the suspect in more than one identification procedure; and (5) whether the lineup administrator gives confirmatory or post-identification feedback, which may afford the witness a false sense of confidence in the identification.

Like the factors discussed in Brathwaite, some of these additional considerations are easy to understand given the expected knowledge of most jurors, whereas others are either unfamiliar or maybe even counterintuitive. For example, most jurors will appreciate that a witness who is intoxicated will be less able to make a reliable identification. But the witness’s memory may also be affected by the actor’s race or ethnicity, the presence of a weapon, or multiple perpetrators, all of which tend to call into question the identification rather than bolster it. Other factors, such as degree of certainty, may correlate with reliability—but that depends on whether the witness was exposed to other opinions or identification results that artificially bolstered his or her confidence.

Educating the jury regarding these factors is obviously of paramount importance to criminal defendants, but prosecutors may also need to explain factors that affect witness memory when confronting a defendant’s alibi witness. These issues may likewise be implicated in civil cases, such as products liability (what product did the witness use and when?) and personal injury (who was at fault for the automobile accident?), or any situation where a party must prove a specific person is responsible for an action that gives rise to damages. Proper instruction on eyewitness testimony is therefore a problem that involves both the criminal and civil bars.

Moving beyond Brathwaite

In its 2014 report on the problems with eyewitness identification, the National Academy of Sciences recommended that courts give specific jury instructions that go beyond Brathwaite:

Jury instructions should explain, in clear language, the relevant principles [of assessing eyewitness identification].... [T]he instructions should allow judges to focus on factors relevant to the specific case, since not all cases implicate the same factors.... [J]ury instructions have tended to address only certain subjects, or to repeat the problematic Manson v. Brathwaite language, which was not intended as instructions for jurors.3

Although other states have adopted more extensive jury instructions based on this recommendation, Minnesota courts have continued to grapple with how better to educate juries on issues surrounding eyewitness identification. In one recent case, the district court was asked to provide a more detailed jury instruction on eyewitness identification that included additional factors shown to affect reliability. The lower court, holding that expert testimony was required before it would give an instruction in those areas, denied the instruction. The Minnesota Court of Appeals affirmed this result.4 Several months later, another district court denied a defense request for an expert to testify about the eyewitness identification issues that were present in that case. Again, the court of appeals affirmed, citing earlier Minnesota Supreme Court decisions that give district courts wide discretion in determining the admissibility of expert testimony.5 The Minnesota Supreme Court denied review of both cases. 

In 2019, another eyewitness identification case came before the court of appeals on the issue of jury instructions. The defense requested the standard jury instruction with additional cautionary language on the procedures used by the police in obtaining the identification, cross-racial identification, and weapons-focus distraction. The district court denied the instruction, the court of appeals affirmed, and the Supreme Court denied review.6

These decisions illustrate a perplexing Catch-22 for litigants. A district court decision to disallow expert testimony on eyewitness identification will not be overturned on appeal absent abuse of discretion, and a district court decision to deny a more case-specific jury instruction may be affirmed because an expert is required to support such an instruction. 

In 2017, the Minnesota Supreme Court issued an order that recognized the unique challenges presented by eyewitness identification evidence. The Court asked its Rules of Evidence Advisory Committee to review studies and literature in the area of eyewitness identification and to recommend potential reforms. The committee published its report on October 1, 2018, and made recommendations related to police practices in identification procedures, admissibility standards for identification evidence, the use of eyewitness identification experts, jury instructions, and appellate standards of review.7 The Supreme Court distributed the report to criminal justice professionals throughout the state and directed additional training for judges, but neither the Supreme Court nor the District Court Judges Association, which promulgates standard jury instructions, has adopted any of the specific recommendations.

In its report, the committee agreed that CRIMJIG 3.19 is inadequate and should be “updated and modernized,” but could not reach consensus on the details of an appropriate jury instruction.8 Some members preferred to simply add a list of criteria to the current rule, which would provide a measure of brevity, but would not explain the counterintuitive nature of certain factors. Others preferred an approach that would explain the prevailing science regarding the factors. This approach would provide the jury with more useful information, but it is longer and some of the factors may be better dealt with by expert testimony than jury instructions.

Example jury instruction9

The example jury instruction appended to this article attempts to address the primary issues surrounding eyewitness identification while acknowledging that a very detailed instruction runs the risk of losing jurors’ attention and being rendered ineffective. The content is highly dependent on the facts of the particular case, and it would be a rare case that would present all the issues covered by the instruction. Accordingly, the instruction would be tailored to include only the factors that are potentially implicated in each identification. We hesitate to claim that it is a model instruction, but there is no reason it, or a comparable counterpart, should not be regularly available for trial where eyewitness testimony is received. Other states have also adopted more extensive instructions that may provide useful language in appropriate Minnesota cases.10

We hope this example instruction provides a starting point for practitioners and judges to open a dialogue and to craft an instruction that provides jurors with useful information, yet does not confuse or overwhelm them. Jurors need this information in order to understand the complex factors that affect identification and memory. A jury instruction is the simplest and most economical way to impart this information. Ultimately, this is an issue that the judiciary should address to ensure that juries uniformly have the tools to evaluate eyewitness identifications and to minimize the risk of incorrect identifications. 

APPENDIX: Instruction on Eyewitness Identification Testimony 

Defendant requests the following special instruction in lieu of CRIMJIG 3.19 —Identification Testimony. Defendant requests a cautionary instruction before testimony by eyewitness(es) and at the close of the case.

The burden is on the state to prove beyond a reasonable doubt each and every element of the crime charged. This burden includes the burden of proving beyond a reasonable doubt the identity of the defendant as the person who [insert activity observed].

Identification testimony is an expression of belief or impression by an eyewitness. The value of identification testimony regarding the defendant depends on the opportunity the eyewitness had to observe whether or not the defendant was in fact the person who [insert activity observed] and to make a reliable identification of the defendant on a later occasion. 

In appraising the identification testimony of an eyewitness you may consider the following:

1. The level of stress the eyewitness was under at the time the eyewitness viewed the person. High levels of stress can diminish an eyewitness’s ability to recall and make an accurate identification.

2. Whether a weapon was present during the commission of the offense. When a visible weapon is used during a crime, it can distract an eyewitness and draw their attention away from the perpetrator. “Weapon focus” can impair an eyewitness’s ability to make a reliable identification and to describe what the perpetrator looks like if the crime is of a short duration.

3. The amount of time the eyewitness had to observe the event may affect the reliability of an identification. A brief or fleeting contact is less likely to produce an accurate identification than a more prolonged exposure.

4. The circumstances of the event, including the distance between the perpetrator and the eyewitness and the lighting conditions. Greater distance between an eyewitness and the perpetrator and/or poor lighting conditions can diminish the reliability of an identification by the eyewitness. 

5. Any intoxication impairing the eyewitness’s ability to later recall persons and events.

6. The lapse of time between the alleged offense and the identification. Delay between the commission of the crime and the time an identification is made by an eyewitness can affect the reliability of the identification. 

7. The ability of the eyewitness from one ethnic/cultural group to effectively recognize distinguishing features of a person of a different ethnic/cultural group.

8. The accuracy of the eyewitness’s prior description of the perpetrator.

9. Whether the eyewitness was exposed to opinions, descriptions, or identifications given by other witnesses or any other information or influence that may have affected the independence of the identification. Eyewitness memories can be altered when other eyewitnesses share information about what they observed. Statements and/or conduct of another witness can also affect the eyewitness’s confidence in their identification. 

10. Whether or not “unconscious transference” affected the eyewitness’s identification. “Unconscious transference” occurs when an eyewitness unconsciously identifies a person as a perpetrator because they may have seen the person in another situation or context.

11. The circumstances of the out-of-court identification by the eyewitness to law enforcement, including:

a. whether the identification was administered in a double-blind or blind fashion. Double-blind administrators do not know who the actual suspect is. Blind administrators are aware of who the suspect is but shield themselves from knowing where the suspect is located in the lineup or photo array;

b.  the quality and specificity of the instructions given to the eyewitness before viewing the photo lineup;

c. whether law enforcement gave confirmatory or post-identification feedback to the eyewitness. Confirmation can reduce doubt and engender a false sense of confidence in an eyewitness, and feedback can falsely enhance an eyewitness’s recollection of the quality of his or her view of an event; 

d. the composition of the photo lineup shown to the eyewitness, including whether the defendant stood out from other members of the lineup or photo array; 

e. whether the lineup was properly administered by law enforcement to the eyewitness; and

f. any statements or conduct by law enforcement after the identification was made that may have influenced the identification or the eyewitness’s confidence in the identification.

12. Any other factors given to you to assess the credibility of a witness.

JULIE JONAS is the legal director for the Innocence Project of Minnesota. She also teaches classes on wrongful convictions and supervises clinical students working on innocence cases at the University of Minnesota Law School and Mitchell Hamline School of Law. 

JEVON BINDMAN is an attorney at Maslon LLP and focuses his practice on construction and appellate litigation. He is a member of the Minnesota Supreme Court Rules of Evidence Advisory Committee and a contributor to Minnesota CLE’s Minnesota Courtroom Evidence Deskbook.

DAVID HERR is an attorney at Maslon LLP and is an experienced appellate lawyer. He is co-author of Minnesota Handbook of Courtroom Evidence and Minnesota Trial Objections, both published by West/Thomson Reuters.





1 Manson v. Brathwaite, 432 U.S. 98 (1977).

2 Minn. Practice Series, Jury Instruction Guide—Criminal 3.19.

3 Nat’l Acad. of Sciences, Identifying the Culprit: Assessing Eyewitness Identification 112 (2014).

4 State v. Thomas, 890 N.W.2d 413 (Minn. Ct. App. 2017), rev. denied (Minn. 3/28/2017).

5 State v. Anderson, No. A16-0565, 2017 WL 1157882 (Minn. Ct. App. 3/27/2017), rev. denied (Minn. 6/20/2017).

6 State v. Davis, No. A18-0758, 2019 WL 2262225 (Minn. Ct. App. 5/28/2019), rev. denied (Minn. 8/20/2019).

7 Report of the Minnesota Supreme Court Rules of Evidence Advisory Committee (10/1/2018), available at http://www.mncourts.gov/mncourtsgov/media/PublicationReports

8 Id. at 29.

9 The example jury instruction was drafted by Connie Iversen, a managing attorney in the Second Judicial District Public Defender’s Office. It has been modified slightly by the authors.

10 See, e.g., Mass. Model Eyewitness Identification Instruction (11/16/2015), available at https://www.mass.gov/files/documents/2016/11/sk/model-jury-instructions-on-eyewitness-identification-november-2015.pdf; N.J. Model Criminal Jury Charges (rev. 7/192012), available at https://www.innocenceproject.org/wp-content/uploads/2017/06/NJ-Jury-Instruction.pdf.