Should Eyewitness Testimony Be Admissible in Court?

Eyewitness testimony is admissible in every U.S. courtroom, and no serious reform movement proposes banning it. The real question behind “should eyewitness testimony be admissible in court” is narrower and more useful: given how often eyewitnesses are wrong, are the safeguards around their testimony strong enough? Roughly 62 percent of convictions later overturned by DNA testing involved a mistaken eyewitness identification. That number is why the debate exists, and it is why the answer is less about keeping witnesses off the stand than about controlling what happens before they get there.

Why Outright Exclusion Isn’t on the Table

A victim who watched an attacker’s face for thirty seconds in good light has evidence worth hearing. The system has never seriously entertained silencing that witness. What it has entertained, and mostly rejected, is blocking questionable identifications from reaching juries.

In the entire history of Supreme Court rulings on eyewitness identification, only one case has resulted in suppression on due process grounds. In Foster v. California (1969), police ran a lineup where the suspect towered over the other participants and wore clothing matching the witness’s description, then staged a one-on-one meeting when the witness still couldn’t identify him, and finally placed him in a second lineup. The Court suppressed the identification 5–4.¹ That is the ceiling of how egregious a procedure has to be before the Constitution forces exclusion. Nearly every contested identification makes it to trial, where the jury decides what to do with it.

The Constitutional Rule That Lets Most Identifications In

The Supreme Court first addressed suggestive identification procedures in Stovall v. Denno (1967), ruling that due process depends on the totality of the circumstances. A decade later in Manson v. Brathwaite (1977), the Court declared reliability the “linchpin” of admissibility: even a suggestive procedure can produce an admissible identification if a judge finds it reliable overall, balancing the corrupting effect of the procedure against factors supporting accuracy.²

In 2012, the Court drew a hard boundary around that framework. Perry v. New Hampshire held that due process requires no preliminary judicial screening of eyewitness reliability unless law enforcement arranged the suggestive circumstances.³ If a witness happens to spot a suspect in handcuffs in the back of a police car and identifies him on the spot, that encounter is inherently suggestive, but because police didn’t orchestrate it, the identification faces no constitutional gatekeeping. The defense is left with cross-examination.

So the constitutional due process test has two threshold requirements before a judge even reaches reliability. The procedure must have been arranged by law enforcement, and it must have been both suggestive and unnecessary.⁴ A procedure is suggestive when it steers the witness toward one person, like a show-up with a single suspect or a lineup where only the suspect matches the earlier description. It is unnecessary when officers had no pressing reason to skip a fairer alternative. When police rush a suspect to a hospitalized victim who may not survive, the suggestiveness is justified by the emergency.

The Five-Factor Reliability Test and Its Weak Spots

When a judge does evaluate a challenged identification, the analysis follows five factors from Neil v. Biggers (1972), reaffirmed in Manson:⁵

Opportunity to view: how much time the witness had and under what conditions.
Degree of attention: whether the witness was focused on the person or distracted.
Accuracy of the witness’s prior description of the person.
Level of certainty the witness showed when identifying the person.
Time elapsed between the event and the identification.

Fifty years of research has eroded confidence in some of these factors, especially witness certainty. A witness who says “I’m absolutely sure” sounds compelling, but confidence inflates easily through investigator feedback, repetition, and the passage of time. By trial, often months or years later, a witness has rehearsed the memory to a point where confidence bears little relationship to what they actually saw.

Research does show initial confidence can predict accuracy, but only when the identification procedure was conducted properly: blind administrator, fair lineup, no feedback before the confidence statement was recorded. Under contaminated conditions, confidence becomes almost meaningless. The Biggers factors don’t make that distinction, which is why state supreme courts in New Jersey (State v. Henderson, 2011), Massachusetts, and Oregon have moved beyond the federal test and required enhanced jury instructions grounded in current science.

What Goes Wrong Before the Police Arrive

Memory is not a recording. Perceptions are encoded in fragments, stored with gaps, and reassembled on recall. Each stage introduces error, and some distortions happen at the moment of the crime, before any investigator can affect them.

The weapon focus effect is well-documented: when a perpetrator shows a weapon, the witness’s attention narrows to the threat, and recall of the face measurably drops. Cross-race identification is another persistent problem. People are substantially better at recognizing faces of their own race, and nearly 40 percent of known wrongful identifications involve cross-race encounters. The effect appears across all racial groups and reflects perceptual experience rather than attitude. High stress narrows focus. Brief exposure limits what the brain captures to begin with. Crimes tend to be fast, frightening, and chaotic, which is exactly the condition under which memory performs worst.

These “estimator variables” can be weighed after the fact but not controlled. What can be controlled is everything that happens next.

What Goes Wrong After, and How Procedures Can Fix It

Post-event contamination is the most preventable damage. Conversations with other witnesses, media coverage, and casual remarks from investigators can all alter what a person believes they saw. The most insidious effect comes from post-identification feedback: when an officer tells a witness “good, you identified the suspect,” the witness’s confidence inflates retroactively, and they later report a better view, more attention, and more certainty than they originally had. Judges and juries rely on exactly those details.

In 2017, the Department of Justice issued a memorandum establishing identification requirements for federal law enforcement agencies conducting photo arrays and lineups.⁶ Three elements of that memo map directly onto what the research says reduces error.

Blind Administration

The officer running the procedure should not know who the suspect is, so no unintentional cue can be given. When staffing makes true blind administration impractical, the memo requires a “blinded” setup where the administrator physically cannot see which photo the witness is viewing, such as separate shuffled folders or a screen angled away from the officer.⁶

Pre-Identification Instructions and Immediate Confidence Statements

Before viewing the array, the witness should be told that the actual perpetrator may not be present and that the investigation will continue regardless. That reduces the pressure to pick someone, which is a leading cause of false identifications. If the witness does select someone, the administrator should immediately ask the witness to describe, in their own words, how confident they are. That contemporaneous statement, captured before any feedback, is far more diagnostic than confidence expressed later at trial.⁶

Lineup Composition

Fillers should resemble the suspect in age, build, and general appearance. A lineup should contain at least five fillers, only one suspect, and no filler the witness is known to recognize. When the suspect stands out, the lineup functions like a show-up in disguise: the witness compares faces to each other rather than to memory, and the person who looks most different gets eliminated, funneling the choice toward the suspect regardless of recognition.

What the Defense Can Actually Do

A defendant who believes an identification was tainted can file a pretrial motion to suppress. The defense must show that the procedure was both suggestive and unnecessary; the prosecution then argues either that it wasn’t suggestive or that the five Biggers factors establish reliability anyway.² Suppression is rare. The standard is forgiving, and most judges find enough reliability in the totality of the circumstances to let the evidence in.⁴ That is why trial-stage safeguards carry so much weight.

Cross-Examination

The Sixth Amendment guarantees the right to confront witnesses. For an identification, that means probing lighting, viewing time, cross-race factors, what the witness was told before and after the lineup, and any inconsistencies between the original description and the defendant’s actual appearance. A prepared cross-examination can surface problems a jury would never see on direct, but it depends on counsel understanding the science.

Expert Testimony on Memory

Federal courts and many state courts allow qualified psychologists to testify about the general factors that affect eyewitness accuracy. Under Federal Rule of Evidence 702, expert testimony is admissible when the expert’s specialized knowledge will help the jury and rests on reliable principles applied to sufficient facts.⁷ These experts don’t opine on whether the specific witness is right. They explain phenomena like weapon focus, the cross-race effect, and post-identification feedback. Judges sometimes exclude the testimony as unhelpful to the jury, a ruling researchers criticize because many of the findings are counterintuitive, especially the weak link between confidence and accuracy under standard procedures.

Jury Instructions

The least intrusive safeguard is the most widely available. Federal model instructions, such as the Ninth Circuit’s, direct jurors to weigh viewing opportunity, lighting and distance, whether the identification was the product of the witness’s own memory or later influence, inconsistent identifications, elapsed time, and the totality of circumstances.⁸ Some state instructions go further and address the cross-race effect, weapon focus, and the limits of confidence directly.

A Rule That Cuts Against the Defense

One procedural rule works in the other direction. Under Federal Rule of Evidence 801(d)(1)(C), a witness’s earlier out-of-court identification is not hearsay as long as the witness testifies at trial and can be cross-examined about it.⁹ The lineup identification comes in as substantive evidence of guilt, not merely to bolster credibility. The theory is that an identification made closer in time to the event is more reliable than one made in court years later. The consequence is that a potentially flawed lineup identification carries full evidentiary weight.

The In-Court Identification Problem

When a prosecutor asks, “Do you see the person who attacked you in this courtroom?” and the witness points to the defendant at the defense table, the identification is largely a formality. The defendant is the only person in the room sitting next to defense counsel and facing charges.

Courts are split on what to do about it. Under Perry, most federal courts treat first-time in-court identifications like any other testimony, because the suggestive setting wasn’t arranged by police in the same sense a lineup is.³ Several state supreme courts have reached the opposite conclusion, treating a first-time in-court identification as inherently suggestive and requiring the prosecution to justify it before the jury hears it. The split reflects a basic disagreement about whether the safeguards built for police lineups should also apply to the most suggestive identification procedure of all: pointing at the only person on trial.

Where the Debate Lands

So, should eyewitness testimony be admissible in court? It already is, and it will stay that way. The useful question is what conditions it should come in under. The constitutional framework set by Manson and Perry leaves nearly every contested identification in the jury’s hands, which puts the pressure on everything that surrounds the testimony: how police ran the procedure, whether an expert explained the science, and what the judge told the jury before deliberations.

The DOJ’s 2017 reforms accept that the science is settled on blind administration, proper instructions, and immediate confidence documentation, and build those requirements into federal policy. State decisions like Henderson take a different route by expanding judicial scrutiny. Both start from the same recognition: the traditional framework asks jurors to evaluate the reliability of a stranger’s memory while that stranger sits in a witness box radiating sincerity, and human intuition is poorly equipped for the task.

1
Justia U.S. Supreme Court Center. Foster v. California, 394 U.S. 440 (1969)
2
Justia U.S. Supreme Court Center. Manson v. Brathwaite, 432 U.S. 98 (1977)
3
Justia U.S. Supreme Court Center. Perry v. New Hampshire, 565 U.S. 228 (2012)
4
Legal Information Institute. Identification in Pre-Trial Process
5
Justia U.S. Supreme Court Center. Neil v. Biggers, 409 U.S. 188 (1972)
6
U.S. Department of Justice. Memorandum for Heads of Department Law Enforcement Components
7
Legal Information Institute. Federal Rules of Evidence Rule 702 – Testimony by Expert Witnesses
8
Ninth Circuit District and Bankruptcy Courts. Manual of Model Criminal Jury Instructions – 4.11 Eyewitness Identification
9
Legal Information Institute. Federal Rules of Evidence Rule 801 – Definitions That Apply to This Article