If you ask most people whether they could identify a stranger who attacked them, most would say yes. Of course. They saw the person. They remember the face. How could they be wrong?
Decades of research - and more than 375 DNA exonerations - say otherwise. Eyewitness misidentification is the single most common contributing factor in wrongful convictions in the United States, present in over 60% of cases where DNA later proved innocence. People who were absolutely certain they had the right person were absolutely wrong.
Understanding why this happens isn't just academic. For families fighting a wrongful conviction, it's one of the most actionable areas of a case - because courts and legislatures have increasingly recognized the problem, and challenges to eyewitness identification have become one of the more productive avenues for post-conviction relief.
Why Memory Doesn't Work the Way We Think
Human memory is not a recording. It's a reconstruction - and every time we reconstruct a memory, we're influenced by what we've learned since the event occurred. A witness who identified someone from a lineup two years before trial has had two years for that memory to be shaped, reinforced, and potentially altered by the investigation itself.
The research on this is not new or fringe. Elizabeth Loftus has been studying eyewitness memory since the 1970s and her findings have been replicated thousands of times. Memory is malleable. It's affected by stress. It's affected by cross-racial identification. It's affected by the confidence a witness expresses - which has essentially no correlation with accuracy. A witness who says "I'm absolutely certain" is not more likely to be right than one who says "I think that's him."
How Lineup Procedures Create False Identifications
It's not just memory that's the problem. The procedures used to conduct lineups and photo arrays have historically been designed in ways that make misidentification more likely.
When the officer administering the lineup knows who the suspect is, that knowledge can be transmitted to the witness through subtle cues - body language, the order in which photos are presented, reactions to the witness's choices. This is called administrator influence, and it's one of the reasons many jurisdictions now require "blind" administration - meaning the officer running the lineup doesn't know who the suspect is.
The instruction given to witnesses matters enormously. If a witness is told "we have a suspect in this lineup" rather than "the perpetrator may or may not be in this lineup," they feel pressure to make a choice rather than being allowed to say "they're not here." That pressure produces identifications that wouldn't otherwise occur.
Sequential lineups - where photos are shown one at a time rather than all at once - have been shown to reduce misidentification rates because they prevent witnesses from making relative judgments ("that one looks most like him") instead of absolute ones ("that's him").
Cross-Racial Identification
People are significantly less accurate at identifying faces of people from a different racial group than their own. This is documented, replicated, and not particularly controversial in the scientific literature. It is, however, frequently overlooked in criminal cases - and in wrongful conviction cases involving cross-racial identification, it's a significant factor that should be raised.
Research suggests the cross-racial effect is strongest when the identification is made under stress, when the exposure time was brief, and when the witness had limited prior experience with people of the other racial group.
What This Means for Post-Conviction Cases
If eyewitness identification was central to the conviction and the identification procedures used were flawed, there are several potential avenues worth exploring.
First, was an expert on eyewitness memory presented at trial? If not, and if the defense attorney didn't request one, this may support an ineffective assistance of counsel claim - particularly in light of how well-established the science is.
Second, did the prosecutor fail to disclose how the lineup was conducted, what instructions were given, or how many times the witness viewed the lineup before trial? If so, this could be a Brady violation.
Third, has the witness since expressed doubt or recanted? Recantation alone is not sufficient to overturn a conviction - courts treat recantations with skepticism - but recantation combined with documented lineup procedure problems is a much stronger case.
If eyewitness identification is at the center of a wrongful conviction case you're fighting, bring this specifically to the attention of any innocence organization or post-conviction attorney you contact. It's one of the areas where scientific evidence has made the most inroads in court, and it's an increasingly recognized basis for post-conviction relief.
Find innocence organizations and post-conviction legal resources in your state through our state directory.