Skip to main content

Verified by Psychology Today

Wrongful Convictions, Memory, and Eyewitness Testimony

If you see something, say something. But what did you see?

Key points

  • The year 2021 brought to light several high-profile examples of wrongful convictions.
  • As a result, a number of stakeholders are getting involved in finding new ways to investigate wrongful conviction claims.
  • One of these is taking steps to make sure witness and expert testimony doesn't do more harm than good.
Andrea Piacquadio/Pexels
A frustrated man.
Source: Andrea Piacquadio/Pexels

"If you see something, say something." That's a powerful message, one that encourages all of us upstanding citizens to come forward and help police prevent or solve crimes. Things that, in years past, you might have thought were none of your business.

Once criticized for raising their eyebrows over eyewitness accounts, even law enforcement seems to have had a change of heart. Their plea for the public to step forward after Gabby Petito disappeared led to two vacationers unearthing photos of the van in which she and her fiancé, Brian Laundrie, were living. These pictures led to the location of Gabby's body.

Eyewitness accounts are the backbone of evidence; we trust few things more than what appears before our very eyes. But how do you know if you're right and if what you're seeing is accurate? And what if you're wrong? A review of wrongful convictions, many involving eyewitness testimony, has sparked some serious heart-searching as well as a slew of bills aimed at preventing faulty testimony behind them.

The power of the eyewitness

Anyone who works in the courtroom knows that eyewitness testimony can be a powerful weapon for justice or the jester in a comedy of errors. Several recent high-profile stories show you just how badly things can go when mistakes happen. We read statistics like this:

Eyewitness misidentifications are known to have played a role in 70 percent of the 349 wrongful convictions which were overturned based on DNA evidence.

There's another wild card: the malleability of memory. Research has shown that, in certain circumstances, a person can falsely remember committing a crime that was actually committed by someone else. Memory can be contaminated by a number of things—suggestion, time, erroneous information, drugs/alcohol.

Some of these wrongful convictions have been so scarring that you might wonder if we should just give up on eyewitness testimony. But don't throw the baby out with the bathwater.

Improper procedures can contaminate any evidence—DNA, fingerprints, blood. But the solution isn't to quit using DNA analysis or stop collecting fingerprints; it's to make sure we follow the proper protocols. And the same is true for eyewitness memory.

As a victim's advocate, it's infuriating to me to read things like, "The trauma of rape makes it impossible to accurately identify crime victims." Sure, we have to understand human memory and the things that impact it.

But there are much better ways of assessing the accuracy of a particular eyewitness other than discrediting all rape victims, such as discerning the circumstances a crime victim is most likely to be accurate and using procedures to help her recall and preserve what she remembers.

The gold standard of eyewitness recollection

I've seen several depositions which, in comparison to the original police report, the confidence of the eyewitness, shaky at first, grew over time until, by the time the trial rolled around, he was 100 percent certain. I've read accounts where the initial eyewitness description morphed into something vastly different from the original version. Neither of these makes logical sense.

A recent study sheds some light. Researchers found that the first time a witness told their story was when it was most likely accurate. This finding was particularly true when the witness was confident in what they saw. The more time went by, and the more they told their story, the more contaminated it became.

These researchers suggest that police only rely on the initial account of the eyewitness and, in the initial interview, ask the witness how confident they are.

Another study suggests there are ways police can help their witness accurately recall information in the initial interview. After the witness gives their freeform account of what they saw (but still in the same interview), take the witness through a series of questions grouped in clusters. For example, focus the witness on what the person looked like—hair, weight, height, clothes, demeanor, etc. Then, when those questions are exhausted, switch to another cluster—the sequence of events, for instance, or everything you remember about where you were. There is something about clustering questions that helped witnesses remember details and improve their accuracy.

When it comes to lineups, the devil is in the details

Research has demonstrated that you don't have to be devious or ill-intentioned to get bad results from an eyewitness. Lineups can be a can of worms when it comes to eyewitness identification.

But there's good news. Investigators can use some simple strategies to up the odds their witness's memories will be error-free:

  • The person conducting the lineup should have no idea who the suspect is. Neither should the eyewitness. (This double-blind strategy guards against giving the witness unconscious or unintentional cues.)
  • The people or photos in the lineup should resemble the original eyewitness description, not the police suspect. (This helps prevent skewing the witness's recall toward a different memory of what the initial suspect looked like.)
  • Tell the eyewitness that the person may not be in the lineup, and they should not identify someone unless they are confident. (This takes the pressure off a witness to pick somebody.)
  • If the witness does identify someone, ask them how confident they are in their choice. Some researchers even argue we should ask a witness to give a confidence rating for every photo or person in the lineup. (This strategy recognizes that not all identifications are equal in strength.)

The bottom line

A wrongful conviction helps no one. It casts a wide net in terms of the damages it does; an innocent man pays for someone else's actions, the victim does not get justice, our confidence in the justice system suffers. It also paves the way for more crimes; a 2019 study found that serial killers committed 114 serial murders after someone else was in prison for a crime they didn't commit.

With so much at stake, we've got to get it right.


If you're interested in more forensic psychology and true crime articles, check out my website, true crime channel and book. Serial Killers: 101 Questions True Crime Fans Ask.

More from Joni E Johnston Psy.D.
More from Psychology Today
More from Joni E Johnston Psy.D.
More from Psychology Today