The New Science of Eyewitness Memory | John Wixted | TED

Name: The New Science of Eyewitness Memory | John Wixted | TED
Duration: 17 min 23 s
Description: We've built a legal system that distrusts eyewitness memory — backed by cautionary science and high-profile exonerations. John Wixted, a leading psychology r...

17:23Added Aug 30, 2026

We've built a legal system that distrusts eyewitness memory — backed by cautionary science and high-profile exonerations. John Wixted, a leading psychology r...

Watch on YouTube →

Contributed by 刘嘉琪

Transcript

00:04Imagine for a moment that you're absolutely certain
00:07about the person you saw commit a crime.
00:10You're so confident,
00:12you'd be willing to testify about it under oath in a court of law.
00:16Your memory is strong,
00:18crystal clear, absolutely unshakable.
00:22But now imagine that that same memory, though it feels 100 percent true,
00:27is actually false
00:29and could send an innocent person to prison,
00:31maybe even to death row.
00:34This is the complex and sometimes heartbreaking world of eyewitness memory.
00:39But for decades, we've been telling ourselves a story about eyewitness memory
00:42that itself may not be entirely true.
00:46Most of you have probably heard cautionary tales
00:48about how wildly unreliable eyewitness testimony can be.
00:53You may have heard about famous cases
00:55like the case of Ronald Cotton,
00:57where Jennifer Thompson, a rape victim,
01:00misidentified him as her attacker.
01:02As she would later recall her testimony from his criminal trial,
01:07"I was absolutely, positively,
01:09without a doubt certain that he was the man who raped me
01:12when I got on that witness stand.
01:13And nobody was going to tell me any different."
01:16The jury understandably found her testimony convincing.
01:20Cotton was found guilty and sentenced to life in prison.
01:25But Ronald Cotton did not rape Jennifer Thompson.
01:28Instead, it was a known rapist prowling her neighborhood that night.
01:32Cotton spent almost 11 years in prison
01:35before DNA testing finally proved his innocence
01:38and revealed the identity of the true rapist,
01:41a man named Bobby Poole.
01:44Jennifer Thompson's testimony was sincere, but her memory was wrong.
01:49DNA exoneration cases just like this one,
01:52involving confident misidentifications,
01:54have happened literally hundreds of times,
01:57leading many to seriously question the reliability of eyewitness memory.
02:03But wrongful convictions like these are not the only reason
02:06why most people think eyewitness memory is unreliable.
02:09For years, scientific research has also painted a damning picture
02:12of human memory.
02:14Starting in the 1970s,
02:15scientists like Elizabeth Loftus
02:18began to show how shockingly easy it is to manipulate memory.
02:22In groundbreaking studies,
02:24she and others implanted false memories in adults
02:27of having been lost in a shopping mall as a child,
02:30or as having been attacked by a vicious animal,
02:33even though these things never actually happened.
02:37Findings like these seem to confirm our worst fears.
02:39Memories are not like video recordings.
02:42They're more like evidence from a crime scene,
02:46collected by people without gloves,
02:49distorting and contaminating it with every touch.
02:53This message from science
02:56reinforced the message from the wrongful convictions,
02:58and the conventional wisdom was set in stone for decades.
03:02The legal system should not trust eyewitness memory.
03:05It's just too unreliable.
03:08But here's where the story takes an unexpected turn.
03:11What if the problem is not so much about how unreliable eyewitness memory is,
03:16because of how easily contamination can create false memories,
03:21and more about how and when we test the witness's memory.
03:26Think about forensic evidence like DNA or fingerprints.
03:30Everybody knows that forensic evidence can be contaminated
03:33and end up implicating an innocent person,
03:37much like contaminated memory can.
03:39But we don't just dismiss forensic evidence for that reason.
03:43Instead, we collect it as early as possible in the police investigation
03:48before it's contaminated.
03:51Why do we do that?
03:53It's because reliable information comes from analyzing uncontaminated evidence,
03:58not contaminated evidence.
04:00And the exact same principle applies to memory evidence.
04:05Collect it early, before it's contaminated.
04:10An eyewitness’s memory of whether or not the police suspect
04:13is the person who they saw commit the crime
04:16can be highly reliable,
04:19but only if the witness's memory is uncontaminated,
04:22not after it's been contaminated.
04:25And scientists now agree
04:26that even the first test contaminates the witness's memory for a given suspect.
04:32If the suspect’s innocent, for example,
04:35it's the first time the witness is seeing his face,
04:39and this is happening at a time
04:40when the witness is actively thinking about the crime.
04:43So even if the witness says,
04:45"No, that's not the guy who did it,"
04:48this is how the innocent suspect’s face first becomes associated with the crime
04:53in the witness's memory.
04:55That's contamination.
04:57You can’t keep that from happening,
05:00and you can’t put the witness’s memory back the way it was.
05:04So focus on the first uncontaminated memory test
05:07early in the police investigation,
05:10not the last, thoroughly contaminated test
05:13that happens at the criminal trial one, two, or even three years later.
05:19Unfortunately, courts tend to do the reverse,
05:22placing their faith in the last test of memory at trial
05:26while all but ignoring the critical first test.
05:30This is a seriously underappreciated problem.
05:37Well, with that in mind,
05:38let's take a closer look at how the police conduct
05:40that all-important first test of a witness’s uncontaminated memory.
05:48In the days or weeks after a crime,
05:50the police might find a suspect,
05:52a person who they think may have committed the crime,
05:55and they'd like to show him to the witness
05:57to see if they have the right guy.
05:59They could just hand the suspect's photo to the witness and ask,
06:03is this the guy who did it?
06:06The problem is, that would be suggestive,
06:08because it would reveal to the witness who the police think
06:10may have committed the crime.
06:12To test memory in a less suggestive way,
06:15the police will often show the witness a whole set of six photos.
06:20It's called a six-pack photo lineup.
06:22One photo is of the suspect,
06:25and the others are of similar-looking individuals
06:28who the police know are innocent.
06:30That way, they can still show the suspect's photo to the witness,
06:34but without revealing who they think committed the crime.
06:37It's a much fairer way to test memory,
06:39and it becomes fairer still when other recommended practices are followed.
06:44Such as letting the witness know
06:46that the perpetrator who they saw commit the crime
06:49may or may not be among these photos.
06:51And the officer who's administering these photos to the witness
06:54should not even know who the suspect is
06:57to avoid unintentionally influencing the witness’s choice.
07:02When it's done this way, it becomes a pure test of the witness's memory,
07:06and this is where things start to get interesting.
07:09About 10 years ago, work from my lab,
07:11published in strong, high-impact scientific journals,
07:17first reported that a confident identification of a suspect
07:21from an initial photo line-up is highly reliable.
07:26Big surprise.
07:28Not unreliable.
07:31For a scientific field that has spent decades
07:33cautioning the legal system
07:35about the unreliability of eyewitness memory,
07:38these findings were not easy to enthusiastically embrace.
07:44But almost all of the recent science finds that an initial,
07:47confident identification is much more reliable
07:50than the field previously thought.
07:52Not infallible, of course, but certainly not unreliable.
07:58These new scientific findings raise an interesting question
08:01about the DNA exoneration cases that I told you about earlier.
08:05The ones where we know that on the last test of their memory,
08:09at the criminal trial,
08:10witnesses confidently misidentified an innocent person
08:14contributing to a wrongful conviction.
08:17The question is: What did those witnesses do
08:20the very first time their memory was tested for that same person?
08:25And when you look into that,
08:27you find they usually did not confidently misidentify
08:30the innocent suspect at that time.
08:32The problem is, nobody listened to them.
08:35Remember Jennifer Thompson?
08:38She struggled with the initial photo lineup a few days after the crime.
08:43Narrowing it down to two pictures,
08:46wavering hesitantly between them for literally minutes
08:49before finally landing on Ronald Cotton's face and saying,
08:52"I think it's him."
08:55It was obviously inconclusive identification,
08:57full of doubt and indecision.
09:01By the time of his criminal trial, after much memory contamination,
09:05that her doubts were gone and she became absolutely, positively,
09:08without a doubt certain
09:09that Ronald Cotton was the man who raped her.
09:12If they had known then what we know now,
09:15focus on the first test,
09:17where, in this case, you find a completely inconclusive identification,
09:21it seems likely that Ronald Cotton never would have been wrongfully convicted
09:25in the first place.
09:26And that's the point.
09:29Alright, here's where the story takes another interesting turn.
09:33On this first test using a photo lineup,
09:35witnesses often don't even tentatively identify the suspect
09:39the way that Jennifer Thompson tentatively identified Ronald Cotton.
09:43Instead, at a time when the witness's memory of the perpetrator is as fresh
09:47and strong and uncontaminated as it will ever be,
09:51the witness looks at the photos,
09:52including the photo of the suspect, and says,
09:56"None of these guys match my memory of the person who committed the crime."
10:00In other words, the witness rejects the lineup,
10:03providing clear evidence that the suspect in the lineup is innocent.
10:09You see, this is absolutely the key point.
10:13The first test of an eyewitness's uncontaminated memory
10:18can provide reliable evidence pointing in either direction:
10:22towards guilt or innocence,
10:25depending on how the test turns out.
10:27And when the witness rejects the lineup,
10:29it provides reliable evidence pointing in the direction
10:34of the suspect's innocence.
10:36Yet many witnesses who reject an initial photo lineup,
10:40after their memory becomes contaminated,
10:43will show up at a criminal trial a year or two later,
10:46unaware that their memory is contaminated,
10:49and now confidently identify the very same person
10:52they initially rejected.
10:54Half the time, they don't even remember doing that,
10:56it was so long ago.
10:58These defendants are often convicted, sentenced to long prison terms,
11:02and are now behind bars.
11:04And unlike Ronald Cotton,
11:08they do not have any DNA evidence to prove their innocence.
11:12But what they do have is a new message
11:16from the world of memory science that can help to do that.
11:19And ironically,
11:21in a complete mind flip,
11:24it's reliable evidence of innocence
11:26from the memory of an eyewitness the first time they were tested.
11:31Consider the case of Miguel Solorio.
11:34He was arrested in 1998 for murder.
11:37But on the first test, four witnesses rejected his photo lineup.
11:43Nobody paid any attention to that.
11:45More than a year later,
11:46two of those same witnesses showed up at his criminal trial
11:49and identified him as the shooter in front of a judge and jury.
11:53He was convicted and sentenced to life in prison
11:56without the possibility of parole.
11:59Miguel spent 25 years of his life,
12:03from age 19 to age 44,
12:06behind bars,
12:07before finally being exonerated in late 2023,
12:12with the help of the new science that I'm telling you about today:
12:16focus on the first uncontaminated test.
12:20As tragic as Miguel's story is,
12:23at least the new science helped to overturn his conviction
12:26before he spent his entire life in prison.
12:30But others with a similar story remain behind bars as we speak.
12:34And some of them,
12:36their lives are hanging in the balance.
12:39Consider the case of Charles Don Flores.
12:42He, too, was arrested for murder in 1998.
12:45There was one witness in this case.
12:48On the day of the crime,
12:49she told the police that she saw a white male with shoulder-length hair
12:53go into her neighbor's house
12:54shortly before the murder occurred.
12:57A couple days later, she goes back to the police station.
13:00They hypnotize her.
13:01It's always dicey, but she stuck to her story.
13:04She said it's a white male with shoulder-length hair.
13:07The police asked her to make a composite sketch, so she did.
13:11Came out to be a white male with shoulder-length hair,
13:13just like she said.
13:15Now keep in mind, this is very early in the police investigation.
13:19This is a time when her memory of the perpetrator is as fresh and strong
13:23and uncontaminated as it will ever be.
13:27And this is what the perpetrator looks like.
13:30But the police suspected Charles Don Flores,
13:32a Hispanic man with short hair.
13:35And for some inexplicable reason,
13:37put his photo in a photo lineup and showed it to the witness.
13:43This is still early in the police investigation.
13:46This is the all-important first test of a witness's uncontaminated memory
13:50for the suspect, Charles Don Flores.
13:53The witness looked at this photo lineup,
13:57didn't see any white males with shoulder-length hair,
14:00and quite understandably rejected it,
14:03providing clear evidence that Flores is innocent.
14:07Fast forward to his criminal trial a year later,
14:10and you can probably guess how this story turns out.
14:13Now the witness is more than 100 percent certain that it was Charles Don Flores
14:17she saw go into her neighbor's house on the day of that murder.
14:22The jury was convinced, he was found guilty
14:25and sentenced to death.
14:27He's been on death row for more than 25 years,
14:30where he remains to this day.
14:33And in all those years,
14:36no jury even heard what the witness did
14:39on the first test of her uncontaminated memory,
14:42much less about the new scientific consensus,
14:45emphasizing the importance of focusing on that first test,
14:48because that's where the reliable information is.
14:54So, you see, this isn't just an academic exercise.
14:57It's about exonerating the wrongfully convicted,
15:00some of whom are on death row,
15:02and about preventing wrongful convictions from happening in the first place.
15:06It's about understanding that memory is not an enemy to be distrusted,
15:11but a complex tool that, when properly used,
15:13can serve the cause of justice.
15:16The reforms that need to be made going forward seem pretty clear.
15:20Number one, in both police investigations and legal proceedings,
15:25prioritize the first test of the witness's uncontaminated memory,
15:29because that's where the reliable information is.
15:32Number two, at the same time,
15:34de-emphasize confident suspect identifications
15:39made by a witness who earlier rejected that same face
15:42because that likely reflects memory contamination, not the truth.
15:46And number three,
15:47to get these reforms underway, educate legal professionals,
15:51police chiefs,
15:53defense attorneys, prosecutors, judges
15:56about the new science of eyewitness memory.
16:00Over the last couple of years,
16:01my colleagues and I have been doing just that,
16:03reaching out to defense attorneys and prosecutors alike
16:06to bring this new understanding to the legal system.
16:09We find that initially, at least,
16:11defense attorneys are alarmed by this new message
16:14because they're uncomfortable with the idea that eyewitness memory
16:17can ever be reliable.
16:20To them, listening to what an eyewitness has to say,
16:23even on the very first test of their uncontaminated memory,
16:26is just a recipe for another wrongful conviction.
16:29And prosecutors,
16:31they're equally alarmed, but for the opposite reason.
16:34To them, not listening to a witness
16:36who makes a confident suspect identification
16:39after previously rejecting that same face
16:43sounds like a recipe for letting the guilty walk free.
16:47So we've managed to bring both sides of the aisle together in a way,
16:52to both the defense and the prosecution our new message sounds dangerous.
16:57But it's not dangerous.
16:58It's just how memory works.
17:01To better serve the cause of justice,
17:03both sides and all of us have to follow the science by listening to memory
17:09when it most reliably speaks the truth.
17:13It will become a more just world if we can find a way to do that.
17:18Thank you.
17:20(Applause)